• 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-838 going into protect, some troubleshooting done

The failure progresses at maximum voltage downwards, thus tests will not reveal until way too late, or in circuit.

leave the diodes, get rid of all the 726's and 1451's and do the 3503, yes do the transistors/
 
Cool... Thanks for all the help Mark.

Another thing. I noticed while replacing the caps in C2 and C4 that the old ones were electrolytic (obviously) but that the replacements weren't. That won't be a problem will it?
 
Cool... Thanks for all the help Mark.

Another thing. I noticed while replacing the caps in C2 and C4 that the old ones were electrolytic (obviously) but that the replacements weren't. That won't be a problem will it?

no. it's an advantage, and was intended to be so.
 
Mark,

I finished replacing all caps and transistors (except drivers and outputs) in AWH-031-A for both left and right channels. No diodes were replaced. I've run the amp at low to medium volume for a few hours now with no problems. I think I can call it good now, knock on wood.

I've gone back and tried to find parts for the rest of the unit, based on mattsd's list and the work we did here. I added diode replacements for everything too. All parts were checked for stock. Hopefully this can become a master reference for the SX-838. The only thing not here is the tuner board and that is probably it's own can of worms.

Highlighting/Color reference:

No color/highlight: Vetted part, OK to order.
Italics: Found elsewhere on AK, hasn't received a MTF stamp of approval
Bold: Part I selected, needs a lot of scrutiny

_______________________________
Chassis

C1: 8,000uF 50V
C2: 8,000uF 50V
C3: 1,000uF 63V ACH-001-C: 647-UPW1J102MHD 1,000uF 63v

_______________________________
Equalizer Amp Board (AWF-011-0)

AWF-011-0: C1: 1uF 25v CSSA: 647-UKL1H010KDDANA 1uF 50v
AWF-011-0: C2: 1uF 25v CSSA: 647-UKL1H010KDDANA 1uF 50v
AWF-011-0: C7: 330uF 6.3v CEA: 647-UPW1A331MPD 330uF 10v
AWF-011-0: C8: 330uF 6.3v CEA: 647-UPW1A331MPD 330uF 10v
AWF-011-0: C11: 3.3uF 25v CEANL: 647-UKL1H3R3KDDANA 3.3uF 50v
AWF-011-0: C12: 3.3uF 25v CEANL: 647-UKL1H3R3KDDANA 3.3uF 50v
AWF-011-0: C17: 100uF 25v CEA: 647-UPW1E101MED 100uF 25v
AWF-011-0: C18: 220uF 16v CEA: 647-UPW1V221MPD 220uF 35v

Q1: 2SA725: 512-KSA992FBU
Q2: 2SA725: 512-KSA992FBU
Q3: 2SC1313: 512-KSC1845FTA
Q4: 2SC1313: 512-KSC1845FTA
Q5: 2SC1313: 512-KSC1845FTA
Q6: 2SC1313: 512-KSC1845FTA

_______________________________
Microphone Amp Board (AWM-064-0)

AWM-064-0: C2: 220uF 35v CEA: 647-UPW1V221MPD 220uF 35v
AWM-064-0: C3: .47uF 16v CSSA: 667-ECQ-V1H474JL3 .47uF 50v
AWM-064-0: C6: 220uF 6.3v CEA: 647-UPW1C221MPD 220uF 16v
AWM-064-0: C7: .47uF 50v CEANL: 667-ECQ-V1H474JL3 .47uF 50v

Q1: 2SC1344: 512-KSC1845FTA
Q2: 2SC1344: 512-KSC1845FTA

_______________________________
Control Amp Board (AWG-029-0)

AWG-029-0: C3: 47uF 35v CEA: 647-UPW1V470MED 47uF 35v
AWG-029-0: C4: 47uF 35v CEA: 647-UPW1V470MED 47uF 35v
AWG-029-0: C7: 10uF 25v CSZA: 647-UKL1E100KDDANA 10uF 25v
AWG-029-0: C8: 10uF 25v CSZA: 647-UKL1E100KDDANA 10uF 25v
AWG-029-0: C13: 100uF 10v CEA: 647-UPW1C101MED 100uF 16v
AWG-029-0: C14: 100uF 10v CEA: 647-UPW1C101MED 100uF 16v
AWG-029-0: C15: 4.7uF 25v CEANL: 647-UKL1H4R7KDDANA 4.7uF 50v
AWG-029-0: C16: 4.7uF 25v CEANL: 647-UKL1H4R7KDDANA 4.7uF 50v
AWG-029-0: C29: 10uF 25v CSZA: 647-UKL1E100KDDANA 10uF 25v
AWG-029-0: C30: 10uF 25v CSZA: 647-UKL1E100KDDANA 10uF 25v
AWG-029-0: C31: 10uF 25v CSZA: 647-UKL1E100KDDANA 10uF 25v
AWG-029-0: C32: 10uF 25v CSZA: 647-UKL1E100KDDANA 10uF 25v
AWG-029-0: C35: 100uF 6.3v CEA: 647-UPW1C101MED 100uF 16v
AWG-029-0: C36: 100uF 6.3v CEA: 647-UPW1C101MED 100uF 16v
AWG-029-0: C37: 47uF 35v CEA: 647-UPW1V470MED 47uF 35v
AWG-029-0: C38: 47uF 35v CEA: 647-UPW1V470MED 47uF 35v
AWG-029-0: C39: 2.2uF 25v CSZA: 647-UKL1H2R2KDDANA 2.2uF 50v
AWG-029-0: C40: 2.2uF 25v CSZA: 647-UKL1H2R2KDDANA 2.2uF 50v

