Welcome! Log In Create A New Profile

Advanced

Laser Cutting Head

Posted by aka47 
Re: Laser Cutting Head
July 29, 2008 09:30PM
Rapid prototyping optics is tricky, optics can't be bumpy, and would require polishing. When light hits a bump, it refracts - bumps, scratches, steps, and so on need to be somewhat smaller than the wavelength we're dealing with. I'd be happy with store-bought optics for the first few generations.

I've created a page up on the objects wiki:
[objects.reprap.org]
where we can jot down ideas (and eventually collaboratively design a such a machine.)

I'm assuming we want a ~45 sealed CO2 laser, capable of cutting thin plywood or acrylic sheet, and a machine based on a 1-10 watt diode laser is more interesting for LOM-type fabrication. Happily, it's a wiki, so we can create a page just by going to that url, for example:
[objects.reprap.org]

Anyway, please go to
[objects.reprap.org]
and write down what you know. If you can contribute old plans and designs, the wiki is a great place for them.

A few points:
* What size workspace do we want? I'm guessing A0 or A1 size, 841 mm
Re: Laser Cutting Head
July 29, 2008 11:10PM
Hurro Sebastien

Some of the below is condensed input/points taken from contributions made in the preceding.

For initial development I was going to suggest a cut down modified Darwin (Just cos all the parts and Cartesian bot development are done/readily available etc)

In suggesting cut down I am thinking of a Darwin minus the Z axis. and the vertical components of the frame drastically shortened.

Probably get rid of the base XY frame. and use a thick board with M8 Holes drilled to accept stub corner posts. (Easy alignment and dimensionally pretty stable). The corner posts can off course now be studding rather than bar and attache to the board with Washers and nuts. If the board is thick enough (Kitchen worktop thick) and the stub corner posts short enough diagonal bracing can be omitted as well.

On Dimensions we can of course stretch the current Darwin Axes but there is a point at which the M8's will flex and bow with the weight of the carriage so this will limit the max size somewhat. For development purposes standard Darwin XY size should be fine and cost effective.

A final design size though I guess would be governed by application. ie

A graphic/paper system is likely to be using feedstock in paper sizing ie A4, A3, A1, A0.

A ply/plastic cutting system is likely to be using feedstock based on standard sheet material sizes or easy fractions thereof ie full sheet 4 foot by 8 foot, half sheet 4 * 4, quarter sheet 2*4 etc.

Enclosure and forced ventilation will be necessary for the fumes and for safety. A box enclosure though could be unnecessary particularly during development. A Tent arrangement would be more cost effective and do the trick.

An eventual larger size unit is likely to end up a gantry bot using lightweight aluminum extrusion and the electronics/motors/software etc from the cut down Darwin development machine.

My thoughts and contribution for what they are worth.

cheers

AKA47


Necessity hopefully becomes the absentee parent of successfully invented children.
VDX
Re: Laser Cutting Head
February 16, 2010 02:42AM
Hi all,

... i've updated the Laser-Cutter-Wiki with some prices for diode-lasers:

- comercial 'naked' diodelaser with 8Watt@975nm and 0.1mm atached glass fibre: ~230 Euros (in quantities above 5 pcs).

- turnkey-system with 10Watts@975nm, a visible red pilot-laser and jack for the glass fibre: ~1000 Euros (in quantities above 5 pcs).

I hope to start next week with our salvaged 5Watt-diodelasers, so the costs of a 'naked' 5Watt-laser could go below 100 Euros and my estimation for a DIY-turnkey-system (without pilot-laser but all electronics and focussing-optics) could be something between 300 and 800 Euros (depending of parameters and components) ...


Viktor
Re: Laser Cutting Head
February 17, 2010 12:55PM
I will follow this message and reprap wiki. Do you have a blog about your works?

I am interested on this subject.


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

[www.3dprinting-r2c2.com]
VDX
Re: Laser Cutting Head
February 17, 2010 02:16PM
... until now there is no blog - my laserworks were a sort of specialized optimizations and workarounds with comercial systems and mostly covered by NDAs eye rolling smiley

The actual project with the salvaged pumping diodes is free as in OS, so i'll try to blog the complete development and publish in the wiki in english and with the GRRF and/or in the german forum in plain german ...

