Post by thread:[EE]: Printer to PCB?

Has anyone tried to adapt a bublejet printer to print ink direct onto PCB?? I'm thinking it would make an excellent fast prototyper to ink a single-sided PCB for easy etching and SMD use. This eliminates all the drilling and other time issues with prototyping through-hole designs and eliminates the photodevelop hassle with expensive consumables. I've checked the bubblejet ink onto a PCB but it beads and contracts. However the ink can be replaced with a different type by the popular refil method so I don't see that as a big issue. In years past I have used everything from nail varnish to waterproof felt tip to do PCBs, the ink just needs to be waterproof and adhere well to copper. Obviously a flatbed plotter or CNC machine could be used, but i'm thinking the speed and fine detail of a bubblejet coupled with ease of having a printer driver to work straight from Protel or Eagle would make it a very fast prototyping system. :o) -Roman -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

Hello Roman, do you know the toner transfer method? In this case you print the positive picture of the pcb on a heat resistant plastic film with a laser printer. Afterwards you take the raw pcb with the copper side upwards, put the plasic film with the picture side down on it. Then ironing the plasic film. The toner will melt and the image is transfered to the copper. Plastic film is removed afterwards. While the melting toner particles are etch resistant you can put you pcb direct in Fe(III)Cl or something else for etching. After etching you can remove the toner with something like aceton, and voila, the pcb is ready for drilling. Regards Thomas -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

Roman Black wrote: > I've checked the bubblejet ink onto a PCB but > it beads and contracts. However the ink can be > replaced with a different type by the popular > refil method so I don't see that as a big issue. > In years past I have used everything from nail > varnish to waterproof felt tip to do PCBs, the > ink just needs to be waterproof and adhere well > to copper. Hi, Roman, some wise man wrote on 2002-07-29 04:11:28 PST: "Re: [OT]:Jet-printer ink that could resist ferric chloride." {Quote hidden}>Mike Singer wrote: > Does anybody know any ink that can be used for printing > onto pcbs by ink-jet printer, ink that could resist ferric > chloride solution? > > > Hi Mike, ANY ink will do. In my childhood years > > of making hobby PCBs I have used dalo pen, nail > > polish, ink from a bottle, and finally settled on > > waterproof felt-tip pen. Any ink that dries on the > > PCB will do, it makes a barrier to the FeCl and > > works great. Alchohol (solvents) etc will dissolve > > the ink but FeCl won't, it will just etch around it. > > -Roman Seriously, I've partially disassembled Epson-IIS, lowered paper feed shaft 5 mm down, so pcb could be transported horizontally. But I had to stop this sort of relaxation between my main win-software works. Even postings to Piclist are now problems because of the lack of time since I've new customers to my program. By the way, you wrote on July 12, 2002: "Re: [OT]: Client, who absolutely trusts in Delphi." {Quote hidden}> I use a similar rule with customers as with dating. > If the girl is keen, then date. If you have to really > spend a lot of effort talking her into it, don't > bother. A "trouble" customer like this is never worth > it, he hasn't given you a cent yet and is already > complaining and telling you what to do. Some people > are just like that and you really don't need to do > business with them, it will only get worse. For the > effort and heartache this guy will cost you, you could > spend the same effort finding a few *good* customers. I answered then: > For the last time this is my approach to customers > too, they are real bastards sometime, at least here > in the Ukraine. But the reason, why I posted my > questions is I doubt this way is most effective. Now I should admit this way is really effective. Regarding ink, you can try deferent options when printing: micro- weave, glossy paper, film etc. Maybe pcb heat or UV treatment before etching will work. Please, if you succeeded with ink, drop a message to the List. Mike. -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

Brad Woods wrote: > > I don't know how you would do this, but... Most bubblejet ink is water > soluable, adding soap to the ink will keep it from beading up. Yes water-soluble is a large part of the problem, both causing bad beading (ie low adherence to the copper) and is totally unsuitable for etching anyway. :o) It has to be a non-water soluble ink, which are usually alcohol or acetone soluble and mostly adhere well to metals. -Roman -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

