Requirements for cascode VAS transistors

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Excellent choice.

Another good option is the KSC1845 - particularly if you can get the higher grades - since it has slightly better Cob than the BC550C.
the cascoded Cob is much less of an issue compared to an uncascoded Cob which sees the varying Vce and thus a varying capacitance (miller capacitance multiplier).
The bc550C when cascoded will be better than an ultra low Cob device that is used uncascoded.

I don't know what effect those options will have on slew rate. A good investigation for the LTspice experts.
 
Vz,
looking at your left "standard" Hawksford, I can see led+led as the cascode voltage from emitter of Q6 to base of Q7. This is ~6*Vbe
Comparing to the right diagram, I see ~3*Vbe from the emitter of Q7 to the base of Q8
That means the left Q6Vce = ~5Vbe whereas the right Q7Vce = ~2Vbe

Does that get the best performance out of the VAS transistor?
Is there any benefit to performance in changing D12 and/or D9 to double diodes to move Q7 further from semi saturation operation?
Or changing D9 to a red LED?
 
Last edited:
Vz,
looking at your left "standard" Hawksford, I can see led+led as the cascode voltage from emitter of Q6 to base of Q7. This is ~6*Vbe
Comparing to the right diagram, I see ~3*Vbe from the emitter of Q7 to the base of Q8
That means the left Q6Vce = ~5Vbe whereas the right Q7Vce = ~2Vbe

Does that get the best performance out of the VAS transistor?
Is there any benefit to performance in changing D12 and/or D9 to double diodes to move Q7 further from semi saturation operation?
Or changing D9 to a red LED?

In fact, on the left picture, those two red LEDs set rather high voltage reference, required for particular design. It will work even with one LED or 2 x 2n4148 diodes (BC550/560 work fine at very low voltages, close to zero CE).

With regards to the right picture - D7, D12 are parts of the enhanced Wilson mirror, so they must provide 1 x Vbe each. D9 actually sets the Vce (and Vbe) for Q7 - this one may be adjusted - more diodes, LEDs, etc., depending on how much voltage you like for Q7. The coolest bit in this design is Q15 - together with D9, it plays the same role, as D10, D12 on the left picture, but doing it with better precision.

Attached is its clipping behavior.
 

Attachments

  • Clipping.png
    Clipping.png
    28.8 KB · Views: 538
We hard to find good through hole devices for VAS as today, most of them change to SMD type. There is only Zetex and their product range that have few necessary requirements. However, the Cob of their transistors seems a bit higher than others.

I'm wondering if i use ZTX694B for EFA current mirror, is that good?
D. Self recommend using high gain transistor to get high DC precision. ZTX694B and its relative that have very high gain (400 as minimum).
 
ZTX694b was specified in a HiFi World ClassA design back in the late 1990s.
It is a very good device where it's 9pF of Cob is not a concern.
But not as good as 2 to 3pF devices for VAS duty.
It's hFE is a bit non linear at lower Ic.
For uncascoded VAS, BC546C (Fairchild version) will perform much better than ZTX694B if BC can handle more power dissipation, since the Cob is 6pF and very straight of hFe graph. Right?
 
Can you please tell what transistor it is?



Perhaps a misunderstanding but, what does "higher grade" means in the context of "slightly better Cob", ie. what qualities are we looking for in order to discerna a KSC1845 with "slightly better Cob" from the inferior types of KSC1845?

The transistor I have mentioned is Russian KT630A. It is much better than 2N1711.
I think that guy wanted to mentioned high-hFe grade of KSC1845 like F or U version. And he wanted to tell that Cob of KSC1845 is lower than BC550.
 
Cascoded structure, like Hawksford's cascode (left picture) will show better results anyway, including rather clean clipping.

The right structure is my modification of the Hawksford's cascode - more complicated, but even more linear, and showing a perfect soft clipping (note, as shown - it's current-driven).

I normally use TO-126 device as a cascode transistor - more flexibility when working with higher (+/-50...75V) rails in terms of power dissipation.

3503/1381 are excellent, however MJE340/350 will also work well enough here.
Wow, it is an eye-opening about better topology. :eek:
My knowledge is so limited :(
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.