DIY printhead

I did extract the foil from a HP11 cartridge, but under the microscope has a backbone running down the middle, I assume the heater part, so made it less attractive for gluing onto a body, other cartridges may be better?, I did scrounge some glass micro pipette's, about 6mm dia tapering down to ???, less than 1mm, haven't measured it, I hope to try them as spark unit bodies, but the long taper is an issue.

Diamond I read that as being a diamond tip, but actually says, diamond tip "geometry"

I wonder if sticking a small pin into alluminium baking foil and cutting it out with a wad punch could work?.

Edited 1 time(s). Last edit at 09/06/2010 05:20AM by johnrpm.

---

Random Precision

... johnrpm pointed me to laser-ejected droplets - black ink in a thin glass-nozzle, heated with a laser-pulse, will explode in a small vapour-bubble and shoot a droplet away through the open end of the nozzle.

The refilling of the bubbles will be soemthing undefined ... but when feeding a consant flow of ink through the nozzle an let the laser pulse in the correct rate (so it will shoot away the ink befor it can come out), you'll receive a constant sequence of droplets shooting out of the nozzle.

Instead of modulating the flow and the laser, you can modulate the free flying droplets with electrostatic - let the droplets fly between two charged plates, so the path will bend to one side.

With modulating the charge you can control the path of the droplets - either analog for drawing lines, or by directing the droplets on a target, where tey'll be caught (and you can suck the recollected ink) ... and only if you release the path, so some droplets will miss the target, they'll fly free to hit the painting area ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Viktor,
Would DVD lasers have enough power or would we need a bigger laser?,
Looks like I am going to get side tracked again.

---

Random Precision

... i think the 220mW-types should work, but you have to use dark ink - black or green (what's absorbing best)

BluRay-diodes should work too - i have a 200mW here but haven't used it.

I have a 'RealBlue' with 1Watt@445nm, but haven't used it too

For perfect coupling the energy into the tube you should file and polish a flat in the side of the tube so you'll have a plane and really thin window ... trying to insert a flat with melting the glass could deform the capillary ...

*** EDIT: - but maybe flattening the red-hot tube with pliers could be the perfect way ... you'll get two 'windows' and the capillary will be flattened too, so you have a bigger surface to absorb the energy ...

Edited 1 time(s). Last edit at 09/07/2010 04:23AM by VDX.

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Would any glass do, or would soda glass be better, I shall also try a mistmaker, see wiki page, it could melt plastics and silicone rubber, so may have enough energy.

---

Random Precision

... for the common laserdiodes from DVD-burner's and other NIR-lasers (808 - 975nm) any transparent material will work.
For BlyRay-diodes you have to look at the transmission-specs ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

VDX Wrote:
-------------------------------------------------------
> ... for the common laserdiodes from DVD-burner's
> and other NIR-lasers (808 - 975nm) any transparent
> material will work.
> For BlyRay-diodes you have to look at the
> transmission-specs ...

I am bit a lost now. Are we still pursuing the inkjet head using piezo tecnhology?

---

---
New cutting edge RepRap electronics, ARM 32 bits @ 100MHz runs RepRap @ 725mm/s:

[www.3dprinting-r2c2.com]

... we're still at the inkjet head, but there are many methods firing droplets out of a nozzle ... some, like piezos, are a bit more complicated, other, like heated micro-bubles (with a resistor or even a laser), are something easier to realize

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

VDX Wrote:
-------------------------------------------------------
> ... we're still at the inkjet head, but there are
> many methods firing droplets out of a nozzle ...
> some, like piezos, are a bit more complicated,
> other, like heated micro-bubles (with a resistor
> or even a laser), are something easier to realize
>

Because after seeing the video registered by johnrpm printing with piezo head, I thought it could be possible to replicate and cheap.

I think we should continue with piezo, since there is already a kind of working prototype. I will wait to see what happens.

---

---
New cutting edge RepRap electronics, ARM 32 bits @ 100MHz runs RepRap @ 725mm/s:

[www.3dprinting-r2c2.com]

... he has problems creating single droplets or modulating a continuous stream, so he's asking for alternatives ... another relative simple DIY-methode is a needleprinter-solenoid 'hammering' on a small box filled with ink ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

So far I have not tried too hard to create single drops, but had the idea to use a laser some time ago, [forums.reprap.org] and want to encompass as many ideas as possible, the mistmaker is a piezo unit and small lasers are reasonably cheap.