Victor Faria wrote: > > I have read some where ''' a pcblist I think where someone had some results > with floor wax. > also steel blue (steel marking ink) Yes the steel blue or "Prussian Blue" is a good ink for this use and is alcohol soluble and waterproof. It might be a bit thin out of the tin but from memory the alcohol dires out quite quickly and it thickens up, it should be possible to use a painters viscosity measure cup, and get it to the right viscosity and then fill a bubblejet cartridge in the normal way. -Roman -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

Thanks Thomas. :o) Yes I have seen the toner method work, and it works with plain paper too if the toner is dark enough. But the idea of ink straight to PCB is tempting. ;o) -Roman Thomas Lehmann wrote: {Quote hidden}> > Hello Roman, > > do you know the toner transfer method? > In this case you print the positive picture of the pcb > on a heat resistant plastic film with a laser printer. > Afterwards you take the raw pcb with the copper side upwards, > put the plasic film with the picture side down on it. > Then ironing the plasic film. The toner will melt and the > image is transfered to the copper. Plastic film is removed afterwards. > While the melting toner particles are etch resistant you can put > you pcb direct in Fe(III)Cl or something else for etching. > After etching you can remove the toner with something like aceton, > and voila, the pcb is ready for drilling. -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

Mike Singer wrote: > Seriously, I've partially disassembled Epson-IIS, > lowered paper feed shaft 5 mm down, so pcb could > be transported horizontally. > But I had to stop this sort of relaxation between > my main win-software works. Shame. :o) I'm interested to see how it turns out. {Quote hidden}> By the way, you wrote on July 12, 2002: > >For the > > effort and heartache this guy will cost you, you could > > spend the same effort finding a few *good* customers. > > I answered then: > > For the last time this is my approach to customers > > too, they are real bastards sometime, at least here > > in the Ukraine. But the reason, why I posted my > > questions is I doubt this way is most effective. > > Now I should admit this way is really effective. I thought you would see the light. Chasing after difficult customers is OK if you have lots of time and very few customers. But in the real world those 5% of people are mongrels no matter how hard you try and please them. The moment they get unreasonable just tell them to get lost and spend that time finding or helping the GOOD customers. Like that 80:20 rule, the bad customers take 80% of your time and return 20% of the income. > Regarding ink, you can try deferent options when > printing: micro- weave, glossy paper, film etc. > Maybe pcb heat or UV treatment before etching will work. > Please, if you succeeded with ink, drop a message > to the List. Ok, will do. The main issue for me is not the ink, I see that as easy, but the feeding of the circuit board. To be really useful for small SMP protoypes it needs to be able to handle 1" x 1" boards, so that points to a flatbed system like moving the printer and the tiny board(s) sit on the flat bed. -Roman -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

> Hi Cristian, how did you get the PCB to pass > through the printer? Will it work with very small > PCB's like 1" square?? Citizen used to make a flat bed ink jet printer about 10 years ago. Would have been ideal for this use. Unfortunately I cannot remember the model number. It also had this other idiosyncrasy that you could print "normal way up" for doing long print jobs which would go over many pages, or it could print "upside down" in which case it would feed to the top of the page, and then print down as you viewed the printout. Designed to be used for tear off invoices at a counter. Come to think of it, this must mean it was a normal dot matrix print head rather than inkjet, because it would do multipart paper :)))) Just needed a ribbon with water resistant ink (not) :))) -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

>Ok, will do. The main issue for me is not the ink, >I see that as easy, but the feeding of the circuit >board. To be really useful for small SMP protoypes >it needs to be able to handle 1" x 1" boards, so >that points to a flatbed system like moving the >printer and the tiny board(s) sit on the flat bed. If you really want to use such small pieces of laminate then I think you will have to get an electrostatic hold down plotter. The only other way would be to tape the piece of PCB to a sheet of paper. Pretty well every printer I have come across relies on the paper being of a minimum width (typically A4 or American equiv) or continuous tractor feed stationary just to pass through the printer without problems. The only other way I see is to be able to punch your small bit out of a sheet, a bit like a cookie cutter making a hole in pastry, or a sheet of Avery labels. Then you can refeed the laminate for the next small pcb in 3 days time, taking care to print on a different spot, and not the hole ;)) -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

