There was a lot of discussion about bootstrapped opamps in a recent headphones amp thread, as a way to reduce common mode distortion.
I tried to see how compact I could make a pcb for an opamp operating into pretty much ideal conditions: it has both bootstrapped supplies and a small class A buffer isolating the opamp from the load. With only 7.5ma max in a SE class A output stage, it's not made to drive heavy loads obviously.
The feedback resistors aren't onboard as the board is made to serve as a direct replacement for a single opamp. Still, the ratio of the dividers driving the transistors has to be matched to the overall gain, as explained here: http://www.analog.com/media/en/technical-documentation/application-notes/742022599AN232.pdf It's not a plug and play solution. Minimum suggested gain is 2.
Comments, suggestions for improvements ?
I tried to see how compact I could make a pcb for an opamp operating into pretty much ideal conditions: it has both bootstrapped supplies and a small class A buffer isolating the opamp from the load. With only 7.5ma max in a SE class A output stage, it's not made to drive heavy loads obviously.
The feedback resistors aren't onboard as the board is made to serve as a direct replacement for a single opamp. Still, the ratio of the dividers driving the transistors has to be matched to the overall gain, as explained here: http://www.analog.com/media/en/technical-documentation/application-notes/742022599AN232.pdf It's not a plug and play solution. Minimum suggested gain is 2.
Comments, suggestions for improvements ?
Attachments
@DPH: I've tried and that's not quite working out. Soldering a 7.5mm pitch cap over the + and - pins would be the best solution. That and restricting the board to slowish opamps like the opa134.
Well, not quite. The best would be to switch to a API sized format, which gives a gnd pin too and would allow for a better layout.
Well, not quite. The best would be to switch to a API sized format, which gives a gnd pin too and would allow for a better layout.
Indeed, that would be a possibility. Otoh, the API format enables stuff like this. Yes, it's bigger but there's a lot more onboard. A clipping circuit at the input, proper decoupling, a powerful opamp as output.
Disregard the opa134 label for the second opamp, it should be a lm6171. Btw, this one favors low value resistors for its feedback.
Disregard the opa134 label for the second opamp, it should be a lm6171. Btw, this one favors low value resistors for its feedback.
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.