I wouldn't mind etching it. While I love to work with pro boards, I also like etching in the morning and testing that evening.
SMD will test you etching skills. You should use 1/2oz copper. The traces are narrow and squeeze through some very narrow areas but I've had good success with it. Liquid tin is a good idea. Solder bridging is more of an issue with home etched boards though.
Don't worry PCBWay is cheep
As you can do it home etching friendly, just do it as an alternative version.
I hope I haven't spoiled your fun at home etching. jkuetemann got me hooked on Chinese boards and I hardly ever etch any more myself. The only problem now is I usually made my first board as a tester and found all my mistakes, then etched good ones. Now I've got one shot at it and am stuck with my mistakes or need to get a second set done.
I want one !!
Actually , this is the best of the CFA's to go SMD - everything
is low current with the super-pair VAS.
"current on demand" does not "feed" saturation on this or the K-C.
1-2 ma Q1-8 , super pairs are "ice cold"... just the 5-6ma Q11/12.
PS - Q1-8 could be run as low as 1.2mA with R4-7 , R18-19 changes.
OS
I still need to pick out some transistors for this one. All my stock is for 12V circuits. I'm loaded up on BC846/856 and MDJ340/350 but would like to go with something better. The footprints are all the same with SMD so designing is easier.
My only concern on these is the 1W resistors. I'm paralleling 2512 1W resistors for all of them. R12/13 and R16/17 might see some 3W peaks with high rail voltage so not a huge problem but I don't want R14/15 browning the boards.
I don't know what composition is acceptable for film caps either. Is Polyphenylene Sulfide (PPS) a good replacement for Polypropylene?
Electrolytics are smaller footprint and cheaper in through hole and don't get knocked off the board as easily so that's a no brainer, but need to be on the opposite side of the board for single side etching.
Last edited:
Count me in for two kypton c smd boards
Did you ever get your Kypton-C boards running?
No,i will continue etching at home. I can't wait this slow "Chinese train" . I like the quickly way.
I have repaired two faulty boards, yes i am Electronics Engineer
all DC voltages are fine, can't wait to test with sinus / MusicIf someone are interested i have new kypton C boards to sell, just remember shipping from Denmark.
Here's something that has been bothering me about these output boards since I first saw them. I think the cap connections in the purple circles are wrong. Current is entering the cap, then needing to reverse direction and exit the same way it entered. This doubles the amount of current flow on that trace. I know the trace is extremely oversized for what it is carrying but my thinking is these traces will be producing EMF at double the frequency the output devices are outputting. 2 sets of 5 cap traces plus the two big cap traces emitting noise just seems like unnecessary trouble. I read a couple of Randy Slone's books years ago and he mentioned this. It's something that I've paid attention to ever since.
Attachments
I still have not been able to get mine working. The bottom half of the sine and square waves is misshapen. I have been meaning to post some pics and measurements but just got busy with other projects.. The Krypton ND worked first try.
Blessings, Terry
The bigger caps could likely be omitted completely, but the smaller ones perform decoupling at the transistor collectors and are almost certainly necessary for stability.
If I were looking to make a change I would consider forgoing the higher value at the power inlet and go with larger ones at the individual output transistor collectors. Done that way I would make a beefier ground return for those large decouplers.
Wether the existing arrangement is actually an issue or not is debatable, but my gut says it probably isn't an issue. What I would consider is making the MOSFET 'option' official and make provisions for the gate protection and bias modification on the board with the SMD pads for easy jumpering of optional components. Just my own thoughts on it.
If I were looking to make a change I would consider forgoing the higher value at the power inlet and go with larger ones at the individual output transistor collectors. Done that way I would make a beefier ground return for those large decouplers.
Wether the existing arrangement is actually an issue or not is debatable, but my gut says it probably isn't an issue. What I would consider is making the MOSFET 'option' official and make provisions for the gate protection and bias modification on the board with the SMD pads for easy jumpering of optional components. Just my own thoughts on it.
I changed the layout a bit when I did my sub amp boards. I don't have two options for the RE resistors but I moved the rails inboard of the output devices and did install larger decoupling caps at the outputs. I was debating installing 10000mF snap ins at the inlets but they would have been too tall.
Attachments
jwilhelm,
If you could make it so you can just edit the number of output devices that would work out great. I can see how small it would look with one or three output pairs and the only thing that changes is that center section. Now everything and the input section on one board with variable output pair's is the ticket.
On the output coil what distance is really good, how little does it really have to move away to have an effect?
If you could make it so you can just edit the number of output devices that would work out great. I can see how small it would look with one or three output pairs and the only thing that changes is that center section. Now everything and the input section on one board with variable output pair's is the ticket.
On the output coil what distance is really good, how little does it really have to move away to have an effect?
I think the coil position is going the depend on the application. It's going to emit noise 13 or 14 (or how ever many turns it is) that the speaker wire does so I think the key is to keep it away from signal. AndrewT's other point was to keep it away from a metal chassis so as not to induce noise into the chassis ground.
Yes the ground is an interesting question. Since I won't have a metal chassis I've been wondering what other issues I will have to deal with. I can always have the 'enclosure metalized if need be for shielding or grounding reasons but not sure that will be necessary at this point. It almost seems that it could be an improvement to have no metal around the circuit and only have the ground wires acting as ground, nothing circulating in the chassis metal?
I think double insulation and no earth ground is the way to fly. No ground loop issues with other equipment that way.