> -----Original Message----- > From: pic microcontroller discussion list > [.....PICLISTKILLspam@spam@MITVMA.MIT.EDU]On Behalf Of Roman Black > Sent: Monday, 4 November 2002 22:21 > To: PICLISTKILLspamMITVMA.MIT.EDU > Subject: Re: [EE]: Printer to PCB? <elsnipo> > Ok, will do. The main issue for me is not the ink, > I see that as easy, but the feeding of the circuit > board. To be really useful for small SMP protoypes > it needs to be able to handle 1" x 1" boards, so > that points to a flatbed system like moving the > printer and the tiny board(s) sit on the flat bed. > -Roman Roman, Try an EPSON CD writer, the PCB could fit into the tray with a cardboard adaptor (dummy cd) Regards, Kat. --- Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.410 / Virus Database: 231 - Release Date: 31/10/2002 -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

Roman Black wrote: > ... To be really useful for small SMP protoypes > it needs to be able to handle 1" x 1" boards, so > that points to a flatbed system like moving the > printer and the tiny board(s) sit on the flat bed. > -Roman Yes, since I've shifted paper feed shaft down about 5 mm, my next step would be to place on it "flat bed" - textolite sheet with dimensions more than A4 format. It requires a couple of small bearings in front of the shaft, and a couple from behind, and some slide guide on each side. Starting position could be set by printer's paper-in mechanical switch, when the printer starts to rotate the paper feed shaft. The bloody old Epson Stylus Colour IIS mechanism lies uncovered on one of my bookshelves waiting for the better times. :-) I tried to print on a horizontal sheet of cardboard helping it with my hands: It worked. Mike. PS: Correct me, please, if I use " paper feed shaft" term wrongly: English is not my native. ( The rubber-coated roller rotated by stepper to feed a paper ). -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

I've got an old citizen color thermal transfer printer that I've been playing with to do something similar. I'm able to pass CD's through the transport but the ribbon breaks when passing over the sudden transitions in media hights. I think I can overcome this by building up the surrounding areas of the cd carrier card to keep everything on the same level. I've seen similar carrier cards on ink jet based cd printers as Kat suggested. Perhaps something similar really could be rigged for small circuit boards. I originally got this printer with proto circuit board production in mind; but the printer overlaps the colors with a clear protective layer that makes etching a bit fuzzy. Need to figure out how to build a custom print driver to kill that function off. -Dal > > {Original Message removed}

On Mon, 4 Nov 2002, Alan B. Pearce wrote: *>The only other way I see is to be able to punch your small bit out of a *>sheet, a bit like a cookie cutter making a hole in pastry, or a sheet of *>Avery labels. Then you can refeed the laminate for the next small pcb in 3 *>days time, taking care to print on a different spot, and not the hole ;)) There is another way: spot-glue the part to a larger sheet. Peter -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

On Tue, 5 Nov 2002, Roman Black wrote: *>Yes water-soluble is a large part of the *>problem, both causing bad beading (ie low *>adherence to the copper) and is totally *>unsuitable for etching anyway. :o) You could use inversion masking. Print with bubble on copper then spray on a nitro or other based laquer while the bubble ink is still wet. The laquer won't catch on the wet parts. Thus you would obtain a negative in laquer after partly drying the laquer and washing with water. This has not been tried but certain kinds of stone printing used in graphics arts use this method (or nearly) and do work (and have for hundreds of years). I am under the impression that the wax transfer jets work directly for etching if enough print density is achieved (i.e. Epson etc). Peter -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

There used to be people on eBay that sold plans to turn various printer models into CD printers, presumably by allowing them to pass the thicker material through the machine. I don't know if these people are still around, but it might be worth a shot, or at least, a google search about home made CD printers. I think the only way to get a small size PCB to pass reliably is with some sort of carrier. If the carrier was cheap enough, then it could easily be remade for each new PCB size. Cardboard of some sort springs to mind, though I suppose it might wear after a while. As a side thought, if you did use cardboard, perhaps you could use little spikey wheels along the edges to move it along. Something like tractor feed on a dot matrix printer, but without preformed holes. Since the carrier would be basically garbage anyways, who cares what it looks like after? I hope you come up with something useful in regards to the ink and then share :) The gun bluing sounds interesting, will it work with regular print heads though? I'd love a machine that could print directly onto PCBs. Part of the reason I hand wire all my projects is it's just easier (in my current situation) than doing a PCB. Of course, the other reason is that I don't know how to design PCBs :) I did download Eagle yesterday though, and I will likely try it eventually. Figures the project I want to use it on will need a bigger board that I can do with the freeware version. Josh -- A common mistake that people make when trying to design something completely foolproof is to underestimate the ingenuity of complete fools. -Douglas Adams Roman Black wrote: > Hi Cristian, how did you get the PCB to pass > through the printer? Will it work with very small > PCB's like 1" square?? -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

