It has good OL gain, but it doesnt specify important parameters such as slew rate, GBP, no THD or graphs. The 3nV/sqrt(Hz) may look like an impressive noise spec when compared to the table of opamps on the webpage, but it seems that something like an AD797 with its 0.9nV/sqrt(Hz) was conveniently ommited from the table, also no LME49720 with its 2.7nV/sqrt(Hz) and 140dB of open loop gain, its also hard to beat the 0.00003% THD and 0.00005% IMD of the LME49720 Class A or not.
Last edited by a moderator:
It is class A up to 15mA output, then it enters class AB up till 65mA output.
But with most loads it will never get above 15mA.
Also no THD graphs, probably because it is pretty well impossible for a discrete to even get close to an integrated opamp in this department.
Jan
Last edited:
Play around with two pole compensation, either in SPICE or on the breadboard using an LM101 (mass market version is LM301 ) as attached below. I predict you'll discover that an amplifier with two pole compensation has a transient response with LOTS of overshoot.
1969 -- a good year for analog electronics.
1969 -- a good year for analog electronics.
Rupert was certainly not a businessman, he got booted out of companies he founded at least twice.
His crowning achievement is probably the Focusrite console. A cost-no-object project instigated by George Martin. Only 10 complete consoles were ever made, each cost 2-3x as much as an SSL.
Sadly these day old consoles are frequently broken up into parts to make mic pres since they are worth much more that way.
I didn't think that many were made. At least one is still running in Air studios
According to them only 3 were actually made. Custom Neve History - AIR Studios all still in use.
According to them only 3 were actually made. Custom Neve History - AIR Studios all still in use.
Thanks! Very interesting. But I couldn't tell---what opamps WERE they using?
Which is why I didn't say that he did commission it but you're right.
With hindsight 'inspired' would have been a better choice than 'instigated'.
my latest uses opamps with valve in feedback loop.
Test on thd are as expected as the ne5532 is control element.
Interesting that looking at fft , same opamp only, waterfall shows dominant third.
With 6sL7 in fb loop the third reduces and second appears, slightly higher than reduced third.
This is a difference that can be heard and measured.
-
Test on thd are as expected as the ne5532 is control element.
Interesting that looking at fft , same opamp only, waterfall shows dominant third.
With 6sL7 in fb loop the third reduces and second appears, slightly higher than reduced third.
This is a difference that can be heard and measured.
-
Hope this helps.
Relative to the thread question, it looks like within the opamp output stage is where some work is required.
That is, if your goal is great sounding audio.
It was mentioned earlier in this thread, the idea of forcing a full non switching Class A output opamp. I did it sort of externally but a good designer could do it on chip and get same results.
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.