What nozzle design would the community like to have machined?

Hello,

On many occasions, I have been contacted about making one nozzle design or another. This evening, I have had somebody inquire about another nozzle design. The biggest problem is that a one-off build is very expensive as I sometimes need to buy materials, tooling, etc. Then, all of the machinery needs to be setup in order to make each part.
There can easily be 4 to 8 hours worth of time involved in making several nozzle parts for an experimental hot end. Once I quote a price, figuring in a low-end shop rate of $60.00 per hour, it gets very expensive.

So, I was thinking that maybe the best way to determine if there is any focused interest in a new hot-end design that I would ask the community. If there is enough interest in one particular design, I'll run off a batch of machined parts for that design and make them available at an affordable price. Please bear in mind that I don't have a 3d printer, so I can't do any testing. But, I can make the parts as per a blueprint.

Some possible ideas...

[reprap.org]
[reprap.org]
[www.thingiverse.com]

Of course, if there is some other idea, please feel free to add a link to it.

Regards,

Brian

From what I have read things that are important in a nozzle are:

1. Short melt zone (oozing control)...
2. Short, slippery transition zone (decrease possibility of jamming)...
3. Mechanically robust...

The "Best Compromise Nozzle" (your first link) has a fairly long melt zone. This is necessary because of the large nozzle and, the need to wrap nichrome and the need to mechanically thread the nozzle into the insulator.

The "Geared extruder nozzle" (your second link) still has a longer melt zone than necessary (again because of the need to wrap nichrome).

The "All steel hot end for Makerbot" (your third link) has a long and not so slippery transition zone and relies on radiating heat from the washers. I would expect its performance to vary a bit with airflow and ambient temperature. and for it to be prone to jamming, particularly with PLA.

Of the three, I like the "Geared extruder nozzle" the best, but I think it could be improved in two ways that spring to mind. One you pointed out in your for-sale thread. The price of PEEK makes the PEEK block that provides the mechanical stability is a costly part of the design. As you suggest, replacing this with a PEEK washer and steel bracket would result in a lower cost design.
The second improvement would be to re-design the nozzle to include a resistor heater block. This would remove another source of un-reliability in hot ends, the nichrome wire that can become brittle and break or break through its insulation and cause a short circuit.

Anyone have a resistor heater block design that can be mechanically held in place with a peek washer?

Edited 2 time(s). Last edit at 08/22/2010 01:49AM by Greg Frost.

Hello,

I just added a picture of a design that was tried successfully. With the exception of the really long PEEK thermal barrier, it is very similar to the 2nd link (Geared Extruder Nozzle) that I posted. With the exception of the dimensions, the nozzle is almost identical. (I have a picture of a standard Mk4 style hot end of the left, to show the size of it.) It has a heater block that is only 0.3 wide and is screwed onto the nozzle. From the point where the filament reaches the brass heat zone, to the tip of the nozzle, is only 0.4 inches.

The "Geared Extruder Nozzle" design also shows M7 threads to attach the nozzle to the thermal barrier. We found out that that the threads are probably not necessary as the plate (not shown in the picture) and rods hold everything together. More importantly, if you try to take it apart, after use, the PEEK threads jammed and broke in the nozzle due to the solidified filament. In addition, it would be cheaper to make if it is not threaded.

The "all steel hot end" should, in theory, have a short melt zone as the turned-down section, of the heater barrel, is supposed to act as a thermal barrier. My idea is to make heat sinks similar to the picture on the lower left of that page. (I am actually making some for testing.) With a resistor heater block, the actual melt zone may be pretty short. Since the new MK5 extruder has an all metal outside, plastic may be completely unnecessary. It would be possible to sleeve part of the heat sink end with PTFE.

Regards,

Brian

The one on the right in your attached picture is getting close to ideal (if it had a much shorter thermal barrier - I presume that the thermal barrier is that long to make the whole assembly dimensionally similar to the one it is shown alongside). It still has the issue of putting the thermal barrier under tension and because the nozzle is threaded into the thermal barrier it is probably slightly longer than necessary. Something like that, but attached under compression with a thermally insulated metal bracket would be ideal.