Hi, Roman, Using a tray you can do that. You have to mill that tray for precision.A double scotch will position that 1" piece of PCB. Cristian {Quote hidden}>Cristian wrote: >> >> I've succeed to pass a PCB through a bubblejet printer. >> The ink problem remains unsolved for me. > > >Hi Cristian, how did you get the PCB to pass >through the printer? Will it work with very small >PCB's like 1" square?? >-Roman -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

There are two ways to search for: 1. Ink to copper 2. Ink to photoresist. The acetone-soluble ink will melt the ink cartridge, for sure. Cristian >Brad Woods wrote: >> >> I don't know how you would do this, but... Most bubblejet ink is water >> soluable, adding soap to the ink will keep it from beading up. >It has to be a non-water soluble ink, which >are usually alcohol or acetone soluble and >mostly adhere well to metals. >-Roman -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

Josh Koffman wrote: > ...I'd love a machine that could print directly onto > PCBs. Part of the reason I hand wire all my projects is it's just easier > (in my current situation) than doing a PCB. Of course, the other reason > is that I don't know how to design PCBs :) I did download Eagle > yesterday though, and I will likely try it eventually. Figures the > project I want to use it on will need a bigger board that I can do with > the freeware version. > > Josh > -- > A common mistake that people make when trying to design something > completely foolproof is to underestimate the ingenuity of complete > fools. Another great reason why use ink-jet, old PCB fighter Roman have not mentioned, is the ability to produce _VERY_ big PCBs very cheaply. Imagine big indicator panel about 0.4m x 1m. Even from Olimex this PCB could cost hundreds $. With A2 ink-jet printer it could cost 10 times less. The cost of the "system-on-pcb" gets reasonable. Mike. If you think of something to be foolproof, the fools are always greater then the proof! Eduard Teller, American Nuclear Physicist -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

Actually I've just remembered there is a printer designed for printing PCBs it was advertised in Nuts & Volts - I'll look it up, th e printer looked like a modified HP photo printer, however this was a commercial enterprise but you could only contact the manufacturer by email, if it helps anyone the contact was a Chinese man. colin -- cdb, .....bodgy1KILLspam.....optusnet.com.au on 05/11/2002 -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

>So that is an Epson PRINTER that prints onto >CD's?? I've never seen those. First one I came across was designed to look like LaserJet Printer as far as the PC was concerned. I do not know what the printing technology was, but assume it was probably inkjet. Came as part of an automated CD writer system. Put a stack of blank CD's in one side, and the autofeed stuffed one into the writer, when finished writing it went into the printer. That would have been about 8 or 9 years ago now. Was used to archive scanned census data in NZ for the statistics department. -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

cdrom2go.com/equipment/ezcd.htm My dad has one o' these in his home recording studio setup. It looks like a standard epson printer with a flat transport and a plastic cd carrier. Prints fairly nice, but needs specially coated CD's to print upon. I've been playing with an thermal transfer printer to see if I can't get it to work as well --it won't need a prepped CD. ----- Original Message ----- From: "Roman Black" <EraseMEfastvidspam_OUTTakeThisOuTEZY.NET.AU> > Katinka Mills wrote: > > Try an EPSON CD writer, the PCB could fit into the tray with a cardboard > > adaptor (dummy cd) > So that is an Epson PRINTER that prints onto > CD's?? I've never seen those. > -Roman -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

Roman, here is the info on a supposed printer designed for PCB making. The information is from the New Products section of the June edition of Nuts & Volts magazine MCU Electrostatic PCB Maker Will take output from most EDA software and Front Panel design s/w such as Autocad trace widths down to 0.006" (note Inches ) precision is to 1200dpi. The blurb then gets confusing as it waffles on aboutcorrect pressure and heat application. The magazine has a photo of the printer Contact : Charles Liu , 9300 Flair Drive, 106, Dept.NV, El Monte, CA 91731 Fax: 626-287-7607 email: charlesliu_88spam_OUTyahoo.com I though of enquiring about its use for prototyping, but was put off by no website and a yahoo email address. Colin -- cdb, @spam@bodgy1KILLspamoptusnet.com.au on 06/11/2002 -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