But this wouldn't be a very fast process spinning smiley sticking its tongue out


Viktor
Re: Laser Cutting Head
February 17, 2010 02:28PM
VDX Wrote:
-------------------------------------------------------

> The actual project with the salvaged pumping
> diodes is free as in OS, so i'll try to blog the
> complete development and publish in the wiki in
> english and with the GRRF and/or in the german
> forum in plain german ...

Why not buying a laser diode instead of trying to salvage?

And why not start with one cheap and not powerful, and going to upgrade when project get more user traction?

Is there any Ebay laser diode relative cheap, I mean less than $50? and what wroks could we do with that?


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

[www.3dprinting-r2c2.com]
VDX
Re: Laser Cutting Head
February 17, 2010 02:48PM
Hi casainho,

... a comercial pigtailed diodelaser with 8 Watts of power costs me 350 Euro when singlecounts and 250 Euros when ordering more than 5 ...

When salvaging a wrecked pumping module, i'll get some ten of pigtailed diodelasers with 5 Watts of power for less than 100 Euros each and optionally they're coupled in pairs and quads, so i can use them as 10 or 20 Watt-modules too ... but then with more needed current too ...

I'm working with a special 1Watt-diode with thinner fiber, so nearly the same energy-density as with a conventional 2Watt-diode and there is some limitation in power for proper acting, so more power would be much better ...

The typical 'burning lasers' as mostly stated in youtube and ebay are mostly below 1Watt (~220mW to 500mW the 'best') and not fiber-coupled, so you can't feed them properly and don't receive enough energy for usefull cutting-speeds or sintering powers ...


Viktor
Re: Laser Cutting Head
February 17, 2010 07:10PM
The actual project with the salvaged pumping diodes is free as in OS, so i'll try to blog the complete development and publish in the wiki in english and with the GRRF and/or in the german forum in plain german ...

But this wouldn't be a very fast process


If it is much faster to publish on the wiki in german, I volunteer to maintain the german->english translation. smiling smiley

And if you prefer to publish on the wiki in english, I volunteer someone else to maintain the english->german translation. ... casainho, how is your german? grinning smiley


-Sebastien, RepRap.org library gnome.

Remember, you're all RepRap developers (once you've joined the super-secret developer mailing list), and the wiki, RepRap.org, [reprap.org] is for everyone and everything! grinning smiley
VDX
Re: Laser Cutting Head
February 18, 2010 12:14AM
Hi Sebastien,

... OK, then i'll start the documentation first in german and with time translate/rewrite in english ...

I'm learning a lot simply by reading and writing here, but i think for plain english i need some more direct interaction grinning smiley

But anyway - my timing problem is lack of time at home, so the development runs only at night (where i can't work noisy) and some spare hours in the weekends eye rolling smiley


Viktor
VDX
Re: Laser Cutting Head
February 18, 2010 03:43AM
... i updated "Laser Cutter" in the Wiki - added sketch how embedding a pilot-laser and focussing the IR- and pilot-beams ...


Viktor
Attachments:
open | download - Diodelaser with pilotlaser.jpg (141.9 KB)
Re: Laser Cutting Head
February 18, 2010 06:01AM
In reply to [dev.forums.reprap.org]

VDX Wrote:
-------------------------------------------------------
> ... i updated "Laser Cutter" in the Wiki - added
> sketch how embedding a pilot-laser and focussing
> the IR- and pilot-beams ...

My English is worst than VDX :-) -- so German even worst :-)

Viktor, with that 500mW red laser diodes from Ebay, what could we do? -- could we at least do engraving on some material? -- I wish there was such an option for RepRap, a cheap laser tool head.

As for your pilot laser optical head, I think I understand the need for the red laser beam. Is that for help user to understand that IR laser is on? and to avoid SAFETY problems? or is just a guide? and if so, is a guide needed?

As for safety, I believe EVERYONE should use googles and that if IR reflects and user can't see it, the read beam will serve for nothing :-( -- read beam giving like a sensation of fake safety.

If red laser bean is not really necessary, than that laser head will be simple to build and cheap.

Can you please suggest googles? what are you using? and put that info on English wiki? Looks to me that this is a very important step before anyone start playing with laser...

As for IR laser, what could we do with this one?

Edited 1 time(s). Last edit at 02/18/2010 06:02AM by casainho.

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

[www.3dprinting-r2c2.com]
VDX
Re: Laser Cutting Head
February 18, 2010 07:30AM
Hi casainho,

... you can 'engrave' (or melt small cavities or grooves in) dark, good absorbing plastic-surfaces even with 220mW, but only slow and not in depth.

