It's a great setup. I was just wondering what values you chose for the chokes and filters. Did you tune it to a specific frequency?
In fact I didnt, I used chokes I had for another project (a tube linestage, actually). Usually, I would calculate the input impedance (voltage / current at nominal operation) and use that impedance to calculate the filter so optimum damping without ringing could be achieved, but it doesn't hurt to make the caps a little larger, since this is not the output of a class-D amplifier. So, gut feeling it was
, and it seems it worked.Forgot to mention, of course when calculating a corner frequency has to be chosen , so that the damping at the "problem frequency" (usually the switching frequency of the DCDC converters) is high enough. There is a great book by Richard Ozenbaugh on EMI filter design that explains that procedure in detail.
Last edited:
No, just because distortion itself can't help to estimate amp's performance.
I suppose, that all we need - is overall amplifier stability, while sounding really depends of components of actual build.
While we are really can reach ~100 dB feedback depth through all audioband we will feel all the ugliness of our measurement setups.
Also with independently biased LTPs could be designed something like this.
Attachments
I just wanted to check if the correction of mismatching that your circuit provides actually changes the distortion profile. I found that mismatch shows up as distortion, maybe 70dB down but still tells us something...
And of course you are right, distortion only is not enough, and with feedback a lot of issues can be covered (and some others brought forward).
I like that circuit, simple and elegant! do you control the corner frequency of the common mode control somehow? It seems it provides a lot of CM gain in the audio band
And of course you are right, distortion only is not enough, and with feedback a lot of issues can be covered (and some others brought forward).
I like that circuit, simple and elegant! do you control the corner frequency of the common mode control somehow? It seems it provides a lot of CM gain in the audio band
Bad case matching obviously results in 1ma for 10 ma devices. This could be mostly eliminated with slightly different source degenerating resistances.
No, i haven't tuned that behaviour now, but there are nothing complicated.
Yes, curent-mirror-loaded LTP could provide up to 60-70 dB gain, so, again yes, loop could be unstable without proper tuning.
Oh, yes, mirrored second stage have much more swing, check attached.
Attachments
Hi!
That's interesting...
Is the material conductive? I'm looking for something non-conductive for the same purpose, but for transistors in metal case where the collector is connected to the case.
greetings
That's interesting...
Is the material conductive? I'm looking for something non-conductive for the same purpose, but for transistors in metal case where the collector is connected to the case.
greetings
Since I received multiple questions for the plates, they were designed with the FPD tool from schaeffer-ag.de and produced there. If you want to build them yourself send me a PM and I can share the files, just in case you want to use the case for your own build
#frontplatte #schaefferag
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.