Hello,

As you suspect, the thermal barrier was made that long in order to be about the same length as the standard hot end. Removing the threading between the nozzle and the thermal barrier is a good idea. I didn't think about the stress factor you just mentioned; but, you have a really good point about it. Also, as you mention, without the threads between the nozzle and thermal barrier, the nozzle could be shorter yet. (Removing the threading also makes it cheaper to manufacture.)

The thermally insulated metal bracket wasn't shown in the picture...but it was used.

Regarding the total length: Should it be about 55mm? The total length of the normal Mendel PTFE thermal barrier and combination nozzle?

Regards,

Brian



Greg Frost Wrote:
-------------------------------------------------------
> The one on the right in your attached picture is
> getting close to ideal (if it had a much shorter
> thermal barrier - I presume that the thermal
> barrier is that long to make the whole assembly
> dimensionally similar to the one it is shown
> alongside). It still has the issue of putting the
> thermal barrier under tension and because the
> nozzle is threaded into the thermal barrier it is
> probably slightly longer than necessary. Something
> like that, but attached under compression with a
> thermally insulated metal bracket would be ideal.

What is the smallest size hole you can do?, I am playing with printheads and although we do not have a final design yet, it will be a small hole.

johnrpm Wrote:
-------------------------------------------------------
> What is the smallest size hole you can do?, I am
> playing with printheads and although we do not
> have a final design yet, it will be a small hole.


Hello,

I have done 0.25mm. But, I don't think going smaller would be a problem. How small would you like to go?

Regards,

Brian

Brian
Is .050mm asking to much, the body can be about m6, 6mm long, but the final dimensions will depend on community effort.

---

Random Precision

That is a very small orifice size! It will take forever to print a large object. Remember that in general, the time to print will go up with the cube of the reduction in orifice size. A 0.050mm orifice will take nearly 1000 times longer to print the same object, although admittedly with very high resolution and very smooth surfaces. Unless you only plan on writing a resist layer onto a pc board or other etching process. But even then 0.050mm line width is much smaller that any pcb fab house that I have heard of. Just curious what the need for the very small orifice size is for.

Mike

Hello,

I think 0.050mm is possible. I just checked and drill bits are made for that size. Drilling a hole that small would be quite expensive, however. The drill bits are $11.71 each, in quantity, and probably wouldn't last very long.

There would, of course be some other problems. The pressure required to extrude would be pretty high and the 3d printer would have to be precisely aligned. Given the composition of a RepRap or MakerBot, I have my doubts that the machine can hold that degree of precision. Most likely, you would have to attach the extruder to the head of a CNC milling machine in order to use it. (I could be wrong as I don't have a 3d printer; however, from everything I have read I think this is the case. If I am wrong, somebody will be along to straighten me out. :-) )

Regards,

Brian

We are trying to develop a printhead here
not molten polymers but resins, waxes etc, still needs a lot of development, but if we succeed most people will need nozzles.

---

Random Precision

Hello,

Ok, now I understand what you are doing. Do you have a blueprint of the nozzle you would like to be made?

Regards,

Brian


johnrpm Wrote:
-------------------------------------------------------
> We are trying to develop a printhead here
> not molten polymers but resins, waxes etc, still
> needs a lot of development, but if we succeed most
> people will need nozzles.

The nozzle at the bottom has few critical dimensions, instead of m6x6, m5x5 would be ok, the only thing that may need
to be accurate apart from the nozzle hole may be the land diameter on the front face, this apparently effects wetting but
I can not give a size as yet, so flat fronted for now.

Brass would be good for hot wax but polymer should be ok for inks I think, when I play with waxes in may need bigger hole sizes depending on viscosity.

