The MONGREL (supersym II)

[ostripper]

Another gem... tied with the "leach" , as a EX candidate. Built it , I know it will not BURNNNN!
OS

[ostripper]

Originally Posted by AndrewT

c89 is in the wrong place.
It should be between the VAS and it's load (the CCS or bootstrap)

I wonder if that accounts for the very high current in the cap when the sim modeled behaviour?

I got that (below) right from Mr. self's book ??? (Page 303 D. Self 1988).
OS

[AndrewT]

you are right.
That is what Self shows.
All I can say is that intuitively it seems more right to couple VAS to it's load. I have no evidence.

[ostripper]

Originally Posted by AndrewT

you are right.
That is what Self shows.
All I can say is that intuitively it seems more right to couple VAS to it's load. I have no evidence.

It does seem to still be coupled to it's load , just through a 10-16R resistor. The self example is the only correct one (he shows 3 examples in the book) out of over 500 examples of Vbe's in my schematic collection. Many have NO CAP at all , and if the wiper breaks in the adjustment... POOF!! mine would just become grossly underbiased . Poor designs (example pix1)

He says the 10-20R resistor "offers first order cancellation of Vbias variations due to current changes".

Changing the value of this resistor changes the correction voltage of the VBE in response to the current changes. Mr. self , using coincidentally , a 6ma current source ... measured 1/10th the voltage variation with the resistor. Thus, you would want the cap AFTER the correction resistor to smooth out the corrected voltage.

On a symasym type amp , this would be a most valuable addition since it does not have a fixed , stable current source like the blameless. Here is a picture (#2)of the whole thing (quasi killer with the better "blameless") , I just showed the Vbe last time.

OS

[AndrewT]

Originally Posted by ostripper

just through a 10-16R resistor. The self example

He says the 10-20R resistor "offers first order cancellation of Vbias variations due to current changes".

one of the first jobs I did when I read that section back in 2001 was to pick out the maximally flat "tops" of the 4 lower plots. 16r suits 8mA, 27r suits 5mA.
Further calculation brought up 10r suits >=10mA and 30r suits 4mA.

It does look like the extra r is moving towards zero as the VAS bias is increased. But where?

[keantoken]

Ostripper, if you're worried use a PNP Vbe multiplier and keep the cap connected between E and C. This will eliminate the 10R resistor from the signal path.

C85 reminds me of:

Based on Hugh Dean's AKSA 55

- keantoken

[AndrewT]

then the extra resistor is in the path to the lower driver.

[ostripper]

Originally Posted by keantoken

Ostripper, if you're worried use a PNP Vbe multiplier and keep the cap connected between E and C. This will eliminate the 10R resistor from the signal path.

C85 reminds me of:

Based on Hugh Dean's AKSA 55

- keantoken

The 10R resistor in the signal path is of no concern , as mr. self said, it is a small local feedback to compensate for current variations. Also we have 100r base stoppers right after the VBE .... so does it matter?? In chapter 12 , he makes "fun" of his own CCS with its 2 diodes which is why he says to never omit the 10R resistor. I don't have these problems with my cap multipliers and 2 more advanced CCS's.

I also have that "magic" cap in your link on my Vbe base, it keeps the Xover point the same (.6ma variation) across the entire audio spectrum , even with the IRFP's.
OS

[keantoken]

Originally Posted by AndrewT

then the extra resistor is in the path to the lower driver.

Yes, but with a singleton VAS no drive comes from that side, it's just a CCS. Not that it matters considering the base stoppers as OS pointed out...

Ostripper, the 10R resistor provides feedforward compensation like in a feedforward regulator (if you don't know what that is, look it up, don't guess). However at AC C85 changes this. If you adjust the 10R for DC bias stability, then at AC it will overcompensate, wasting energy through C86 and decreasing AC impedance.

Using a PNP Vbe multiplier like I said would help because then no AC would need to come through the 10R resistor.

- keantoken

[AndrewT]

Originally Posted by keantoken

Yes, but with a singleton VAS no drive comes from that side, it's just a CCS. Not that it matters considering the base stoppers as OS pointed out...

you even forced me to draw it out.

There are two audio signal outputs from the Vbe multiplier.
The upper feeds the upper driver.
The lower feeds the lower driver.

NPN multiplier has the extra resistor in the top feed.
PNP multiplier has the extra resistor in the lower feed.

Whether it's a CCS or a bootstrapped resistor does not affect the two audio feeds to the driver stage.
Replace the lower CCS transistor with a complementary VAS. It looks exactly the same if you travel from +ve rail to -ve rail.