Considering a new extruder design

mung Wrote:
-------------------------------------------------------
> Sublime Wrote:
>
> > 1) For PLA you need the entrance of the heater
> to
> > stay below 60c all the time.
>
> Why do you say this?
> What happens if the entrance is not below 60C?
> Where are the calculations and experimental
> evidence that set this limit of 60C

PLA has a glass transition temperature of 60c. If we go above 60c it turns soft and expands so it no longer fits in the barrel and increases the pressure required to get it to the tip.


> > 2) Needs some thermal mass to deal with the
> > cooling effect of pushing filament through it at
> a
> > high rate.
>
> Why do you need thermal mass, and what function
> does the thermal mass do to effect the temperature
> of the filament?
>
> I could answer your question for you, but I think
> if you could answer it you may better understand
> thermodynamics.
>
> input = output

We want a consistant temperature with little to no fluctuation. But filament takes heat from the hotend EXTREMELY quickly. As soon as you start feeding filament into the hotend it takes away heat. And Yes if you pump a huge amount of heat into the hotend you can keep the material soft but not just at the right temperature for building stuff. We want the extruded material within 5 degrees of the set temperature ALWAYS.

> > 3) Needs to be safe. We use 6.8ohm resistors
> > because they should never be able to start a
> > hotend fire @ 12volts.
>
> What makes you think 6.8R resistors would never be
> able to start a fire????
>
> I think in the right circumstances any resistors
> could start a fire, but in most cases hotend fires
> are rather unlikely if there is nothing inflamable
> nearby. Of course there are always the unknown
> unknowns like a magnitude 9 earthquake at the
> fukashima nuclear power plant, or melting polymer
> producing a flammable gas and a build up of
> pyrophoric carbon igniting the gas in a confined
> space causing an explosion???

Yes and the sun will kill everyone on earth some day but for now lets deal with what's at hand and not be STUPID.
PLA and ABS both can catch fire (not the metal) BUT a 6.8 ohm resistor in a heatsink (metal heater block) @ 12volts can only draw 1.7amps so it never reaches above the ignition temperature of PLA, ABS, WOOD, etc. YOURS on the other hand is the scariest thing I have seen to date in the reprap world, it will burn anything that gets close enough.

> > 4) Needs to be able to handle the pressures
> inside
> > the hotend @ 50mm/s @ 200+c for 4hrs to 24hrs
>
> I doubt this is a problem, but I am beginning to
> think speed could be a problem as many machines
> are running much faster than 50mm/s, I imagine
> that 500mm/s could be the upper limit.
>
> E = (t2 - t1) * M * S.H. + any latent heat
> required for phase change
>
> The speed problem is due to speed of heat
> conduction within the polymer, and this is why
> smaller diameter feed stock are better and I
> initially did some calculations that showed 1.25mm
> could be about optimum.

Once again you have not read up on RepRap enough yet. MOST hotend failures are from pressure causing the hotend to fail by leaking or causing fatigue. And yes smaller feedstock does mean lower pressures but defeats your idea to lower costs becuse the filament costs more per kg.

Most people run there machine BELOW 50mm/s. Above this is considered fast. High quality prints are done closer to 10mm/s

>
> > 5) Can't transfer a large amount of heat to the
> > plastic carriage it has to be mounted to.
>
> I think thermal management is quite important in a
> heating device, I really have no expertise in
> mechanical engineering or thermodynamics, but I do
> remember a few things from A level physics. heat
> in = heat out, heat moves from hotter places to
> cooler places until everything is the same
> temperature, heat will travel mainly by conduction
> or convection, surface area and airflow are
> important in convection, thermal conductivity of a
> material is important in conduction.
>
> Ideally you want all your heat to go into the
> polymer you are melting the closer to this ideal
> the closer to the calculated heating energy for a
> given mass of polymer your extruders power
> requirements will be.

My hotend runs at 225c and I can hold it with my bare hands at the cold end 65mm from the hotend. @ 10watts (21watts max) and it maintains the temp to within 3 degrees no matter how fast I extrude. This required 1)thermal mass 2) insulating the heater block from all air flow (no open resistor design will handle a fan blowing right below the tip)


> Without the right tools you will not be able to
> make the most efficient machine.

My machine is made of found junk and built with a hand drill, vise grips, and screw drivers and it prints faster in PLA than any other machine I have seen. (120mm/s)

> I would like to see someone make a better
> extruder, but I think I do not have the time or
> the expertise, and to actually make a self
> replicatable extruder is although possible not
> feasible with present technology.

If your hotend is so good and you look to have many of them send me one or two and I will show its failings. But it sounds like you saw the RepRap project and decided you could do it better than the 10,000 other engineers that have developed them this far. We have many good hotend designs available already the only reason one is not standard is because WE DON'T all want the same thing.

PLEASE READ EVERY PAGE OF NOPHEADS BLOG [hydraraptor.blogspot.com] then read all the relevant data on the WIKI [reprap.org] then maybe you will be caught up to where the project is.

AND most importantly having parts to break for the seller to make more money on later is a NO NO in opensource. If you are doing this for profit you are in the wrong place. We would rather spend twice as much now and never replace the heater than pay less for something known to have planned obsolete parts as you spoke of earlier.

EDIT:
I just read through some of the more heated moments on the previous page and would like to note that I also come across a little angry as you yourself do. But don't want this to turn into a fight or discourage you from developing further I originally was saying you need to test it while pushing filament through it for hours and hours because they act differently when under heavy use.

Edited 2 time(s). Last edit at 07/11/2011 03:30PM by Sublime.

---

FFF Settings Calculator	Gcode post processors	Geometric Object Deposition Tool Blog
Tantillus.org	Mini Printable Lathe	How NOT to install a Pololu driver

Sublime Wrote:

> PLA has a glass transition temperature of 60c. If
> we go above 60c it turns soft and expands so it no
> longer fits in the barrel and increases the
> pressure required to get it to the tip.

Generalisation?

> We want a consistant temperature with little to no
> fluctuation. But filament takes heat from the
> hotend EXTREMELY quickly. As soon as you start
> feeding filament into the hotend it takes away
> heat. And Yes if you pump a huge amount of heat
> into the hotend you can keep the material soft but
> not just at the right temperature for building
> stuff. We want the extruded material within 5
> degrees of the set temperature ALWAYS.

Has anyone made controlled tests of extrusion, interlayer cohesion, and physical properties of plastic after extrusion, at different temperatures?

> Yes and the sun will kill everyone on earth some
> day but for now lets deal with what's at hand and
> not be STUPID.
> PLA and ABS both can catch fire (not the metal)

You could burn metal given the right conditions if you wanted!.

> BUT a 6.8 ohm resistor in a heatsink (metal heater
> block) @ 12volts can only draw 1.7amps so it never
> reaches above the ignition temperature of PLA,
> ABS, WOOD, etc.

What makes you think that, I can see a number of possible failures where temperatures could rise or where fire could occur at lower temperatures??

YOURS on the other hand is the
> scariest thing I have seen to date in the reprap
> world, it will burn anything that gets close
> enough.
>
If I wanted it to yes.


> Once again you have not read up on RepRap enough
> yet. MOST hotend failures are from pressure
> causing the hotend to fail by leaking or causing
> fatigue.
If the hot end fails due to pressure then you are using pressures that are too high, or you are building the hotend from insufficiently strong materials?

And yes smaller feedstock does mean lower
> pressures but defeats your idea to lower costs
> becuse the filament costs more per kg.

Why does it cost more??
Proportionally how much more does it cost?

> Most people run there machine BELOW 50mm/s. Above
> this is considered fast. High quality prints are
> done closer to 10mm/s

Why?


> My hotend runs at 225c and I can hold it with my
> bare hands at the cold end 65mm from the hotend.
> @ 10watts (21watts max) and it maintains the temp
> to within 3 degrees no matter how fast I extrude.
> This required 1)thermal mass 2) insulating the
> heater block from all air flow (no open resistor
> design will handle a fan blowing right below the
> tip)

I don't have a fan right below the tip, I don't have any fans, where did you see a fan of my design???



> > Without the right tools you will not be able to
> > make the most efficient machine.
>
> My machine is made of found junk and built with a
> hand drill, vise grips, and screw drivers and it
> prints faster in PLA than any other machine I have
> seen. (120mm/s)
>

Maybe your machine is not the most efficient?.

> If your hotend is so good and you look to have
> many of them send me one or two and I will show
> its failings. But it sounds like you saw the
> RepRap project and decided you could do it better
> than the 10,000 other engineers that have
> developed them this far. We have many good hotend
> designs available already the only reason one is
> not standard is because WE DON'T all want the same
> thing.

I do not think my hot end is 'so good', I think it is probably rather poor compared with commercial machines (I have never seen a Reprap or a commercial machine in the flesh for real comparison).

I doubt there are 10,000 engineers working on development of these types of extruders, maybe somewhere between 10 and 100, I think most people probably just assemble kits of follow assembly plans of other peoples developments, would be interesting to see some real data about it??

If there were 10,000 developers we would see more designs in the extruders wiki?

> PLEASE READ EVERY PAGE OF NOPHEADS BLOG
> [hydraraptor.blogspot.com] then read all the
> relevant data on the WIKI [reprap.org]
> then maybe you will be caught up to where the
> project is.