Q3: 2SA725: 512-KSA992FBU
Q4: 2SA725: 512-KSA992FBU
Q5: 2SC1312: 512-KSC1845FTA
Q6: 2SC1312: 512-KSC1845FTA

----
_______________________________
Power Amp Board (AWH-031-A)

AWH-031-A: C1: 1uF 25v CSSA: 667-ECQ-V1H105JL3 1uF 50v
AWH-031-A: C2: 1uF 25v CSSA: 667-ECQ-V1H105JL3 1uF 50v
AWH-031-A: C3: .47uF 25v CSSA: 667-ECQ-V1H474JL3 .47uF 50v
AWH-031-A: C4: .47uF 25v CSSA: 667-ECQ-V1H474JL3 .47uF 50v
AWH-031-A: C7: 33uF 16v CEA: 647-UPW1C330MDD 33uF 16v
AWH-031-A: C8: 33uF 16v CEA: 647-UPW1C330MDD 33uF 16v
AWH-031-A: C11: 330uF 10v CEA: 647-UPW1A331MPD 330uF 10v
AWH-031-A: C12: 330uF 10v CEA: 647-UPW1A331MPD 330uF 10v
AWH-031-A: C13: 100uF 50v CEA: 647-UPW1J101MPD 100uF 63v
AWH-031-A: C14: 100uF 50v CEA: 647-UPW1J101MPD 100uF 63v

Q1: 2SA726: 512-KSA992FBU
Q2: 2SA726: 512-KSA992FBU
Q3: 2SA726: 512-KSA992FBU
Q4: 2SA726: 512-KSA992FBU
Q5: 2SC1451: 512-KSC3503DSTU
Q6: 2SC1451: 512-KSC3503DSTU
Q7: 2SD358: 512-KSC2073TU
Q8: 2SD358: 512-KSC2073TU
Q9: 2SB528: 512-KSA940TU
Q10: 2SB528: 512-KSA940TU
Q11: 2SC1079: 863-MJ21194G
Q12: 2SC1079: 863-MJ21194G
Q13: 2SA679: 863-MJ21193G
Q14: 2SA679: 863-MJ21193G

D1: 1S2473: 512-1N4148
D2: STV-4H: 621-UF1004 x4 or 512-1N3595 x3 or 512-1N4148 x3
D3: STV-4H: 621-UF1004 x4 or 512-1N3595 x3 or 512-1N4148 x3
D4: 1S2473: 512-1N4148
D5: 1S2473: 512-1N4148


_______________________________
Protection Board (AWM-025-C)

AWM-025-C1: .22uF 10v CSSA: 667-ECQ-V1H224JL .22uF 50v
AWM-025-C2: .22uF 10v CSSA: 667-ECQ-V1H224JL .22uF 50v
AWM-025-C3: 330uF 6.3v CEA: 647-UPW1A331MPD 330uF 10v
AWM-025-C4: 330uF 6.3v CEA: 647-UPW1A331MPD 330uF 10v
AWM-025-C5: 4.7uF 25v CEA: 647-UPW1E4R7MDD 4.7uF 25v
AWM-025-C6: 100uF 16v CEA: 647-UPW1E101MED 100uF 25v

Q1: 2SC869: 512-KSC2310YBU (nothing available in KSC2310) or 512-KSC1845FTA
Q2: 2SC869: 512-KSC2310YBU (nothing available in KSC2310) or 512-KSC1845FTA
Q3: 2SC945: leave original
Q4: 2SC945: leave original
Q5: 2SA733: leave original
Q6: 2SC945: 512-KSD1616ALBU (I had 512-KSD1616ALTA, doesn't seem to matter)
Q7: 2SC1384: 512-KSC2690AYS
1N4004 flyback diode is added between pins 9 (anode) and 10 (cathode): 512-1N4004

D1: 1S2472: 512-1N4148
D2: 1S2472: 512-1N4148
D3: 1S2472: 512-1N4148
D4: 1S2472: 512-1N4148
D5: 1S2472: 512-1N4148
D6: 1S2473: 512-1N4148
D7: 1S2473: 512-1N4148
D8: 1S2473: 512-1N4148

