I finnished my etching this weekend, and it was a mess, and with a useless result.PokeMon wrote: ↑Thu May 11, 2017 11:19 pmIt depends on how much money you want to invest in your projects or hobby. This is a value of a small evening dinner only. And you have to buy copper boards and chemicals as well and maybe need a few attempts to get a suitable result. So it is simply a matter of choice I would say.
Maybe I'll try it again if I need a very small pcb, but for boards larger than 4 x 4 cm I'll order them looking forward.
ZX81, Lambda 8300, Commodore 64, Mac G4 Cube
This is a schematic used to drive the Xilinx tools, it's not for PCB design. ISE can be driven using text files in a number of hardware description languages or by using a more literal description of the circuit schematic as it appears to be in this case.Tezz wrote: ↑Mon May 15, 2017 10:56 amI thought I'd start learning to use Eagle which I've just installed. RetroTechie posted the zip file "ZX81_ULA-in-a-CPLD_v1.10.zip" here: viewtopic.php?f=8&t=1862&start=12 which includes .sch files for each of the schematics. Are they from an earlier version of Eagle? They won't load in the latest version.
- Posts: 2897
- Joined: Mon Sep 26, 2011 10:56 am
- Location: Looking forward to summer in Somerset, UK...
I don't know if you have seen this article and the follow ups. Lots of help and advice.
And yes, it is the way of the world for your first, and maybe 2nd, 3rd attempts to fail for some reason. Back in 1985/6/7 (I think, well, one of those years anyway... ) in a college lab, me and my fellow classmates made mistakes even though we were under the supervision of a teacher...
Still occasionally dropped a clanger in later years at home (the embarrassment of an inverted footprint ).
There are a lot of variables, and it takes experience from trying it before you get to crack it. Then there is the drilling...! But it's a great feeling when you have finished a board and hold it glinting in the light
So keep going
Thanks for your encouragement and link to the article. I'll let it be for now, and try with smaller pcb.
ZX81, Lambda 8300, Commodore 64, Mac G4 Cube
So. The major difficulty for me however is that outside of the USA we don't tend to have things like Kinko's or other mainstream printing bureaux. The UV method also involves an initial plus ongoing material outlay. Developing chemicals, UV exposure rig, transparencies and the per-board cost of pre-sensitised pcb material. You need to experiment _a lot_ to find the right combinations of board type vs exposure time, developer mix vs developing time, board type vs developer mix/strength, ... I gave up in the end. It would take an entire evening to make a board. There are so many 'diy' type suggestions too - like developing with salt, etching with vinegar, etc etc. Ignore them all. Use peroxide & hydrochloric acid. Fast clean etch and it cleans your drains too.
If you're serious about making PCBs at home (I am - it's great fun!) then buy a laser printer and a laminator. I got my samsung laser in a sale for less than the replacement cartridge I bought a few weeks ago. The 'starter' cartridge that came with the printer lasted 2 years of very light use. The laminator needs some research. Don't just get the cheapest. I spent about double the price of the average ebay low end laminator price. It was worth it though, I've got a robust unit which has served me well. It's a GBC and I highly recommend the brand. So initially you'll spend more (though you can do without the laser printer - use the one at work) but you don't need to buy developer or pre-sensitised boards.
One thing that took the time for me was finding a reliable transfer medium. I started out using a shiny laser printer paper from HP which worked but it:
[*] was thick and therefore took much longer to transfer
[*] didn't separate cleanly and therefore introduced a risk of lifting traces
I experimented with lots of things, such as the 'magazine' paper but mainly these were difficult to:
[*] align, esp. for double-sided boards, because you can't see your printing
[*] handle because the ink would rub or the stuff would crinkle
I have the almost perfect medium now. It's the waxy, shiny backing paper for the laser labels I use when posting things. It's:
[*] FREE!! I use the labels anyway.
[*] plain and translucent, so you can align things really easily
[*] thin - so you don't need much heating
[*] so slick the toner transfers _completely_ with no post transfer cleanup
I say it's almost perfect ... It takes the toner beautifully but if you bend it in any appreciable way retrieving it from the printer then the toner flakes off. You have to be careful. Toner is another variable - I bought wasted money on a 'compatible' cartridge and I never will again. The official Samsung toner is BOSS. Always buy official. Anyway.
It makes gorgeous boards though. The second link there went from printer to drilled board in just under an hour.
Thats how my prototype looks like. The Programming pins will not be soldered in a future version and the Quartz needs to be SMD type, but otherwise you can get the Idea of it I hope.