For changing the colour from light to dark you need much higher powers than 500mW - this is depending on the material-absorption, the energy-density (spotsize at specific optical power) and duration-time. The bad thing is the nonlinear behaviour of bright materials (even wood): - you have to heat the spot some millisecond before it went brown and then you instantly burn a hole with uncontrollable depth in some microseconds eye popping smiley

The absorption-rate depends highly on the wavelength - 975 is slightly better than 808nm, and the 10064nm from a CO2-laser are perfect for plastics and for our eye bright or transparent materials ...

In metalls it's the other way - shorter wavelengths results in better absorption, but you need much more power ... some 100 Watts to Kilowatts. Or a better laser with smaller spot and so higher energy-density - my 50Watts-fiberlaser can be focussed down to 5 microns, so for micro-cutting comparable to a NdYAG-laser with 1 Kilowatt power and a spot of 200 microns (the energy-density with 50Watt@5microns is much higher, but the thermal distribution around the small spot is much worser)


For the pilot-laser: - it's a guide for adjusting the exact position and only in second a beam-indicator ...


And for laser-goggles: - i have old/outweared goggles with the specified blocking wavelength between 600nm and 1400nm ... functionally, but not producible.

New ones will cost between 200 and 400 Euros, so this is a major price-factor too ...


Viktor
Re: Laser Cutting Head
February 18, 2010 09:10AM
VDX Wrote:
-------------------------------------------------------
> Hi casainho,
>
> ... you can 'engrave' (or melt small cavities or
> grooves in) dark, good absorbing plastic-surfaces
> even with 220mW, but only slow and not in depth.

So maybe is a good bet to go with a 1W IR laser of $35, where we may be able to cut small thin sheets of some plastics or paper?


> For changing the colour from light to dark you
> need much higher powers than 500mW - this is
> depending on the material-absorption, the
> energy-density (spotsize at specific optical
> power) and duration-time. The bad thing is the
> nonlinear behaviour of bright materials (even
> wood): - you have to heat the spot some
> millisecond before it went brown and then you
> instantly burn a hole with uncontrollable depth in
> some microseconds eye popping smiley

Today I went to a very small shop that have a laser machine of 50W. The guy there told me that it is difficult when he get different materials, he need to go experimenting...


> The absorption-rate depends highly on the
> wavelength - 975 is slightly better than 808nm,
> and the 10064nm from a CO2-laser are perfect for
> plastics and for our eye bright or transparent
> materials ...
>
> In metalls it's the other way - shorter
> wavelengths results in better absorption, but you
> need much more power ... some 100 Watts to
> Kilowatts. Or a better laser with smaller spot and
> so higher energy-density - my 50Watts-fiberlaser
> can be focussed down to 5 microns, so for
> micro-cutting comparable to a NdYAG-laser with 1
> Kilowatt power and a spot of 200 microns (the
> energy-density with 50Watt@5microns is much
> higher, but the thermal distribution around the
> small spot is much worser)

Ok, I would focus on plastic and paper for now.


> For the pilot-laser: - it's a guide for adjusting
> the exact position and only in second a
> beam-indicator ...

That laser machine also had a pilot laser. And also had 2 warnings signal, one for the invisible laser and other for the visible laser. That machine is all in metal but have a transparent protective cover which I presume does the protection, just as the goggles (the warning signals tells that it acts as protection and machine stops if the cover is open).

The guy don't use the pilot-laser for nothing, just for guide. And he told me that he need to adjust the height (Z) every time he needs to cut different thick material. I saw a plate of acrylic molten because of bad Z adjustment... which guy told me that it started to burn inside machine...

So I think we should avoid for now that pilot-laser, and just maybe use a big RED LED for user indication.

See here this video, where we can see an IR laser working, without a pilot-laser. I would love to have such results :-)
That guy have more interesting videos about this subject.


> And for laser-goggles: - i have old/outweared
> goggles with the specified blocking wavelength
> between 600nm and 1400nm ... functionally, but not
> producible.
>
> New ones will cost between 200 and 400 Euros, so
> this is a major price-factor too ...

Would you trust on a cheap of $29 like this?

If we buy that relative cheap laser of $35, what more else do we need to build a laser head for Mendel? would be possible to repstrap the laser head? and electronic control circuit? any lenses?


