PVC hopper

If the extruder can be used for PVC then a hopper can be installed for feed. Using the the feed empty signal can alert to a hopper refill. Small pellets flow easier than rods.

That's true. The problem with pellets is that they require a larger extruder. I tried using powder, but then I got what I suspect was electrostaticly driven clumping of the feed powder which caused lots of problems.

if the speed is 4mm/sec and these rods are 3mm dia and 100mm long?
4mm/sec0.5mm)**2/4XpiX3600sec=2827MM*3/hr, (3)*2/4XpiX100=706mm*3 so 2827/706=4 or every 15 minutes add a new rod. I am guessing about the 100mm length. The nice thing about the drill machine was I could walk away and let it drill the board all by it self. I don't think the steppers could move a lot of mass anyhow. The hoppers had an auto vacuum feed right from the box. Of course the extruder wasn't moving.

I can see having rods if we're talking about material that we use very little of, for example, printing solder on HDPE to make circuit boards. Beyond that, however, I tend to support the use of continuous filament in the Mk II and its derivatives. That's the reason why I've not gone over to the new Mk II design with the cable link between the pump and motor.

Dr. Higgs, of all the people who are working on this you are the only one pushing forward with new ideas. I was delighted with your catch phrase, "if it wasn't made to be disassembled, it wouldn't have screws". I have a three axis drill machine that I built and used 11 years ago. The X and Y axis have a plate to which a dremel drill press is mounted. The lever from the drill press is removed and replaced by a large timing sheave and a small piece of angle Iron serves as a mount for the third 12VDC stepper. It has a small timing sheave and a timing belt to lower the drill press. When the press is fully up, this is Z home, (mechanical switch).I used 1/4" 20 threaded rod, coupling and a welded nut as the driver. The sides of the plate ride on desk draw bearing slides The drill press has two circular clamps about 4" apart. I could just use duct tape to fatten up the extruder to the clamp. The assemble is built on a piece of plywood. What kind of surface would be best, (non stick, maybe a cutting board?) The range is x about 6", Y about 6" and Z about 3". I could probably make most of the parts within those dimensions. Does the flexible wire rope interfere with continuous filament feed?
I've read all of these problem with the thermistor. If you correctly measure and calculate the Rz and the beta and determine the Cap size, the software shouldn't be a problem. It is just measuring discharge time and finding a temp from this. If any interrupts disturb this timing process, you would think the cap would be re-charged and re- timed again. This Kit coming out, I'm not sure what type of material would work best in it, but I need an extruder assemble to get started!

englewood Wrote:
-------------------------------------------------------
> Dr. Higgs, of all the people who are working on
> this you are the only one pushing forward with new
> ideas.

LOL! That is SO wrong in SO many ways.

While it is true that there is a big push for getting Darwin out the door right now, this effort reflects more of a determination to get an operational RepRap that can work into the field for a lot of people to kick off rapid replication and open source 3D printing than it does a lack of creativity and new idea generation. Those people were geysers of new ideas before they undertook this grinding piece of work and I am certain that they will resume that role as soon as they get Darwin far enough along to be able to catch their collective breath. There are times when even the most creative people, which by and large the RepRap team are, have to grit their teeth and stop being creative long enough to get something actually working. That's what is happening right now. If this wasn't happening RepRap would be little more than a talking shop.

> What kind of surface would be best, (non stick, maybe a cutting board?)

I made my working surface of float glass and have been duct taping foamboard of the sort that you can get in any good art or office supplies store on top of that. HDPE, which I am using, won't stick to float glass. I recently bought an HDPE cutting board which I plan to see if can be used as a working surface. I haven't had time to try using it, though.

> Does the flexible wire rope interfere with continuous filament feed?

I don't use the flexible wire rope approach in the Mk I AEM (a modification of the original Mk II design). It rubs against the filament but does not cause feed problems. I understand that the whole point of the wire rope idea was to clear the feed so that rigid rods of materials like Wood's Metal could be fed into the Mk II.

Edited 2 time(s). Last edit at 07/19/2007 09:32AM by Forrest Higgs.

If they want positive feed, twin screws with one shaft driven by the motor and the other through a top idler gear. This would surely grab the material. With the price of BLDC motors and encoders dropping like a rock, you would think it would have been part of this design(I have a few brushed dc servo with encoders). This is the USA and we live in a foot, pound and Gc world here. I have a hell of a time trying to find metric at the local hardware! I am going to make connectors for the steppers on the drill machine tonight and see If I can move it, it has sat for ten years!

The wire rope is smaller diameter (3mm) than the M5 studding (it fits inside it) and it bends away from the filament so they should not be in contact.

---

[www.hydraraptor.blogspot.com]

OK, I'll bite.

> This is the USA and we live in a foot, pound and Gc world here.

Really? To quote Wikipedia on the SI system:

>> The system is nearly universally employed, and most countries do not even
>> maintain official definitions of any other units. A notable exception is the
>> United States of America, which still uses many old units in addition to SI.

