"My design is a bit complex, it has large amounts of open loop gain and without a cautious layout will be a pain to get stable,..."
Yes indeed. I just finished a Slone variation onf the Self amp. For reason's I won't go into, I wanted to see how small a "footprint" I could squeeze it into. As a consequence I got to careless about letting the input section be toio close to the output section. The circuit was wired correctly with regard to the schematic but went up in smoke on the first try. It took three re-layouts before the result were satisfactory.
What aspects did you find most difficult to solve to get both the desired performance while maintaining stability?
Unlike you, I have never done well with perf board, I'm simply too likely to wire up something wrong way around. I have much better success using software that lets me go consistently from a schematic to a board layout and then etch my own PCBs. It seems to takes more time, but avoids stupid mistakes I'm prone to when wiring direct on a perf board. So, for me the results are bit quicker.
Yes indeed. I just finished a Slone variation onf the Self amp. For reason's I won't go into, I wanted to see how small a "footprint" I could squeeze it into. As a consequence I got to careless about letting the input section be toio close to the output section. The circuit was wired correctly with regard to the schematic but went up in smoke on the first try. It took three re-layouts before the result were satisfactory.
What aspects did you find most difficult to solve to get both the desired performance while maintaining stability?
Unlike you, I have never done well with perf board, I'm simply too likely to wire up something wrong way around. I have much better success using software that lets me go consistently from a schematic to a board layout and then etch my own PCBs. It seems to takes more time, but avoids stupid mistakes I'm prone to when wiring direct on a perf board. So, for me the results are bit quicker.
Well I' don’t have any specific rule on that mater, because it's design dependent, it works on some designs and in other dont, the best way is to do like you did, just try and try till it's right,
but as a general practice, I like to keep my gain stages very close together and compact, the output stage is easier to get stable, or at least emitter followers
the R/C network before the predriver makes wonders on very high current gain output stages(triple followers)The feedback path very small to avoid parasitic capacitance and the consequent phase rotation...
also I like separate grounds, one for the input and feedback ground, one for the power ground(current sources and that stuff) and one for the zobel network .
Your Sloane amp used 2 pole compensation? I Had a lot of troubles with that technique on my prototype...
I use perf board because it's much easier to make modification on the design...of course PCB look nicer!
Chris, you have Mail
Cheers
Ricardo
Ps: one more thing, the boards on the counterpoint amp look symmetric right? but they aren’t, they are different on very small details, that difference means that one channel has much worse settling time(easily seen on a 50khz square wave) than the other...on high feedback amps, every detail counts...
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.