---

Random Precision

I wanted to note that I spent a good portion of Monday (a holiday around here) constructing and testing a duplicate of Johnrpm's acrylic print head design. I simplified a couple thing in order to make them easier for me to make - specifically, I did not set up a threaded nozzle but simply glued the nozzle to the body. For testing purposes, a threaded nozzle is a much better design, but I do not have the skills to make the threaded nozzle in a reasonable period of time.

The head appears to work in drop on demand mode using water as the liquid. I say appears because I was running a 5 ms pulse once every 500 ms, so it was not quite a true drop on demand test. There are a couple of obvious differences between the devices that Johnrpm and I constructed:

1) The driver for this device uses a "high" voltage power supply and a mosfet to drive the piezo directly, so the pulse profile is likely to be significantly different from the profile generated by a driver design that uses a transformer to drive the piezo. High voltage here means 50 to 80 volts.

2) The 20mm piezo disk used has a resonant frequency of 3 kHz, not 6 kHz. I believe the difference is that the metal disk is thinner in the 3 kHz piezos than in the 6 kHz piezos of the same diameter.

3) The nozzle aperture was constructed differently. A small square from a thin sheet (0.016 inches) of aluminum was cut out and then a hole was drilled in the square with a #80 drill bit. The hole was then manually tapered from one side using a 60 degree etching bit that is normally used for milling pcb boards. The square was then glued to the body with the taper expanding toward the inside of the body.

The differences may help to explain the differences in behavior between the two heads, or they may be red herrings. More testing will tell.

The drops are far too large to be particularly useful for anything I want to do, so a smaller nozzle aperture is definitely needed. It was interesting to note that the presence of a small air bubble in the chamber (although not directly in line with the nozzle opening) did not appear to significantly affect the production of drops.

I will write up more details and provide some pictures later this week.

Great news, looking forward to seeing the pictures, it seems to take me 2 hours to document every 1 hours work,
can you give some detail about the power supply also please.

edit, is that pure water, if so well done, I could not get water to work

Edited 1 time(s). Last edit at 09/07/2010 05:55PM by johnrpm.

---

Random Precision

madscifi Wrote:
-------------------------------------------------------
> 1) The driver for this device uses a "high"
> voltage power supply and a mosfet to drive the
> piezo directly, so the pulse profile is likely to
> be significantly different from the profile
> generated by a driver design that uses a
> transformer to drive the piezo. High voltage here
> means 50 to 80 volts.

Well, this means is a simple switch ON/OFF of that 50/80 volts? -- if so, that means a really simple and cheap hardware driver, also simple firmware :-)


> The drops are far too large to be particularly
> useful for anything I want to do, so a smaller
> nozzle aperture is definitely needed.

Ok, but can you guess the size of hole you did? and the size of drops?

What size of drops would be useful for you? -- for me, 0.25mm, to be able to print tracks of 0.25mm for LQFP pads of 0.5mm pitch.


It was
> interesting to note that the presence of a small
> air bubble in the chamber (although not directly
> in line with the nozzle opening) did not appear to
> significantly affect the production of drops.
>
> I will write up more details and provide some
> pictures later this week.

So, do you think we have project? do you think we will success with this kind of piezos and inkjet print head?

Yes, please share and document. I think some parts of the wiki can be deleted if you have new information. Maybe we could put a section for historic and put there a link every time we change important things of the wiki. Well, I mean, we should not be afraid to loose all the work.

---

---
New cutting edge RepRap electronics, ARM 32 bits @ 100MHz runs RepRap @ 725mm/s:

[www.3dprinting-r2c2.com]

The water was plain, straight out of the RO machine. It is possible that a minisule amount of isopropyl alcohol was mixed in since I did some tests with the previous print head (using the same reservoir and tubing) with alcohol instead of water, but I'm pretty certain it would have been completely flushed out by the time I started with the new head. However, I will flush everything out and test again just to be certain.

I used this design to covert the output of a variable dc power supply to a high AC voltage, and a standard diode bridge and filter capacitor to convert back to high voltage DC.




