Chuck Guzis wrote: Registration becomes the issue. In my experience, it's just not worth it, nowadays, unless you're making a HiFi amplifier or something equally "non-dense". Saving a couple of dollars on a bare board just doesn't justify the cost (in time) E.g., why not make your own memory SIMM/DIMMs too? :> Years ago we had our own etch tank, wave, etc. and *still* ended up pushing board fab to outside vendors (messy process).

Roger Merchberger wrote: Top to bottom. Sure! But, if you aren't prepared to take up MAKING BOARDS AS A HOBBY (i.e. if you're just going to try it once or twice), you'll probably be dissatisfied with the results. And, if you take THAT up as a hobby, you may end up being discouraged with how it forces you to approach projects (e.g., having to avoid using certain packaging technologies). I know that when a client "forces" ("strongly encourages"?) me to use only certain technologies (e.g., for a board that will have to be repaired in the field by folks who may not be particularly skilled with a soldering iron :> ), it has a noticeable impact on the quality of the design that I can produce. Certain parts just aren't available in thruhole technologies. Certain other parts require (very) expensive sockets (e.g., don't even try to do a BGA design!). Board size increases (this can also have mechanical implications -- e.g., a tiny board rarely needs a stiffener or other support). Etc. As I said, if you are making a HiFi amplifier or something with reasonably low density parts (e.g., 1 DIP per sq in) it's "doable". But, if you are using this as a *tool* so that you can make things that you want to *use* (i.e. the "hobby" is the design and/or USE of the item and not it's FABRICATION), then it just seems like a big timesink for very little return. It's not even that... it just doesn't make sense to build things like these (commodity parts) unless someone GIVES you a small sh*tload of *chips* that you could otherwise not use. I went this route building 4MB memory "modules" for an old Compaq 386 I had 15 years ago (compaq was charging $150/MB). It just wasn't worth the effort in the long run :< The etch tank was for prototypes. For production, it didn't make sense to do boards in house (the chemicals involved are really nasty!). But, we found that the time required to make even small lots of prototype boards didn't justify the hassle. If you left the board in the etch tank too long, you ended up with gaps in traces. If you didn't leave it in long enough, you ended up with solder bridges between them. We even tried using metallic *tape* for some designs (a 2KV, 2KW switcher) and found it didn't save you anything. Nowadays, with 3 day boards (order *one*... they'll probably make 3 - 6 and GIVE you the others! :> ) it's easier to get the design out of the way early and spend those few days getting together the components you'll need, etc. If you want to do it to say you've *done* it, by all means, try it! Just don't think of it as a way to save any real time/money. Better to spend your time finding others that want a similar design and go in together on a "lot". --don

On 7/31/2006 at 2:11 PM Don Y wrote: ...one benefit that's not been mentioned is the advantage of having an outside-produced board returned to you tinned with solder mask and plated-through holes (and possibly silkscreen). I couldn't imagine what it takes to do a one-off of one as a "hobby".

Brad Parker wrote: Sure. "Therapeutic" to some degree. If you want "bragging rights" (i.e. to say you've done it), then rational arguments are useless. ;-) But, it's sort of like mowing the lawn when you are a kid; the first time is *cool*. After that, you quickly understand why they call them CHORES! :> I'd just rather spend my time doing something *creative* instead of mundane! :-/ I've got *really* nice wirewrap tools -- sitting in a box in the garage! :>