• 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

SX950 restoration issue

I
merlynski said:
Monitor the dcv at the + side of C14. That should be completely stable there. Q9 and Q10 form a capacitance multiplier to kill ripple on their output. Somehow you are getting the opposite effect. Did you change any bulbs to LEDs on the pins 2 & 3 circuit? (AM, FM, Phonos, Aux, Stereo lamp).
Click to expand...
I know your schematic is fuzzy ;). When you asked about C17, for Q8 and Q9, were you referring to C12? C17 is between Q1 and Q3. The voltage on pins 4 & 5 were stable. I will check dcv at C14. All the lamps on the input selector switchs are oem.
 
jboerger3 said:
I

I know your schematic is fuzzy ;). When you asked about C17, for Q8 and Q9, were you referring to C12? C17 is between Q1 and Q3. The voltage on pins 4 & 5 were stable. I will check dcv at C14. All the lamps on the input selector switchs are oem.
Click to expand...
Yup, looking closer it IS C12. Thank you! When these problems arise I edit the schematic in Paint as they are noted.
 
Love to hear that relay click. DCV @ + side of C12 is steady at 21.2. DCV @ + side of C14 is steady at 9.9. Pins 2&3 still moving and pins 4&5 stable at 20.1vdc. C12 and c14 are examples of components that were changed after publication of the service manual. C12 is 2,200mF and C14 is 47mF
 
My schematic shows C14 being 220µF.
Check collector of Q9, + of C13, and base of Q9 for stable dcv.
 
Base of Q9 sb 6vdc, 1.27v is a problem. Look for an open trace between the emitter of Q10 and the base of Q9.
 
This is embarrassing. Q10 was installed incorrectly. Put in a new 2383 (correctly). Pins 2&3 have 8.9vdc. The collector on Q9 is 9.1 and the base is 9.5. Now getting 20.5 on pins 4 &5. All readings pretty stable.

duh
 
We have all been embarrassed that way before :rolleyes:.
Pins 2 and 3 should be 5.4vdc, you may have damaged something else (Q9, circle of failure). Pins 4 and 5 sb +13.5 vdc. Check that zener again (ZD11) I bet it is fried again.
You may have excessive voltage on C14 now, and damaged it, measure before changing anything else.
The +13.5v supply feeds the tuner muting circuit, don't leave excessive voltage on there long.
 
M'ski. You nailed it on the zener. Replaced it and pins 2&3 are steady at 5.5vdc and pins 4&5 are steady @13.0! All other readings just under spec under the dim bulb. I feel better. I guess it is on to the power amp?
 
You have now been around the power supply circle of failure for several laps. Good news that it is now working :thumbsup:.
Power amps have the same tendency to take out closely dc coupled components. And output transistors are much more expensive. When an output stage fries it often takes out the driver transistor as well, and several resistors when it lets out the magic smoke. Watch orientation carefully. The DBT is your friend when changing parts in a power amp.
:bigok:
 
I will go over all the transistor installs, in the amp board, before doing anything else. It appears that the right channel was working. So, I will take a good look at the left. Thank you! Stay tuned.
 
Since my last posting, I got out the magnifying glass to check orientation of all the transistors on all the visible boards (power supply, power amp, protection and phono boards). I adjusted the offset and bios, listened to some compressed tunes via an echo dot plumbed through the aux input (sounded good) for a while and at some point, fried the 14v zener again. I took a few days off before replacing the zener and all the transistors on the power board except for Q’s 2,4,6. The negative circuit has been rock solid through throughout this process.

Q1 and Q8 each had one pad missing and the corresponding traces had lifted. I watched a few youtube videos and cut a couple of legs from a pulled 2073, scraped the green from the subject traces and soldered in the legs. Doing that kind of repair on a vertical board is a bit of a pain. But, I am happy with how it turned out as both components appear to be stable.

Yesterday, I powered up and got reasonable voltages across the board, but, after some warm up, the lamp circuit started acting up and voltages moved up from the initial 5.8vdc to 11.9vdc. I set the offset, which was a rock stable 0vdc on both channels on the mV setting of the Fluke 117. Multi-turns make such a difference. The bios floats a bit, but, I set both channels at 20.5vdc. If I try to reduce it anymore, the meter jumps down to 18vdc. I played tunes through the aux again and likely fried the aux bulb on the front due to the pins 2&3 being at 10.8vdc. I have some replacements on the way and wanted to be sure that both channels were working.