Some notes on the parts:
C1 - should be 200 volt or better. The specific value is not too important.
R4 - should be at least 1/2 watt for up to 100 volts.
T1 - supplies 1/2 amp at 12 volts. Prefer a lower rating to a higher rating for now. A 24 VCT, 230 V primary would be equivalent.

If you have a lower current transformer, and you keep the voltage output under 100 volts, you can probably use any standard TO92 style switching transistors (2n2222, for example).

You'll want a variable low voltage supply feeding it - 3v input generates about 45 volts out. It is easy to accidentally exceed the voltage rating of the output capacitor if one is not careful as this circuit can easily supply over 200 volts.

Johnrpm, I know you know this, but since I have no idea who will read this in the future I think the following warning is appropriate: Don't let this circuit fool you, the output is quite capable of killing you. If you don't know what you are doing you should not attempt to build it.

Here is the driver circuit that I'm using:




You can take the 5 volts from the Ardurino. VCC is from the high voltage supply. The mosfet must be designed for a logic level input, other mosfets will not work. The IRL540 is only rated for up to 100 volts.

Q3 and Q4 (sorry, I did up the schematics together, so the component number is all nonsense) can be any standard complementary small signal switching transistors, 2n3094/2n3096, etc. I'm using some old generic devices from radio shack which is why I have not identified the parts. I don't know that these are actually necessary, it might work just as well without Q3 and Q4, but I'm simply documented what I did here.

R6 is not a misprint, 10K is a good value, 1K would be too small (it would load down the power supply unnecessarily).

I just went through and checked everything and discovered that I'm actually using 220 ohms for R5.

Casainho, I'm not ignoring your questions, but I simply don't have time to respond to them this evening. I've spent all the time I have on the writeup above. Tomorrow.

Edited 1 time(s). Last edit at 09/08/2010 02:14AM by madscifi.

madscifi Wrote:
-------------------------------------------------------
> I used this design to covert the output of a
> variable dc power supply to a high AC voltage, and
> a standard diode bridge and filter capacitor to
> convert back to high voltage DC.

Nice! This are cheap and simple electronics :-) It's very good to couple to any other board, because we can just use any microcontroller/board which runs from 5V or 3.3V. This electronics are great to couple with RepRap Mendel :-)

The only question here is the dangerous voltage, but maybe this can be minimized in future.

Good work madscifi :-)

---

---
New cutting edge RepRap electronics, ARM 32 bits @ 100MHz runs RepRap @ 725mm/s:

[www.3dprinting-r2c2.com]

Thank you Madscifi, for showing the circuit, I think you are right about not needing q3 and q4 if a signal level mosfet is used, but belt and braces never hurts.

A thought about the high voltage problem, by having 2 piezo disc's back to back, (mentioned earlier) and mounted on a thick walled tube with the feed and nozzle on the periphery, in theory it should double the displacment for a given voltage, or have the same displacement for half the voltage, (in theory)
also the piezo transformrs are very interesting, not sure how they work but will try to glue 2 together and see what happens.

---

Random Precision

q3 and q4 are to increase the gate charge and discharge current to make it switch faster. MOSFETs have a high gate capacitance, hundreds of pF, so to switch it quickly you need more current than you can get from a logic output.

---

[www.hydraraptor.blogspot.com]

Nophead is, as usual, correct. I added the q3 and q4 to the circuit while testing my first print head attempt, hoping that a faster rise time on the pulse might help things (it seems I have an unholy obsession with faster rise times for some reason). While the sub-circuit definitely decreased the rise time of the pulses, it did not have any apparent effect on the operation of the first print head.

I did get a chance to do a very quick test last night and the new print head works without the q3/q4 sub-circuit, so I now know that really fast rise times are not needed with the bimorph print head design. However, I doubt that will stop my quest for faster rise times ;-)

casainho Wrote:
-------------------------------------------------------
...
>
> Ok, but can you guess the size of hole you did?
> and the size of drops?
>

A #80 drill bit is about 343 um in diameter, so the hole will be somewhat larger than that. I did a quick test and captured some drops on plastic - the smallest are around 1/2 mm in diameter, the larger around a mm. It is possible that the larger captured multiple drops, or that the drop generator is creating a small number of drops for each pulse.

