Hi
The VAS will only swing as high as the cascode voltage, about +/-39V. Also you need to put a reverse bias diode from C to E across the cascode transistors to prevent them from reverse biasing during clipping. If you use a cascode VAS with a voltage following output stage, you need boosted rails for the VAS to drive the output to full swing.
The VAS will only swing as high as the cascode voltage, about +/-39V. Also you need to put a reverse bias diode from C to E across the cascode transistors to prevent them from reverse biasing during clipping. If you use a cascode VAS with a voltage following output stage, you need boosted rails for the VAS to drive the output to full swing.
zerohead_ak47 said:
Yea, when you aren't cutting off the peaks of the signal, harmonic distortion decreases.
But now, the green LED's will produce around 4 - 6V. The maximum VAS voltage swing is about 55V. The problem is, you still can't get the output up to full swing, rail to rail. If you have no voltage gain in the output stage, nothing wrong with that, but you need to have boosted voltage rails for the cascode VAS. The good news is that the boosted rails require realitivily little current and can be easily regulated for better supply rejecton.
On the flip side, you might could use a small bootstrap from the output to supply the voltage boost to the regulators to raise the rails when the output swing is full.
If you do something like this, you might want to replace the 12K resistor with a current source.
My , My... this does look familiar.
Another Holman topology amp.. they do work very good,
better than a blameless. You can place a 200R trimmer/1k3
in place of the top 1K5 to get perfect LPT Ic.
They beat me to it , but a standard "blameless" led or
2 diode CCS is the "way" (80db PSRR -1k).
As far a being a high class amp, it can be as good as
any fully balanced VAS topology. I have one built just like
it (fa3) , and it sounds so good , I'm basing my next 2 amp's
on it (6 device triple OPS BJT and MOSFET).
The only issue this particular design has is that a high
gain input pair (LPT) must be chosen (zetex 696 or high
gain group ksc1845/KSA992) your 970's with no degen might just
give you 55db OLG max..
Attached is a similar amp , "steal" any idea off of it
it does work ,by the way.. very reliable.
OS
Another Holman topology amp.. they do work very good,
better than a blameless. You can place a 200R trimmer/1k3
in place of the top 1K5 to get perfect LPT Ic.
They beat me to it , but a standard "blameless" led or
2 diode CCS is the "way" (80db PSRR -1k).
As far a being a high class amp, it can be as good as
any fully balanced VAS topology. I have one built just like
it (fa3) , and it sounds so good , I'm basing my next 2 amp's
on it (6 device triple OPS BJT and MOSFET).
The only issue this particular design has is that a high
gain input pair (LPT) must be chosen (zetex 696 or high
gain group ksc1845/KSA992) your 970's with no degen might just
give you 55db OLG max..
Attached is a similar amp , "steal" any idea off of it
it does work ,by the way.. very reliable.
OS
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.