• 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

SX-828, won't come out of protection.

I just bought a heathkit IT-18 in circuit transistor tester on the auction site. That should help me eliminate a lot of things. I have most of the part here for a rebuild of the unit, but I just have a feeling that it's a mistake to replace everything without it being in a know good state beforehand. According to hifi engine, the 828 has 6 x FET, 4 x IC, 55 x transistors, 36 x diodes. That's a lot of replacing of transistors that may not be faulty and it won't help me identify where to look for anything else that may have been damaged around it.
 
Ok, I have gotten back to this a little bit. I have not powered it on since the smoke happened. Here is the status.


Power Supply

Changed all electrolytic caps.

Spot checked, in circuit, resistors with DMM, plus visual inspection.

Changed Q1 and Q3 to 512-KSC2073TU and refit heatsinks with metal screws, isolation washers and mica. Continuity test shows no connection from collector to heatsink.

Changed D5 to 1N4004 (changed due to low voltage on pin 17, figured it couldn't hurt)


All old transistors tested good after removal. All transistors than had a different pinout were taken into account and inserted according to the schematic and the screen print on the board.


Protection Board

Changed all electrolytic caps.

Spot checked, in circuit, resistors with DMM, plus visual inspection.

Diodes tested with DMM in circuit.

Changed Q1 and Q2 to 512-KSC1845FTA

Changed Q5 to 512-KSA992FBU

Changed Q6 to 512-KSD1616AGBU

Changed Q7 to 512-KSC2690AYS


All old transistors tested good after removal. All transistors than had a different pinout were taken into account and inserted according to the schematic and the screen print on the board.


Main Amp

Changed all electrolytic caps.

Spot checked, in circuit, resistors with DMM, plus visual inspection. Compared to the side that wasn't railed and they seemed to match.

Changed out the 2SA726 (Q2, Q4) to KSA992FBU (matched hFE really close)

Changed out the 2SC1124 (Q6, Q8) to KSC2690AYS

Changed out the 1S1212 (D5) to 1N4148. The old diode tested fine out of circuit.

Tested Q14 and Q16 out of circuit and they tested good.


All old transistors tested good after removal. All transistors than had a different pinout were taken into account and inserted according to the schematic and the screen print on the board.

_________________________________________________________


So I'm not sure what I've accomplished so far other than swapping some parts. But before I proceed I have some questions.


Question #1 Now I'm ready to remove and replace 2SC484 (Q10) with KSC2073TU and 2SA484 (Q12) with KSA940TU. Do these need to be hfe matched?


Question #2 If I remove the Pre In/Out jumpers on the back of the unit, does this electrically isolate the AF board and the Head Amp board? I haven't done anything to these and would like to get the Power Supply, Protection and Main Amp boards working before I get into those and don't want there to be any harm going either direction if something isn't correct.


Question #3 So far everything I have replaced has tested good with my new transistor tester (Heathkit IT-18) which tests both gain and leakage (granted off of a D Cell battery). I have no confidence that any of my parts swapping is going to help the railed condition on that side of the Main Amp board and shotgunning this isn't really ideal to me. Could a transistor appear fine out of circuit, but not in? What else could rail that channel?


Question #4 I am confused about Right and Left channel. In the service manual I downloaded from HiFiEngine, pdf pages 50 and 51, It shows pins 12 and 18 on the Main Amp board going to pins 6 and 5 on the Protection board. It also shows pin 8 on the Main Amp board going, through pin 20 and out 19 on the Power Supply board, and then to pin 1 on the Protection board. It labels these as Left Channel on the schematic at the bottom of page 51 for the protection board. Yet at the Pre In/Out jumpers, Pin 2 on the Main amp board is clearly connected to the Right channel and that goes to Pin 13 on the AF board as expected. Is this just a case of it not mattering what is Left and what is Right on the Protection board? I'm really confused by this. Thoughts?


Thanks again to anyone who can help.


Cheers,

Matt
 
I know everyone is getting used to the forum being back up now, so I want to see if anyone has thoughts on this now. I don't want to fire it back up if I'm taking an obviously bad approach.
 
Ok, I'm going to build one tomorrow. Should I leave the pre/main jumpers out? Does that isolate the boards from each other? Also, should I remove Q14 and Q16 while I check voltages if the bulb stays dim?
 