I find I can make so many misunderstandings and misinterpretations of other peoples work, and I don't want to constantly badger people asking for help and reiteration of their explainations, I also do not want to end up whispering to a chinese mans ear

I don't want to just reiterate other peoples comments without knowing if they are correct or not, and, without understanding the methods and physics behind them.

I will try and make the effort to read all the information you suggest, though as I said I do find it hard understanding it all.


> AND most importantly having parts to break for the
> seller to make more money on later is a NO NO in
> opensource. If you are doing this for profit you
> are in the wrong place. We would rather spend
> twice as much now and never replace the heater
> than pay less for something known to have planned
> obsolete parts as you spoke of earlier.

Why?

Would you refuse to buy a car that had break pads that wear away?

Would you buy a car that cost ten times the price of other cars and weighed 100kg more with worse fuel economy just so you can have expensive tungsten carbide brake shoes and discs that will never wear away??

Entropy is a fact of life, things become disordered and wear out, the question you should be asking is probably something else?


> EDIT:
> I just read through some of the more heated
> moments on the previous page and would like to
> note that I also come across a little angry as you
> yourself do. But don't want this to turn into a
> fight or discourage you from developing further I
> originally was saying you need to test it while
> pushing filament through it for hours and hours
> because they act differently when under heavy use.

You mention heated moments, am not sure what you are talking about do you mean my hot end extrusion tests, I think from context you mean this as angry moments???


I am not angry and have not been at all angry at any point of my development of this extruder, I find it is a nice relaxing hobby and when I have any spare time I like to spend it doing hobbies, I enjoy this but I also have other hobbies that I find more enjoyable.

I sometimes do get annoyed at the hobbies I play with usually because they break or I find they cost too much, but not with this one so far.

If you are saying that I seem angry and want to fight, this may seem the case to you but it is not how I feel and I wish I could make it so people do not think such of me.

Maybe my English is not so good and people misunderstand what I say, before someone said this thread was getting 'fighty' I did not understand what had been said wrong then, and I tried to explain myself. I am only trying to ask questions about what I do not understand.

I would be pleased if you could help me to explain myself better and edit what I say so it does not give people the impression that it is angry or fighting.


I am pleased of your comments and do not want to make a Chewbacca defence of my experiments and prototype, I only want to show what I have done so that people can comment and help me improve it, and ask questions about what I do not understand.

As far as your suggestion to test extrusion for hours and hours I know this is a good idea, but I do not have enough filament to do this test, and I do not currently have a machine to print with and even then I do not have anything I want or need to print.

Unfortunately these types of tests are at least a year away.

I am sure there are many problems with my extruder design that will not be seen until it is actually used on a real printer, but until then I am doing what I am able to in the hopes that solutions to the unforeseen problems can be found.

mung Wrote:
-------------------------------------------------------
> Sublime Wrote:
>
> > PLA has a glass transition temperature of 60c.
> If
> > we go above 60c it turns soft and expands so it
> no
> > longer fits in the barrel and increases the
> > pressure required to get it to the tip.
>
> Generalisation?

If you had read up on RepRap you would know that most of the PLA sold in the world is made by a company called natureworks in the US, they make the resin and companies all over the world extrude it into filament. The main one used is called 4043 [www.natureworksllc.com]

This video shows what happens when the filament expands before it is supposed to. Remember the filament sits in the heat for minutes or hours during a real print so preheating the filament is a real concern.

>
> > We want a consistant temperature with little to
> no
> > fluctuation. But filament takes heat from the
> > hotend EXTREMELY quickly. As soon as you start
> > feeding filament into the hotend it takes away
> > heat. And Yes if you pump a huge amount of heat
> > into the hotend you can keep the material soft
> but
> > not just at the right temperature for building
> > stuff. We want the extruded material within 5
> > degrees of the set temperature ALWAYS.
>
> Has anyone made controlled tests of extrusion,
> interlayer cohesion, and physical properties of
> plastic after extrusion, at different
> temperatures?

Yes the information is available but not laid out the way you want. YOU have to do some work to sort through it all. Not all the testing is done the way you would want either. But Data is Data, if you know what to do with it.

>
> > Yes and the sun will kill everyone on earth
> some
> > day but for now lets deal with what's at hand
> and
> > not be STUPID.
> > PLA and ABS both can catch fire (not the metal)
>
> You could burn metal given the right conditions if
> you wanted!.

But like I said lets not be stupid and please think of what actually happens. Our microcontroller is reading a thermistor or thermocouple to sense the temperature, but sometimes they fail or other issue and the board sends a constant un-regulated 12v to the hotend. In this case we will not have a fire and the hotend still works later, but if a constant unregulated 12v went to yours it would burn out and ???

I also notice that you are using a variable current power supply to do your testing but this is far more than we use. Most people use Bang Bang control (turning it on and off +/- 5c of the target) which needs the thermal mass to maintain the temp. Some people have moved on to PID control which gives us closer to a constant current but is still controlled by the thermistor or thermocouple via the microcontroller.


> > BUT a 6.8 ohm resistor in a heatsink (metal
> heater
> > block) @ 12volts can only draw 1.7amps so it
> never
> > reaches above the ignition temperature of PLA,
> > ABS, WOOD, etc.
>
> What makes you think that, I can see a number of
> possible failures where temperatures could rise or
> where fire could occur at lower temperatures??

Yes unforeseen issues, but under normal use it won't. On the other hand yours we can predict to cause fire because it will burn up if left unregulated (you proved this by turning up the amperage). If you listen to the people you asked for help from instead of assuming you know better you would have one less safety concern.

> YOURS on the other hand is the
> > scariest thing I have seen to date in the
> reprap
> > world, it will burn anything that gets close
> > enough.
> >
> If I wanted it to yes.

Mine can't.............................. I have had a computer crash during a over night build were the hotend sat in plastic for 10hrs without issue and you won't trust yours alone for 5 minutes.


>
>
> > Once again you have not read up on RepRap
> enough
> > yet. MOST hotend failures are from pressure
> > causing the hotend to fail by leaking or
> causing
> > fatigue.
> If the hot end fails due to pressure then you are
> using pressures that are too high, or you are
> building the hotend from insufficiently strong
> materials?
>
> And yes smaller feedstock does mean lower
> > pressures but defeats your idea to lower costs
> > becuse the filament costs more per kg.
>
> Why does it cost more??
> Proportionally how much more does it cost?

Lets see, 3mm stock is 50m per pound (approx) but 1.75mm is more like 85m so it takes more energy and time for the manufacturer to make it which costs us money. In the US and Canada (where I live) it costs $4/lb more for 1.75 and I use 5lbs a month just for personal use. That's $240 a year extra. If I was one of the people making stuff to sell it would be more like $2000 a year extra.

>
> > Most people run there machine BELOW 50mm/s.
> Above
> > this is considered fast. High quality prints
> are
> > done closer to 10mm/s
>
> Why?

Because that's the maximum they can move without loosing positioning accuracy. And for a better outer finish.


>
>
> > My hotend runs at 225c and I can hold it with
> my
> > bare hands at the cold end 65mm from the hotend.
>
> > @ 10watts (21watts max) and it maintains the
> temp
> > to within 3 degrees no matter how fast I
> extrude.
> > This required 1)thermal mass 2) insulating the
> > heater block from all air flow (no open
> resistor
> > design will handle a fan blowing right below
> the
> > tip)
>
> I don't have a fan right below the tip, I don't
> have any fans, where did you see a fan of my
> design???
>

Once again you need to understand what we are doing before doing it better. We need a fan under the tip to cool the plastic as it comes out so it stays the right shape. We don't just heat up some plastic and squirt it out, the temperature is being changed all the time depending on how we want the plastic to behave. For bridges ( spanning free air ) we need it 5+c cooler and our first layer is 5+c hotter than the rest (none of these are set rules they change with room temp, build surface, plastic types, etc) but the main thing is we need to control the temperature accurately all the time. Even after it comes out of the nozzle. We try and cool the last layer to its glass transition temperature but also want to keep the first layer that warm too, so we don't have uneven shrinkage/warping.


>
>
> > > Without the right tools you will not be able
> to
> > > make the most efficient machine.
> >
> > My machine is made of found junk and built with
> a
> > hand drill, vise grips, and screw drivers and
> it
> > prints faster in PLA than any other machine I
> have
> > seen. (120mm/s)
> >
>
> Maybe your machine is not the most efficient?.

Damn this is why no one likes you. Its insulting because you can't print at all and I know for a fact I print faster than most people out there.


>
> > If your hotend is so good and you look to have
> > many of them send me one or two and I will show
> > its failings. But it sounds like you saw the
> > RepRap project and decided you could do it
> better
> > than the 10,000 other engineers that have
> > developed them this far. We have many good
> hotend
> > designs available already the only reason one
> is
> > not standard is because WE DON'T all want the
> same
> > thing.
>
> I do not think my hot end is 'so good', I think it
> is probably rather poor compared with commercial
> machines (I have never seen a Reprap or a
> commercial machine in the flesh for real
> comparison).

Commercial machines take all day to make what a fast RepRap prints in hours.


