Pioneer SX-680: What kills those STK0039's?

dbxdx5

Super Member
A little less than a year ago, I was working on an SX-680 that needed a new module for the right channel. Tracked down an NOS one, popped it in, and the unit was good to go.

Fast forward to today, and the same receiver is back on my bench needing a new module for the left channel. Or at least so it appears based on the blown 5A fuse and these voltage readings, which move with the module when swapped to the right side.

Q2


1 -1.33V

2 -34.5V

9 +34.5V

10 +1.33V


Q1

1 +31.6V

2 -34.6V

9 +34.5V

0 +29.4V

Besides ensuring good thermal conductivity between module and heat sink, as well as checking for any shorts in the speaker wiring, what else can/should be done to prevent module failure?
 
So I guess some of the Chinese copies must actually work, for at least a little while. I have read of failures straight out of the box.
 
I've actually had several pairs of the aftermarket modules work in 780s, and they seem to hold without releasing any magic smoke at 5W continuous with higher peaks, and that's pretty noisy with efficient speakers. Of course, 8Ω only speakers, and only one set at a time. The replacement modules from rcs16 are pretty bullet-proof, but would need to be configured for the 680's Q-C arrangement from the C-S design. I'm thinking a little engineering would make it happen. Then again, modules are available but their quality and durability are generally questionable.
 
I have a 680 and 690 and wouldn't mind upgrading them to modules.
If anyone has done any research into this I'd love to know.
Cheers :thumbsup:
 
Thanks for the replies thus far. To phrase my original question a little differently, are there any "must-check" components before installing a new STK in an SX-680?
 
Thank you Sir.
Would've been nice to see some pics for the "680-690" as they differ slightly to the 780.
Not a big deal though. Sockets will have to go obviously.
Also, unless I missed it too, the bias and offset data.
Looking at the SX-680 schematic I don't see any dc offset adjustment, and the custom module does not add it, it uses the SX-780 (or whatever unit) adjustment, if any. The module does add bias adjustment, which the original STK-0039 packs do not include. Setting bias is discussed in the custom module thread. Should be the same process, as the 680 has the same emitter resistor values as the 780. Good luck. I did use them myself in an SX-780 with excellent results.
 
Last edited:
Thanks for the replies thus far. To phrase my original question a little differently, are there any "must-check" components before installing a new STK in an SX-680?

THe sx-680 amp is a bit messy due to the tone controls being in the feedback loop.

Q1 wants +/- 1.3v at pins 1 and 0

pin 0, +1.3v at the intended pin is hard pulled (no transistor) down from the +42v supply through R311 (2.2k) and R313 (4.7k)
pin 1 has to be pulled by enough current to set -1.3v through R317 (100) Q303 and R345 (100) to -33.3 volts.
Thus either q303 (2sc1438) is dead or it isn't being driven on it's base connection.
a ksc1845 should work here.

Think of this analogy.

Two railroad rails (the +42v and -33.3v)
module q1 is a block of wood with 2 screw eyes on opposite sides.
2 springs, one spring between each screw eye and a rail.
one spring is R313 and is unvarying
the other spring is the transistor Q303 and we can control it's pull...

Q303 's pull is increased or lessened to adjust the position of the wooden block to a point in between the rails,
working against the fixed strength of the other spring R313.
since Q303 "snapped" = sproing >>> the block blows on over to the rail.
 
Mark, thanks for taking the time to share the information and explanation. The analogy helped a lot. I'll take a look at Q303.
 
From MTF, a rare and welcomed analogy (Most are for the engineers :D). Made my crappy day better.
 
Back
Top Bottom