> What size of drops would be useful for you? -- for
> me, 0.25mm, to be able to print tracks of 0.25mm
> for LQFP pads of 0.5mm pitch.
>

At this point I don't know what size of drops would work for what I'm interested in. I do know that the current 1/2 mm drops are too large for one use of interest, but I have not done any research to determine what a usable drop size would be. It is not something I think I need to worry about for a while...

>
> So, do you think we have project? do you think we
> will success with this kind of piezos and inkjet
> print head?
>

Do we have a project? Yes, several, in fact. I welcome your enthusiasm but I think we are a long ways from having a useful and safe design. I am, of course, speaking only for myself here. The executive summary of the remaining text in this post is that it suggests that I think it will take several years to develop a really safe and generally useful inkjet style print head, so unless the reader is really masochistic, the reader should probably just skip the remainder of this post and move on to the next.

Will the current design turn into a useful Reprap component? Probably not. The current design requires the user to handle the print head in order to remove any air in the chamber, either by turning the nozzle over or by opening and closing a bleed valve. This is going to be very messy and will inevitably get whatever liquid is being used all over the place. In addition, the liquid is currently in contact with a not-insubstantial voltage. Try getting that past any sort of sanity check. It may be possible to insulate the piezo on both sides without seriously impacting the performance, but that is conjecture at this point.

There is another problem that looms ahead once someone attempt to actually put the print head on a Reprap. The print head depends on the liquid level in an external reservoir, and depends on a flexible connection between the reservoir and the head. Any movement causes pressure changes in the print head chamber that affect the operation of the head. Tap the feed line lightly with your finger and liquid goes flying out the nozzle. That will probably not a problem on my Repstrap as the tool head only moves along the Z axis, but it will likely be a serious problem on a Mendel or a Cupcake. There might be solutions, but again it is all conjecture at this point and nobody has even begun to look into the issue.

On the other hand Johnrpm's design has demonstrated that it is possible to construct an inkjet style single nozzle print head. I think that makes it a worthwhile project. At the moment we have exactly one print head in existence that drops on demand using water and we don't even know why it works and the others of it kind do not. I happen to suspect that it is the nozzle construction that makes the difference, but I have not had time to determine if that is the factor, or one of several factors, or has any effect at all.

I think a tubular design will improve the situation as it naturally insulates the liquid from the piezo and might, at most, require the operator to tap the tube to eliminate any trapped air. Both Johnrpm and I are looking at tubular designs, but here we start to diverge a bit. He is investigating an unique design using small piezos designed to generate sparks in lighters. I'm looking at using more traditional design using ring piezos, a design that is known to work but that uses harder-to-source, at least in small quantities, piezos. If his design ends up working I'll probably switch to that as the spark piezos are much easier to obtain than the ring piezos, at least in small quantities that don't cost an unreasonable amount of money.

On the other hand the spark piezos appear to require much higher voltages, so there are tradeoffs. The problem of movements affecting the operation of the print head are likely to be smaller in a tubular design as the reservoir can be part of the tube or connected rigidly to it. I don't know if the effects will be small enough to make operation on a Mendel viable. In any case, I think the tubular design is going to be limited in use to a very small number of people, mainly because I think it is going to be useful for making micro objects, not macro objects, as a single nozzle with anything approaching a small drop size is simply going to print at an extremely slow rate (Johnrpm may have a different view on this due to his work with continuous drop generation). I view it more as a lab tool rather than as a generally useful tool. It might, however, be a useful tool that can be used to help build a different multiple-nozzle print head.

Johnrpm's laser based concept is, I think, the most likely way forward for a inkjet style print head that will have a lot of value for a large number of Reprap users. A multiple nozzle print head that can be used for power printing and that can be built by a reasonably dedicated person would be very valuable indeed. However, the basic design seems to be completely under patent here in the US, so I'm not going to be able to do much more than root those outside of the US on.

Is this a project you should spend your time on? That depends on what you want to accomplish.

If you want to do powder printing with an inkjet head in the next couple of months, look into reusing an HP cartridge or the print head from an inexpensive printer. Reverse engineering the drive signals to the print head will be much easier and quicker than figuring out how to build a multiple nozzle device from scratch. In addition, it will be cheaper and it will work better.

If you want to produce PCB boards, look into milling, or using a laser to pattern a spray painted etching mask, or simply do photo etching from inkjet printed transparencies.

