Could be thinner actually, Measured again and more like 3/32" ..good eye! I used a cheap plastic ruler..and assumed a lot!, also the walls vary a bit as the extrusion isn't completely consistent. The other thicknesses I'd be more confident on.. The dimensions other than thickness are clearly easy, even, numbers..
When I was a commercial fisherman we used fathoms all the time.. and everyone still use knots on boats it seems!
I have been studying the BA-2 among other Pass amps trying to learn. I have a question concerning the bootstrap current source on the BA-2. Isn't this the same concept as the on found in the PLH article. Also, if the feedback is taken from that drain of the VAS fet and not the output, doesn't this limit the performance of the entire amp to that of the ouput stage since it is not included in the feedback loop?
Yes, "as the on found in the PLH".
Well, Yes the FB will be correcting the Diff Amp and VAS stages, if it can but, will not effect any distortion from the output stage. Apparently N.P. thought the "Au Natural" output stages can do just fine with their intrensic (100%) FB. I don't want to get into that argument. The limiting of FB to only the 2 voltage gain stages avoids excessive phase shift etc. of all 3 stages, making it likely better sounding than a total global FB approach.
I like the first answer a lot. I am interested only because i am considering X-ing the BA-2 and wondering if anything is gained by doing so. SUSY is amazing in its ability to reduce THD and it would seem, if what you say is true about the output stage, SUSY may not be of benefit here since the output stage would determine the final numbers and sound. I understand this may be simplistic in explanation, but does it make since Is the output performance as shown in Nelson's graph based on this standard FE? If so, i guess it would stand to benefit from SUSY. I know he speaks of its standalone quality. I was considering, and still am, using an OPA 1632 to drive F4 outputs, but this seems more interesting and challenging. SSR085 ZV4 currently on the work table.
I am interested only because i am considering X-ing the BA-2 and wondering if anything is gained by doing so. SUSY is amazing in its ability to reduce THD and it would seem, if what you say is true about the output stage, SUSY may not be of benefit here since the output stage would determine the final numbers and sound. I understand this may be simplistic in explanation, but does it make since Is the output performance as shown in Nelson's graph based on this standard FE? If so, i guess it would stand to benefit from SUSY. I know he speaks of its standalone quality. I was considering, and still am, using an OPA 1632 to drive F4 outputs, but this seems more interesting and challenging. SSR085 ZV4 currently on the work table.
Last edited:
I have rread the A75 article, and barring missing something, i do have not gained any insight into whether or not to include it in the feedback loop. He talks about the possibility of adding it, but doesn't seem to provide any comments or performance data on it while included. If not including it is OK, it would seem that a SUSY pre-amp, like your Pumpkin, ZenMod, is all that is needed. run that into an F4 with or without the FE, which is what the BA-2 output stage is, correct. FLG, you seem to be of the opinion that including it will be beneficial.
I accidentally wired my first version of the Krell Klone KSA100 without including the output stage in the GNFB loop.
The design of the PCBs specifically allows this "accident" to slip through.
I then connected the outputs into the GNFB loop.
Very difficult to stabilise and had to use a large Cdom which killed the SQ. I gave up, to await the "experts' guide" on how to set up the Klone. That help never came. Too few completed the build.
But with the GNFB loop closed around the drivers the amp was more stable. It did not blow fuses when connecting a 100nF across the speaker terminals.
The design of the PCBs specifically allows this "accident" to slip through.
I then connected the outputs into the GNFB loop.
Very difficult to stabilise and had to use a large Cdom which killed the SQ. I gave up, to await the "experts' guide" on how to set up the Klone. That help never came. Too few completed the build.
But with the GNFB loop closed around the drivers the amp was more stable. It did not blow fuses when connecting a 100nF across the speaker terminals.
I would try it both ways and of coarse what it sounds like would be a major determining factor. My guess would be The Non-Global FB would be the better choice. This is not the first time I've seen this configuration, including the optional connection.
Hi Andrew,
that sounds pretty strange. My Krell-clone, the MiniK, drives caps up to several uFs and does not complain at all. Only the square wave becomes a little smoothed as expected.
It uses, however, a different compensation scheme, not the standard Miller-compensation that seems to work not so well with modern high speed transistors in this amp.
I do not know the 100 clone as mine is based on the 50, but the compensation does not appear to work right.
Hannes
Apologies for a serious OT here but Hannes' post caught my eye. I need to drive for an optics experiment a piezoelectric element. According to the spec sheet it will show as a 0.4 uF capacitor and I need to put a 30 V sinewave across it at up to 50 kHz. Could someone point me towards a buffer design that could handle this load?
Thanks and apologies again for the OT