Mike Singer wrote: > Another great reason why use ink-jet, old PCB > fighter Roman have not mentioned, is the ability to > produce _VERY_ big PCBs very cheaply. > Imagine big indicator panel about 0.4m x 1m. > Even from Olimex this PCB could cost hundreds $. > With A2 ink-jet printer it could cost 10 times less. > The cost of the "system-on-pcb" gets reasonable. Good point! You could do very long PCB provided your printer feed was reliable and the page length was long enough. I worry a bit that the PCB would move like the problem with stick scanners that roll the page through. -Roman -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email RemoveMElistservTakeThisOuTmitvma.mit.edu with SET PICList DIGEST in the body

Josh Koffman wrote: > > There used to be people on eBay that sold plans to turn various printer > models into CD printers, presumably by allowing them to pass the thicker > material through the machine. Interesting. > I think the only way to get a small size PCB to > pass reliably is with some sort of carrier. > if you did use cardboard, perhaps you could > use little spikey wheels along the edges to move it along. Something > like tractor feed on a dot matrix printer, but without preformed holes. I think this is in the right direction, what would be better still would be to move the printer along using a tractor feed type system and have all boards placed on a flatbed. The positively registered feed would save all dimensioning problems as there could be no slip. Since the paper feed is usually a stepper motor maybe you could make the printer movement as a leadscrew/threaded rod etc then just make sure that the steps/inch is the same so it would use the same printer drive hardware. > I hope you come up with something useful in regards to the ink and then > share :) The gun bluing sounds interesting, will it work with regular > print heads though? It's not actually "blueing" as that is a nasty chemical process. The "prussian blue" is simply an alchohol based INK used in metalwork to aid in marking out the metal with a scriber. It's a dark blue/purple and adheres well to the metal and is cheap in large tins. Viscosity should be fully modifyable by adding alcohol or drying it out a bit. > I'd love a machine that could print directly onto > PCBs. Part of the reason I hand wire all my projects is it's just easier > (in my current situation) than doing a PCB. Yep I do a lot of veroboard (stripboard) work for one-offs and being able to do small boards in minutes would be very handy. I'm thinking in conjunction with the $20 hot air SMD solderer we discussed here a year back you could do nice small apps in minutes. -Roman -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email spamBeGonelistservspamBeGonemitvma.mit.edu with SET PICList DIGEST in the body

cdb wrote: {Quote hidden}> here is the info on a supposed printer designed for PCB making. > > The information is from the New Products section of the June edition > of Nuts & Volts magazine > > MCU Electrostatic PCB Maker > > Will take output from most EDA software and Front Panel design s/w > such as Autocad trace widths down to 0.006" (note Inches ) precision > is to 1200dpi. > > The blurb then gets confusing as it waffles on aboutcorrect pressure > and heat application. > > The magazine has a photo of the printer > > Contact : Charles Liu , 9300 Flair Drive, 106, Dept.NV, El Monte, CA > 91731 > > Fax: 626-287-7607 email: TakeThisOuTcharlesliu_88EraseMEspam_OUTyahoo.com > > I though of enquiring about its use for prototyping, but was put off > by no website and a yahoo email address. Thanks Colin, i'll check that out. Not sure I like the sound of electrostatic operation though... Laserprinters etc rely on the paper NOT scratching the photographic film on the drum. Can't imagine small PCBs with hacksawed edges being good there. :o) -Roman -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email RemoveMElistservTakeThisOuTmitvma.mit.edu with SET PICList DIGEST in the body

Hello Roman, as a veroboard user this link might be of interest for you. Is something like "veroboard CAD". They have a working demo, which is only limited in the number of parts to use. http://www.abacom-online.de/html/lmaster.html They also have some ohter useful electronic related software. Hope this helps until your pcb-printer is ready.;-) Regards Thomas >Yep I do a lot of veroboard (stripboard) work for >one-offs and being able to do small boards in >minutes would be very handy. I'm thinking in >conjunction with the $20 hot air SMD solderer >we discussed here a year back you could do nice >small apps in minutes. >-Roman -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email RemoveMElistservEraseMEEraseMEmitvma.mit.edu with SET PICList DIGEST in the body