Note the "in addition to" there. The USA signed the "Convention du Metre" more than a century ago, back in 1875!

Can you (here in the USA!) buy a ruler at any office supply store that does *not* have a millimeter scale on it? Can you find a packet of cereal or even a loaf of bread or bottle of shampoo at your local grocery store that lacks weight in grams on its packaging? Can you buy a gallon of Pepsi? No, but you can buy a two liter bottle of it! What dosage of acetaminophen does one Tylenol tablet contain (Tylenol is a US brand)? Is it measured in milligrams, or in grains or scruples, on the packaging? :-) Do your light bulbs specify their power usage in fractions of a horsepower, or do they use Watts? Even my Ford van uses "metric" (SI) fasteners, and it is over a decade old.

Face it, the world has decided to use SI units. The scientific world made that choice several decades ago (or earlier). The engineering world has, in some areas of the world, most notably the USA, been slow to follow them. Nevertheless, SI units are becoming more and more common here in the USA, it is just that people always resist change, even when the new thing/idea they are changing to has clear advantages over the old one they are familiar with.

Do you know, without looking it up, how many scruples there are in an ounce? How many grains in a scruple? Probably not. But you probably know how many milligrams there are in a gram :-)

Incidentally, I don't even know what a "Gc" is, and I live in the USA. I tried looking it up in the definitions file of the units conversion program, and failed to find it there either. What *is* a Gc (other than being 10^9 times the speed of light, which I don't think is what you intended, and that's not exactly a commonly used unit in engineering anyway, unless you work for a startup designing faster-than-light spaceships!!)?

Jonathan

Gc is the correction factor to allow a lbm to equal a lbf. it is 32.2 ftlbm/sec*2lbf. An example would be 1lbmX 32.2ft/sec*2/32.2 ftlbn/sec*2lbf=1lbf
Did you find a Rz and Beta for the thermistor? Also does anyone know the intial clearance the extruder nozzle should be at above the work surface?
I like the SI system, but go to lowes or the hardware store, you don't get far in threaded rods and fasteners in the SI system.

Once you've lived and worked a few places you'll discover that the places that to outsiders look like bastions of SI are anything but. In Sweden, for example, floor plans had gone through a number of permutations and most of them still used metres for large measurements and Swedish inches for details . Those would give you headaches.

Similarly, engineers in all of Europe used kw-hr as a measurement of energy. That was bizarre, to say the least. In Sri Lanka, residential stands were sold in perchs, a delightful English measure right out of the Tudor era. Land in South Africa was sold in acres, hectares and morgen depending on, well, I was never able to quite say. In China, you bought fish by the catty and gold by the tael.

In all of them, speed over water was measured in knots. Yet in all of these places, serious intellectuals would hammer the idea of a non-SI America.

After a while, you just get used to it.

Edited 1 time(s). Last edit at 07/20/2007 01:04AM by Forrest Higgs.

I am not a software guru, just a mechanical engineer who used to run a reactor for the Navy. Anybody know about that initial height clearance of the nozzle above the work surface? I am going to lay in a section of float glass for the work surface. I guess everyone else is a dedicated scientist? I do have my master's and state P.E. license! Everybody picks on the poor engineers!

englewood Wrote:
-------------------------------------------------------
> I am not a software guru, just a mechanical
> engineer who used to run a reactor for the Navy.
> Anybody know about that initial height clearance
> of the nozzle above the work surface? I am going
> to lay in a section of float glass for the work
> surface. I guess everyone else is a dedicated
> scientist? I do have my master's and state P.E.
> license! Everybody picks on the poor engineers!

Okay, okay... I duct tape a piece of standard foamboard onto my float glass working surface. The foamboard's thickness depends on who I bought it from. Some of it that I get is 5 mm thick while other pieces are 0.25 inch. The area measurements of both types are in inches.

With HDPE being extruded at about 140-145C and laid down at about 2.4 mm/sec I use a nominal layer thickness of 0.55 mm. For the first layer which needs to bind to the paper face of the foamboard I tend to use a somewhat thinner first layer, usually about 0.4 mm, especially if the paper surface is slick rather than matt. The HDPE tends to have more trouble sticking to slick paper surfaces.

I understand that CAPA sticks fairly well to glass. HDPE won't stick to it at all.

Edited 2 time(s). Last edit at 07/19/2007 08:49PM by Forrest Higgs.

Thin initial layer for adhesion to the foam board. Does the nozzle have to be very high above the workpiece. I will have the extruder on the Z axis. If I can keep the extruder at height above the workpiece, then I don't have to move the Z axis, just lay another layer down till it is done. 2.4mm/secX (.55)*2Xpi/4X3600 sec/hr= 2053mm*3/hr. or 2.053cc/hour

As a starting point, set your nozzle head to have about 1.2mm clearance over the table - you can adjust for the best height from there. In my experience, a good rule of thumb for the height is to have a clearance of about 1.5 times the height of your extrusion - that seems to work best for me.

englewood Wrote:
-------------------------------------------------------
> Thin initial layer for adhesion to the foam board.
> Does the nozzle have to be very high above the
> workpiece. I will have the extruder on the Z axis.
> If I can keep the extruder at height above the
> workpiece, then I don't have to move the Z axis,
> just lay another layer down till it is done.
> 2.4mm/secX (.55)*2Xpi/4X3600 sec/hr= 2053mm*3/hr.
> or 2.053cc/hour

I'm missing something here quite obviously. I reset xy and z axes after each layer to correct for cumulative encoding errors since I use shaft encoded DC motors rather than steppers.