>
> I doubt there are 10,000 engineers working on
> development of these types of extruders, maybe
> somewhere between 10 and 100, I think most people
> probably just assemble kits of follow assembly
> plans of other peoples developments, would be
> interesting to see some real data about it??
>
> If there were 10,000 developers we would see more
> designs in the extruders wiki?

Almost everyone who owns a RepRap is a developer and as I said earlier we have lots of good extruder designs, why do you assume we don't. And why do you think we need a better one.


> > AND most importantly having parts to break for
> the
> > seller to make more money on later is a NO NO
> in
> > opensource. If you are doing this for profit
> you
> > are in the wrong place. We would rather spend
> > twice as much now and never replace the heater
> > than pay less for something known to have
> planned
> > obsolete parts as you spoke of earlier.

> Why?

In opensource WE SEE when you cheap out to make more money and call you on it. Then we take your design make it the way it should be and tell you to go F yourself with your planned obsolete part, at which point you are stuck with 249,990 left over unsold crappy extruders.

Remember this project is about taking the need and power away from big business/corporations and putting it in the hands of all of us. ITS NOT ABOUT FINANCIAL PROFIT.

>
> Would you refuse to buy a car that had break pads
> that wear away?

I refuse to buy the cheapest ones made in china from old discarded shoes. I buy the ones known to last the longest and work the best even at a higher cost.

>
> Would you buy a car that cost ten times the price
> of other cars and weighed 100kg more with worse
> fuel economy just so you can have expensive
> tungsten carbide brake shoes and discs that will
> never wear away??

Once again you are just being insulting. I own a Subaru over a Lada because its a better car. I paid ten times more for it than I would have for a Lada and it costs more in gas.....so....


> You mention heated moments, am not sure what you
> are talking about do you mean my hot end extrusion
> tests, I think from context you mean this as angry
> moments???

Yes in English heated means close to a fight.


> I am not angry and have not been at all angry at
> any point of my development of this extruder, I
> find it is a nice relaxing hobby and when I have
> any spare time I like to spend it doing hobbies, I
> enjoy this but I also have other hobbies that I
> find more enjoyable.
>
> I sometimes do get annoyed at the hobbies I play
> with usually because they break or I find they
> cost too much, but not with this one so far.
>
> If you are saying that I seem angry and want to
> fight, this may seem the case to you but it is not
> how I feel and I wish I could make it so people do
> not think such of me.
>
> Maybe my English is not so good and people
> misunderstand what I say, before someone said this
> thread was getting 'fighty' I did not understand
> what had been said wrong then, and I tried to
> explain myself. I am only trying to ask questions
> about what I do not understand.
>
> I would be pleased if you could help me to explain
> myself better and edit what I say so it does not
> give people the impression that it is angry or
> fighting.

First off don't answer questions with riddles. If you know the answer better than us, TELL US. Don't just act like you know better than the people with working machines that have taken tens of thousands of hours to collectively design. We have practical experience and have dealt with the issues that have arisen during actual real world use of the device. Also commercial companies like Stratsys and Dimension have been building these for 25 years, millions into research and development and they don't use anything like your hotend, why?

What works on paper doesn't always work in the real world because you can be missing a variable somewhere.

> I am pleased of your comments and do not want to
> make a Chewbacca defence of my experiments and
> prototype, I only want to show what I have done so
> that people can comment and help me improve it,
> and ask questions about what I do not understand.
>
> As far as your suggestion to test extrusion for
> hours and hours I know this is a good idea, but I
> do not have enough filament to do this test, and I
> do not currently have a machine to print with and
> even then I do not have anything I want or need to
> print.
>
> Unfortunately these types of tests are at least a
> year away.
>
> I am sure there are many problems with my extruder
> design that will not be seen until it is actually
> used on a real printer, but until then I am doing
> what I am able to in the hopes that solutions to
> the unforeseen problems can be found.

Keep going for sure, innovation takes doing your own thing despite what others say.
I would also suggest contacting one of the Plastic filament companies and ask for some samples. ( they have pieces to short to sell ).

As for the type of documentation you want. Unfortunately with thousands and thousands of people contributing a little all the time from all over the world we do not have a set of guide lines or procedures because it changes on a daily basis. The best thing to do is follow it on a daily basis for a few months before starting ( I followed RepRap for 2 years before building one ). NopHead has done an amazing job documenting all of his testing along with providing the Scientific Data to back up his claims.


EDIT: what takes all your time? A hotend takes 2hrs to build from scratch and you have a milling machine to attach it to, print something already!

Edited 2 time(s). Last edit at 07/13/2011 03:07PM by Sublime.

---

FFF Settings Calculator	Gcode post processors	Geometric Object Deposition Tool Blog
Tantillus.org	Mini Printable Lathe	How NOT to install a Pololu driver

nice post sublime.
seriously nice.
calling spades spades while still in the end remaining rational and I guess a bit supportive.

Mung: please take sublime's advice and organise a working printer. much of what he is saying and what you seem to misunderstand or misconstrue will become much more ovbious when you start to use your extruder in a printer.

I love seeing the way you are challenging the status quo and trying new things, but when you are provided with anecdotes based on peoples real printing experience that suggest some of your ideas will have problems, you need to counter them with experiments of your own that demonstrate that your extruder doesnt suffer the same fate as many that have gone before you. If is does have failings, then you can make improvements and the whole community will benefit from your experimentation.

Never a dull moment in the RepRap world...

@Mung - If you do decide to get a printer up and running also think about extruding other materials, we have LOTS of plastic extruder and hot-end designs, seriously almost every single reprapper decided to try and make a better extruder and/or hot-end design but we have few Paste, ceramic and other materials being investigated. A different way of thinking may result in a whole other type of extruder or building material.

You have over 4,500 hits on this post, some are visiting for entertainment value that's for sure and I stopped posting advice to you pages back. but take Sublime's words and work with them. It may give us all something good to think about in the long run.

---

[richrap.blogspot.com]

I thought I would show another video of the nozzle with hand extrusion.

The thermometer has poor thermal contact with the nozzle so shows a few degrees off what the actually temperature is.

The cold end of the extruder is about 30' celcius.

The power was removed at the end of video to show the thermal storage of the hot end, it seems to be able to extrude about 166mm of 0.52mm polyamide filament from the nozzle after heating power is removed (the nozzle was drilled with a 0.35mm drill).




I think it is not good I comment further about what is said here, as I do not want to annoy people further, but someone tells me I should look at commercial printer and I find good links shown below may help others:

[haveblue.org]
[www.incredilution.com]
[www.ebtx.com]

Good test.

One question, how long does the cold end stay cold enough to hold?

---

FFF Settings Calculator	Gcode post processors	Geometric Object Deposition Tool Blog
Tantillus.org	Mini Printable Lathe	How NOT to install a Pololu driver

Sublime Wrote:
-------------------------------------------------------
> Good test.
>
> One question, how long does the cold end stay cold
> enough to hold?


This is a question like 'how long is a piece of string?', there are so many factors involved you have to be a little more specific.....

Maybe we could define it as thermal resistance at a given temperature and calculate on the temperature difference but it would take me too long to explain the theory and calculate the maths for the hotend so I will not.

Also I think I read somewhere that most polymer printers work in a heated chamber at 100'C, so I think maybe there would be some active refrigeration required to keep the cold end at the correct temperature.

There are a number of variables involved and I think you should look into the whole subject to get a better idea then I hope you will be able to produce the best extruder!!!


There are always improvements to be made, but I fear I have finished on this project for a while now.

I shall say that given the conditions in the video the hot end does not and probably will not get any hotter than in the last 1 minute of the video.

The key is making sure that convection and radiation of heat over the surface area is greater than the conduction up the shaft of the extruder.

Note that the hot end in the video is 26mm long, and also that I made a mistake about the nozzle orifice drill size it actually was done with a 0.45mm drill.

I also should mention that people seem to think that this could be developed further and quicker, as I saw a comment saying 'making an extruder only takes 2 hours', this is not the case. Most things take a lot more time to consider and design than they do to assemble. This extruder is just a hobby project and I do not have time to throw at it.

The current extruder may only take a few minutes to assemble from parts, but each part takes time find the material, to machine, to measure and test, and with the extruder barrel takes over 12 hours of treatment to improve thermal characteristics(although unattended).

I hope that others can improve on this extruder as it is something that anyone with some time and a little money could work on (You really do need a small metal lathe to make better quality parts and manufacture things quicker, but this should cost £100 or less for a second hand lathe).

I also have decided not to provide any plans for this, I hope that I can show others the methods used but they can make their own experiments as this will hopefully lead to new improved extruders rather than many identical copies.

I have now a few old experimental extruders that I will send if people want to test them, as I know someone suggested I send them a sample to test.


Finally I will give people some extra information that maybe useful and say that there are many Tellurium salts that are available that can be useful for electrochemical deposition onto the extruder barrel.

The web site below shows some good science, though I was taught at school about the development of HSS as being a first great example of industrial experimental science.

[www.zts.com]




Finally I should say I will hopefully make some editorial to the wiki about this extruder and I hope to make a final video about general development of an extruder in the next month, but I think this extruder design will not probably ever become finished now.

