PCB printer using inkjet cartridges

[jeremy]

http://www.kickstarter.com/projects/cartesianco/the-ex1-rapid-3d-printing-of-circuit-boards

What do you think? I think if it can pull off QFP (or SOIC at a stretch) then it will be worth it for me.

[chicken]

Interesting concept. I wonder what the 3rd dimension is. 

[BurtyB]

I wonder what the 3rd dimension is. 

I'd guess multiple layers.

[con-f-use]

Both material and the printer seem too expensive to me as it is.

[rthorntn]

The first thing I thought was it's expensive, then I thought it's probably better to get a $770AU 3020 chinese CNC from ebay as it would be more versatile, I love the way the EX1 cost goes up $300 at a time:

5 Super Early Bird's at $899 all gone
20 Early Birds at $1199 all gone
100 at $1499 with 92 of those left

$1499 for what they mention is a groovy inkjet...

[BBQ]

I find it exciting, but i can't justify the price $1,499 not assembled.

[carloscuev]

Yes it's very expensive. I really find no use for that, for one-time personal gadgets I think it may have reliability issues, and for prototyping a production design I would want it with actual copper and via plating capabilities.

Making PCBs at home with toner transfer or photoresist film is way better and cheaper, a little bit more tedious though.

I wonder how nobody has came with a 2 side photoresist + via plating kit carefully tweaked to ensure a 2 or 3 hour max process time for a 10cm x 10cm board.

[wraper]

Yes it's very expensive. I really find no use for that, for one-time personal gadgets I think it may have reliability issues, and for prototyping a production design I would want it with actual copper and via plating capabilities.

Making PCBs at home with toner transfer or photoresist film is way better and cheaper, a little bit more tedious though.

I wonder how nobody has came with a 2 side photoresist + via plating kit carefully tweaked to ensure a 2 or 3 hour max process time for a 10cm x 10cm board.

Via plating is not easy. Includes some hard obtainable/expensive chemicals. Need a really big bath, big copper electrodes (special electrode copper, you can't use whatever if you want good results), high current PSU, pumping liquid/shaking the board. And yet it will be to hard to get stable results. Actual PCB cost may be even higher than order from China and you waste a lot of time. So only some very rare people do it at home.

[carloscuev]

Via plating is not easy. Includes some hard obtainable/expensive chemicals. Need a really big bath, big copper electrodes (special electrode copper, you can't use whatever if you want good results), high current PSU, pumping liquid/shaking the board. And yet it will be to hard to get stable results. Actual PCB cost may be even higher than order from China and you waste a lot of time. So only some very rare people do it at home.

Well, I can tell from actually doing it that the chemicals are not hard to find, the hardest is a suitable surfactant (I've used Polyethylene Glycol which can be bought at eBay easily, is not a hazardous substance so the international shipping went like a breeze). My bath is less than 4 liters and I've succesfully done 10cm x 10cm boards. The copper electrodes are not big, in fact they are pellets inside a small titantium mesh basket (this tiny one works just great http://www.ebay.com/itm/141107835459) but the phosphorus-copper pellets are hard to find, had to contact a local art-electroplating shop to buy them. Not needed a high current PSU, I've done, again, 10cm x 10cm boards with a 1.5Amp LM317 power supply and several high power ceramic sub-10OHM resistors to limit current. In fact the current just affects the time required to finish the plating. I've made it in about 90 minutes. I use standard Indian Ink as the hole activator, works great. And yes it takes a lot of testing and attention to detail to get stable results. Total cost of materials is low but the work and time to start getting decent results is ridiculously big. You've also have to invest time creating some tools in order to save some time on each board which is a bummer.

Right now it's definitively better to plan ahead and get the PCB designed ASAP and send to a cheap chinese fab. I regret wasting a lot of time, effort and money for all the testing I did (but the final recipe was cheap) in order to be able to do this in my own room, it's just not practical.

The biggest 2 issues I've found are: Alignment of the layers and holes. With a chinese CNC the hole alignment could be worked out but there's still the layer alignment problem. The other one is the time required per board, it's a very slow process that requires constant attention so you can't have enough time to concentrate in other things while etching, plating, etc. Listening to a nice podcast ( as TAH ) or flushing your Watch Later list on YT is a nice workaround killing 2 birds with one stone haha.

But I believe that the two issues, and others that may arise, could be worked out and optimized so that we may have a worthy Kit+Procedure (with cheap consumables of course) able to guarantee acceptable results in an acceptable time for an acceptable price that could take home prototyping to a new level. Or at least I like to think so.