> As a starting point, set your nozzle head to have about 1.2mm clearance over the > table - you can adjust for the best height from there. In my experience, a good > rule of thumb for the height is to have a clearance of about 1.5 times the
> height of your extrusion - that seems to work best for me.

I was into fine tuning z-axis levels some months ago but I finally automated all of that when I got positioning on Tommelise up to a dependable level of repeatibility.

When I start printing for a day I do a eyeball set for distance down to the working surface to account for swelling or warping of the foamboard. I tap the extruder head lightly on the foamboard surface as a start z axis point. If I want a firmer smear of HDPE onto the foamboard for the print raft I tap it a bit harder. This is an experience sort of thingy and realistically is a matter of maybe .2-.3 mm down from a nominal tap. When you extrude from less than about 0.55 mm the HDPE squirts out to the side(s) and covers more area than it would otherwise. You have to be careful because if you set the extruder orifice too close to the foamboard you can cause the polymer pump to backpressure and stall. You don't always get a clutch click to warn you of this. Sometimes the threaded drive rod on the pump simply abrades off the contact surface on the filament rather than jamming so you don't get a click.

You can set the orifice a bit higher higher than the top of the level you wish to extrude. This makes a more square or rounded extrusion thread with a narrower width. The problem with that is that you have to run the extrusion barrel hotter in order to get a good weld to the previous layer. The problem with that is that radiation from the extruder barrel and head heats the print surface a bit more than you really want if you are trying to do smaller features and these tend to melt and/or smear.

LOL! There's a lot to this stuff. At this point there is a lot more art than technology to getting a good print out of HDPE.

Edited 2 time(s). Last edit at 07/19/2007 11:06PM by Forrest Higgs.

As far as I know, you can't drop molten plastic onto the base from a great height and expect accuracy, so I'm confused by the idea of keeping the extruder high above the table. You'll need the Z axis height to change a minimum of once per layer, as far as I can tell, no matter what design of robot you are using.

> I guess everyone else is a dedicated scientist?

No, not at all. I'm just a programmer and network admin, I have no official connections with science/research/etc. at all, and my higest educational qualification is a Bachelor's degree. I don't think Zach, or Eric, or many others here, would think of or describe themselves as being "dedicated scientists" :-) I'd say that most Reprappers are interested hobbyists, rather than scientists, dedicated or otherwise!

I grew up in Germany, which is a pretty thoroughly "metric" place. I can't think of a single totally non-"metric" unit of measure I came across there. (Hours days and weeks are permitted in SI, so kilowatt-hours, though a little peculiar, are not quite as far away from SI as, say ft-lbs!). The UK has multiple major lapses, but even there things seem a bit further along than in the USA regarding use of SI units. Kilowhatt-hours, sure. Fluid ounces? Gallons? Nope, mL, or (for historical reasons) pints when dealing with milk or beer! In the Philippines as in Germany, you buy gas in liters, you travel kilometers in your vehicle, and you buy food and household stuff in kilos. In Russia, it's the same (when you can find the things you want available for sale... when I was over there, that was often an issue!). The nm and knot are also specifically permitted by the BIPM, and so are much less of a deviation from the overall SI approach as pounds and feet are.

Of course many places have a few local exceptions, but in all of the places I have lived in, across multiple continents, common things people use every day use liters/kilos/meters -- except in the USA! I really do think, based on personal living experience in several cultures, that the USA is subjectively and probably also objectively significantly "less metric" than the rest of the world, for no apparent reason other than inertia, and perhaps that its sheer scale makes interoperability with other nations seem less important. But then, I make no claim to be a serious intellectual, either :-)

[BTW, I've not worked with the thermistor yet]

Jonathan

To determine the max height I suppose we could use newton's law of cooling. This is an exponential (with time and convection,h) law to find the time to cool from one temp to another in still air. using 0.5gt*2=s we can find the distance to cool from Tnozzle to Tmelt. I thought maybe this had been done already. Measure the resistance at and room temp. I believed I used 383F for my final measurement, and it was verified by two digital sources. If biot number is < 0.1, you have a uniform temp that varies with depth into the wall. I can easily put my thermistor behind the acorn nut nozzle and wrap it up in fiberglass. I just don't want to work on my heating tube, because I believe the kit will have a better one. I need to get female molex pins and female bullet connectors to mate the drill machine to the reprap control boards. After that I can move this thing around! I think I will light oil the threaded rods and desk drawer bearing sliders, since it has been 10 years since the last moved. This is my repstrap!