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

[www.3dprinting-r2c2.com]
VDX
Re: Laser Cutting Head
February 18, 2010 12:07PM
Hi casainho,

... the 1Watt-'burning'-diode is usable for first tests and slow cutting of black plastic-foils ... for cutting paper you need more power.

Another flaw is that it's not pigtailed - the emitted light isn't uniform, as you have a nearly rectangular/oval shape of the conical beam, what's resulting in a poor focus-quality. Or you have to apply first a bar-lens and an aperture-filter, what's reducing the effective output power by some ten percents.

With my pigtailed 1Watt-diode i have a glass-fiber with a core of 60 microns diameter, emitting a circular beam with 1 Watt output power, whar can 1:1 or slightly better refocussed to a perfect round spot with 50 microns in size.

The 5Watt- and 8Watt- pigtailed diodes have fibres with 105 microns in diameter, so the resulting focus will be around 100 microns in diameter, but circular too.

With the 'naked' diodes you'll receive elliptic spots or stripes what's around 200 to 300 microns in the 'fast' or 'bad' orientation ... or with bar-lens+aperture a circular focus around 100 microns and maybe 0,5Watt power left.

The diode in the video "from a CD-burner" could be even weaker - they're mostly 220mW - but it's equipped with a good collimating and focussing optic, what's not included with the stronger diodes.


For the cheap goggles:
"This is a new protection glasses to view diffuse 808-850nm Infrared laser."
... it's intended for use with diffuse lasers, could be enough for the 1Watt-diode with 808nm, but i don't think it's good enough for more power and focussed beams.

Testing of the blocking quality is easy: - as web-cams can see the IR-beam, you can direct the laserbeam through the goggles on the camera ... with good goggles you can't see any light emitting from the laser ...


For completing a laser-toolhead you need the laser, two lenses, a power source of maybe 12 Volts and the needed Amps, a current-driver (e.g. one LM317 as current-stabilizer, or more in parallel for more than 1.5 Ampere), a SolidStateRelais capable of the current and a TTL- or PWM-signal for the SSR switching the power through the current-driver and laser ...

The guy with the 50Watt-signing laser gave you a short glimpse in the problems with engraving different (bright) materials with 50Watts of power!

Was it a CO2-laser with 10064nm or a NdYAG-laser with 1070nm?

With the 808nm-laserdiode and less than 1Watt this problem is even worser ...


Viktor
VDX
Re: Laser Cutting Head
February 18, 2010 01:18PM
... atached is a photo of 'good' laser-goggles with full blocking at 700-1100nm (diode-lasers and NdYAG-lasers) and 10600nm (CO2-lasers) ... but for CO2-lasers any goggles are good, as the 10600nm-IR-light can't go through glass or plastic ...


Viktor
Attachments:
open | download - Laserbrille.jpg (80.5 KB)
Re: Laser Cutting Head
February 18, 2010 02:25PM
VDX Wrote:
-------------------------------------------------------
> Hi casainho,
>
> ... the 1Watt-'burning'-diode is usable for first
> tests and slow cutting of black plastic-foils ...
> for cutting paper you need more power.

It could be a start for RepRap. We need to start with simple things and it is more easy to go further if we get more users/developers traction.


> Another flaw is that it's not pigtailed

Ok, but the prices you gave for pigtailed lasers are much more! Isn't there other cheap option? or something in the middle?


> The diode in the video "from a CD-burner" could be
> even weaker - they're mostly 220mW - but it's
> equipped with a good collimating and focussing
> optic, what's not included with the stronger
> diodes.

Well, if is more cheap and have that advantages, would be great for a start, even if just for a proof of concept.

Well, let's not forget this as an option. I understand that you have much more experience and want more advanced laser/power.


> For the cheap goggles:
> "This is a new protection glasses to view diffuse
> 808-850nm Infrared laser."
> ... it's intended for use with diffuse lasers,
> could be enough for the 1Watt-diode with 808nm,
> but i don't think it's good enough for more power
> and focussed beams.
>
> Testing of the blocking quality is easy: - as
> web-cams can see the IR-beam, you can direct the
> laserbeam through the goggles on the camera ...
> with good goggles you can't see any light emitting
> from the laser ...

Nice to know. It's something we can later put on wiki, on the section for the safety. I am also interested on registering on video and photos this systems working, as for doing a kind of art - please see this my short video of printing.


