I am not sure I comprehend; are you refering to your AD797?
There are others. Solomon jumps right to the answer. You can present the same op-amp as two cascaded voltage gain stages and show unconditional instability, then introduce the pole spliting effect of Ccomp. It is then possible to work in the right plane zero effects in the second stage. I would hope the goal is for folks to learn more than the "right" name for things.
As has been demonstrated a single folded cascode stage can, with enough bootstrapping, have DC gain of >5*10^6 and the same gm-C AC characteristics with Ccomp to ground.
Had to laugh, Solomon shows the 10.7kHz full power BW on the 741.
Last edited:
Surely you jest!Had to laugh, Solomon shows the 10.7kHz full power BW on the 741.
LM301 was a favourite of mine in Jurassic days cos it was easy to disable the i/p stage to incorporate a nice LN BJT instead. All dis BS got swept into the Don't Recycle Bin once the TDA1034 (original name for NE5534) came out from Mullard.
I would hope the goal is for folks to learn more than the "right" name for things.
I think knowing the right name for things is an excellent starting point to knowing how they really work.
As has been demonstrated a single folded cascode stage can, with enough bootstrapping, have DC gain of >5*10^6 and the same gm-C AC characteristics with Ccomp to ground.
The only snag is shunt compensation to ground increases the non-linearity of the stage it compensates.
Be that as it may, it would be fascinating if you would produce a discrete amplifier for diyaudio based on your AD797.
Be that as it may, it would be fascinating if you would produce a discrete amplifier for diyaudio based on your AD797.
Luxman preamp C02 circa 1983 , at least for the topology.
What about the compensation capacitor "Cdom"?
Shouldn't we call it "Ccomp" ?
We could call it Douglas . It is only right .
The only snag is shunt compensation to ground increases the non-linearity of the stage it compensates.
Don't see that in practice, or I'm not sure what you mean. All else being equal the input error signal is identicle, and without the symmetry you can't recapture the base current error of the second stage (in the two stage case) without a Hawksford cascode or equivalent.
I have built discrete single stage amps with vanishingly low THD (Ccomp to ground).
Luxman preamp C02 circa 1983 , at least for the topology.
Not even close, that's an ordinary folded cascode tied to V- on the output side. The term and patents appear well prior to the 1983 date.
Last edited:
Not even close, that's an ordinary folded cascode tied to V- on the output side. The term and patents appear well prior to the 1983 date.
Where did i say that there was no prior art.?.
The question was about a discrete design , indeed.
To say that it s not close is somewhat exagerated.
Attachments
An ostensible improvement to the folded cascode arrangement is apparently due to Shinichi Kamijo:
http://www.ne.jp/asahi/evo/amp/J554K2955/x6amp.pdf
P.S. Wahab, I see you use Simetrix. Good man!
Did you purchase the full version?
http://www.ne.jp/asahi/evo/amp/J554K2955/x6amp.pdf
P.S. Wahab, I see you use Simetrix. Good man!
Did you purchase the full version?
Last edited:
Mullard was Philips, UK for a VERY long time. It was one of the concessions extracted by Michiel de Ruyter for the withdrawal of his forces from the Medway.TDA1034 came out from Mullard or was it Philips?
TDA1034 was one of the very last Mullard branded designs. But the huge Mullard, Blackburn factory made valves for the Philips empire throughout the Golden Age of Valves under several names.
For many years, my favourite OPA was Philips NE5532/4. When I again got serious about low level electronics circa 1990, I found these were no longer made in the Southampton factory and weren't as consistent and quiet as the old ones.
Some people swear by NOS Mullard EF86s & EL34s. I treasure NOS Philips NE5532/4 and Mullard BFW11s
Last edited:
An ostensible improvement to the folded cascode arrangement is apparently due to Shinichi Kamijo:
http://www.ne.jp/asahi/evo/amp/J554K2955/x6amp.pdf
P.S. Wahab, I see you use Simetrix. Good man!
Did you purchase the full version?
Unfortunately this talented creator deceased last year IIRC ,
i downloaded most of his site to keep his work alive if ever
his adress was to be downed , i guess that many dyiers
did also back up his datas.
As for Simetrix i ve only the free version that is node number
limited but it s still enough for quite large schematics ,
i dont like LT Spice whose interface is surely the less
user friendly one could "design" , and after all the size
limitation will mandate extreme care to extract
all that is possible when adding a component...
For many years, my favourite OPA was Philips NE5532/4. When I again got serious about low level electronics circa 1990, I found these were no longer made in the Southampton factory and weren't as consistent and quiet as the old ones.
NE series were from Signetics that was bought by Phillips
who inherited the IC portfolio.
Last edited:
Ok. With GNF it's a TIS and if a ZGNF it's a VAS. I can sort of live with that.
But, that pesky Cdom, even with ZGNF still seems to me to be a TIS . . .
You usually don't need a Cdom in a ZGNF amp, since the (so called VAS) Trans Impedance Stage (TIS) is loaded by a R, but as you say it is still a TIS.
As for Simetrix i ve only the free version...
The only problem with Simetrix is you have to use Rosenstark's method to determine loop gain, and this requires a copy of the circuit whose loop gain you're simulating unlike LTspice.
The only problem with Simetrix is you have to use Rosenstark's method to determine loop gain, and this requires a copy of the circuit whose loop gain you're simulating unlike LTspice.
Can you please elaborate about this ?.
Do you mean that LT Spice can accomodate pure
mathematical analytic models of a given circuit.?
NE5532
He really had a far-sighted vision for its time.
edit: Look here for a fascinating story abot the collaboration between Mullard and Philips during WW-II: http://www.dos4ever.com/EF50/EF50.html (btw, Gerrit Alma was my uncle)
Cheers, E.
LOLMullard was Philips, UK for a VERY long time. It was one of the concessions extracted by Michiel de Ruyter for the withdrawal of his forces from the Medway.
...........
He really had a far-sighted vision for its time.
edit: Look here for a fascinating story abot the collaboration between Mullard and Philips during WW-II: http://www.dos4ever.com/EF50/EF50.html (btw, Gerrit Alma was my uncle)
Cheers, E.
Last edited:
Mullard was Philips, UK for a VERY long time. It was one of the concessions extracted by Michiel de Ruyter for the withdrawal of his forces from the Medway.
TDA1034 was one of the very last Mullard branded designs. But the huge Mullard, Blackburn factory made valves for the Philips empire throughout the Golden Age of Valves under several names.
For many years, my favourite OPA was Philips NE5532/4. When I again got serious about low level electronics circa 1990, I found these were no longer made in the Southampton factory and weren't as consistent and quiet as the old ones.
Some people swear by NOS Mullard EF86s & EL34s. I treasure NOS Philips NE5532/4 and Mullard BFW11s
I would love to meet the guy(s) who actually designed the NE5532/4. Every time I comes up Neve gets brought up. He may have spec'd it, but he did not do the detailed design. Was it done in the UK, Eindhoven or in USA under Signetics?
Where are you 5532/43 designer?
A lot of people don't know it, but for many years Philips was a top tier analog company globally in revenue, if not technology. How so? UOC-TV and then all the other analog consumer stuff.
Last edited:
You usually don't need a Cdom in a ZGNF amp, since the (so called VAS) Trans Impedance Stage (TIS) is loaded by a R, but as you say it is still a TIS.
Yes, on reflection you are right. But who in there right mind operates an amp without feedback. Nuts.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
- Home
- Amplifiers
- Solid State
- Audio Power Amplifier Design book- Douglas Self wants your opinions