Originally Posted by Krow
I am using the Toner Transfer method using Photo Paper. It works, but the problems I have are that it loves to really hold on to the toner once it's transferred it. Peeling it off is a pain in the ass. When I started using this paper it turned out to be waterproof, which royally sucked the big one when trying to peel it because it was like separating plastic layers. I solved that by ironing on a regular piece of paper so the back plastic layer glued itself to that and was easier to separate.
I've found it to be wildly inferior to using an exposure mask. I suggest you try that method, as I said, you can produce 0.25mm (10 mil) tracks with an amazing consistency. Since I tried using a mask and directly exposing the board instead of going with the photo transfer method, I've never looked back.
The toner transfer method takes a long time and is very sensitive to the paper you use and how you peel the paper. The exposure method is immune to that, the quality of your boards only depends on the opacity of the masks. And since I've found I can use tracing paper... I don't even bother to get them printed anymore. Inkjet with ink weight set to Heaviest, do a small printing first so the heads clean up, and done with it. Side note: You can also reuse the transparencies, and if you get them laser printed on acetate (make sure they use the heaviest option available) you can reuse them basically forever.
Also, I see you're using SOIC MOSFETs. Why not use packages with larger heat transfer capacity? D-PAK (TO252) comes to mind. Large legs, and you can solder them to a common power plane for maximum current capacity, as well as using that pad as a heatsink. Soldering it gets a bit tricky if your iron is small though (less than 35W).
Originally Posted by Thermo Electric
Sounds like your missing the pull down resistor to me or its WAY to high. That was the problem i found. People going on about 10k or 2k but actually it needed 220 ohms
With 3A drivers capable of doing full bridge or half-bridge obtainable at less than 30 cents a piece, I find it silly to even consider going driverless for an application with such power, by the way.
Krow, should you require help with the power side, I will glady help you
And yes, cutting the boards with a table saw is fine. Just make sure the teeth hits the copper from above, else you'll just rip it away. Keep in mind that FR boards severely wear down HSS blades though. After all it's glass what you're cutting.
Originally Posted by Krow
HA! so here's something just found out. Since I designed the controller as ALWAYS ON when the controller isn't hooked up (as if it gets unplugged or something) the logic is reversed. 100% means 0% and the opposite is true. I can either try redesigning, or work the code in reverse, which might be easier. Not sure.
A MOSFET gate that's left floating will almost always end up making the D-S junction go into conduction state if there's power present across it.Edited by Artikbot - 3/31/14 at 5:03am