> For completing a laser-toolhead you need the
> laser,
- buy laser (how much? $50, $100, $200?)

two lenses,
- buy lenses... how much they cost? is difficult the assembly of them? can a support for laser and lenses be printed?

a power source of maybe 12
> Volts and the needed Amps,
- get a PC PSU, should be more than enough


a current-driver (e.g.
> one LM317 as current-stabilizer, or more in
> parallel for more than 1.5 Ampere),

- Cheap and very easy to prototype :-)


a
> SolidStateRelais capable of the current and a TTL-
> or PWM-signal for the SSR switching the power
> through the current-driver and laser ...

- I guess it should also be relative cheap and easy to prototype.

So the big problem here is the cost of laser and lenses. Would be good if we could make the electronics in a way that works with up to maybe 5W laser (I guess that laser Vf = 2V and If = 2.5A, looks nothing impossible for our cheap and simple technology).

We could then just use the same electronics and head for a variety of different laser powers. The users that want to spend more money would buy a more powerful and expensive laser. The head, electronics and firmware/software could be the same or near the same - this would easy the life for everyone :-)

Viktor, did you made already some laser head using a DVD laser? or we can make one for start or document what you have already done...


> The guy with the 50Watt-signing laser gave you a
> short glimpse in the problems with engraving
> different (bright) materials with 50Watts of
> power!
>
> Was it a CO2-laser with 10064nm or a NdYAG-laser
> with 1070nm?

I don't know. But the laser is not on the head, head is just a mirror and support.
The machine have a fume extraction system.


> With the 808nm-laserdiode and less than 1Watt this
> problem is even worser ...

:-)

Were you that wrote this blog message with title first tests building a powder-based stereolithography printer: part 1?
If not, what do you think?

Edited 1 time(s). Last edit at 02/18/2010 02:45PM by casainho.

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

[www.3dprinting-r2c2.com]
VDX
Re: Laser Cutting Head
February 18, 2010 03:06PM
Hi casainho,

> It could be a start for RepRap. We need to start
> with simple things and it is more easy to go
> further if we get more users/developers traction.
>
>
> > Another flaw is that it's not pigtailed
>
> Ok, but the prices you gave for pigtailed lasers
> are much more! Isn't there other cheap option? or
> something in the middle?

... salvaging CD- and DVD-burners would be the cheapest start, but it's only good for some fuming, not real work.

The 'cheap' burning diodes with 20$ to 30$ are only slightly more powerfull, but need all the aditional equipment, as the 5Watt-lasers eye rolling smiley

So you can start with a wrecked DVD-burner, or go the 'powerfull' way - for me the 'burning' diodes aren't worth worrying about when i have either really cheap CD-lasers complete with optics or the pigtailed 5Watt-types.

The cost-estimation of 100 Euros isn't final - maybe i can sell the pigtailed 5Watt-diodes for 50€ or even less ... this depends on the conditions and reworking-issues ...


> > Testing of the blocking quality is easy: - as
> > web-cams can see the IR-beam, you can direct
> the
> > laserbeam through the goggles on the camera ...
> > with good goggles you can't see any light
> emitting
> > from the laser ...
>
> Nice to know. It's something we can later put on
> wiki, on the section for the safety. I am also
> interested on registering on video and photos this
> systems working, as for doing a kind of art -
> please see this my short video of printing.

... i'll make more photos and videos when the 5Watt-prototypes are working ...


> > For completing a laser-toolhead you need the
> > laser,
> - buy laser (how much? $50, $100, $200?)

... let's see ...



> two lenses,
> - buy lenses... how much they cost? is difficult
> the assembly of them? can a support for laser and
> lenses be printed?

... for a similar 8Watt-diodelaser i lathed some plastic-tubes and put two lenses with 15mm diameter in a 20mm-tube with some distance-fittings, so the complete optical head was 20mm in diameter and maybe 60mm in length - should be really easy to print ...


> a power source of maybe 12
> > Volts and the needed Amps,
> - get a PC PSU, should be more than enough

... yes ...


> a current-driver (e.g.
> > one LM317 as current-stabilizer, or more in
> > parallel for more than 1.5 Ampere),
>
> - Cheap and very easy to prototype :-)

... yes, maybe we can find an even simpler solution later ...


> > a SolidStateRelais capable of the current and a
> > TTL- or PWM-signal for the SSR switching the power
> > through the current-driver and laser ...
>
> - I guess it should also be relative cheap and
> easy to prototype.