Maybe next year I will finish the next generation design?

You are keeping the cold end pointing down, thats what is wrong with it. In a printer will be on the upper side, which will make the heat go there much faster. Heat travels upwards. And since there is no radiator on the (upper!) cold end, the heat will just accumulate there untill will reach the equilibrium point with the medium, IN TIME. Take into account that average print takes 8hrs. I had my longest print of 32+ hours straight.

For a test, put the hotend with the cold side upwards, and keep it heated for like 4 hours or more and see what happends.

Sublime said the entry point should be below 60C, but i would rather be inclined to say below 50C, for the safe side and long life of design. Also consider when you put the thermocouple on the surface of the object the temperature it reads lower than it actually is. Use a thermocouple mounted with kapton tape or silicon of motor engine gasket. The temperature probe in the video i believe its for liquids and i dont think its adequate for this job (compare and see).

The upper side needs pretty much an aluminium plate for carriage, or some other heat sinking form, otherwise is very much susceptible to fail on the cold side not being cold enough - in the first hour(s). And with an aluminium radiator / plate carriage, (e.g. with a heat sink big enough in any form) - there are many extruder designs, very much since the old times of reprap. Search terms "all steel hot end" on thingiverse. So you will see its nothing new. Also, i believe that once you mount the heat sink, the requirements for the heater will change slightly.

One direction of study i never got around to, is to try make the filament not come from above, but instead from a side, exactly because heat travels upwards. I think this could ease the insulation requirements somewhat, but i never got around to see how would that be done, and how exactly it would work. But why not you might test it - that is if you find it a worthy thought.

Edited 2 time(s). Last edit at 09/11/2011 08:19PM by mung.

Thanks for the help Noobman, your ideas have been very useful, but have left me with more questions.


NoobMan Wrote:
-------------------------------------------------------
> You are keeping the cold end pointing down, thats
> what is wrong with it. In a printer will be on the
> upper side, which will make the heat go there much
> faster. Heat travels upwards.

Do you have a good set of equations for calculating the heat transfer that I could use for this upward travelling heat?

I really know nothing about thermal transport or thermal engineering as I studied electronics in college, I remember when I used to fly there were exams on weather involving tephigrams, adiabatic lapse rates, and humidity and I know gravity is important in convection within fluids where thermal expansion and mass motion occurs.

How can I workout how much heat will travel upwards by radiation and conduction???

Is gravity a very big factor in conduction and radiation of heat?

I seem to remember something about phonons and electron transport and maxwells equations, are there engineering methods for working out the heat transfer in metals?

How can you calculate the speed of transfer?


> Sublime said the entry point should be below 60C,
> but i would rather be inclined to say below 50C,
> for the safe side and long life of design.

How much difference in MTBF does the reduction in temperature make?


> The upper side needs pretty much an aluminium
> plate for carriage, or some other heat sinking
> form, otherwise is very much susceptible to fail
> on the cold side not being cold enough - in the
> first hour(s). And with an aluminium radiator /
> plate carriage, (e.g. with a heat sink big enough
> in any form)

How can I find out what sort of heatsink I should use?

Must it be aluminium, or could I use something else like copper?

Do you have calculations I could use to find out how much surface area the heatsink needs?



Also, i believe that
> once you mount the heat sink, the requirements for
> the heater will change slightly.

How will the requirements change, what else will I need if the requirements change?



Unfortunately I have not done any further work on the extruder, but I am still considering it and hope to do further work in a month or two.

The feedback and explanations of the best design methods have been VERY useful and I am grateful for everyones help.

The sharing of information and all the design calculations really help to speed up the development process, and I feel the help from experts and specialists in the different engineering disciplines is invaluable in bringing together a design with multidisciplinary engineering requirements.

mung Wrote:
-------------------------------------------------------
> Do you have a good set of equations for
> calculating the heat transfer that I could use for
> this upward travelling heat?
No, not really an equation anywhere, and wasnt of any concern to me mathematically, only in practice like this. On a blind guess i would flavor the normal distribution e.g. ~70-80% upwards and rest downwards but i also think its not as simple as that (again blind guess). I dont think it matters to us and i dont think it is even possible to make an heat energy model for something like this, as the calorimetrics as science is incredibly hard to make accurate models and predictions. e.g. making a tool to measure heat energy accuratelly is something almost impossible / incredibly hard to do even in a static system, not to mention a dynamic one, e.g. extruder will loose heat in filament being melted downwards. Also open systems are even more complicated, and even this will be an open system: there will be air current traveling upwards from heater to the top of carriage, which will increase the heat transfer upwards, and even some of the heat from heated bed will get into the carriage/top of the assembly.
Basically there are almost no closed systems in calorimetrics, this is why if you make a fire on the gound, from its center 10 cm horizontally you might not feel it, but same 10 cm on the height you will be right in the middle of the flame. I think its coz the air current pulled from down and moving upwards coz of a lower density, etc, so the system is basically open with the environment and cant be considered closed, as if it will be like that than the flames will be a circle around the center and not the typical flame going upwards aspect. I wouldnt be surprised if the typical "stylized" aspect of a flame, e.g. the sectional look through a "cooking machine flame" is the starting point for the mathematical model you are looking for, but again i dont think its that simple. Again, i dont think anybody could be able to come with an accurate caloric model of this system, and it is tones of times more complicated than it looks at first glance.
Bottom line I would prefer to do things like these empirically: think about it, put best guess in practice, measure and see if it works.

> Must it be aluminium, or could I use something else like copper?
I guess anything should work, some even use a fan to cool the heat barrier.
But copper will add weight. The X and Y weights are very important for several reasons:

1) First, the static deformation of the 8mm bars. Dunno the values for standard mendel or prusa. For a new machine i am making, i calculated few days ago the whole X carriage assembly has about 1,5 kg (probably same as the mendel). On a single 8mm bar (e.g. 2 vertical bars one 180 deg wont take weight on it), the deformation with steel coef (the bars are hardened so will be lower, but its ok to under~approximate instead of being over-optimistic) on length of 60 cm gives about 0.0336 mm in the middle - this is probably high value, normally should be like 10 times (this was [static] value) lower than minimum semnificative distance (~0.01 in our case). So its best that i dont use 8mm vertical, but make it horizontal bars, like in prusa, which even with one 180 deg constraint, will still distribute weight on both bars instead of just one, and then the deformation decreases exponential (or logarithmic, something like that). Otherwise i would of needed 12mm bars. Conclusion: a Y with 2 vertical bars might (at limit) be ok for a standard mendel size, but doesnt seem to be ok if bigger.

2) I think the bed and carriage have to be equal in weight, and this will impact on acceleration. Its general opinion that X and Y should use same make and model of motors, probably for same reasons. Basically X carriage and Y (bed) need to be minimal weight. And probably only the biggest value matters for combined acceleration. At this point i honestly dont quite understand why ppls use bowden cable extruders with something that looks like a ~3-4 kg heated bed, i dont see how the overall acceleration could improve this way, but anyways thats my point of view so it cant be anybody's elses.

> Do you have calculations I could use to find out
> how much surface area the heatsink needs?
No, as i said, i would rather experiment and see if it works or not. But i will attach a picture of my own solution for it. My older mendel carriage trapped heat and i think it was pretty bad for a number of reasons, i looked on thingverse but none of the other solutions inspired me confidence, so i decided to make it different. What i did is to use a small alu plate (~5.5mm width) on both roles as a carriage and as a heat dissipator. Most of screws are only to one side (dunno english name for this screw setup, in my language its called "prezon"). To make it like that, tap lower diameter to take out all play space, e.g. to fit a M4*0.7 screw like that, tap M3.5*0.7 and it will be a good slightly forced fit. On the side with the single bearing i had horizontal holes through material (like the horizontal ones for bearings), so a fan could be mounted on the opposite side of where belts fit. And the carriage is for mendel, i think for prusa the belt grips have to be different. The ptfe can be unscrewed from those 2 screws, leaving the extruder on. Also the extruder can be dismounted individually if its screws have access (std wade only with a slim 7mm key for the motor side - difficult access). Overall, my carriage looks bad, but I consider it to be best solution. I consider it better than any other printed carriages at least. This is why i would recommend it. Maybe I should of done a few big holes in it to decrease the weight, but again my bed is heavier than that, so at next change i will make my bed lighter.

> Also, i believe that once you mount the heat sink, the requirements for the heater will change slightly.
Of course, but on the overall i guess it still depends on the overall picture, e.g. if you use a thermal barrier or not (like all-steel-hot-end). I used both old carriage and the new one with same std 6r8 resistors, but both had ptfe. I have no experience without ptfe, so dunno if, or how it would work without a designed thermal barrier, like in an all steel hot end it would probably need active heat dissipation - like a fan mounted to one side.

Edited 1 time(s). Last edit at 09/12/2011 06:51AM by NoobMan.

with the occasion of the post above of this i also putted the information about carriage on a page here [www.reprap.org] and as a second timer, maybe i gave a better presentation / explanations there.

I am sorry to say I have gained no further help from your post, could you tell me some details about your age employment and education so I can better make judgement on weather to follow your ideas.