Picture...... [forums.reprap.org]

---

Random Precision

Could the nozzle be made out of 8mm hex stock? Also, how many nozzles would you need, initially?



johnrpm Wrote:
-------------------------------------------------------
> The nozzle at the bottom has few critical
> dimensions, instead of m6x6, m5x5 would be ok, the
> only thing that may need
> to be accurate apart from the nozzle hole may be
> the land diameter on the front face, this
> apparently effects wetting but
> I can not give a size as yet, so flat fronted for
> now.
>
> Brass would be good for hot wax but polymer should
> be ok for inks I think, when I play with waxes in
> may need bigger hole sizes depending on
> viscosity.
>
> Picture......
> [forums.reprap.org]
> lename=test_heads.jpg

MakerBot just came out with a totally new design which they claim can print for days or weeks without problems.

I saw it and it looks great! I have a MK5 extruder ordered and I already have the raw materials and tooling ordered in order to make nozzles for it. I will first make brass nozzles ranging in size from 0.25mm to 0.70mm. Once I have my turret lathe setup for coolant I can start making stainless steel nozzles, as well.

Regards,

Brian


freds Wrote:
-------------------------------------------------------
> MakerBot just came out with a totally new design
> which they claim can print for days or weeks
> without problems.

8mm hex would be ok, as its early days in development its hard to say how many would be needed, I realize you need to know if its worth tooling up for, Can I turn it round as ask what is the minimum quantity you would be prepared to do, and a ball park price?.

---

Random Precision

Hello,

With the exception of the 0.050 drill bits, I have all of the tooling. However, due to the cost of the drill bits and the precision required, I am going to quote $15.00 each for a minimum quantity of 10.

I do have a concern regarding the blueprint: Since the nozzle threads are M6x1 and the ID is 3.0mm, there will be only 0.020" between the thread root and the inner diameter. Would it be possible to reduce the ID to 2.0 or 2.5mm?

Regards,

Brian


johnrpm Wrote:
-------------------------------------------------------
> 8mm hex would be ok, as its early days in
> development its hard to say how many would be
> needed, I realize you need to know if its worth
> tooling up for, Can I turn it round as ask what is
> the minimum quantity you would be prepared to do,
> and a ball park price?.

You asked the question: What nozzle design would the community like to have machined?

I just found a very decent design that I'd like to see added to the community. If you wish to offer the community a choice then build these and let the best design out there win.

[reprap.org]

As I understand it you manufacture the hybrid insulators? So take this design and add your hybrid to it and make it your own design. It might be worth a go.

Hello,

Are you suggesting that I make hot ends of every design or that I build only the one you have linked to?

Adding a hybrid thermal barrier, to most designs where there is a plastic thermal barrier, is always a possibility.

Regards,

Brian


Prober Wrote:
-------------------------------------------------------
> You asked the question: What nozzle design would
> the community like to have machined?
>
> I just found a very decent design that I'd like to
> see added to the community. If you wish to offer
> the community a choice then build these and let
> the best design out there win.
>
> [reprap.org]
>
> As I understand it you manufacture the hybrid
> insulators? So take this design and add your
> hybrid to it and make it your own design. It
> might be worth a go.

Quote
Are you suggesting that I make hot ends of every design or that I build only the one you have linked to?

The design i linked to looks to be a well thought out design, and worth getting out in the community.

Prober: Why do you think that design is easier to make? There is a lot of thread cutting/tapping going on with that design and more parts to machine. I prefer Adrians design, but the PEEK block is expensive. Replacing that with a PEEK washer and a steel bracket would be cheaper (wont provide quite as much thermal resistance though). If Brian is going to be moving to a new design, adding further simplifications like a resistor heater block should also be considered.

Edited 1 time(s). Last edit at 08/25/2010 05:07PM by Greg Frost.

Hello,

Something to consider is that the fewer the parts, the cheaper it is to make and purchase.

Personally, I like the version below, with the modifications I have listed above. I have had somebody successfully test something very similar to this.