... switching with currents below 2 Amps could be done with Darlington-transistors or MOSFETS, what's eventually in the box ...


> So the big problem here is the cost of laser and
> lenses. Would be good if we could make the
> electronics in a way that works with up to maybe
> 5W laser (I guess that laser Vf = 2V and If =
> 2.5A, looks nothing impossible for our cheap and
> simple technology).

... for 5 Watt output power you need more than 10 Watts of electrical energy!

AFAIK the 5Watt-diodes consumes max. 5.2Amps ...

But it's better for lifetime, when not exceeding 4.8Amps, so i'm thinking of 4 LM317 á 1.2Amps (or 5 á 1Ampere) in a parallel configuration, so you can popilate and switch in discrete steps and fine-tune with PWM ...


> We could then just use the same electronics and
> head for a variety of different laser powers. The
> users that want to spend more money would buy a
> more powerful and expensive laser. The head,
> electronics and firmware/software could be the
> same or near the same - this would easy the life
> for everyone :-)

... yes ...


> Viktor, did you made already some laser head using
> a DVD laser? or we can make one for start or
> document what you have already done...€

... i made some tests with fuming, but got some stronger with 1Watt and 8Watt. The 1Watt-diode is pictured in the documentation, later i'll post images with the 5Watt-types ...


> I don't know. But the laser is not on the head,
> head is just a mirror and support.
> The machine have a fume extraction system.

... sound like a CO2-laser - this is the typical laser for cutting organic materials, and better suited for plastic, paper and wood, than a diode-laser ...


Viktor
Re: Laser Cutting Head
February 18, 2010 03:41PM
> ... salvaging CD- and DVD-burners would be the
> cheapest start, but it's only good for some
> fuming, not real work.
>
> The 'cheap' burning diodes with 20$ to 30$ are
> only slightly more powerfull, but need all the
> aditional equipment, as the 5Watt-lasers eye rolling smiley
>
> So you can start with a wrecked DVD-burner, or go
> the 'powerfull' way - for me the 'burning' diodes
> aren't worth worrying about when i have either
> really cheap CD-lasers complete with optics or the
> pigtailed 5Watt-types.
>
> The cost-estimation of 100 Euros isn't final -
> maybe i can sell the pigtailed 5Watt-diodes for
> 50€ or even less ... this depends on the
> conditions and reworking-issues ...

50€ I think would be good. The final price of this head would not be much more than an extruder or heated bed module.

> ... i'll make more photos and videos when the
> 5Watt-prototypes are working ...

Please do it :-)


> > two lenses,
> > - buy lenses... how much they cost? is
> difficult
> > the assembly of them? can a support for laser
> and
> > lenses be printed?
>
> ... for a similar 8Watt-diodelaser i lathed some
> plastic-tubes and put two lenses with 15mm
> diameter in a 20mm-tube with some
> distance-fittings, so the complete optical head
> was 20mm in diameter and maybe 60mm in length -
> should be really easy to print ...

Were can we get that lenses? can you send a link were we can read more about them?


> ... for 5 Watt output power you need more than 10
> Watts of electrical energy!

Not a real problem, actual extruder heater uses about 30W and heated bed uses about 80W.


> AFAIK the 5Watt-diodes consumes max. 5.2Amps ...
>
> But it's better for lifetime, when not exceeding
> 4.8Amps, so i'm thinking of 4 LM317 á 1.2Amps (or
> 5 á 1Ampere) in a parallel configuration, so you
> can popilate and switch in discrete steps and
> fine-tune with PWM ...

Maybe a circuit for a power LED would be ideal for this task.

We need to be able to control the current of the laser and to be able to turn off and on, at which frequencies? which duty cycles? which rise up times?
Isn't there any Open Source project already over Internet were we can reuse the schematic and information?


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

[www.3dprinting-r2c2.com]
VDX
Re: Laser Cutting Head
February 19, 2010 12:38AM
... here are some LM317-applications for constant current-driving and a simple workaround for generating geometrical defined pulses for the laser from the XY-clocks instead of calculating the needed power from the moving speed ...