Essentially, if your goal is to use the print head to make something else in the next couple of years, then you will most likely be disappointed. I think it will be some time before one will be able to produce much of interest beyond simply a technology demo.

If your goal is to contribute to exploring and developing a variety of different print head designs, and you think it an interesting enough project in and of itself to keep you interested for the next year or two, then yes, it is likely to be a very interesting project to work on.

Anyway, that about exhausts my thoughts on the matter. Lots of words and worth every penny paid. I will point out that much more progress will be made by trying things then by spending a lot of time worrying about problems that are only imagined. My concern about moving the head around during printing might fall into that category, but as I don't actually have a Mendel yet I cannot test it on one myself. Johnrpm and others may completely disagree with my assessments, I don't know. In all likely hood my assessments and thoughts on this topic are deeply flawed and utterly wrong - it would not be the first or last time.

If, after reading all my babble, you still want to work on this, I would suggest the following: try to construct a print head. Follow the basic design for the acrylic print head that Johnrpm published in this thread a while back, scaled for whatever piezo you have. If you don't have a lathe or a mill and some acrylic, make one up as a 3d model and print it on your Repstrap/Reprap in ABS. In fact, do it on the Reprap regardless. The one important thing is that the chamber must be well sealed. Use epoxy or acrylic to seal the chamber after printing, if necessary. Print the chamber without a feed hole and create the feed hole by drilling after the fact. Glue the piezo, nozzle and a feed tube using acrylic. Again, no leaks allowed. If you cannot make up a nozzle similar to the one I described in my earlier post, let me know and I'll mail you one that is known to work (as soon as I can produce another known to work).

It is not known that a Reprap printed head will work but I suspect it will. If it cannot be made to work then we all will have learned something, which is, after all, the point of all this, at least from my perspective. If it does work than you will have made it much easier for the next 3d printer owner that wants to get involved. You'll also get bragging rights for having designed and printed the first Reprap-printed inkjet head. But don't delay too long or I might beat you to it. ;-) Just remember that it is likely just a stop along the way, not an end point. Or prove me wrong and make it useful on a Reprap. Either is good.

Some test results:

Running at 45 volt pulses (repeated every 500 ms), I tested with the following pulse widths:

10 us - no drops observed
20 us - no drops observed
30 us - no drops observed
40 us - no drops observed
50 us - drops
60 us - drops
70 us - drops
100 us - drops
1 ms - drops
5 ms - drops

A transformer based pulse generator that is appropriate for the bimorphs should have no trouble producing a 100 us or longer pulse at a stable 45 volts, so I don't think this print head works with water and "single" drops because of the drive circuit. I'll need to rebuild my transformer based driver and try it just to be certain. Note that the transformer based driver that I posted earlier that derives from Microdrop Generation is still not appropriate as it is designed for pulse widths down below 10 us. My best guess at this point is that the nozzle is the responsible party. As far as I know, it is the first that we have tried that was essentially tapered right up to the outside edge (or at least most of the way). Also, it is a known, rather short, length. While I cannot rule out the different piezo yet, I think it is much less likely to be the cause.

Here is a snapshot of the device and reservoir. I'm testing with the device oriented exactly as seen in the picture, with the nozzle pointing up and to one side, the piezo on the bottom.




In the image below I took the backside of the first print head and passed in slowly and unevenly about 1/4 to 1/2 inches above the nozzle. You can see the drops (I count 12) from the print head in the lower part of the image. An American quarter is included for scale. I'll find something more useful for future pictures (like, say, a ruler?).

Thank you once again Madscifi,
I can only agree with all you have said, For ink a hacked printhead would be faster to deploy, thats why I have tended to concentrate on higher energy units for higher viscosity materials, I found myself collecting the parts for the laser printhead, but realized that it may not cope with higher viscosity ???, so will play with laser stuff at some point because its interesting but must attempt to stay on track with spark units.

Thanks again on behalf of us all, we are lucky to have you as a developer, the effort you have put in and time it takes to document stuff is not insignificant.