[reprap.org]

However, this version below is all metal and, assuming that the heat is bled off by the time the metal touches the plastic, it would probably be very reliable over the long term. (I am looking in particular at the lower left-hand picture with the machined aluminum heat sink.)

[www.thingiverse.com]

Regards,

Brian

>>>>However, this version below is all metal and, assuming that the heat is bled off by the time the metal touches the plastic, it would probably be very reliable over the long term. (I am looking in particular at the lower left-hand picture with the machined aluminum heat sink.)

[www.thingiverse.com] <<<<<

Ok for a repstrap, not ok for a production mendel etc. Weight is an issue.

It's a lot lighter than what I have on my Mendel at the moment and not much compared to the weight of the extruder motor.

---

[www.hydraraptor.blogspot.com]

I have several thoughts:

Quote

With the exception of the 0.050 drill bits, I have all of the tooling. However, due to the cost of the drill bits and the precision required, I am going to quote $15.00 each for a minimum quantity of 10.

reifsnyderb:
-I would suggest that you make the .5mm holes an optional service rather than bundling them in to the part.
My gut reaction is that your costs associated with those holes are going to exceed the added value to the customer. This is going to make them a significant source of lost profit.

My reason for saying this is I've never been good with holes smaller than 1/16. They just make my scrap rates go through the roof. Then again: my reaction to the problem is to max out the spindle RPM and baby the feedrate, so if there's a better way: maybe I'm off base.

Do high spindle RPMs wear out the bearings faster?

How much do you think you can cut the unit price by block skipping the drilling cycle? By increasing lead time, and doing group orders: you may be able to get more parts out the door per set up time invested. Depending on the price adjustment: this may go in to a positive feedback loop of order volumes which makes everyone happy.

In regards to this design:
[reprap.org]

I'm not sure if I understand the reasoning behind the many segments. The drawing isn't very clear about materials. Is this supposed to minimize material costs on certain components? My first reaction is that the design may be based on some misconceptions about areas of manufacturing difficulty.

I see a lot of parting operations on that nozzle. I have trouble getting parting tools to reliably face a surface both square and to dimension. Frequently, to get an accurate length: I end up having to chuck the part up, face, measure, touch off, and face the part again. CNCs are better at this, but all the same: I think parting tools are generally a pain.

It's not to say that it's especially difficult, but that I hope there is design reasoning behind it other than making it easier to manufacture. Longer holes of multiple depths and diameters are easier than pancake construction IMHO.

Finally:
If the weight savings is an important design feature for repraps, and not for repstraps: it may make sense to mass produce 2 versions which are appropriate for their tasks rather than 1 version which is a compromise.

Edited 1 time(s). Last edit at 08/26/2010 02:50PM by JohnnyCooper.

nophead Wrote:
-------------------------------------------------------
> It's a lot lighter than what I have on my Mendel
> at the moment and not much compared to the weight
> of the extruder motor.

Your a special case Nophead

far, far too many mdf made repraps that would not hold up IMHO

Hello,

JohnnyCooper Wrote:
-------------------------------------------------------
> I have several thoughts:
>
> With the exception of the 0.050 drill bits, I have
> all of the tooling. However, due to the cost of
> the drill bits and the precision required, I am
> going to quote $15.00 each for a minimum quantity
> of 10.
>
> reifsnyderb:
> -I would suggest that you make the .5mm holes an

This is for the print head with 0.05mm holes.

> optional service rather than bundling them in to
> the part.
> My gut reaction is that your costs associated with
> those holes are going to exceed the added value to
> the customer. This is going to make them a
> significant source of lost profit.

I see your point. However, drilling 0.05mm holes is probably beyond the capabilities of most small home shops and some machinists. I consider it a challenge. :-) In this case, and with my tooling, it makes more sense for me to drill these holes on the milling machine. My smallest milling machine is a Sherline Model 2000 and I have a drilling attachment that is made just for this purpose.