Viktor
Attachments:
open | download - LM317-Stromkonstanter.JPG (19.3 KB)
open | download - LM317-60V250mA.jpg (16.6 KB)
open | download - LM317-12V[N]x1A.JPG (41.2 KB)
open | download - XY-Pulsgenerator.jpg (39.4 KB)
Re: Laser Cutting Head
February 19, 2010 07:09AM
VDX Wrote:
-------------------------------------------------------
> ... here are some LM317-applications for constant
> current-driving and a simple workaround for
> generating geometrical defined pulses for the
> laser from the XY-clocks instead of calculating
> the needed power from the moving speed ...

I understand that LM317 designs. I also understand the XY-Pulsgenerator, which will make a pulse on his output everytime happens a X or Y step of stepper motors.

I am curious about the shape of current wave we will put on laser... is just an ON/OFF of certain time?

How do we control a different currents to laser, for engraving?

Edited 1 time(s). Last edit at 02/19/2010 07:50AM by casainho.

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

[www.3dprinting-r2c2.com]
VDX
Re: Laser Cutting Head
February 19, 2010 02:10PM
... for coarse controlling the output-power of a 8Watt-diode i built a current-driver with 6 parallel LM317 with switches, so i could select the current by the count of switched LM317s in steps of 0.7 Amperes.

For finetuning i used a microcontroller generating either constant PWM or a defined pulse-form when activated by an external puls (with a foot-switch for soldering a pad).

So i could melt a droplet of gold-paste 60 microns in diameter and 20 microns high with 'slow' evaporating the flux, or burn a hole through the goldpad and in the ceramic-layer beneath ...

When driving the laser with XY-step-pulses you don't need PWM, so you can simply switch the needed amount of current for burning a hole or melting a spot. For better adjusting you can define the LM317s with different currents - e.g. 0.2A, 0.3A, 0.5A, 0.7A, 0.9A, 1.1A, 1.3A ... and so on ...

So you can use and add the currents from 0.2Amps to 3Amps in many configurations ...


Viktor
Re: Laser Cutting Head
February 19, 2010 04:41PM
Ok. Since LM317 is a linear regulator and power is dissipated by it, maybe we could do the same with a power transistor or mosfet?

Like using a current sense resistor and an opamp as a comparator of voltage across sense resistor. Like doing a linear variable current source for the laser. What do you think?

(I need to take time for reading projects about this DIY laser systems)


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

[www.3dprinting-r2c2.com]
Re: Laser Cutting Head
February 20, 2010 11:07AM
The most efficient way to do it will be with a switch mode current regulator but I read you have to be careful with a laser diode as a slight overshoot, even for a very small time will destory them. So the feedback control loop would need to be spot on.

I have a small 1W IR diode laser with a regulator circuit, so I will try it and blog the circuit sometime. It has a relay in it so I think I will be redesigning it once I have reverse engineered it.


[www.hydraraptor.blogspot.com]
VDX
Re: Laser Cutting Head
February 20, 2010 01:28PM
... here is another driving and modulating circuit with a transistor ... you have to adjst the values for your diode ...


Viktor
Attachments:
open | download - ttl_treiber.jpg (20.6 KB)
Re: Laser Cutting Head
February 20, 2010 05:54PM
nophead Wrote:
-------------------------------------------------------
> The most efficient way to do it will be with a
> switch mode current regulator but I read you have
> to be careful with a laser diode as a slight
> overshoot, even for a very small time will destory
> them. So the feedback control loop would need to
> be spot on.

Ok, and so let's avoid that solution. I guess we can try to have a few LM317 with good dissipation and use the +3.3V of PSU.


> I have a small 1W IR diode laser with a regulator
> circuit, so I will try it and blog the circuit
> sometime. It has a relay in it so I think I will
> be redesigning it once I have reverse engineered
> it.

Why did you have such 1W IR laser? :-) are you interested also on laser for Mendel?

What would you prefer to have/work on, a laser cutting head for you or powder-based stereolithography printer?


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

[www.3dprinting-r2c2.com]
Re: Laser Cutting Head
February 21, 2010 12:44PM
I don't think 3.3V will be enough. Viktor's 5V circuit looks good though.

I bought the laser to experiment with, but have not had time yet. Mainly for cutting SMT stencils, etc. Maybe some LOM or sintering experiments.


