Quad 522 equivalent transistors

Jump to Latest
#3
No, the was the output transistors (3 of them) that fails. But I wanted to be sure that not other transistors was bad. I desoldered and tested all the transistors. And like you say all other transistors was ok. Now I want to replace with new ones ...
 
#8
This subject transistor replacement isn't exhausted yet. I recently had occasion to replace some burnt-out transistors (40872 with BD244C and 2SD424 with MJ15033) in the output stage of one of my Quad 405-1 amplifiers (serial number 35092; PCB M12368 ISS 9). The repair worked, but the smell of burnt phenolic quickly showed a fault: R39 = 10 Ohms in the Zobel network was overheated. I suspected parasitic oscillation and checked some Hfe values at 1mA base current of other transistors from the batch I bought. The MJ15003 came out at 55. Since the original Toshiba 2SD424 R is specified as a selected Hfe range of 40-80 this seemed to be OK.

I then discovered that I had ordered BD244C in error, instead of the BD242C recommended by Quad (and supplied by dada electronics) as a replacement. I removed
the two working 40872 transistors to measure their Hfe, then I measured the Hfe of
several possible replacements for the 40872, in each case these are the averages
of four measurements:

40872 Hfe = 82 RCA (only two, original)
BD244C Hfe = 99 CDIL
BD244C Hfe = 178 ON Semiconductor
BD242C Hfe = 165 ON Semiconductor
TIP42C Hfe = 181 ON Semiconductor

From this it appears that the most suitable replacements are the CDIL (Indian-made) BD244C. These were the transistors I installed that gave the trouble!

As a workaround I installed a 2W 10 Ohm resistor to replace R39, but connected to
the amplifier 0V ground (junction C15/C16) instead of the op amp gound used by Quad on this PCB. Keith Snook has already written about this error, which was corrected by Quad in later PCBs from PCB MI12563 ISS 3. I haven't yet established whether I've cured the problem.

My question to the forum is: Have the transistors now made a higher Hfe? Or is it
that the RCA devices were selected for a lower part of the range? The samples I
have are all consistently in the same range. Jon Finch may have a view on this
since he reported in post #407 https://www.diyaudio.com/forums/solid-state/206460-quad-909-clone-41.html.on selecting devices to avoid parasitic
oscillation.
 
#9
The recommendation for BD242C doesn't originate with Dada. I first saw it on geocities.com on a now vanished Quad site, in a post attributed to Mike Coatham. I've been using them for 15 years in everything from 405 to 909 without problems. The main gain and voltage amplification comes from Tr2, not this one, so the Hfe doesn't need to be large.

Esmond Pitt
Dada Electronics Australia
 
#11
ENCR said:
My question to the forum is: Have the transistors now made a higher Hfe? Or is it
that the RCA devices were selected for a lower part of the range? The samples I
have are all consistently in the same range. Jon Finch may have a view on this
since he reported in post #407 https://www.diyaudio.com/forums/solid-state/206460-quad-909-clone-41.html. selecting devices to avoid parasitic
oscillation.
Click to expand...
Assuming your question refers to me, I'd suggest that maintaining the specifications of semis will be done differently from one manufacturer to another and the processes themselves seem to be always in transition from one state to another in order to meet market and regulatory requirements. Products that were interchangeable in their recommended fields of application such as power switching, may no longer perform as well or consistently in others any more, such as audio or other forms of linear operation. Meantime, some changes may be indeed be improvements, so if you're trying out substitutes, take care to segregate and keep records of your stock and also any replacements that are already fitted.

Add to that, 40972/2N6111 was introduced some 50 years ago and processes have changed markedly over that time, to remain competitive as much as improve the reliability and performance of other, new products. This just means that RCA power semis that were used in the early Quad models, may no longer be identical to those produced by other fabs and all will probably be using different, modern and cheaper processes anyway. Those who copy it or otherwise compete with products, may meet the basic specs too but could have other issues with Ft, hFE and linearity, since many medium size power transistors were developed for power switching and applications where linearity is generally undesirable.

A leading parameter for susceptibility to oscillation, by the way, is the transistor's Ft; the gain-bandwidth product. hFE is just DC gain and whilst these are clearly related in typical small-signal transistors, it may not necessarily be so for large, high current types.
 
You must log in or register to reply here.
Top Bottom