Hi everyone. I have a design for an analog vocoder that I've modified to suit my needs and would like to make a prototype of. However, this will be by far the largest prototyping project that I've undertaken, and I'm interested in hearing tips on how best to systematically go about it. I plan on breaking down each channel of the vocoder onto separate prototype boards, but even so each board will still need to have several op amp ICs and dozens of resistors and capacitors. I'm concerned that trying to prototype this on perfboard is going to turn into an impossible tangled mess, and would like some pointers on how to arrange things in a systematic manner. Is there any specific type of protoboard that you guys would recommend that might make this process easier? Is there a general rule on the size of board to use vis a vis the number of components involved? Any advice on what kind of per-board decoupling arrangement would be sensible would be appreciated as well.
If you're only ever going to make the one prototype then go ahead and use perfboard, but if you intend to repeat the exercise then teach yourself to make photoetched pcbs. I made perfboard prototypes for years before I bit the bullet then kicked myself for not learning to make pcbs earlier. There are plenty of good cheap or free pcb design programs about so invest the time in learning how to use one of them. You'll get the investment back tenfold in time saved making prototypes. The chemicals needed for developing and etching pcbs are cheap and readily available. You'll need a UV light box for exposing your artwork. Commercial ones are a bit pricey but you can easily make your own.
I peronally would never buid anything on the perfboard that has more than 4 parts 
No, seriously, it's a big pain on you back.
I have been building prototypes by photoetchingfor different years by now.
It is much, much better.
Nevertheless your first boards won't come out as nice as you think.
At first you need to build a bromograph that works nicely; second but not less important, you need to find the kindof paper gives you best contrast and most important the right printer.
You also need to find good presensitized board , I think you will need two sides board.
When you have done these things then you need to find the exposure time the developing time and the etching time.
It takes several attempts to do so.
At firt it can seem to be descouraging, but then you'll see that it is really cool to be able to make your own board.
Of course, you need a cad to design your layout.
For the decoupling scheme, it all depends on your design and layout.
How far is the power supply, and the ground strategy...and so on.
It is not so immediate to throw a decoupling scheme.
You need to build your prototypes and check your own layout.
Look for oscillations or spikes.
All in all, if you need specific tips on type of papers and so on...you can write here or you can also drop me an email.
Hope this helps.
No, seriously, it's a big pain on you back.
I have been building prototypes by photoetchingfor different years by now.
It is much, much better.
Nevertheless your first boards won't come out as nice as you think.
At first you need to build a bromograph that works nicely; second but not less important, you need to find the kindof paper gives you best contrast and most important the right printer.
You also need to find good presensitized board , I think you will need two sides board.
When you have done these things then you need to find the exposure time the developing time and the etching time.
It takes several attempts to do so.
At firt it can seem to be descouraging, but then you'll see that it is really cool to be able to make your own board.
Of course, you need a cad to design your layout.
For the decoupling scheme, it all depends on your design and layout.
How far is the power supply, and the ground strategy...and so on.
It is not so immediate to throw a decoupling scheme.
You need to build your prototypes and check your own layout.
Look for oscillations or spikes.
All in all, if you need specific tips on type of papers and so on...you can write here or you can also drop me an email.
Hope this helps.
I get mine done in China for a third of the price.
It can take a while for them to travel half way around teh world but worth the wait.
Got my last batch from China for the price of tooling in the UK !
I also agree with the above comments, althoug, when you have to make a simple prototype and you want to do it within few hours, you can do it your own if you know how to do it.
The only thing is...if clearance is very tight, you have a lot of parts, if you need vias...then...it's best to have the board professionally done.
Cost depend much upon dialer.
There are intereszting deals out there.
I know a lot of people on the forum make their own PCB by photoetching...it's up to you.
