Hi all, preparing and leading up to a modding project in the new year, I was making the PCB's for part of the case's features (cant give too much away, hehe). Heres my experience in short version... I mirrored the pcb track image without thinking it through properly, I mirrored it thinking that when i flipped the transfer paper over, it would be the right way round and all would be ok, only once it was all finished and I had the finished PCB in my hand was it that I realised that i would mount components on the bottom of the board (not the copper side) and if I did that the inverted copper traces would then be the wrong way round again... So yeh, in my absent mindedness, I inverted an image that never needed to be inverted because I would be mounting the components from the inverse side to the copper, so in effect, the correct image of the copper traces would be inversed anyway... grrr @ me What a waste of 3 hours On the plus side, the board was remarkably well done, all traces were nice and sharp and there was no edge bleeding at all, it was a perfect board, just backwards... For the record, this was my first ever attempt at making a PCB, and I failed dragon2309
I was just trying to think this through and nearly got a headache, but anyway, if your image is "merely" mirrored one way that is un-needed, can you not assemble the components onto the track side of the board? Thought stream: You design the board looking 'top down' at the component side, and it was ok (suppose you have an IC with pin1 on the top-left of it's rectangle), you then mirrored it unnecessarily so pin1 is now on the right, then transferred it to the board, so pin 1 is still on the right. if you now put the component in, the pins are all in the wrong place, but if you turn the component on it's back and put it through from the other side of the board, pin1 will now go through the correct hole. yeah here comes that headache, spacial visualisation isn't my thing hehe
haha, yeh, it confused the hell out of me too, at least 30 mins sitting at ym computer chair, PCB in one hadn, the schematic drawing next to my keyboard, the generated 3D on the left screen, the gerber file on the right screen... It's a pickle I know, it would be fine if i didnt mirror it to begin with lol.... And yes, you are right, the only way this board I have now will work is if i mount the components on the copper side, which would be quite hard... hmm, seeing as the baord is useless to me, im gonna try that, nothing else better to do, how hard can it be, lol, here we go dragon
i once screwed up 2 times making the same board the same day. i was making my 2nd ? 3rd board i think the first time i made it ,i completed everything (etching->drilling) and then at the end when i went to assemble components i noticed it was inverted. second time the ***ing guy at the printing shop photocopied the traces into a transparency inverted. trusting the guy so much i didnt bother to check and it ended just like the first one discovered it was ****ed up in the end, ofcourse i never and will never ever go there again !!. but at the end it came out great as well as many more PCBs, the rule that you should remember is to check and double check everything and to have a plan before you even print. it will make sure that you dont do stupid mistakes like i did. good luck
thanks const_ . I just tried to solder components onto the copper side of the board, its quite hard and very prone to solder spillage over tracks (they are uber-small) so i think its a case of writing this board off and starting again when I get another day off work (probably next wednesday or something) Need to brand it into my head DO NOT INVERT THE IMAGE WHEN PRINTING
Hiya jake, long time no speak, yup, it is the baord you made for me, it did take me a little while to figure out what you were doing with the grounding to the baord... It took at least 5 mins before i realised that you just used the blank space as one massive ground... quite a smart use of limited space. I didnt take any pics of it in progress, as it was my first time, wanted to concentrate, but when i redo it I will take progress shots, the only pic i can get for you now is of the board after I've been playing around with it mounting components on it with no holes and on the wrong side, lol... Ah well, here goes... It should be about 5 days until my next day off when I can redo this board, oh and jake, am I right in thinking that it would be 1mm holes? or is it 0.8mm? dragon2309
Invert the side components are on, leave the side components arent on normal, assuming youre doing a through hole board, with the traces on the side opposite of the components, and you designed it component side up. I like to use mainly SMD components on my board, so top layer gets inverted, bottom layer, if present doesnt.
all of the holes are 35 mils (0.889 mm) excepet the holes for the LM7805 which are bigger at 46.67 mils (1.185 mm, close to a number 56 drill) i'd drill them small at first, test the part to see if it fits, then if necessary, go back and make it bigger. heres a handy drill chart with the many different drill type conversions: http://bobmay.astronomy.net/misc/drillchart.htm edit: looking at your pics, some of those traces and pads look small (looks like you lifted some of the pads up). i can go back and beef them up if you want me to. let me know.
I did notice that, which is why i was a bit worried about the hole sizes, driling too much might get rid of the pad entirely... If its not too much hassle then yes if you could plump them up a bit that would be fantastic. On that board there, looking at it, there was two shorts that i found and cut with my knife, so all in all I'd say it was pretty good. Don't put yourself out or anything, if you're busy then dont worry becuase it works fine the way it is. Only if you've REALLY got nothing else to do... And yeh, the pads lifted when i was desldering my two chips (the PIC16F628 and the 74HCT164N) that I was messing about with... hopefully wouldn't happen with the real deal. Thanks jake, Dave
i'd transfer the ink onto the board, and then before you use the chemicals to remove the copper, i'd drill all the holes. that way you won't lift the pads later. ya, if i find some time this weekend, i'll do it. when do you plan on trying to do it again (so i know how much time i have and if i can get it done)?
This is why you should always put text on the chopper side of the board the first times. Also, stop using sandpaper on the chopper! use chemicals to remove it. It's easier, and better. Oh, and don't drill holes before etching. If you so the etching chemicals will start to eat into the chopper from the drilled hole (exposed chopper). The best way to make holes is to etch tiny holes in the center of the pads, and then drill afterwards with sharp drill bits. (this way the chopper will also help keep the drill perfectly centered on the pad.
It wasn't sand paper, and I don't have a whole array of chemicals at home to play with, that was soft wire wool and it came off in about 30seconds...
No matter what you used it cuts into the traces. (clearly visible on the pictures) You don't need any more chemicals than you already have. I assume you use UV developing for the traces. When you are done etching, just put the entire board under UV light (without mask) and dip it into the developing fluid. It might take 1,5 minutes longer, but it's better for the PCB, and less messy. (since you already have everything handy from the initial developing)
Ok, i see what you mean, i actually used the laser toner transfer method for etching though. Oh and jake, I have a day off on wednesday next week, so I will probably have a bash at it then, unfortunately I have to use my works laser printer (I dont have one) so it would be tuesday that I would be printing onto transfer paper. Cheers mate dragon2309