Today, I got the following measurements after the unit was warmed up:
Pin 1 SB 0 vdc 0
Pin 2 SB 5.4 vdc 11.6
Pin 3 SB 5.4 vdc 11.6
Pin 4 SB 13.9 vdc 13.3
Pin 5 SB 13.9 vdc 13.3
Pin 6 SB 51.5 vdc 49.2
Pin 7 SB 51.5 vdc 49.2
Pin 8 SB 36.5 vdc 35.4
Pin 9 SB 36.5 vdc 35.4
Pin 10 SB 36.5 vdc 35.4
Pin 11 SB 28.0 vdc 27.3
Pin 12 SB 0 vdc 0
Pin 13 SB -13.5 vdc (13.4)
Pin 14 SB -51.5 vdc (51.3)
Pin 15 SB -19 vdc (19.0)

Pin 22 7.5 vac 8.1
Pin 23 7.5 vac 8.1
Pin 29 7.5 vac 8.1

VDC
B C E

Q1 49.8 64.9 49.25
Q3 14.3 49.6 13.7
Q5 36 39.7 35.4
Q7 27.9 34 27.3
Q8 13.9 26.4 13.3
Q9 6.1 26.4 11.2
Q10 6.4 26.4 6.1


Q1-Q8 seem to be functioning properly, although they all seem to be just a little off spec - mostly a little low. I don’t know enough about transistors to speculate on what is going on at the collector junction of Q9 and Q10 - but I see it is a problem. I’ve changed Q9 and Q10 a few times now and I seem to have the same issue of inconsistent voltage at Pins 2&3 each time. Should I pull and test any of the upstream components? I replaced all the diodes and caps on this board when I started the project a few weeks back.

Another thing to note - Q9 emitter voltage (and thus pins 2&3) tends to wander as the unit warms up. i.e. it might be 5.8vdc at startup and slowly move up to 14vdc. And when it is high, it might start falling back. I listened through headphones, for about 15 min (I don’t want to fry the speaker lamp), before taking the above measurements. I shut it down, went upstairs, and started this posting. I wanted to make sure I had everything lined up right in the BCE columns (I know typos hurt) so I went downstairs and started it up. Before checking Q9 and Q10, I checked pins 2&3. They were 5.8vcd and heading north. Vdc for BCE on Q9 was 6.0, 26.5 and 7.2 respectively and 6.4, 26.5 and 6.04 on Q10.
 
The collector voltage of Q9 and Q10 is too high. Verify R20 is 100Ω, brown black brown. Its job is to drop some of the voltage getting to Q9, Q10 so they don't have to work so hard. It should have about 7.3vdc across it. Is Q9 getting real hot? Too hot to touch? From your measurements R20 has no voltage drop. As long as Q9 and Q10 overheat they will keep failing and killing the zener. Check the resistance of R21, 3.9kΩ, orange white red.
 
Last edited:
Check for a solder bridge near Q8 collector that maybe causing a short across R20. Measure pad to pad across R20 for a short.
 
I looked at the traces between Q8 and R20. Close, but no cigar. Q8's collector was one of the two repairs that I made. No touching, but, I took some excess tin off that joint anyway. This is where it gets interesting. I pulled one of the legs of R20 off the board. It is a big boy with nylon standoffs and the body is wrapped in cloth. It tested at 97.9 ohms. But, when the unit was running, I checked voltage at both ends of R20 to ground. Same voltage at each end (26.1vdc). Double checked pad to pad - it goes from +.1 to -.1 I thought that resistors either go open circuit or increase in resistance when they fail? Is this one of those deals where the DMM doesn't put enough current through the component relative to actual usage?
 
Last edited:
Resistor rarely fail by losing resistance or shorting, but it is not completely unknown.
jboerger3 said:
it goes from +.1 to -.1
Click to expand...
Is that the voltage across the pads? Or resistance? Lift (unsolder) one end of R20, and measure pad to pad again. It seems to me that you have a short on the board across R20 pads.
 
I pulled the incoming leg of the resistor and tested resistance end to end with the DMM. It tested at 97.9 ohm. I put it back in circuit. I turned on the unit and measured vdc from each end to ground. It was 26.1vdc on both ends. When I tested vdc across the pads, I got a negligible difference that probably was caused by how I held the leads on the pads. I took a long look at the traces to see if there were any shorts and checked for continuity between traces/joints. All I found was a bit of flux that I needed to clean off. It would be nice if someone was to produce new power supply boards for the 950 like they did with the 1010. But, the 950 does not have the same cache' Hopefully, once it get this unit humming, I won't need to do anything else on it during my last 25 years.
 
Last edited:
Did you measure resistance from pad to pad with the resistor removed? That is how to tell if there is a short across from pad to pad. Lift one end of the resistor so it is out of the circuit and measure resistance from pad to pad. Also, with the resistor out of the circuit, power up and measure the DCV directly from pad to pad, not from ground to each pad. Post results. The resistance of the resistor tells you the resistor is good, but it does not tell you anything about the circuit on the PCB, that has to be tested separately, without the resistor in place.......
 
You must log in or register to reply here.
Back
Top Bottom