Comments within Josh -- A common mistake that people make when trying to design something completely foolproof is to underestimate the ingenuity of complete fools. -Douglas Adams Roman Black wrote: {Quote hidden}> I think this is in the right direction, what would > be better still would be to move the printer along > using a tractor feed type system and have all boards > placed on a flatbed. The positively registered feed > would save all dimensioning problems as there could > be no slip. > > Since the paper feed is usually a stepper motor > maybe you could make the printer movement as a > leadscrew/threaded rod etc then just make sure that > the steps/inch is the same so it would use the same > printer drive hardware. The only problem I see here is that this project suddenly starts to get very hard to make for the average joe with no machining experience. If it was just a simple modification of an existing printer that allowed the thicker material, plus perhaps the cardboard carried, and of course the ink modification, I think most people could handle it. Once you start changing the printer too much, why not just take the controller electronics and a couple mechanical subsystems from the printer, and fashion a totally new machine. I know I don't have the skills/time to do that (not that a machine like that wouldn't be useful). > It's not actually "blueing" as that is a nasty > chemical process. The "prussian blue" is simply > an alchohol based INK used in metalwork to aid > in marking out the metal with a scriber. It's a > dark blue/purple and adheres well to the metal and > is cheap in large tins. Viscosity should be fully > modifyable by adding alcohol or drying it out a bit. So this would be the stuff that flows into the scratches to increase contrast? I was wondering about the gun blueing, IIRC it's actually an acid of some sort. -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email RemoveMElistservspam_OUTKILLspammitvma.mit.edu with SET PICList DIGEST in the body

Another option might be to buy a cheap hobbyist CNC machine and change the routing tools to Draughtmans Ink pens. There's a German firm who have a kit about 700 Euros, for a record player platter style 3 head machine - put together bit like an Airfix kit -now that dates me, I reckon the drill bits could be replaced by pens. Colin -- cdb, RemoveMEbodgy1TakeThisOuTspamoptusnet.com.au on 07/11/2002 -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email EraseMElistservspamspamBeGonemitvma.mit.edu with SET PICList DIGEST in the body

Of course in that case, why not just mill the board directly. For small prototypes this would almost surely be quicker. To me, the big attraction to something that would print right on a board is the fact that I could do artwork on my computer, print it onto the board, and etch. If the printer was made/modified cheap enough, I think there would be a huge response. I'd love to have a machine that milled the boards directly, as that would cut out the etching phase, and the associated chemical problems. I just don't think I could ever afford/make a milling machine. Josh -- A common mistake that people make when trying to design something completely foolproof is to underestimate the ingenuity of complete fools. -Douglas Adams cdb wrote: > Another option might be to buy a cheap hobbyist CNC machine and > change the routing tools to Draughtmans Ink pens. -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email RemoveMElistservKILLspammitvma.mit.edu with SET PICList DIGEST in the body

I'm attempting to make one now. The only problem I see is the milling bit would have to be small, probably 0.3mm. I plan to make my milling machine so you can change from a milling bit to a resist pen or even a larger milling bit for engraving metal. The plotter is constructed entirely from brass and stainless steel for strength, at the moment I'm waiting on some toothed pulleys from RS so I can complete the mechanics. But like you say, I plan on sending the drawing from my PC to the plotter which will give me an almost completed prototype. Justin {Original Message removed}

I was thinking of this too. I definetly think there would be some sort of heat tolerant issues. I also wonder about the avilability of suitably substrate. Etching all the copper off a blank PCB just to end up with a copper free board to print on seems like a waste :) Josh -- A common mistake that people make when trying to design something completely foolproof is to underestimate the ingenuity of complete fools. -Douglas Adams SM Ling wrote: > Would it be possible to use conductive ink on a thin sheet of insulator? > > I only did a 5 minute search, there might be such an "ink", but then there > are adhersion and heat-tolerant issues. Would have to dig deeper when time > allows. -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email spamBeGonelistservSTOPspamEraseMEmitvma.mit.edu with SET PICList DIGEST in the body

