I built a bunch of these amps I designed and is not like anything posted here. Runs mono or stereo. Much lower distortion and higher gain and better sound than 3886. Composite amplifier with 225dB gain each channel. 6800µF/35volt caps. You will receive the amp in the picture without the 3886 modules soldered on but the modules are included. The $20 bill is for size reference. The proprietary schematic is also included but not to be posted here or distributed please. DC servo loop controlled. JFet input. $125 each plus shipping.
Attachments
Last edited:
Much lower distortion than 0.005% as the LM3886 has, how much is "much lower"? Something says to me that you can't have 225 dB in the audio band and then apply feedback.
225dB at DC to about 20Hz and then rolling off to 100dB gain at 10kHz is correct. Did not mean to be misleading. Spec sheet say .01% and I am getting .0032% at 10 watts 8 ohms ±30V, a factor of 3. Looks much lower to me. Maybe a different sense of much for each of us which is fine.
You might check here:
http://www.diyaudio.com/forums/chip-amps/163385-so-just-how-good-can-chip-amp-9.html
Nice kitty!
Real world measurements indicates 0.005% so 0.0032% is a minor improvement I would say.
When you talk about open-loop gain you can't simply add two elements and not apply global feedback. I'll guess you have more than one feedback loop?
If you try to draw a Bode plot with your values you can't get it to work with real parts. Impossible.
When you talk about open-loop gain you can't simply add two elements and not apply global feedback. I'll guess you have more than one feedback loop?
If you try to draw a Bode plot with your values you can't get it to work with real parts. Impossible.
Walt Jung did write a piece in the subject:
http://waltjung.org/PDFs/Composite_Line_Driver_with_Low_Distortion.pdf
http://waltjung.org/PDFs/Composite_Line_Driver_with_Low_Distortion.pdf
.005% is not the spec at 10kHz, it is .01%- please read the data sheet. That makes my .0032 10dB less distortion. So are you saying a reduction from 32% to 10% THD is also insignificant?
Realizing an amplifier is a simply a filter with gain then the sketch below is correct and the two filter elements exactly do simply add up. Do you not agree with this practice? Accepting that is the case then the numbers I give are conservative. The sketch does not show the global feedback loop as that was the discussion.
I have linked you to the real thing built with real parts. Simply because you could never imagine such a thing does not mean I cannot build one. I do not mean to be rude or call you names please.
I have known Walt and we have exchange information and ideas on several occasions. Good guy.
-Sum
Realizing an amplifier is a simply a filter with gain then the sketch below is correct and the two filter elements exactly do simply add up. Do you not agree with this practice? Accepting that is the case then the numbers I give are conservative. The sketch does not show the global feedback loop as that was the discussion.
I have linked you to the real thing built with real parts. Simply because you could never imagine such a thing does not mean I cannot build one. I do not mean to be rude or call you names please.
I have known Walt and we have exchange information and ideas on several occasions. Good guy.
-Sum
Attachments
0.0032% is very good at 10 kHz but I think 32% down 10% makes a bigger difference than 0.01% to 0.0032%
I would be interesting to see what you mean by saying "225 dB" gain but this is nothing you want to show us. I have no problems with that. This thread was about selling, not show or discuss. Feel free to show us though.
I would be interesting to see what you mean by saying "225 dB" gain but this is nothing you want to show us. I have no problems with that. This thread was about selling, not show or discuss. Feel free to show us though.
Please read your own measurement results.😀
According to your posted data the measured distortion at 0.1W at 10kHz actually gets worse when bandwidth limited to 80kHz.
Oops!
By the way, that is an obvious typo- should be .014- yes oops!
Well this discussion was at 10 watts limited to 80kHz very similar to the 3886 data sheet at ±28 volts where mine are ±30. Mine is 10dB less distortion. Important for sonics, probably not. Important to distinguish a better design, probably so. Some may disagree with the last statement and that is fine.
At low power distortion runs into the noise floor. An obvious limit to total distortion. Certainly this design could be improved with a lower noise input stage which would exactly result in lower low power distortion. That is obvious from the raising the power 20dB and the distortion drops exactly 20dB. Noisy bad design flaw... but -93dB below full output was quiet enough for my use.
- Status
- Not open for further replies.