You say such things as 'on a blind guess', I have to say no guess is really blind as everyone bases their guesses on previous experience and if you could tell me your previous experience I could better judge how good your guesses may be.

I must tell you that I have NO REAL WORLD FORMAL TRAINING OR EXPERIENCE IN MECHANICAL ENGINEERING, so I assume that I do not really know much about the engineering involved in designing the reprap. This is really just a hobby for me, thought, I do have college education in electronics, and did maths and physics to advanced school level, which should give me some grounding for working things out from first principals.

You say that designs can only be made by experimenting, but I have always been taught differently, and I think that your view is not an efficient one or the current engineering status quo.

The empirical method is the basis of science and is the final and ultimate test in engineering, but engineering is a practical application of scientific rules/theorys/formulae/phenomena to attain real world objectives.

I think we all guess about things but our guesses are based upon mental visualisations of how the real world works, and often our internal visualisation are not in line with the way the world really works, especially when we try to visual things that may not be visible or are non visual phenomena.


I think I would describe engineering as an educated guess based on scientific knowledge.

Now, I think where simple experimental engineering can fall down is if it is based on poorly informed guesses.

Sometimes, peoples visualisation of the world is wrong, or the way that physics operates is counterintuitive.

In these cases mathematical analysis can be very important, even though it is often tedious hard work.

I believe in a real engineering situation people consult specialists who have spent many years obtaining very good knowledge of short cuts and rules of thumb used in solving engineering problems, and also have well tuned mental visualisations of the way the physical world works for their area of expertise.

I know that most things are generalisations and nothing is exact when trying to calculate things, but normally calculations will help to give useful approximations of the magnitudes of energy distance and time.

Basically, what I am saying is that I am not totally sure of my own understanding of the physics involved in thermal transport, but what you are saying is TOTALLY AGAINST EVERYTHING I HAVE EVER BEEN TAUGHT AT SCHOOL.

As far as I know only convection is effected to any great degree by gravity(due to adiabatic thermal expansion and bouyancy), radiation is an EM wave and although effected by gravity Quantum Electro Dynamic theory shows negligable effect under earths gravity, heat conduction as far as I am aware is treated as being phonon transmission in most materials or in metals it is more similar to electrical conduction.

From those descriptions I ask myself does sound travel differently upwards than downwards and do wires conduct electricity differently if they are horizontal or vertically aligned??

I think I have said before that I am making this forum thread to show what I have been doing and also to ask questions about what I am unsure of in the hope experts could give advice (I could look the information up but mostly I would guess it takes at least a few hours to do this, I am lazy???, I also assume I am probably correct in my guesses).

I have also unfortunately noticed that there really has been very little expert help on this forum and people have not even pointed out simple mistakes in some of the things that I have said.

I know that most experts have spent huge amounts of time and expense gaining their expertise and do not have the time to waste on an open source project, but this being open source and based in universities I would have expected some expertise?

Unfortunately I really do not have time to fully describe all these thermal transport phenomena, and if I did try to I am sure some of my understanding would be shown to be wrong, but please NoobMan unless you can show me some valid reasons for what you have said I think you should remove your previous postings from this thread and move them to a thread of your own as I believe they may well be irrelevant or misleading.


I do not want to be a troll or be rude, and it is good that you NoobMan are trying to contribute to the reprap project (there are many reason why it is good but I will not digress further).

This thread has seen some people saying there are fights starting and I do not want to start a fight, I know people try to give encouragement by posting comments, and some people just want to troll, and others may even have commercial reasons for trying to guide the development.

I would be pleased though if people try to keep on topic and only answer questions if they are sure they have good scientific knowledge of the area involved.

Or if they just want to add encouragement then please keep it short, and if you want to add something slightly off topic start a new thread and add a link to it.

Sorry this post is so rambling and long winded, I hope I do not cause any offence.

I also still hope to add some further information on the first full prototype and information for would be developers, but never find the time yet.....

FINALLY A NOTE ON FURTHER DEVELOPMENT:

I am still sort of working on my design, though only subconsciously, but I think it is working towards a new simplification, thought really the whole design will still always just be the same as the standard pinch wheel concept, but built with more finess and accuracy to make it smaller, lighter, and more efficient.

I will make an educated guess here based on what was made for the first full prototype, I think the weight will be at least half, though dimensions will stay fairly similar there is a possibility for maybe 15% size reduction on the first prototype. I will also make sure the resistors cannot burn out (blow out or fuse) when driven by 12V. I am guessing that I will ditch the multiple nozzle changer as I think this would not be something anyone would use much. I think the build time could be over 200 hours as there will most likely be some precision machinng and I think the geometry and topology may need a fair amount of analysis and adjustment to get all the shafts, pivots, and linkages in the smallest possible volume.

mung Wrote:
-------------------------------------------------------
> I am sorry to say I have gained no further help
> from your post, could you tell me some details
> about your age employment and education so I can
> better make judgement on weather to follow your
> ideas.

What are you, some kind of robot or a copier machine? Dont follow my ideeas, the posts here are to exchange impressions, from which you form your own ideeas and follow those. Not mines...

I will stop posting here before i am asked about the length of my most sigificant bit (/organ) for contest purposes. By all means if such a question does arise, consider that i have none.


> You say such things as 'on a blind guess', I have
> to say no guess is really blind as everyone bases
> their guesses on previous experience and if you
> could tell me your previous experience I could
> better judge how good your guesses may be.

"On a blind guess" its a phrase from my language which means literally that - what one would do / choose if no other / further data would be available. And about you considering how good my guess would be ... well i guess thats up to you isnt it? Ironically, you will just have to take a "blind guess" on this.

> I must tell you that I have NO REAL WORLD FORMAL
> TRAINING OR EXPERIENCE IN MECHANICAL ENGINEERING,
> so I assume that I do not really know much about
> the engineering involved in designing the reprap.
> This is really just a hobby for me, thought, I do
> have college education in electronics, and did
> maths and physics to advanced school level, which
> should give me some grounding for working things
> out from first principals.

Its a hobby for us all. This is a forum. The peoples posting here can be doctors or clean toilets for a living - i couldnt care less since it would be same thing. Dunno about you but I can learn from anybody who has a good tip, doesnt matter if he is poor or rich. AFAIK the only thing required to join the forum and post here is ... to know to write.

> You say that designs can only be made by
> experimenting, but I have always been taught
> differently, and I think that your view is not an
> efficient one or the current engineering status
> quo.

I didnt said that. I said i would rather do one extruder and see how it works, instead of ... well wondering for a half of year about what what the heat transfer model would be and still having nothing. If i want to get results i get em that way. Build one of the damn things, then see if it works and how would i improve it if needed. For example I believe that could probably make 100+ extruders with hotends like mine before someone can finish a somewhat realistic computer model with full variables about all temperatures and transfers in a reprap machine.

> [... passing over some generalisations about science and stuff ...]

> In these cases mathematical analysis can be very
> important, even though it is often tedious hard
> work.
[...]
> Basically, what I am saying is that I am not
> totally sure of my own understanding of the
> physics involved in thermal transport, but what
> you are saying is TOTALLY AGAINST EVERYTHING I
> HAVE EVER BEEN TAUGHT AT SCHOOL.

Dont yell. Oh, wanna do that caloric modelling for a mendel? By all means dont let me stop you. I cant help you but i can wish you good luck with it. Which i do.

> As far as I know only convection is effected to
> any great degree by gravity(due to adiabatic
> thermal expansion and bouyancy), radiation is an
> EM wave and although effected by gravity Quantum
> Electro Dynamic theory shows negligable effect
> under earths gravity, heat conduction as far as I
> am aware is treated as being phonon transmission
> in most materials or in metals it is more similar
> to electrical conduction.

Since i cant belive that you havent seen a fire in your life, seems like you never thought of it. Read again what i posted about the shape of it. When you light up a match, how come 90% of the flame is above the match, and only a small part under it? Why did i said that carriage gets heat from extruder and even from heated bed? Air is very good insulator, but sadly is not perfect. Flashnews, the heated air has lower density, travels upwards, gets heat on the upper parts. Baloons fly this way. So no, actually heat and electricity seems to be kind of different, at least in the universe i live in. In the universe i live in, also when you go up in the air, the temperature decreases. When you go down in a frozen lake, temperature rises. Maybe pressure is a factor (p*v/t = constant) but also maybe there are others. Because nothing is not a closed system.

> From those descriptions I ask myself does sound
> travel differently upwards than downwards and do
> wires conduct electricity differently if they are
> horizontal or vertically aligned??

And just like some other guy said, the light travels faster in expensive optic cables than it does in cheap ones.

> I think I have said before that I am making this
> forum thread to show what I have been doing and
> also to ask questions about what I am unsure of in
> the hope experts could give advice (I could look
> the information up but mostly I would guess it
> takes at least a few hours to do this, I am
> lazy???, I also assume I am probably correct in my
> guesses).

Oh well, sorry here it is my mistake. Apparently i was trying to give my advice to someone who doesnt actually want to receive any. Deffo my mistake.

You see, I thought this is just a thread like any other. Dunno about the general perception, but my perception about forum threads is that basically its an open invitation for anybody to post there. And forum threads dont "belong" to someone, but do belong to everybody. What is important to respect is their topic (e.g. dont post off-topic), and that is to pay respects to all the other readers (not the original poster), but to those readers looking for information and just reading the thread names should get what the thread is about. Maybe my understanding about it is wrong tho, so maybe threads do belong to private ppls.