I am willing to help if you need any tip as...if there is somebody ready to help you out...you will learn much sooner.....
back when I first started, you could barely ask this questions on the forum without being hang but time has passed....and things have changed....and lately I see people on this forum being very kind.....so... that's good!!
just let me know in case you need anything!
The only thing is...if clearance is very tight, you have a lot of parts, if you need vias...then...it's best to have the board professionally done.
Cost depend much upon dialer.
There are intereszting deals out there.
I know a lot of people on the forum make their own PCB by photoetching...it's up to you.
I am willing to help if you need any tip as...if there is somebody ready to help you out...you will learn much sooner.....
back when I first started, you could barely ask this questions on the forum without being hang
but time has passed....and things have changed....and lately I see people on this forum being very kind.....so... that's good!!just let me know in case you need anything!
I used one of these for rapid prototyping. Very handy, but a bit pricey.
http://www.lpkf.com/_mediafiles/1076.pdf
http://www.lpkf.com/_mediafiles/1076.pdf
For prototyping, perf board is **, for all the reasons you have stated.
Doing a PCB for a prototype is hugely time consuming, and if you make the normal errors, you have to cut and retrace anyway.
I use terminal board for protypes, the one about 1"(2 1/2 cm ) wide with a row of terminals down each side. Works perfectly with tubes and regular wired components. Transistors fit perfectly into 3 adjacent terminals, DIP opamps etc get their leads bent out a bit and also fit. And as everything is, (or should be) on the same side (the top) you can see all the wires/parts and logically trace/follow things and not get lost like one does on perf board. If you haven't figured it yet - I HATE using perfboard!
You need to draw it out before you start building to optimise placement, but it's a doddle if you want to change something - and it's BY FAR the quickest way to RELIABLY prototype anything that's not SMD sized. We have made & sold very small runs of quite complex products using this terminal board procedure, only going to a pcb if making more than (say) 5 off and then only if expecting to make even more.
Regards, Allen (Vacuum State)
Doing a PCB for a prototype is hugely time consuming, and if you make the normal errors, you have to cut and retrace anyway.
I use terminal board for protypes, the one about 1"(2 1/2 cm ) wide with a row of terminals down each side. Works perfectly with tubes and regular wired components. Transistors fit perfectly into 3 adjacent terminals, DIP opamps etc get their leads bent out a bit and also fit. And as everything is, (or should be) on the same side (the top) you can see all the wires/parts and logically trace/follow things and not get lost like one does on perf board. If you haven't figured it yet - I HATE using perfboard!
You need to draw it out before you start building to optimise placement, but it's a doddle if you want to change something - and it's BY FAR the quickest way to RELIABLY prototype anything that's not SMD sized. We have made & sold very small runs of quite complex products using this terminal board procedure, only going to a pcb if making more than (say) 5 off and then only if expecting to make even more.
Regards, Allen (Vacuum State)
For something as repetitive as a vocoder, it seems worthwhile to make a PC board. Once you've laid one out, you can array it and get several channels on a single board.
Toner transfer is tricky; if you can get it to work, great. I suspect that technique is as critical as materials, since some people report better luck using regular paper, or Staples photo paper, than the expensive stuff sold for the purpose.
Photosensitive stock is more consistent and controllable. I've had good results from penplotting onto frosted mylar, then using two copies overlaid to increase contrast. Inkjet onto "overhead transparency" film and doubled-up also works. And inkjet printers are a bit more available than pen plotters, and work with scanned artwork, which makes using magazine layouts easy.
Toner transfer is tricky; if you can get it to work, great. I suspect that technique is as critical as materials, since some people report better luck using regular paper, or Staples photo paper, than the expensive stuff sold for the purpose.
Photosensitive stock is more consistent and controllable. I've had good results from penplotting onto frosted mylar, then using two copies overlaid to increase contrast. Inkjet onto "overhead transparency" film and doubled-up also works. And inkjet printers are a bit more available than pen plotters, and work with scanned artwork, which makes using magazine layouts easy.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.