(fast rise times, oooh shiny)

---

Random Precision

madscifi Wrote:
-------------------------------------------------------
> casainho Wrote:
> --------------------------------------------------
> > Ok, but can you guess the size of hole you did?
> > and the size of drops?
> >
>
> A #80 drill bit is about 343 um in diameter, so
> the hole will be somewhat larger than that. I did
> a quick test and captured some drops on plastic -
> the smallest are around 1/2 mm in diameter, the
> larger around a mm. It is possible that the larger
> captured multiple drops, or that the drop
> generator is creating a small number of drops for
> each pulse.

0.343mm? -- it's nice to know you can drill such small hole, because before we would need to pay $150 for 10 nozzles of 0.25mm just for try make the same you just did :-)

Now the question is, could we make smaller drops other than that 0.5mm, by changing the electronics drive pulses or diameter of the piezo? -- I would be very happy with 0.1mm drops.

How about using a laser for making a hole on acrylic? we could then have smaller holes than 0.343mm, right? (I do not have a laser....)


> > What size of drops would be useful for you? --
> for
> > me, 0.25mm, to be able to print tracks of
> 0.25mm
> > for LQFP pads of 0.5mm pitch.
> >
>
> At this point I don't know what size of drops
> would work for what I'm interested in. I do know
> that the current 1/2 mm drops are too large for
> one use of interest, but I have not done any
> research to determine what a usable drop size
> would be. It is not something I think I need to
> worry about for a while...

For me, I think 0.1mm would be good for PCB printing, clothes, paper and wood -- that should be enough for people start creating things that they can't do now with cheap commercial printers.


> > So, do you think we have project? do you think
> we
> > will success with this kind of piezos and
> inkjet
> > print head?
> >
>
> Do we have a project? Yes, several, in fact. I
> welcome your enthusiasm but I think we are a long
> ways from having a useful and safe design.

For me, the project need to be able to print with 0.1mm (or less) drop sizes and for using the cheapest ink I can find in stores. The inkjet may have some tweaking/adjust every time before print, just like happens now with the plastic extruder that needs to get hot and extrude a bit before start printing.

As for safe design, we should investigate more about using lower voltage, maybe we can minimize the "high" voltage usage. Also since I understand, just one face of the piezo is in contact with the paint, that face could have the GND (relative to Earth) potential.


> Some test results:
>
> Running at 45 volt pulses (repeated every 500 ms),
> I tested with the following pulse widths:
>
> 10 us - no drops observed
> 20 us - no drops observed
> 30 us - no drops observed
> 40 us - no drops observed
> 50 us - drops
> 60 us - drops
> 70 us - drops
> 100 us - drops
> 1 ms - drops
> 5 ms - drops

Could you please verify the drop size for each of that time values? it's important to find if the drop size changes accordantly with the pulse width.


> If, after reading all my babble, you still want to
> work on this, I would suggest the following: try
> to construct a print head. Follow the basic design
> for the acrylic print head that Johnrpm published
> in this thread a while back, scaled for whatever
> piezo you have.

What is the size of the piezos you are using? do you have a link and datasheet for them? Maybe you could quickly write on wiki this informations so we can try replicate your working print head.


If you don't have a lathe or a
> mill and some acrylic, make one up as a 3d model
> and print it on your Repstrap/Reprap in ABS. In
> fact, do it on the Reprap regardless. The one
> important thing is that the chamber must be well
> sealed. Use epoxy or acrylic to seal the chamber
> after printing, if necessary. Print the chamber
> without a feed hole and create the feed hole by
> drilling after the fact. Glue the piezo, nozzle
> and a feed tube using acrylic. Again, no leaksI don't remember how did you made
> allowed. If you cannot make up a nozzle similar to
> the one I described in my earlier post, let me
> know and I'll mail you one that is known to work
> (as soon as I can produce another known to work).

I don't have my 3D printer working now, but soon I will and I will print this print head. Could you please do a 3D design and share with us? with the correct dimensions for piezo, etc?

Maybe I can go to a local laser shop and ask for them do a hole on a little piece of acrylic and use that piece as a nozzle. What do you think?

---

---
New cutting edge RepRap electronics, ARM 32 bits @ 100MHz runs RepRap @ 725mm/s:

[www.3dprinting-r2c2.com]

... i have flute-millbits with 0.3mm, drillbits with 0.3, 0.2 and 0.1mm and can burn holes down to 0.02mm with my laser, but only in coloured materials.