_______________________________
Power Supply Board (AWR-055-0)

AWR-055-0: C8: 470uF 80v CEA: 647-UHE2A471MHD 470uF 100v
AWR-055-0: C9: 470uF 80v CEA: 647-UHE2A471MHD 470uF 100v
AWR-055-0: C10: 100uF 50v CEA: 647-UPW1J101MPD 100uF 63v
AWR-055-0: C11: 100uF 50v CEA: 647-UPW1J101MPD 100uF 63v
AWR-055-0: C12: 220uF 16v CEA: 647-UPW1V221MPD 220uF 35v or 647-UPW1E221MPD 220uF 25v
AWR-055-0: C13: 2200uF 10v CEA: 647-UPW1C222MHD 2200uF 16v
AWR-055-0: C14: 220uF 50v CEA: 647-UPW1H221MPD 220uF 50v or 647-UPW1J221MPD6 220uF 63v
AWR-055-0: C15: 470uF 16v CEA: 647-UPW1E471MPD6 470uF 25v

Q1: 2SD313-E: 512-KSC2073TU
Q2: 2SD313P E or D: 512-KSC2073TU
Q3: 2SC945-Q or R: 512-KSC2383YTA (512-KSC945LBU end-of-life, but available - many other KSC945 parts available)

D1: SR31AM-8: 512-1N5404
D2: SR31AM-8: 512-1N5404
D3: SR31AM-8: 512-1N5404
D4: SR31AM-8: 512-1N5404
D5: 1S1886: 512-1N4004
D6: 1S1850: 625-BY251GP-E3/73 x2
D7: 1S1885: 512-1N4004
D8: WZ-130: 78-TZX13C
D9: WZ-130: 78-TZX13C
 
change:

The d1616 in the PROTECT circuit DISCHARGES a 100uf capacitor upon turn-off and the d1616 is only MARGINALLY better SOA than the original C945. awm-025 Q6.


Use THE C2383 for q6

512-KSC2383YTA to-92L ecb 160v 1a .9w 50mhz 160-320hfe $0.17
for the protect discharge transistor. It's a much larger die.

awm-025 q6 512-KSC2383YTA
 
NOTE: All of the following has been copied over to a new thread, "SX-838 rebuilding, woofers fluctuating, erratic behavior", because the issues are confusing the two subjects.

I was re-reading thru this thread last nite getting ready for a minor parts upgrade of the power amp board and the protection board on my 838. As I was marking up the power supply board schematic so I could check voltages before and after, I cannot find the +45v position? In an earlier post in this thread the OP mentioned checking +45 and -45 rails, yet I cannot find the +45v point? So, I found 6 measurement points, are there 7?

I need to do this because although I don't go into protection, the power output is "erratic or fluctuating" such that the woofer cones on the test speakers fluctuate in and out after warm-up?

Although I have not decided to do a total recap at this time, I was looking for the trim pot numbers and whether to just change them out or just do the whole thing?
 
Last edited:
yes, do the trimpots!!

sx-838 offset voltage adjust
652-3296P-1-103LF 10Kohms 3296P Multi Turn : 12 3/8" Sealed parallel to pcb

652-3386H-1-101LF 100ohms 3386H Single Turn 3/8" ? 10% parallel to pcb, triangle of pins


The +- 45 (44) volt rails are the BIG capacitors, connected to AWR-064 power supply board pins $(+45.0Vdc) and 5(-45.0Vdc) .

You want to ask about 6 versus 7 (really 8) measurement points, you GOTTA detail which six you are LOOKING at.
AWR-055,-064 pin 4 +45vdc unregulated
AWR-055,-064 pin 5 -45vdc unregulated
AWR-055,-064 pin 7 -45vdc unregulated (a DIFFERENT -45v!!)
AWR-055,-064 pin 9 -13vdc regulated
AWR-055,-064 pin 13 +35.5vdc regulated (don't sweat pins 10, 11, 12 until this voltage is correct)
AWR-055,-064 pin 14 +13vdc regulated
AWR-055,-064 pin 21 8.5vdc unregulated

woofers fluctuating are more likely caused by the DREADED 2SC1451 and 2SA726 transistors in the power amps - if the cones are moving TOGETHER, only THEN does the power supply becomes suspect.

And instabilities enough to push infrasound at protection levels through the amp are more likely due to caps ESPECIALLY the damn CSSA sky blue Sanyo "capsistors" conducting DC they are supposed to BLOCK.

I haven't checked the thread, but does it do this with the pre-out / amp in jumpers pulled, if not - it's a preamp problem rather than a power supply or power amp problem. NOTWITHSTANDING all of this the power amp transistors 2sa726 and 2sc1451 are ticking time bombs that can destroy the output transistors.
 