You say that instead of getting advice you could look it up but you are lazy ... by all means, look things up instead. Just a question, can you pls tell me where do you exactly look up for answers about your reprap issues ... because i would aslo love to have such a place, since i never have "experts" to answer me.


> I have also unfortunately noticed that there
> really has been very little expert help on this
> forum and people have not even pointed out simple
> mistakes in some of the things that I have said.
>
> I know that most experts have spent huge amounts
> of time and expense gaining their expertise and do
> not have the time to waste on an open source
> project, but this being open source and based in
> universities I would have expected some
> expertise?

By the way, what i would consider an expert is Sublime, the very guy which you chased away few posts ago. I consider him an expert as i have seen it being very active, both about printing stuff and also on these forums and seen many good opinions of his. I would personally take any advice he would of give me, but well, thats just me and probably for ppls like me. Ppls smarter than that would clearly treat the matter differently.


> Unfortunately I really do not have time to fully
> describe all these thermal transport phenomena,
> and if I did try to I am sure some of my
> understanding would be shown to be wrong, but
> please NoobMan unless you can show me some valid
> reasons for what you have said I think you should
> remove your previous postings from this thread and
> move them to a thread of your own as I believe
> they may well be irrelevant or misleading.

Dunno about which parts of my post you refer to. If its about heat travelling upwards, as i said, light up a match and compare the flame portions that are above or below the match. Or read about how baloons work.

If its about the carriage, and the way it gets heated, well you cant consider the hot end being separated from the carriage itself, as i said, its open system. But if you do need a second better indication, in fact, most of the way i did that carriage is inspired from [hydraraptor.blogspot.com] a blog to which i should give full credit not only for that carriage piece i did, but for most of my opinions that i have around the reprap.

And yes, that is a carriage that is supposed to dissipate the heat, yes that heat coming from extruder. Its right dead-on on-topic. You just simply dont know it because you havent been confronted with the issue yet.


> I do not want to be a troll or be rude, and it is
> good that you NoobMan are trying to contribute to
> the reprap project (there are many reason why it
> is good but I will not digress further).
>
> This thread has seen some people saying there are
> fights starting and I do not want to start a
> fight, I know people try to give encouragement by
> posting comments, and some people just want to
> troll, and others may even have commercial reasons
> for trying to guide the development.

What you thought, that i am an aluminium industry magnate trying to convince you to buy a 10 cm piece of aluminium? By all means, dont. Discard everything i said, and do it different. Do it even better, so on a sunny day i would have a reason to take my hat out. Contrarily to how it appears at the moment.

> I would be pleased though if people try to keep on
> topic and only answer questions if they are sure
> they have good scientific knowledge of the area
> involved.
>
> Or if they just want to add encouragement then
> please keep it short, and if you want to add
> something slightly off topic start a new thread
> and add a link to it.

Peoples will behave like they want to behave, not like what you want them to. Sad, but fact of life.

> Sorry this post is so rambling and long winded, I
> hope I do not cause any offence.

You do lack some of the finesse that only wisdom can offer. Wisdom comes only after pain and suffering, and grows in detriment of ego. By these 2 signs, I would say you are at the start line. Cant hold it against yourself though, we all had starts, more or less different, but for sure we all had starts.


> I also still hope to add some further information
> on the first full prototype and information for
> would be developers, but never find the time
> yet.....
>
> FINALLY A NOTE ON FURTHER DEVELOPMENT:
>
> I am still sort of working on my design, though
> only subconsciously, but I think it is working
> towards a new simplification, thought really the
> whole design will still always just be the same as
> the standard pinch wheel concept, but built with
> more finess and accuracy to make it smaller,
> lighter, and more efficient.

Dunno about the rest, but I would just love to be able to work subconsciously on something - anything. Sadly in my case doesnt work as i need both my neurons to do even the simplest tasks. If you can do it like that, i can only congratulate you.


For the closing line, again sorry didnt realized this thread is exclusively just for you to post your work.

I will stop posting here to not disturb your important work any further.

Well good luck with your work and all rest.

Quote
NoobMan
I will stop posting here to not disturb your important work any further.

This is the best thing I have read in a long time. All of it was good, my stomach hurts from laughing.><

And thanks but i'm no expert.

@Mung
Please keep educating/entertaining us.

---

FFF Settings Calculator	Gcode post processors	Geometric Object Deposition Tool Blog
Tantillus.org	Mini Printable Lathe	How NOT to install a Pololu driver

Dear Mr. Noobman.

I really really am sorry for you as you seem unable to make any reasoned or relevant comments to this thread and you make yourself look foolish by contradicting yourself.

You tell me my design is wrong as heat travels upwards and I should design an extruder nozzle that the filament enters from the side.

Then a post later you tell me that I should not follow your ideas?

I have to write a short fictional anecdote to illustrate how your behaviour seems to me:

Imagine you were walking down the street talking to someone saying how your arm is very itchy today and some stranger walks up to you and says "I think you should cut your arm off"

You would probably look at them and think the person is mad, what sort of idiot would think cutting their arm off is a good idea???

Now you may laugh at him or tell him he is mad or call the police, or alternatively you may tell him you would never cut your arm off as you see no point in doing so, or you could start talking to him about how you once cut off someones finger (which has a very loose link to cutting off your own arm but really in the situation is totally irrelevant and pointless to the current subject), finally you could do what I think would be the most usefull option available to you which is ask the stranger who he is and why he thinks you should cut your arm off.

1) Now suppose the stranger says 'well I am a skin cancer specialist and I noticed when I saw your arm that there seem to very serious melanoma lesions on your arm and if you do not get your arm cut off very soon you are very likely to die within 12 months"
You would probably take his advice seriously and go straight to the hospital for tests and spend many hours researching cancer treatments and cures.

2) On the other hand if the stranger were to say something along the lines of "You will have to guess that, maybe I sell Prosthetics, maybe I know, you will just have to judge that for yourself" then if the stranger started babbling incoherently about how they will take their hat off when it is sunny and that they have no peenis and then says "don't follow my ideas you should exchange impressions and have you own ideas you are a robot!!", you would probably call the police and tell them there was an escaped mental patient bothering you. You would probably ignore everything the stranger had said and try to forget his comments as soon as possible.

So from my perspective based upon the way you have conveyed your ideas I have to say I would put you in category 2.




Anyways its all good fun and I hope you can take this joke seriously.

I still am not sure my extruder design will work, but unfortunately you have not helped me improve it and have not helped my understanding of heat conduction within metals.

I am not sure this forum is the place for too much humor, there are better places for such things?

Kind regards

Mr. Mung

mung Wrote:
-------------------------------------------------------
> Sublime Wrote:
> > 2) Needs some thermal mass to deal with the
> > cooling effect of pushing filament through it at
> a
> > high rate.
>
> Why do you need thermal mass, and what function
> does the thermal mass do to effect the temperature
> of the filament?

I suppose this helps mantain constant temperature as extrusion rate changes (or some other conditions like air flow) because additional thermal mass adds more inertia to the temperature. (Like a capacitor in parallel helps to fight high-freuqency components of voltage, but with temperature.) It acts like an additional feedback mechanism. Another possible beneficial effect from "thermal mass" is that it equalizes temrepature across extruder. Temperatures in different points of the extruder can be different due to thermal flow through it: heat from heater enters extruder in some region and exits in other regions (due to cooling by air or by extruded filament). The difference in temperature is inverse proportional to thermal conductance. (Faraday cage is a good electrical analogy to this situation.)

The above is only a suggestion since I don't have a RepRap. Also I'm a theoretical physicist.

I can also comment on a photo of your heater where metal plate had contact with resistors only on one resistor side. It doesn't look like a good thermal contact. This may lead to large thermal gradients across resistor (it's a guess either). Probably metal should at least wrap around resistor to pick heat up from all its surface and redistribute it without large temperature difference.

> I could answer your question for you, but I think
> if you could answer it you may better understand
> thermodynamics.
>
> input = output

This is true only in a stationary situation, which is not the case.

> > 4) Needs to be able to handle the pressures
> inside
> > the hotend @ 50mm/s @ 200+c for 4hrs to 24hrs
>
> I doubt this is a problem, but I am beginning to
> think speed could be a problem as many machines
> are running much faster than 50mm/s, I imagine
> that 500mm/s could be the upper limit.
>
> E = (t2 - t1) * M * S.H. + any latent heat
> required for phase change

About phase change heat. I've found some data on temperature dependence of specific heat capacity of ABS plastic, and there's only a small bump near "glass transition temperature" of 110 C, i.e. it requires virtually no heat, and specific heat capacity doesn't change much (only twice) across the entire 50 C -- 300 C range. See http://imtuoradea.ro/auo.fmte/files-2010-v2/TCM/Peydro%20Rasero%20Miguel%20Angel%20L1.pdf . I don't know about the other plastics.

mung Wrote:
> NoobMan Wrote:
> > You are keeping the cold end pointing down, thats
> > what is wrong with it. In a printer will be on the
> > upper side, which will make the heat go there much
> > faster. Heat travels upwards.
>
> Do you have a good set of equations for
> calculating the heat transfer that I could use for
> this upward travelling heat?