So if you need specific hole-plates, i can make them from different materials ... only the shipping is a PIA

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

> > casainho Wrote:

> 0.343mm? -- it's nice to know you can drill such
> small hole, because before we would need to pay
> $150 for 10 nozzles of 0.25mm just for try make
> the same you just did :-)


0.050mm not 0.25

---

Random Precision

johnrpm Wrote:
-------------------------------------------------------
> > > casainho Wrote:
>
> > 0.343mm? -- it's nice to know you can drill
> such
> > small hole, because before we would need to pay
> > $150 for 10 nozzles of 0.25mm just for try make
> > the same you just did :-)
>
>
> 0.050mm not 0.25

Yeah, sorry my mistake/forgot. Even that, it's good to know that we can print using 0.343mm hole, at least we saved that $150 for now.

VDX Wrote:
-------------------------------------------------------
> ... i have flute-millbits with 0.3mm, drillbits
> with 0.3, 0.2 and 0.1mm and can burn holes down to
> 0.02mm with my laser, but only in coloured
> materials.

Ok, plastic I guess, like a piece of acrylic or ABS -- what depth?. And metal and depth?


reifsnyderb can make to us nozzles with 0.25mm!! for the same price of actual extruder nozzles. So right now or we ask to reifsnyderb for a nozzle of 0.25mm or to VDX for a nozzle with a smaller value up to 0.1mm or up 0.02mm done using his laser :-)

If with a 0.343mm we got a 0.5mm drop, with a 0.1mm we may get a 0.15mm drop?? -- 0.15mm drop would be just good for me :-)

Edited 2 time(s). Last edit at 09/09/2010 06:09AM by casainho.

---

---
New cutting edge RepRap electronics, ARM 32 bits @ 100MHz runs RepRap @ 725mm/s:

[www.3dprinting-r2c2.com]

casainho Wrote:
-------------------------------------------------------
> VDX Wrote:
> --------------------------------------------------
> -----
> > ... i have flute-millbits with 0.3mm, drillbits
> > with 0.3, 0.2 and 0.1mm and can burn holes down
> to
> > 0.02mm with my laser, but only in coloured
> > materials.
>
> Ok, plastic I guess, like a piece of acrylic or
> ABS -- what depth?. And metal and depth?

... i've already milled 2mm POM, 0.5mm aluminium and 0.3mm brass with the 0.3mm-millbits ... and drilling should work until maybe 3mm (max. length of the threaded piece).

The laser can cut/'drill' plastic until some centimeter thickness with repetitive shots - shape and quality is depending of beam geometry ...
Cutting 0.3mm steel wasn't a problem (it burned/oxidized at air, so i have to apply inert gas) but i hadn't tested with thicker metal sheets, so should be possible for thicker sheets and other materials too ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

VDX Wrote:
-------------------------------------------------------
> ... i've already milled 2mm POM, 0.5mm aluminium
> and 0.3mm brass with the 0.3mm-millbits ... and
> drilling should work until maybe 3mm (max. length
> of the threaded piece).

Would you be able to provide us a nozzle of 0.1mm diameter hole, in brass, just like the one johnrpm did for his acrylic piezo housing? (sorry, can't find the picture he took).

I guess we just need to print with ABS the piezo housing, and tapping it for the nozzle.

---

---
New cutting edge RepRap electronics, ARM 32 bits @ 100MHz runs RepRap @ 725mm/s:

[www.3dprinting-r2c2.com]

... i'll try with the 0.3mm-sheet - i can mill squares or disks out of the sheet, what's better for you, and which size/diameter?

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

VDX Wrote:
-------------------------------------------------------
> ... i'll try with the 0.3mm-sheet - i can mill
> squares or disks out of the sheet, what's better
> for you, and which size/diameter?

0.3mm sheet? I would prefer some brass or aluminum of 2 or 3mm thickness. Since it is to glue to print head body of ABS plastic, that both materials may not be well glued, right? anyone knows if gluing is a good idea?

For 3 or 2mm thickness sheet, I think you don't need to mill it, just use some saw. And maybe a square sheet of 5x5mm will be good, with a 0.1mm hole at center.

The sheet could be the same and/or glued as on this picture from madscifi:

---

---
New cutting edge RepRap electronics, ARM 32 bits @ 100MHz runs RepRap @ 725mm/s:

[www.3dprinting-r2c2.com]