• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

SA-8800 protection mode help

petehall347 said:
i know its not critical but it should however be present . q1 should still have been off .
Click to expand...
Yes, the key spec here is Q1 should be off. As long as the base voltage is more negative than about +0.35vdc Q1 will be in cutoff (unless it is bad/leaky CE).
 
Absolute max Ic for the 2SC945 is 100mA according to the datasheet. If C24 charges to the +16vdc, that zener D12 and the BE junction of Q2 limits it to, the Ic of Q1 instantaneous current flow through R12 can be as high as 160mA. This will eventually damage Q1 (if 2SC945). Markthefixer advises a KSC2383 for this current dump application because it can handle the peak current better.
 
@petehall347

Removed Q1 for testing. Measured resistance from B to E and B to C

B(-) to E (+): OL
B (-) to C (+): OL
B (+) to E (-): 100kohm <-- these two are probably too high
B (+) to C (-): 100kohm
Tested BAD

D13: removed and measured 0.567V forward voltage - probably ok. Reinstalled it.
C22 - removed and measured: 1.7uF. Spec'd as 2.2uF/50V so probably ok? Reinstalled it.
Measured Anode D13 to gnd: -99mVDC (with Q1 removed)

@merlynski
Cathode D13 to gnd: 5.45VAC
Anode D5 to gnd: 37.25VAC

Anode D13 to gnd: -99mVDC
C7+ to gnd: 50.8VDC

Looks like Q1 definitely needs to be replaced... I'll start a mouser order. Might as well replace Q2 while I'm at it...I'll grab KSC2383s for Q1,3 and 4, and KSC2690 for Q2!
 
Taking voltages and resistance measurements on an HP34401A benchtop unit, diode measurements on an cheapo innova 3320 handheld, and capacitor measurements with a Mastertech MS8910 tester that's meant for SMD components, but works well enough with leaded caps too.
 
I guess now that I look at it, I can do the diode measurements on the HP benchtop unit as well!

Similar result on D13 - looks ok.
 
i was wondering about the voltage measurement on d13 with it having little smoothing . just 1/4 wave rectified and the 2.2uf cap . i would think your meter should cope with it .
 
Q3 and Q4 are the dc detectors for the amp outputs, they don't need the robustness of Q1, because they will almost never get triggered, unless the amp fails. I would replace them with 512-KSC945CYTA for the higher gain and speed. NOTE: Pay attention to the exact part number you are ordering, the 512-KSC945CYTA transistor has a CENTER COLLECTOR (C) leg, the other versions do not have a center collector leg.
 
petehall347 said:
i was wondering about the voltage measurement on d13 with it having little smoothing . just 1/4 wave rectified and the 2.2uf cap . i would think your meter should cope with it .
Click to expand...
The meter will read an average voltage, unless it is a TRUE RMS meter. The circuit itself does not need a large current drive, as its job is to keep the Q1 base from being forward biased, with only a 100kΩ pull up.

alexhastings said:
D13: removed and measured 0.567V forward voltage - probably ok. Reinstalled it.
C22 - removed and measured: 1.7uF. Spec'd as 2.2uF/50V so probably ok? Reinstalled it.
Measured Anode D13 to gnd: -99mVDC (with Q1 removed)
Click to expand...
A 2.2µF 20% cap is on the edge of spec at about 1.7µF, and if it is leaky or high ESR it may not be charging and staying negative enough. SM shows it as a CEANL, or low leakage. If you are going to make an order anyway order a film replacement, that should resolve some of the possible issues. 505-MKS22.2/50/5 is a 2.2µF, 50v 5% film cap I have used to replace low leakage electrolytics.
 
Sounds good - generally, are the CEANL caps the only ones I should consider subbing in film caps for? I'll buy replacement for the other electrolytics on the power supply board while I'm in Mouser (my experience with 40 year old electrolytics isn't great!). In the mean time, I may be able to find an oscope (I'm a mostly working from home EE right now) to take a look at the D13 voltage... I'll take a look at what I can grab later this week.
 
If you don't replace with film, at least use a low leakage electrolytic (low noise). I replaced all the electrolytics (including tantalums) in the signal path on my SX-1050 I am refurbishing, with film caps, up to 10µF. It took a bit of creative mounting to get the 10µF ones to fit in a couple of places because of size. The 4.7µF were also a bit of a squeeze in a couple places. Some won't go above 2.2µF because of size. The 10µF power supply decoupling caps I replaced with aluminum electrolytics. Peruse a bunch of markthefixer's posts for more insight on film caps.
 
Last edited:
ok replaced Q1 and C22, and the relay kicks in now when I turn it on! I started going thru the adjustments in the service manual - the DC balance voltages register and adjust ok now (I also deoxited the pots while I was waiting for the mouser order). Before I start adjusting the idle currents on the power amp transistors, should I clean and replace the heat sink compound on the 2SA1075s? Curious if there's anything else I should consider doing before adjusting everything and doing an audio test... (I bought replacements for all the electrolytics on the power amp and power supply board, so I can eventually recap it, but I want to make sure it's back to normal first!)
 
I like the idea of clean and refresh the heatsink compound. The heatsink compound on my SX-1050 power supply transistors was dried to dust, and the transistors were loose on the heatsinks. I like to add heatsinks to TO-220 and TO-126 packages that don't have them. Probably overkill, but it can't hurt, does not cost much. Most times you don't even need an insulator.
 
speaking of insulators, I got the heat sink off, and it appears all 4 2SA075s and all 4 2SC2525s have insulators to the sink... are the backs of those actually hot? (yikes!) The datasheets don't really say anything either way... I'll definitely be putting the insulators back!
 
So the amp is working again, and sounds wonderful when it's on! I'm noticing that the time from power-on to relay kick is pretty variable. Usually kicks on in 5secs or so, but sometimes takes upwards of 5-10 minutes! Any ideas what could be causing this?
 
Update - it appears to switch back into protection mode after operating for a minute or two (and then continues to occasionally go in and out) - perhaps other components going bad now that the initial failed components are ok? Q1b voltage is still good (-1.48VDC).
 
Q2, the relay driver, (2SC1384) in the protection circuit is known for becoming flaky as it gets older, replace with 512-KSC2690AYSTU. The leads are a bit fatter, I drill out the pcb by hand with a 1mm drill bit in a pin vise.
 
You must log in or register to reply here.
Back
Top Bottom