Obviously, the general statement that heat travels upwards is nonsence. Condution and radiation are not affected by gravity. Convection is affected since hot air is lighter and travels upwards.

As for the equations, I don't know the right set of governing equation for this case. It must take into account turbulence, and I'm no big specialist in it. It's easier to use ready formulae for heat loss by thermal boundary conductance from some book on thermophysics. For example, I've seen such formalae for different object shapes and orientations in the Russian translation of H. Y. Wong. Handbook of essential formulae and data on heat transfer for engineers. London & NY: Longman, 1977. But unfortunately this itself doesn't help estimate heating by hot air from an underlying hot object.

Well Personally I can't wait until tomorrow, I'm expecting some trans-interdimensional-anti-gravity-extruder to come out of this development, it can't simply be something as 'low-tech' as we already have - an isolated heating element that melts plastic by being forced down a section of cooler material by a stepper motor towards the hot-end by electronics we already have. That would be too easy...

Seriously, why do I keep on clicking on this thread, it's making us all look bad. - > my very last comment on this particular topic.

Good luck Mung, on everything.

---

[richrap.blogspot.com]

Okay, I am sorry, I feel I have been mean spirited to Noobman and some of the other commenters to this thread, and I should have explained my intentions better.

I have reviewed this thread and see the many errors and poor explanations that I have made and some of the silly questions that I have asked.
When writing about technical subjects I know it is very important to double check data and formulae and make sure a constant well defined terminology is used consistently and accurately in descriptions.

I think in some posts I have blurred the line between what I am describing considering the hotend as part of the extruder and visa versa and using the term extruder when I really mean hotend.

Unfortunately making good quality accurate technical descriptions takes a lot of time and thought, and I think I may need to go back and edit some of my mistakes.

These mistakes may have led people to think I was refering to the entire extruder carriage and drive system, and I am sorry for allowing this confusion and then blaming people (noobman) for posting off topic.

With regards to conflicts of ideas.
I can tell you I know I am not the sharpest tool in the box, but I also know I am not totally blunt, I always assume I am somewhere just slightly about the median. I feel that there are some trying to troll and make me look stupid, but I am perfectly willing to accept my mistakes if you point them out to me rather than mislead me in some other direction.

When I ask questions it is truly because I am not sure about a subject and if I disagree with a person I am not making a personal attack against that person, I am merely saying I have been taught differently and I require further explanation to change my view.

I have spent a little time thinking about how heat moves in the hot end and have assumed some factors can be considered to be negligable.

Thinking about factors such as where is the best point to place a heat sink on the hotend and what exactly do certain physical characteristics mean, has unfortunately left me less certain about what I am doing than when I started thinking on the subject.

When people then start giving me conflicting facts to those I have been taught then I become even more confused.

I know everything effects everything else to some extent, but influences have differing magnitudes, I think there is an art in science and engineering in knowing which factors can be ignored.

Starting to question the very basis of physics with questions like 'if gravity is energy does gravity effect energy if e=mc^2 ?', 'when does energy become mass or mass become energy ?', 'what exactly is a dimension?' , 'does temperature and ionisation energy of molecules change from gas to solid?'.

Anyways, I would like to thank Alexey Kruglov for his good book recomendations, I will check them out of the library and hopefully get some simplifications for the heat equations.

I hope there has become some agreement that thermal conduction is not greatly effected by gravity in comparison with the effect of gravity convection within a fluid (I think I may also understand 50% of the reason why, though it would take a rather long explanation).

When I talked about my extruder having more finesse, I was not trying to use the word as some inference of elitism or superiority, I simply meant that I hoped to create something more precision made and exactly engineered to the exact requirements of the job to be done.

If you compare a model T ford and a citroen C5 I would say the difference between the two is finesse, both are essentially the same thing (Cars) and both are brilliantly designed and engineered for the era in which they were created.

Anyways, in finishing I will say I am starting to become unsure what this forum is here for, and like richrap and noobman and others I wonder if I should read or post here again for fear of annoying people if I do not understand their view point or ideas.

I have made some severe and costly mistakes on other projects recently so though a change may be as good as a rest so spent a few hours on assembly of the extruder this past weekend.



There are a few additional components required but the weight will stay around what the scales show (72.1 grams).

The motor is a pm20 stepper and the chip will run it in quadrature mode.

The extruder nozzles are dual mounted size for 3mm and 2mm filament, so the extruder can run 4 filaments 2 of each size.

I tested one dual nozzle and it works well for 2mm filament but needs a resistor changed to improve performance of 3mm extrusion at 12volts (extrudes 3mm well at 15watts but this requires 13.5volts)

I will not explain further than to say the extruder is in rough fit and assembly, there are probably 50-60 hours work required before it can have initial tests.

It is possibly too complicated around 30 parts, but slightly better built than the previous prototype so should be smoother in operation.

Edited 1 time(s). Last edit at 10/24/2011 04:34PM by mung.

Unfortunately I have again canned and dumped the current design of hotend as something jumped through my subconscious and I hope to maybe order some parts for christmas to work on a new concept over the holidays.

I am hoping to try something that should again reduce parts count of the extruder and improve performance, from my initial guestimates I think the hotend should be down to five parts and weigh around 1.3g for 1.75mm feedstock and 1.9g for 3mm, heat losses should be cut by about 40% which could mean sub 5W power required for 300mm per second of 0.35mm filament extruded. Also my new concept should be far more easily produced by automated means, probably only pennies per hotend.

In the next few days I may upload a video of the last of the pervious hotends to show what I felt to be disappointing results for the 3mm feedstock.

As I have said almost since the start it seemed smaller is more efficient, but I will be trying to get a reasonable 3mm feedstock hotend working despite what I see as its flawed conception. Some people seem to think it is less efficient to use 1.75mm as it costs more, but this is only the case because suppliers want to charge more and people are willing to people pay more for 1.75mm there is little reason for the premium price.

If a full energy analysis is done of all phases of the production and usage of the filament 1.75mm works out about 42% more energy efficient than 3mm.

So if I get the parts by christmas I may show some test video of the new hotend in the new year?

mung Wrote:
-------------------------------------------------------
. Some
> people seem to think it is less efficient to use
> 1.75mm as it costs more, but this is only the case
> because suppliers want to charge more and people
> are willing to people pay more for 1.75mm there is
> little reason for the premium price.

I don't know why I am trying to help again, but here I am.

It takes almost THREE times the time to extrude the same volume of 1.75mm as it does 3mm which means it costs THREE times as much for the service which is most of the cost of the filament. The resin it self costs almost nothing.

---

FFF Settings Calculator	Gcode post processors	Geometric Object Deposition Tool Blog
Tantillus.org	Mini Printable Lathe	How NOT to install a Pololu driver

Quote

It takes almost THREE times the time to extrude the same volume of 1.75mm as it does 3mm which means it costs THREE times as much for the service which is most of the cost of the filament. The resin it self costs almost nothing.

I don't know why you are trying to help, and I don't know why I am bothering to defend my statements?

I could spend hours trying to explain and define every minutia and detail to avoid people telling me I am a liar and still they would tell me I am wrong??

I may have made some mistakes in my calculations??

does
time = energy
time = money

??

The answer is it depends

For machines time does not cost that much more money unless there are other more valuable jobs waiting to be done?
If you are paying a person to sit and watch a machine maybe??

There are essentially two energy costs (excluding transport weight and volume considerations, and possibly product longevity)
1) energy cost to produce (manufacture) the feedstock.
2) energy cost to use the feedstock in your 3D printer.

****Now, suppose a large scale filament extruder had to extrude 3mm and 1.75mm filament lets look at the costs.....

1) Heating of the plastic would cost the same for equal weight if it took the same time.
2) extruding of the filament would take the same time if the same weight was extruded in the same time.
3) the only extra cost in energy terms is the difference in friction losses due to surface area skin friction and fluid viscosity internal friction and vorticity/turbulence

So what is the difference in friction losses (I dont have time to show you the calculations), want to do the math for yourself?

I should mention you can easily run the extruder with multiple filaments extruded from the same die so that the extrusion time is the same for both sizes and there are other ways to boost efficiency which are better/worse for the different sizes of filament e.g. cooling is quicker with smaller filament and if using heatpump heat recovery heat can be feed back into the plastic melting system more efficiently. Large industrial plant can be better insulated made more efficient and has larger economies of scale than losses involved with small machines such as e.g. 3D printers and toaster ovens

**** Now what are the costs and disadvantages at the 3D printer side?

1)melting 3mm filament costs 74% more energy to have the same weight of plastic extruded in the same time as 1.75mm
2)the power required from the extruder drive motor is 22% greater to extrude the same weight in the same time.
3)the extruder hot end will break and clog more frequently due to higher stresses and required geometry.

Everything has an optimal size and geometry, I could give you some examples but I have already wasted 20 minutes with this reply.

If you want to refute what I am saying show me your calculations, its very easy to tell someone they are wrong but I would like to see you proove that you are right based on all energy factors involved in the entire process with some hard numbers.

