Looking to build the Nelson Pass AB100 which uses TIP147 / TIP142 Darlington Complementary Silicon Power Transistors.
How important is it to match transistors?
What parameter should I be matching and how close should they be?
By what means do I take the measurements?
Any other hints about transistor matching?
Thank you,
David.
How important is it to match transistors?
What parameter should I be matching and how close should they be?
By what means do I take the measurements?
Any other hints about transistor matching?
Thank you,
David.
H
HAYK
Ask moderator to transfer your post here.
https://www.diyaudio.com/community/threads/ab100-class-ab-power-amplifier.227503/
https://www.diyaudio.com/community/threads/ab100-class-ab-power-amplifier.227503/
One usually does not try to match NPNs to PNPs for a push-pull ops. One can match like polarity if used in parallel for equal current sharing or balance is achieved
Thanks for starting this thread. I need to buy Darlingtons for a so-called "five-device" amp board (low power but very realistic). TIP122/127; TIP142/147.
Yes,
you could try,
see what the differences are in Vbe (using diode test function on DMM) and if you have more than one pair to try, you could try matching for Vbe's.
You can also compare THD measurements manybe other tests to determine how much matching really matters.
You want close to zero DC offset and close matching of collector currents is your ultamate goal.
you could try,
see what the differences are in Vbe (using diode test function on DMM) and if you have more than one pair to try, you could try matching for Vbe's.
You can also compare THD measurements manybe other tests to determine how much matching really matters.
You want close to zero DC offset and close matching of collector currents is your ultamate goal.
For this TYPE of transistor (bog standard epitaxial base) the beta curve vs. Ic is different for NPN and PNP. Getting a match over a useful range is impossible. Being darlington, that problem is amplified compared to a single. Even the vbe vs. Ic curve is different for NPN and PNP, too. The planar types and triple diffused can be made to match, but to get those in a darlington you need to step up to expensive, hard to source in some parts of the world devices from Sanken.
Who better to ask than Nelson himself for the AB100 at :
https://www.diyaudio.com/community/...b-power-amplifier.227503/page-36#post-5930691
Patrick
I do all the time on the few occasions I use BJTs.
I can get good matching for the likes of 2SC4883/ 2SA1859, or 2SC4886/2SA1860.
Darlingtons are difficult for N-P match, even with Sankens.
So best to make your own Darlington.
That in any case, especially if you want to use low value emitter resistors.
Patrick
Can we use Sanken STD-03P/N. pair as ordinary darlington transistors , not use the temperature sensory doodle pin
You need to connect the compensation diodes with a resistor to be able to switch off the driver BJTs.
They are obsolete anyway.
Patrick
They are obsolete anyway.
Patrick
My guess just a diff pair, boostrapped vas and a pair of darlingtons. Not hard to figure out on your own.
Such designs are usually limited enough by other performance factors that trying to match output transistors is like raising sea level by ******* in it. That doesn’t stop one from trying, however.
Such designs are usually limited enough by other performance factors that trying to match output transistors is like raising sea level by ******* in it. That doesn’t stop one from trying, however.
Well the last TIP142/7 amp I just tried to build (just a couple weeks ago) went into massive cross conduction and blew itself to smithereens when fed from a dirty digital source. Simply changing them out for Sanken audio darlingtons (on a new board) worked propertly the first time, by only adding the flyback diodes that were no longer there…..