amplifierguru said:
The idea was to have one opamp with +ve offset and the other to have equal but opposite - ve offset (Im not sure if this is the correct way to do this) so current flows between the two opamps forcing them in to class A
Gain is easy, just add resistors
Capacitors could dubble as input caps which are a nessercery evil anyway, thats if you think capacitors are bad in the first place
Anyway just some brian food
theChris said:
The idea is to have current flowing between the opamps on purpose so current sharing resistors may not be nessercery?
My thinking is comming from over biossed AB amps where one transistor is sorcing current and the other is sinking current at the same time
In this case opamps take the place of the transistors and having the bois at the input each opamp would force class A all the way though (again just a thought)
BVDBOS,
I see what you're trying to do. The problem is most OPamps have poorer PSRR on the miller comp side - or one side anyway and better NPN side stability so the tendency is to favour the upper half and load it down to the negative rail. Even better to the unreg negative supply!
Having said that the best way is with a distortion meter and residual showing for each different one. That's the way to optimise.
I have to agree with jacco that FET input chips is the more noticeable improvement . Both is excellent. I modified some
Meridian DACs this way with good results. Those damn 4 layer boards.
I see what you're trying to do. The problem is most OPamps have poorer PSRR on the miller comp side - or one side anyway and better NPN side stability so the tendency is to favour the upper half and load it down to the negative rail. Even better to the unreg negative supply!
Having said that the best way is with a distortion meter and residual showing for each different one. That's the way to optimise.
I have to agree with jacco that FET input chips is the more noticeable improvement . Both is excellent. I modified some
Meridian DACs this way with good results. Those damn 4 layer boards.
Hi guru
Hi amplifierguru,
The 2N5551 is rated at 160V, 500mA, 625mW and I think this rating is not exceeded because the output load resistor is around 2 kiloOhms in practical conditions.
THD of Stallion 2400W [2 X 1200]
1000Hz--- 0.012%
20000Hz--- 0.04%
at half rated power.
Regards,
Kanwar
amplifierguru said:
Hi amplifierguru,
The 2N5551 is rated at 160V, 500mA, 625mW and I think this rating is not exceeded because the output load resistor is around 2 kiloOhms in practical conditions.
THD of Stallion 2400W [2 X 1200]
1000Hz--- 0.012%
20000Hz--- 0.04%
at half rated power.
Regards,
Kanwar
Hi guru
Hi amplifierguru,
In PA use these figures are best among the amplifiers.
In case of studio reference amps for critical audiophile listening the THD level must be at around 0.0003 % for highest audible frequency or at the frequency of the interest.
One question PLZ?
Have you ever tried your hands on N-channel IGBT amp, I was asking because we are just testing the performance of our newly developed [yet to be launched] N-IGBT amp using APT100GF60LR devices in TO-264 package.
Do not joke with professionals because you are also a professional
regards,
Kanwar
amplifierguru said:
Hi amplifierguru,
In PA use these figures are best among the amplifiers.
In case of studio reference amps for critical audiophile listening the THD level must be at around 0.0003 % for highest audible frequency or at the frequency of the interest.
One question PLZ?
Have you ever tried your hands on N-channel IGBT amp, I was asking because we are just testing the performance of our newly developed [yet to be launched] N-IGBT amp using APT100GF60LR devices in TO-264 package.
Do not joke with professionals because you are also a professional
regards,
Kanwar
That was a misunderstanding.
It sounded like you think you could bias all stages into class A which you (obviously) can´t unless you open it
.Assuming the first stages are class A you´re right of course.
I tried it and it did nothing (for my ears) to a OPA627.
(What is much easier audible with some preamps, esp. long cables or headphones is the limited current output of most opamps.
Applying a buffer often seems sensible.)
IMO
greets
AB outputs are "class A" for sufficiently high impedance loads, most monlithic op amps skimp on outputstage bias current, usually a lot less than 1 mA - but when used with a high Zinput buffer/output amp in the loop they will be operating in class A
some DSL drivers in high bias mode are probably class A to 10 mA or more
some DSL drivers in high bias mode are probably class A to 10 mA or more
Kanwar,
There was a Class A Opamp produced in the 80's although I can't recall whose it was, maybe Signetics or AD. It had about 10MA in the output stage and was I think a TO39 with integral
(round with lots of fins) heatsink to dissipate the extra 300mW. It may have also been high voltage as a driver for power stages.
I think.
There was a Class A Opamp produced in the 80's although I can't recall whose it was, maybe Signetics or AD. It had about 10MA in the output stage and was I think a TO39 with integral
(round with lots of fins) heatsink to dissipate the extra 300mW. It may have also been high voltage as a driver for power stages.
I think.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.