Page 1 of 1
New Transistor Tester design
Posted: Wed Jan 23, 2013 6:16 pm
I blew up my LCD on my old transistor tester, as I didn't use a standard 16 pin header and plugged it in wrong. So while I was waiting for my new LCD's to come I designed this
I added an ISP Header, the auto power off function, and hardware that will hopefully allow it to test inductance, once I figure out how to modify the code. I don't know if it will work, we'll see. I'll post the files later, still need some cleaning up. Firmware will be much later, as currently it's the same as the old post here.
http://translate.google.com/translate?s ... stortester
I seem to be having trouble adding more than 1 picture, probably my connection.
Re: New Transistor Tester design
Posted: Wed Jan 23, 2013 9:42 pm
Very cool! I need to get figured out why my toner keeps separating from my boards.no sand them, clean them with paint thinner and let them dry, apply the paper and laminate it until the cows come home. Then it peels off halfway through my etch.
I'm going to try different paper, then maybe try a different printer, or not using the paint thinner or something. It always used to work just fine.
I'll definetly want to build one of these once I get that working good again.
You set the board in then use the aquarium bubbler to agitate right?
Re: New Transistor Tester design
Posted: Thu Jan 24, 2013 12:34 am
Yea, but I try to keep the bubbles from directly hitting the board, it seems to increase undercut (at least in the cold). Down side is longer etch times. I was reading somewhere that some printers use one type of toner and others use another (wax and plastic I think). One works well and the other not so good.
Try something other than paint thinner. It's probably stronger than you need and different brands contain different chemicals, could be it's leaving some sort of residue behind.
I was doing .012 traces fairly reliably, but had trouble with my last board. I upped them to .020 and dropped to .012 or .008 to sneak between pins as needed.
I've been trying to use the white parts of the paper, but on the transistor tester I had a fit (ruined 3 boards). It was cold (<40F) so I think that was at least part of it. I ended up using a very colorful and noticeably shinier McDonalds ad. I had significantly less pitting of the copper.
Brief rundown of my current etch process:
1. print alignment sheet (pcb on blank paper)
2. cut PCB to size (roughly)
3. dry sand PCB with 400 wet or dry (I have a lot of it)
4. wet sand PCB with 400 wet or dry (same piece) with a drop of soap on board (basically washing the board, but may work soap in better, IDK)
5. rinse board and dry with paper towel
6. pour a few drops of rubbing alcohol on PCB and wipe around good. allow to dry (I know people have trouble with 70%, but it seems to work for me. I need to pick some 90% up).
7. start laminator warming up, print PCB (I'm using newspaper adverts with especially glossy pages, but planning to try wax / parchment paper)
8. carefully align and run through laminator 5-6 times (I usually fold the sides over to help keep everything lined up.
9. drop in bowl of cold water, allow paper to saturate
10. carefully peel paper off. Usually it peels for me but sometimes sticks, if it peels I lightly rub the board until it's gone (some residue remains on the mask, copper should be clean) if it sticks I have to rub harder but try to avoid letting it ball up under my fingers.
11. carefully inspect board, touch up any nicks. I currently use a sharpe but may try a fountain pen at some point (finer tip, variety of inks)
12. place board in holder and put in etch tank. (I've put it directly in the bubbles and it's quicker, but if you have trouble with undercut I'd keep it out of them.)
13. remove from etch tank, rinse thoroughly, sand with 400 grit (wet) to take the toner off, inspect for bad traces.
14. drill a hole on each end and align silk screen, laminate, soak, etc.
15. drill holes, assemble, test
16. clean flux residue from board, spray copper side of board with clear lacquer to protect from oxidation (haven't actually done this yet, but seems like a good idea)