Strictly speaking they don't need to be. However you'll end up with a lower DC offset if you match Q1 to Q2, & Q3 to Q4. They don't need to match to each other though (IE Q1/2 to 3/4).
E2A:- It always helps if the transistors are kept at the same temperature. Bonding Q1 to Q2 & Q3 to Q4 will result in a more stable DC offset over all temperature gradients.
Bests, Mark.
Last edited:
AC coupled the offset shouldn't be a problem if everything is working right - replacing the DC blocking caps with bipolar Al electrolytics, larger values and higher V ratings gives lower measured distortion according to Bateman, Self - and the distortion is at very low frequencies - Al electrolytic cap distortion is unmeasurable away from the corner frequency
if trying for DC coupling then hfe matching input Q would help - and you do want upper and lower pair to have the same (opposing direction) base currents
even with laterals I'd run the VAS hotter since its directly driving the fets - the totem pole/cascade of the VAS Q is curious - even a small heatsink would let you use just one per side - although the cascade does cut the Miller multiplication of Ccb by half - I'm pretty sure better is available in VAS Q selection
and then there are topology changes...
if trying for DC coupling then hfe matching input Q would help - and you do want upper and lower pair to have the same (opposing direction) base currents
even with laterals I'd run the VAS hotter since its directly driving the fets - the totem pole/cascade of the VAS Q is curious - even a small heatsink would let you use just one per side - although the cascade does cut the Miller multiplication of Ccb by half - I'm pretty sure better is available in VAS Q selection
and then there are topology changes...
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.