> My reason for saying this is I've never been good
> with holes smaller than 1/16. They just make my
> scrap rates go through the roof. Then again: my
> reaction to the problem is to max out the spindle
> RPM and baby the feedrate, so if there's a better
> way: maybe I'm off base.

It sounds like you are on the right track. You need to think more along the lines of "peck drilling." Since a "deep hole" is any hole that is more than about 3 to 4 times the diameter of the drill bit, you need to withdraw the drill bit frequently in order to clear the flutes. Believe it or not, I have found that the best way, on a lathe, is to use a hand-held pin vice in order to drill the tiny holes. Then, you can quickly and easily peck drill, clean the flutes, and can feel how the drill bit is performing. I didn't believe it either when I was told to use a pin vice. But, I tried it and it works great.

> Do high spindle RPMs wear out the bearings
> faster?

It depends on how high the RPMs are, the type and quality of bearings, etc. With plain and babbit bearings you normally don't want to exceed around 800 RPMs or you will destroy them. One way to tell if you are running the lathe too fast is if the headstock is too hot. It should feel just warm to the touch. My Hardinge DSM-59, on the other hand, has a top speed of 3500 RPMs. But, I don't run it that fast as the bearings are old and the headstock starts to warm up around 2000 RPMs. The Sherline 4400, on the other hand, can probably run all day at 2800 RPMs. The Sherline also can be "upgraded" to run up to 10,000 RPMs; but, they tell you to adjust the bearings in order to do so. So, I guess in summary the answer is yes but there are many variables as to what "high RPMs" really are.

> How much do you think you can cut the unit price
> by block skipping the drilling cycle? By
> increasing lead time, and doing group orders: you
> may be able to get more parts out the door per set
> up time invested. Depending on the price
> adjustment: this may go in to a positive feedback
> loop of order volumes which makes everyone happy.

What I normally do is make up a batch of parts at a time. Depending upon demand, I'll make anywhere from 10 to 50 identical parts and put them in a bin. Then, when I need them, I just put the order together. For nozzles, I do everything but drill the orifice and stamp the size on them. That way I don't need to keep a large quantity of every size available. Doing it this way does increase the risk that I'll be stuck with unsaleable parts; but, I consider the risk to be rather small.

> In regards to this design:
> [reprap.org]
>
> I'm not sure if I understand the reasoning behind
> the many segments. The drawing isn't very clear
> about materials. Is this supposed to minimize
> material costs on certain components? My first
> reaction is that the design may be based on some
> misconceptions about areas of manufacturing
> difficulty.

Yeah, I think it has too many parts as well. Many times simple is much better for many reasons.

> I see a lot of parting operations on that nozzle.
> I have trouble getting parting tools to reliably
> face a surface both square and to dimension.
> Frequently, to get an accurate length: I end up
> having to chuck the part up, face, measure, touch
> off, and face the part again. CNCs are better at
> this, but all the same: I think parting tools are
> generally a pain.

I try to avoid parting. The only machine where parting really works nice is the Hardinge DSM-59. Even then, the end is either convex or concave, depending upon which end you are looking at. For most parts, I need to follow up with a facing operation. If I can't part it on the Hardinge, I usually use a band saw.

> It's not to say that it's especially difficult,
> but that I hope there is design reasoning behind
> it other than making it easier to manufacture.
> Longer holes of multiple depths and diameters are
> easier than pancake construction IMHO.
>
> Finally:
> If the weight savings is an important design
> feature for repraps, and not for repstraps: it may
> make sense to mass produce 2 versions which are
> appropriate for their tasks rather than 1 version
> which is a compromise.

I don't think weight is a major issue as the aluminum is very light.

Regards,

Brian

reifsnyderb Wrote:
> Personally, I like the version below, with the
> modifications I have listed above. I have had
> somebody successfully test something very similar
> to this.
>
> [reprap.org]

This one gets my vote.

---

--
-Nudel
Blog with RepRap Comic

I got one of your hybrid extruder kits in the mail today from ebay reifsnyerb. Can't wait to get this first Mendel together and see how it works.