Are you using a custom written driver to interface directly to like AutoCAD or something, or are you using one of the available packages out there that translate a standard file into the proper control signals for your machine? Josh -- A common mistake that people make when trying to design something completely foolproof is to underestimate the ingenuity of complete fools. -Douglas Adams Justin Grimm wrote: > pulleys from RS so I can complete the mechanics. But like you say, I plan on > sending the drawing from my PC to the plotter which will give me an almost > completed prototype. -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email KILLspamlistservspamBeGonemitvma.mit.edu with SET PICList DIGEST in the body

There are machines that do directly router a board. They don't remove all the copper, they just cut it free from the surrounding material. What you end up with is a board that from a distance looks like unetched but up close you can see the fine lines of missing copper that makes trace 'islands'. These machines are great for prototyping and one offs, and could probably be built by a home user from a solid flat bed plotter and a Dremel (r) cableflex tool. The routing software would be a pain though... Robert Josh Koffman wrote: {Quote hidden}> > I was thinking of this too. I definetly think there would be some sort > of heat tolerant issues. I also wonder about the avilability of suitably > substrate. Etching all the copper off a blank PCB just to end up with a > copper free board to print on seems like a waste :) > > Josh > -- > A common mistake that people make when trying to design something > completely foolproof is to underestimate the ingenuity of complete > fools. > -Douglas Adams > > SM Ling wrote: > > Would it be possible to use conductive ink on a thin sheet of insulator? > > > > I only did a 5 minute search, there might be such an "ink", but then there > > are adhersion and heat-tolerant issues. Would have to dig deeper when time > > allows. -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email EraseMElistservEraseMEmitvma.mit.edu with SET PICList DIGEST in the body

I've seen a more expensive machine that mills the boards to look just like they'd been etched... except that the traces are a little higher (thicker vertically) when it's done... and the traces look a little "squared off" is the best way I can describe it. I have a little board that was made on this machine that a friend made up that uses an 8-pin PIC one-shot with A/D to measure automotive voltage and shut down my ham radio gear if the battery falls too low. It also has a five minute "turn off the radios" timer that kills the power to the radios via a hefty 30A relay after I get out of the vehicle, and it monitors an accessory line from the key-switch in the Jeep to put it in "turn them back on" mode. It's kinda nice... I just get out of the vehicle knowing that the little PIC board will turn everything off a few minutes later... and it waits a little bit so you can get that last "talk to you later" in when you arrive at your destination. And if you're sitting there in the vehicle waiting for someone and want to listen to the radio, you just turn the key-switch back on for 3 seconds and then off, and it "latches" on... but if you stay too long like that and run the battery down, it kills everything... giving you a large hint that you ought to start the vehicle and charge things back up again. (GRIN) (One of the guys here found his alternator failed on the highway, not by any indication from the vehicle itself, but when all his radios dutifully turned off one night driving down the road... heh heh...) Useful little gadget. (By the way, he wrote the code in PICBasic Pro, and no, I don't have it... but it'd be pretty easy to do again...) The machine made nice boards, but darn expensive machine I think... not exactly hobbyist stuff. (The person who gave me the board said the machine was owned by a local electronics college at the time he had access to use it, so they can't be super-expensive, but I'm sure they're not cheap...) Nate On Wed, 2002-11-06 at 21:44, Robert Rolf wrote: {Quote hidden}> There are machines that do directly router a board. They don't remove > all the copper, they just cut it free from the surrounding material. > What you end up with is a board that from a distance looks like unetched > but up close you can see the fine lines of missing copper that makes > trace 'islands'. > > These machines are great for prototyping and one offs, and could probably > be built by a home user from a solid flat bed plotter and a Dremel (r) > cableflex tool. The routing software would be a pain though... > > Robert -- http://www.piclist.com hint: To leave the PICList @spam@piclist-unsubscribe-request@spam@spam_OUTmitvma.mit.edu

>There's a German firm who have a kit about 700 Euros, for a >record player platter style 3 head machine - put together bit >like an Airfix kit -now that dates me, I reckon the drill bits >could be replaced by pens. That sounds rather like the machine that was described in a series of articles in Elektor magazine. IIRC it was fairly complex mechanically with epicyclical gearboxes and required a PC for position computation. -- http://www.piclist.com hint: To leave the PICList .....piclist-unsubscribe-requestspam_OUTmitvma.mit.edu