[www.hydraraptor.blogspot.com]
VDX
Re: Laser Cutting Head
February 21, 2010 12:55PM
... with the LM317 as current-regulator you need 3 to 4Volts mor than you diode will consume - so for the highpower-diodes with 3Volts or more you should best use a 12Volts-source ...
VDX
Re: Laser Cutting Head
March 10, 2010 02:00PM
... marginal update: - the shipment of the 5Watt-diodes delayed, so i tested soldering and wire-bonding with my 50Watt-laser at 5 Watts and 10 Watts with defocussed beam to get nearly the same conditions as with a diodelaser.

5Watts CW at 1070nm (diodelaser=975nm) is enough to melt a 0.5mm thick solder-wire and fuse the solder with the copper-plating of a PCB - 10Watts is slightly overkill ... it burns the underlying PCB and oxidizes the solder grinning smiley

Clean copper-wire with 0.1mm diameter can be soldered with a droplet of previous positioned solder or a blob of solder-paste positioned over the wire.

The coating of isolated copper-wire can be burned away, here the type and thicknes of the coating is to be tested, but shouldn't be a big problem.

With a proper feeding-methode a wire-wrap-head feeding thin isolated copper-wire from a presoldered pad to another would be a great tool for fabbing PCB's ... only the cutting of the wires is to be solved.

With the 50Watt-laser and a spot smaller 50microns i can evaporate/cut the wire, but with a diodelaser and much bigger spot (between 200 and 500 microns when bundeling more 5Watt-diodes to a stronger module) i have no chance ...


Viktor
Re: Laser Cutting Head
March 10, 2010 02:10PM
VDX Wrote:
-------------------------------------------------------
> ... marginal update: - the shipment of the
> 5Watt-diodes delayed, so i tested soldering and
> wire-bonding with my 50Watt-laser at 5 Watts and
> 10 Watts with defocussed beam to get nearly the
> same conditions as with a diodelaser.
>
> 5Watts CW at 1070nm (diodelaser=975nm) is enough
> to melt a 0.5mm thick solder-wire and fuse the
> solder with the copper-plating of a PCB - 10Watts
> is slightly overkill ... it burns the underlying
> PCB and oxidizes the solder grinning smiley

Good idea and work :-)

Couldn't we mill a PCB using a 10W (or other) diode laser?


> Clean copper-wire with 0.1mm diameter can be
> soldered with a droplet of previous positioned
> solder or a blob of solder-paste positioned over
> the wire.
>
> The coating of isolated copper-wire can be burned
> away, here the type and thicknes of the coating is
> to be tested, but shouldn't be a big problem.
>
> With a proper feeding-methode a wire-wrap-head
> feeding thin isolated copper-wire from a
> presoldered pad to another would be a great tool
> for fabbing PCB's ... only the cutting of the
> wires is to be solved.
>
> With the 50Watt-laser and a spot smaller 50microns
> i can evaporate/cut the wire, but with a
> diodelaser and much bigger spot (between 200 and
> 500 microns when bundeling more 5Watt-diodes to a
> stronger module) i have no chance ...

If a diode laser could melt ABS or other plastic, I would prefer to make a printer using ABS powder, like this one: [homemade3dprinter.blogspot.com]

Edited 1 time(s). Last edit at 03/10/2010 02:11PM by casainho.

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

[www.3dprinting-r2c2.com]
VDX
Re: Laser Cutting Head
March 10, 2010 02:52PM
casainho Wrote:
> Couldn't we mill a PCB using a 10W (or other)
> diode laser?
... no - copper is highly reflective and heat-conducting, so you'll need much more energy-density and when melting through the copper the material of the PCB will absorb to much energy and explode eye popping smiley

Some ten years ago i tried this: - with an Excimer-laser at UV-wavelengths you can engrave copper, but this lasers aren't avaliable in DIY-range eye rolling smiley

With a 100Watts-NdYAG-laser i made a trick - bought PCB's with 50% thicker copper-plating, engraved only 75% of the copper-thickness and etched 30% of the copper away, so the engraved paths were cleared and the residue of the copper-plating was slightly thicker than with conventional PCB's ...

For direct engraving you should better try with coating the copper with a thin layer of black plastic, burn the plastic away and etch the free'd copper ...


> If a diode laser could melt ABS or other plastic,
> I would prefer to make a printer using ABS powder,
> like this one:
> [homemade3dprinter.blogspot.com]
... yes, powder-sintering is the other interesting fabbing process i'm going with the diodelaser-modules grinning smiley

Edited 1 time(s). Last edit at 03/10/2010 02:54PM by VDX.

Viktor
Sorry, only registered users may post in this forum.

Click here to login