We have progress! I built the dim bulb tester. Plugged it in and turned it on (Forgot to take out Q14 and Q16:oops:) with a 150 Watt bulb and it never got bright. We have power, but no click. I was testing around for some voltages that I remembered being off and all of a sudden, CLICK, relay clicked over and then went right back off. No Smoke and a relay click, this is progress. I still need to do a full accounting of voltages, but had to share my excitement with someone and there isn't anyone home.
 
So the good news, is that this amp is no longer railed. When testing all of the following voltages, it was going in and out of protection, which is far more than it's ever done before. I think these voltages indicate that it's time to do the same transistor changes to the other side now. Let me know what you think.

Power Supply
1. 0
2. +12.6
3. 0
4. +4.5
5. +31.6
6. +33.2
7. +35.2
8. -0.25
9. +44.9
10. -.28, bounces around
11. 0
12. 0
13. 0
14. +39.6
15. -40.5
16. 0
17. -40.2 bounces around
18. 0
19. 10.1mv
20. 10mv
21. low volts/mv, bounces all around
22. low volts/mv, bounces all around
23. 0

Protection
1. +20.5mv
2. -13.32
3. -2.5 volts , bounces around
4. 0
5. +4.9mv
6. +4.2mv
7. bounces around
8. bounces around
9. +8.81
10. +35.3
11. +35.3
12. 0 ( I know this is AC, where should I test this against?)

Main Amp
1. -21, but dips to very low
2. +12mv, dropping
3. 0
4. 0
5. 0
6. 0
7. all over the place
8. -18mv, but bounces around
9. +40.4
10. +40.2
11. low volts, but bounces around
12. 0
13. low volts, but bounces around
14. -0.312
15. bounces around
16. -0.316
17. -0.59
18. -0.339
19. -40.2
20. -40.2
21. +40.2
22. -13.2
23. -39.9

Also, does anyone have an idea of what voltages to expect on the AF Board while it's not connected to the Main Amp board. That is where I am going to go next after the Main amp is settled.

Cheers,
Matt
 
Ok, the other channels transistors have been changed out and the relay clicks after a few seconds and stays that way. I haven't taken voltage readings yet, but I will soon.

I'm going to move onto the AF board and Head Amp recap and transistor replacements before I do an audio test. I'm worried about what the rail voltage at pin 13 on the AF board might have done, so I think it's a safer bet to do the rebuild there before I hook them back together via the jumpers. Wish me luck.
 
I guess I can't dive into the rest of the recap job right away as I forgot to order the caps for the AF Board and Head Amp Board (what is wrong with me, order transistors, but not caps? :dunno:). I have most of them in my spares, but the smaller values like 1.5uf, 2.2uf, 3.3uf and 4.7uf aren't here. I found most of them on mouser, but the 3.3uf/25V that's been recommended here in the past is out of stock. I'm not familiar enough with series numbers to know what a good replacement is. Being that this is on the AF board, I'm assuming this is somewhat critical. Anyone have a suggestion?
 
Its best to get the set running and then hit-up the recap/restore.
Let's hear some sounds and check out the function of the set. You might need to add to the order.
This makes for a good baseline to begin the restore. Know it's working. Amps and PS are solid and operational.
No hurries!
 
Normally, I'd be right with you, but the power amps negative rail was going straight into the preamps output. That means I'd be taking a risk connecting them, right? Is there a way to test the preamps output before hooking the jumpers back to the power amp?
 
Ok, I just went out and checked voltage on the outputs of the AF board and it was in the mv range. So I attached the jumpers and tried some headphones. It was a bit noisy and a couple of crackles, but I was using the FM radio with no antenna, so who know. I guess it's not a bad test. I did tune in a local station. I'll button it up and try some other functions.

Also, that board is full of those old baby blue Sanyo caps that are known problems and C46 took the negative rail from the power amp. It's a 3.3uf rated at 25V and got -37 or so.
 
Last edited:
I re-capped the head amp board with the exception of the 4.7uf caps because I didn't have any. I also changed out all of the 2sc1344 transistors to KSC1845's. No change, so it's looking like the crackle is coming from the AF board. I'm thinking baby blue Sanyo caps, 2SA725's or 2SC1312's are the culprits. I'm going to have to wait for some more parts before I can get that one complete.

Also note, that the controls have all been bathed in deoxit and worked over a bunch of times. So it shouldn't be that either.
 