True, but (at least for me) that presents a rather large construction challenge. Plus, much of the work should be precision so that you end up with repeatable accurate results. This basicaly ends up scaring me away from trying to construct such a beast. Josh -- A common mistake that people make when trying to design something completely foolproof is to underestimate the ingenuity of complete fools. -Douglas Adams "Alan B. Pearce" wrote: > The trick would be to start with the mechanics of a small mill that > companies like Proxxon (and possibly Dremel) make to go with their hand tool > range. Then all you would need to do is provide motors to drive each axis. -- http://www.piclist.com hint: To leave the PICList RemoveMEpiclist-unsubscribe-requestspamBeGonemitvma.mit.edu

>True, but (at least for me) that presents a rather large construction >challenge. Plus, much of the work should be precision so that you end up >with repeatable accurate results. This basicaly ends up scaring me away >from trying to construct such a beast. Well that is why I suggest using a small commercial mill assembly, as the close tolerance work is already done. It should not be a big deal to attach motors to the knobs that drive the X/Y axis, and get repeatable accuracy. Something less on the Z axis to lower the motor against a preset stop to get the cutter depth correct should be easily done with a stepper motor. -- http://www.piclist.com hint: To leave the PICList spamBeGonepiclist-unsubscribe-request@spam@spam_OUTmitvma.mit.edu

Josh Koffman wrote: > > I was thinking of this too. I definetly think there would be some sort > of heat tolerant issues. I also wonder about the avilability of suitably > substrate. Etching all the copper off a blank PCB just to end up with a > copper free board to print on seems like a waste :) > > Josh > > SM Ling wrote: > > Would it be possible to use conductive ink on a thin sheet of insulator? > > > > I only did a 5 minute search, there might be such an "ink", but then there > > are adhersion and heat-tolerant issues. Would have to dig deeper when time > > allows. Sorry to restart this discussion. I've been off-net for a few months (new job, moved from VA to CA, etc...) and I'm just now getting to the 12,000+ PICList messages that have piled up. A few or 6 months ago I came across a research paper talking about making micro 3D structures using modified inkjet printers. One of the things mentioned in the paper is the fact that standard inkjet ink is electrically conductive. Naturally I whipped out a multimeter and applied it to the printout of the paper, and sure enough the printed lines were conductive. However, I think the resistance was about 300K per letter (vertically, 10-12pt font). A few brief experiments revealed that the resistance would drop considerably if the same image was printed repeatedly (more layers), I'm guessing it has to do with the increased x-sectional area of the "trace". After playing around a bit more I decided to finish reading the article, which went on to mention that there are special inks readily available that deposit metal particles (silver?) after drying, which greatly increases the conductivity. Unfortunately I can't remember exact numbers or where I saw the article, but I'm sure I found it with google. I'm thinking such inks could be used to print traces and pads directly onto veroboard, with or without predrilled holes. That would solve the substrate heat problem, but I have no idea if the resulting pads will hold up to soldering. Anyone know of a conductive glue that could be used? Low temp solder paste? If would be really nice if you could print conductive traces in different colors. Then the board would look like the layout on the screen. -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

> If would be really nice if you could print conductive traces in > different colors. Then the board would look like the layout on the > screen. Even better, if you are printing the conducting layers on- you could coat the printed board with a layer of non-conducting lacquer/varnish and print another layer of 'track' over the top of that. As big a multilayer board as you wanted, and no vias needed, just gaps in the insulating layer where you want conducting layers to contact... it would be murderous to debug though! Nigel -- Nigel Orr, Design Engineer @spam@nigelRemoveMEEraseMEaxoninstruments.co.uk Axon Instruments Ltd., Wardes Road,Inverurie,Aberdeenshire,UK,AB51 3TT Tel:+44 1467 622332 Fax:+44 1467 625235 http://www.axoninstruments.co.uk -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

Nigel Orr wrote: > > > If would be really nice if you could print conductive traces in > > different colors. Then the board would look like the layout on the > > screen. > > Even better, if you are printing the conducting layers on- you could coat > the printed board with a layer of non-conducting lacquer/varnish and print > another layer of 'track' over the top of that. > > As big a multilayer board as you wanted, and no vias needed, just gaps in > the insulating layer where you want conducting layers to contact... it > would be murderous to debug though! I like the sound of that. -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads