Which is the best hobby do-it-yourself method for making PCB's?

Thats pretty impressive ... I thought SMT were off use for hobby work without some really expensive oven/manufacturing equiptment.

Again, NICE work

Heres my second to last attempt ... its been redesigned since then and will be placed on a .062 FR4 copper board.

With vias I just drill holes ... place a small wire on one end, solder, solder other end ... snip off the remainder

I write low level code for a living ... so this is just hobby work

"samIam" schreef in bericht news:0pngf.27744$ snipped-for-privacy@fe12.lga...

IMHO there is not such a method as the "best" hobby DIY. Both method and result depends highly on the experience and tools available. I ever started using nail polish and nitric acid. Did several types of photografic methods and also tried out toner transfer. The latter worked well as long as I had access to an HP Laserjet 5000. This one put enough toner on the sheet to work as an etch resist. Other laserprinters I tried, including the Laserjet

4Si, did not. Now I'm back at the photo method using two overlayed transparancies for the positive. The best site on the net I'm aware of is: (Almost) all you need to know about etching PCBs at home.

petrus bitbyter

I have found that using pre-clad boards with positive resist work quite well. With the spray on positive resist, it is difficult to get a nice even coating without any bubbles and dust.

have very good quality pre-sensitised PCBs. (Both single and double sided with a long shelve life) Print the layout on a transparency using an inkjet printer. Clamp the transparency against the PCB (photo resist) side with a piece of transparent glass. Expose in sunlight for about 45 seconds. (Experiment a bit). Develop PCB, and etch. I have had very good results with even 10mil tracks and 10 mil spacing. When doing double sided, one must make very sure of the alignment of the 2 sides.

Regards Anton Erasmus

One thing to note when making your own PCB's.. be careful not to mix chemicals.. some cleaners can react with the chemicals used to produce some rather hasty gasses... A friend of mine spent two nights in hospital doing this.

I think the best bet is to use a proto house.. if they can do a panel it can be cheap... for me it was AU $270 for 3 x A4 worth of PCB's .. as much as you can fit on a single A4 sheet and 3 of them. No fuss, no mess, no chemicals or EPA requirements.

Simon

Nice ! That was one of my goal, a xilinx XC95xxx, now I know it's possible to do :-)

For some reason that one didn't work so well. Some kind of scum left on the outside of a few rows of the pins, and I didn't restrict the copper pour underneath, so there's an entire row of pins all bridged to each other under the chip. I'll be carefully re-doing that one, as well as tweaking the Eagle printout by 1/128 of an inch (0.8%) to correct for the undersized artwork. The row of 22 pins causes the board to curve quite badly, so I had to break them up and file off a gap between groups of pins for this particular board. Besides, there are several broken traces ;-(

Yeah, it's interesting trying to get the right amount on the SSOP/QFP packages. I have what's probably too big a needle to dispense, so I've been using a dental pick to shove the paste around and try to get a consistent amount.

I haven't seen any chips move, but it would surprise me if they didn't move a little. Then again you can see that the CPLD didn't center itself very well. I've definitely seen 1206 parts shuffle around a little, probably as the solder melts from one side to the other.

Yeah, that's where the cleanup comes in. Small-tip temp-controlled iron, very small solder wick, dental pick, and an acoustic continuity checker with stickpins for probes. Oh, and lots of patience.

No, something went wrong in the transfer process and I foolishly figured it wasn't a big enough problem. Issues like that seem to stem from a few causes:

- not a consistent enough scratching up of the board before laminating

- fingerprints etc. at various stages

- bits of dust on the board between the toner and the copper

- impatient removal of the backing paper from the toner while soaking

You righ I didn't check this picture in details . To bad for this board, yours others ones were good !

Here some questions form a noob side ;-) (who wish to learn stuff before making to much errors ) I guess you used to much paste for the back. Is it true that when you oven board chips place themself correctly over traces ? Might not work I guess if to much paste is used and if it flood few traces togedther...

why do you have broken traces, is it because of heating in the oven ? Expension and contraction of copper traces?

A smily noob :-)

I've only done this once, a thousand years or so ago, but I made my originals on plastic with tape and a little bit of india ink. They were 2X. I took them to a photo shop and got positives, six-up, registered for double-sided. I stapled the negatives together, got a piece of photosensitive board at Radio Shack, slid the board between the negatives, and exposed it to the sun for about a half- hour on each side. I don't remember what the next step was - developing the resist and washing it, or some such, but then I etched it in a cake pan with about an inch (2.5 cm) of Ferric Chloride solution (also from RS - in their etch kit -- Hey! I said it was a long time ago!), and did get a little bit of undercutting, and drilling it right was very, very interesting, but it did work. :-) (the product died on the vine, however. )-; )

Good Luck! Rich

Thanks :-)

Another question,

Are you putting something on you board after it is finished, vanish of some kind ?

:-)

Erik Walth>

Before you laser print anything definite do a test print! My experience with (cheapish, I admit, Lexmark & Minolta) laser printers is that they do not print 1:1 in both directions whereas my inkjet does.

--DF

Yup. I found that the laser I'm using (NEC Superscript 1400, retired from office use due to feed problems) is 0.8% off horizontal, and I haven't calibrated it for vertical yet. Makes a row of 20 pins across not quite fit, though you can cheat by snapping them in a few places and filing down period gaps.

Nope, just the Liquid Tin. Not sure I want to go the route of putting paint on the board that I'll just have to smell as it burns off later... These are all prototyping boards anyway, I'll be using a board house to do later consolidated boards.

[snip]

Something worth trying as alternative to photoresist pens that leave blobs like this: Staedtler Lumocolor. Nice to touch up the bits where the positive resist or the toner transfer didn't stick. In my experience the black one is more resistant to FeCl than the (somewhat) ancient Dalo photoresist pen I have here (with the blob habit.) The red one is apparently even better. They are available in 4 tip sizes too. (Google is good) I've tried some other overhead marker/permanent marker type pens that didn't work well at all.

When drilling/cutting the board, use a respirator, like when spraypainting. Saves you from waking up with an evil splitting head the next day. One more reason to go with SMT as far as possible. Less drilling.

L

Hello Sam,

But please tell them honestly what it is that you bring. Else it can cause accidents and contamination.

Regards, Joerg

If you are working with postscript, it's reasonably easy to put a wrapper around your print file that fixes up the scale. I'll fish out the hack I've used if anybody needs it.

--
The suespammers.org mail server is located in California.  So are all my
other mailboxes.  Please do not send unsolicited bulk e-mail or unsolicited
commercial e-mail to my suespammers.org address or any of my other addresses.
These are my opinions, not necessarily my employer\'s.  I hate spam.

Early on, I was written up for dispensing chemicals into smaller containers and not labeling them. Since then, I've become a bit of a fanatic about labels.

I like to get an almost-used-up roll of the 2" transparent tape they use in the shipping dept and put that over the label so solvents, bleaching agents, or stains won't have such an easy time making them unreadable.

Lumocolor (I assume you mean the permanent type not the dry erase) are sometimes sold as "cdrom pens". OHP pens are hard to find, there's not much market for them any more.

Bye. Jasen

Eagle has the X and Y scale in the print dialog, I just have to do some larger-scale test prints and measurements. Wonder if I can print out a reference rectangle bigger than the 100x80mm limit of the free version, as long as it's e.g. the tDocu layer and not top or bottom...