Being the impatient guy that I am, I went ahead and recapped the AF Board with what I had on hand (half of the caps, none of the baby blue Sanyos) and swapped out the 2SA725's with KSA992's. The loud crackle is gone, so I'm guessing it was one of the 725's. There is still a static, whirring, buzzing sound. I pulled the preamp jumpers and the sound is gone, so it's not in the main amp. It's also present while I have the Aux on, so it can't be the head amp. This must be either the remaining transistors and caps on the AF board. I have the transistors, but the caps are on order (mouser had Elna Silmic II's for all of them). Lack of parts is going to force me to stop for now I'm afraid. I want to change those transistors while I have the caps out to give me some more room.

Also, the relay turns off briefly when I engage the low filter and there is a boom. I've read this is those little blue sanyo caps, so that should be resolved once my mouser order gets here.

I feel like I've mostly been talking to myself in this thread, but I think it's been useful to type this stuff out and re-read everything. I hope it's useful to some other member at some point.
 
Last edited:
I changed out the remaining caps and transistors on the AF board and surprisingly, it's still the same. Big boom when I engage the low filter and some whirring, buzzing kind of sound on one side that it's not effected by change in volume. Time to go read more posts by Markthefixer.
 
I "Thought" I had changed out the rest of the caps. C37 and C38 were hiding from me behind the switch and I didn't even see them. The good news is that I can now verify, without any doubt, that C37 and C38 can cause the boom from using the the low cut switch. This also increases the DC sent out of the AF board and can cause the protection board to cut off sound momentarily.

I think I need to pull the outputs and treat them to new Mica and heatsink compound on there, then adjust offset and bias. Does anyone have a good step by step procedure for this?
 
I found the procedure for setting DC offset and Idle Current in another old thread. If anything is wrong with what I'm posting, someone please correct me. I am putting this information here for other peoples future reference. Using this method, I set everything to spec with it drifting about .1 mV in either direction. Considering how far off they were, I am really happy. And please, attach test clips while power is off. We don't want to short anything out when we're about to cross the finish line.

DC offset
Notes: I did this with the top and bottom of the case off and the receiver standing on it's side. That way I was able to attach the test clips to the pins on the bottom while still having access to the trimmers on the top. One of the Trimmers access is under the cover for the phono transformer socket, you'll need to remove the 3 screws that hold it down to access it. I used clips like these to connect to the pins while I adjusted the trimmers with a small insulated screw driver with tape on all but the very tip.

1. Let the receiver warm up for 15 minutes.
2. Set vol to minimum.
3. Hook DMM red lead to Pin 7 and black lead to ground. Be sure the DMM is set for mV range.
4. Adjust VR1 for zero DCV (offset).
5. Hook DMM red lead to Pin 8 and black lead to ground.
6. Adjust VR2 for zero DCV (offset).

Idle Current
Notes: I used the sockets on the back side of the outputs as test points for this. On mine the Blue wires going to the sockets were connected to Pins 11, 12, 17, and 18. If you are going to do this, be sure you know which wire is which. I used clips like these to connect to the pins while I adjusted the trimmers with a small insulated screw driver with tape on all but the very tip.

1. Let the receiver warm up for 15 minutes. (This wasn't in what I read, but can't hurt)
2. Set vol to minimum. (This wasn't in what I read, but can't hurt)
3. Hook DMM red lead to Pin 17 and black lead to pin 11. Be sure the DMM is set for mV range.
4. Adjust VR3 for 20mV. Continue to watch to be sure it settles at 20mV(mine continued to creep up for a while and I had to lower it back down a bit).
5. Hook DMM red lead to Pin 18 and black lead to pin 12.
6. Adjust VR4 for 20mV. Continue to watch to be sure it settles at 20mV (mine continued to creep up for a while and I had to lower it back down a bit).

With all of that said, I am going to be reassembling pretty soon. Thanks for watching.
 
I swear, this unit has had just about every single SX-828 issue ever posted about on Audiokarma. Now that I've gotten everything adjusted, I decided to hook up something to the aux and give it a try. However, I have just noticed that there is a pretty loud pop on the input selector when switching between each inputs. The FM has the loudest pop. I've resurrected another thread from last year dealing with that issue, but there was no conclusion. I'm having fun doing this, but I swear. This thing has ALL of the problems.
 
You must log in or register to reply here.
Back
Top Bottom