Once you show me your proof then maybe I will spend the time to give you all my calculations, as this is just a hobby and a bit of fun, I really can't be bothered putting a full technical report on this forum.

I'm done...........

You have not looked into having filament extruded so you are saying shit again with no knowledge of the process. You assume you just squeeze a bunch of plastic out of a hole and it ends up at the exact correct size and shape. Guess what you are wrong its a huge process to achieve a tolerance of +/- .01mm. Please go back to buying lasers without looking into what needed. I was looking forward to seeing you saying more stupid shit without thinking on a new subject.

GOOD BYE FOREVER.................

---

FFF Settings Calculator	Gcode post processors	Geometric Object Deposition Tool Blog
Tantillus.org	Mini Printable Lathe	How NOT to install a Pololu driver

Its the weekend and I blew £12 on parts for my new concept extruder hotend (the parts arrived yesterday). I thought I would just update that I hope maybe to try an experiment tomorrow. I have trepidation about success, but its probably a 50/50 chance of working and if I do not try I will never know. May send a video if it is unsuccessful, but if it works I think I may patent the idea first

Anyways I felt I should comment on sublimes previous outburst:

Sublime said

"I'm done...........
You have not looked into having filament extruded so you are saying shit again with no knowledge of the process. You assume you just squeeze a bunch of plastic out of a hole and it ends up at the exact correct size and shape. Guess what you are wrong its a huge process to achieve a tolerance of +/- .01mm. Please go back to buying lasers without looking into what needed. I was looking forward to seeing you saying more stupid shit without thinking on a new subject.
GOOD BYE FOREVER................."

I really don't know why you seem so angry and defensive Sublime.

I am not trying to attack you I am just stating my opinion based on calculations I have done and enquiries I have made. I am perfectly willing to admit I am wrong (I am wrong quite often).

What I like is the way you avoid the real issues and do not fully explain how you come to your conclusions. I know you are an intelligent person and you obviously work hard and have done some excellent work. I have taken the time to look at you web pages and I can see you have created useful stuff for the reprap community and are obviously highly politically minded advocating anti censorship and socialism. But it seems to me that what you are doing in this thread is a rather insipid hidden form of censorship. By refusing to explain and proove your position you are hiding the facts and trying to force people to make their decisions on validity based on your previous reputation.
Nothing of use will be gained by simply saying "screw you guys, I'm going home".
I am not sure what you hope from the reprap project, but I would have thought that the highest hope of a project is to achieve the goal in the fastest most efficient way?

Again I find everything is complex and the more one analyses the problems both social and technical the more complexity is realised.

I remember someone saying to me that trying to get highly technical experts to work most efficiently together to achieve a single task was like herding cats. But I think the way you are handling your correspondence is not being productive and is really making me feel like never sharing any of my hobby with the community. If you make it so that people do not want to share then there will be no more community projects. I wonder whether maybe this project has too many experts to work together efficiently?.

I will say that I really have no idea about all the issues involved in creating filament but I did ask a plant manufacturer about it and they said that their plant would extrude the same amount of plastic in a given time regardless of the filament crossection although higher HP of motor and extra bobbin winding stations would be required. As I said before economys of scale are also involved and although I know reprap is cheap for other reasons than because people buy 1 million items rather than just one, it has to be said that small batch production costs vastly more. Maybe it takes a few hours to setup the machine for a batch run so the larger the batch the less percentage of total cost is the setup time cost. I think also small producers have less expertise than the large companies or plant manufacturers so maybe cannot get the best machine calibration?

I got an interesting video link from one enquiry showing their plant that gave me some idea about the process:




I can show my very simple calculations if you really want, though I doubt they are much use to this problem and it will take me a long time to try and input the mathmatical notation as text.

I will finish saying that I fully admit that I am no expert, I play with reprap extruders as an interesting hobby, I also have no idea about the entire reprap system as said I do not and probably never will have a reprap. This has some disadvantages for other reprappers but also has some advantages, as I specialise only on the extruder I spend more time and therefore gain more expertise and insight into the extruder than if I spent alot of time building the entire reprap machine.

Specialisation is the key to getting the best results, but this brings the problem of integrating all the specialised parts together, and I think this could be a problem.

I also just had a postscript thought about your statement about accuracy of filament +/-0.01mm has anyone done research into how important this is?
I feel the filament does not need to have this accuracy and also maybe there are ways and means to cheaply measure the extruded filament from the reprap nozzle??
maybe:
1) laser measurement at the nozzle orifice (like the sensor from an optical mouse)
2)torque measurement from the motor
3)pressure measurement (a strain gauge on the nozzle)
4)weight of plastic extruded on the object (strain gauge weighing the object on the print bed)
5)other means??

Update on the new hot end concept...........

As usual everything is more complicated than expected and the first test of the new concept failed (sort of worked briefly).

I was hoping that a mechanical clamped joint would suffice for the electrical connections but unfortunately this had some problems.

I now have to think about alloying some higher temperature solder from lead or more likely zinc (unfortunately I don't have any zinc here now) or thinking about possibly spot welding some components.

I am still hoping that the new concept may work as the hotend heated for a few minutes but I think as it required higher temperatures due to the larger feedstock used the heat possibly cause oxidation or sooting of the electrical contacts (my first guess at reason for failure though this is not a conclusive or fully tested and analysed guess, I give it 50/50 as to being the cause)

The metal parts and heating element have all been machined and assembled and the 3mm feedstock hotend weighs in at 1.5grams, it looks nice and I think could be the absolute minimum possible to create a working hotend (higher accuracy machining could shave off maybe 10% on weight).


I am still hoping to get time to video some of the older work and show it here, though its probably pointless as the old work is probably obsolete?

Maybe tomorrow.

Just another poorly filmed and badly planned video showing the dual extruder hotend being tested by hand pushing the filament.

The video should have been filmed more close up but you can probably see what is going on.

I measure the two feedstock filaments so everyone can see what size they are (blue is 3.2mm green 2.0mm)

I think it should be fairly obvious that the blue takes far longer to melt and extrude (The simplest explanation I can give for this is that the smaller filament has a higher surface area in contact with the heating barrel compared to the volume of plastic that is being heated, there are other factors involved though)

I was going to show a review of all previous extruders and hot ends in the video, but once filming it seemed best to finish filming once I had done the test extrusion.

I may possibly film this again tomorrow??






EDIT: Forgot to mention.....

Resistors are 0.5W rated and were running at 9W, the double extruder was the last revision of the pervious design and weighed 7.4grams.
The cold end of the extruder runs about 55 degrees C when heated to extrude the nylon filament, both the 2mm and 3.2mm nozzles are drilled to extrude 0.4mm filament.

Edited 1 time(s). Last edit at 12/31/2011 02:42PM by mung.

I have got back to considering the extruder again as I am hoping to order some special materials for soldering (I think I am going to go with nitric acid for pre cleaning, very thin zinc foil for the solder, and a special J M stainless flux powder).

As the last video looked rather crap I rerecorded it a while back and have finally managed to upload it ( It still looks rather crappy video to me )

The hotend was running as 12Watts and as should be seen both 3mm (blue) and 2mm(green) nylon extrude reasonably well, the extruder does not jam as there is no possibility for filament thermal expansion locking up the hotend, the dual hotend could run at least 200 hours at 15Watts.

So I can say that the previous hotend worked ( I am sure someone will tell me that it does not work but I don't really care what other people think here )

The basic reason why it was dumped was due to complexity and cost, the dual hotend consists of 22 seperate parts all of which have to be manufactured.

All parts are off the shelf from hardware stores but they all must be modified which takes a long time by hand and does need some tools which are expensive. I think the barrel is the hardest part as it must be done by lathe and I think if attempted by hand would be very very time consuming.
everything except the barrel was made with a thread tap, hand file, tin snips, locking pliers and an electric hand drill (you could use a manual hand drill if you really want as I know some people like to avoid these new fangled devices).

The video below should show to some degree how the dual hotend functions.






The new concept design should have 5 parts, only the barrel requiring machining and should be far lighter and more efficient.

Hopefully I will make an order some time this week and expect to get the parts next week so maybe the weekend after next I will start soldering up the new prototype.

Unfortunately I feel that sharing this new concept hotend design with the Reprap community will not happen as although this is just a hobby for me and I feel there is unlikely to be any money in selling printer extruders. I feel that really there is no interest from the community to develop a new hotend for future reprap printers as there are so many good designs already. I do not want to splinter the market and also I hope that I may possibly be able to sell the patent for my new design to a commercial 3D printer manufacturer.

I will continue with sharing the development of the extruder as a whole once I get back to that part of this hobby, and anyone is free to copy the hotends I have shown already.

Also I should say that I feel even if I were to patent and sell the rights for my new concept I would allow people to use the design provided they did not profit from sale or manufacture of the design.

There are only really three things that need changing on the extruder to make it complete :

1)the electrical connections for the hotends.
2)the filament drive tensioning springs.
3)the quick mount extruder changing system.

I know how everything should be made and function but actual manufacture of the parts is a different matter, and there are always unseen or unconsidered factors that can crop up once things are tested.

Edited 1 time(s). Last edit at 01/24/2012 03:22PM by mung.