Last edited:
OK, I need to put another mouser order together, for at a minimum new trim pots, and then maybe the rest of the caps.

I have the replacements for Q1,2,3,4,5,6 already as well as some protection board parts C6, Q7, D8 and the 1n4007 to add across pins 9 and 10.

So far all I have done is released the protection board from it's clips in prep to remove and replace the above. Reading about not removing the wires from the pins makes me think that the amp board, awh-031A cannot be moved such to make it accessible, there just isn't enough available slack from the power outputs transistors. So I need to study that a little more.

I did not test the unit by pulling the pre/main but I did switch to another pair of small woofer speakers with inline .25A fuses. I did not see these woofers moving like the Minimus 7's did?? Perhaps the rubber surrounds of the Min 7's accentuated issues?

The points you added re voltages are simply numbers on the awr-055 schematic; I wrote the values next to the numbers on the schematic.

As I already plan on replacing the ticking timebomb parts, I will not be turning the unit on again until I have done the replacements. I need to set up a new Ayoue desoldering gun I bought so I can proceed, once I resolve removing the amp board!

Note: I seem to recall that before checking power points and across the speaker connection that the amp should warm up for maybe 30 minutes? Well, this hasn't been powered on for 7-8 months. When I did turn it on, with a set of speakers for a load, I could hear the relay? clicking on and off for maybe 4 minutes before it stopped. The right channel fuse was shot, so I replaced it, and powered up again, and let it run FM for about 20 minutes before checking dcv across the speaker connections, which gave me a reading of .004 and .005?

Thanks Mark, once again you are a fountain of knowledge and present it in an understanding manner!
 
Last edited:
And thus I inadvertently enable a thread hijacking... :tears:

Please start a new thread, we WILL be talking a lot.

Re- not undoing wire wraps - the PS board on the SX-1980 is very bad like this, BUT there is enough room to desolder some of the wire wrap pins and this gives PLENTY of slack. Going back together may require heating the pins to get them past any solidified flux blocking the SQUARE hole!!

So, remember - DESOLDER - not unwrap for slack.

and... document document document whatever is taken apart ON paper!!! so it can go back together correctly. memory fails and plays tricks... I have FILES of all my paper notes on everything I do like that - for future reference, by model number. I NEVER toss them.

Helps get you back to that frame of mind and train of thought. Documentation is a "note to your idiot future self" (and I include myself in that group) where nothing is left to chance.
 
Jim; Here's an optional routine for the higher numbered x3x series. As Mark noted, take and make notes like crazy, and a lot of pictures of each step as you move forward. The 838 is very similar to the 737 and 939 as to location and how the amp board and heatsinks are mounted. You should be able to lift out the amp board enough to work on without desoldering any wires.

Remove the screws holding the heatsink to the chassis.

If the STV-3 (stv-4) is mounted on the heatsink, loosen the screw and EXTREMELY CAREFULLY move the assy's away from the heatsink and toward the board.

VERY CAREFULLY Rotate the assembly up and back enough to separate the amp from the heat sink.

Remove the screws holding the board to the heatsink.

Place a towel across the top of the 838 and CAREFULLY rotate the heatsink up and forward (toward the front panel) on to the towel. The amp board will follow along. The solder side of the board is now accessible.

Loosen all tiewraps and twist ties holding wire bundles to the amp board 1st. This will provide slack in the wires necessary to move the board and heatsink.

Larry
 
Last edited:
This info re the last 4-5 posts has been copied over to a new thread, "SX-838 rebuilding, woofers fluctuating, erratic behavior" because my questions are confusing the OP's original thread intent.

Thanks
 
Just a thread bump to keep this thread close to the new one re 838 rebuilding.

The recap list is in this thread and I plan on using it!
 
Frustrating!!!!! Trying to cut and paste C1, C2 and Q1 from the Equalizer board on PE comes up "Sorry, no direct match". I can't imagine the amount of time to redo this list?

Unless I'm on the wrong site or I'm doing something wrong??
 
Wondering if someone could point me in the right direction. Was swapping out transistors on the power amp (AWH-031-a) board and made a foolish mistake: when swapping out the 2SC1451 transistors for KSC3503's, I accidentally crossed the base and collector leads on both sides. I fired up through DBT and the no relay engagement. I quickly realized what I did and swapped in 2 new 3503's the correct way. Fired up and nothing/no relay engagement. Any idea where to go from here? Not sure if anything else was fried. The DBT stayed about half lit (a little brighter than normal) both before and after the mistake. Any insight is much appreciated.. :(
 
which all transistors you replaced in power amp? Check the dc offset, maybe its too high now and needs adjustment
 
Dial down the bias/idles and try it on the DBT. Watch the meter as you power up and see what the idles are doing. If you can, monitor the DC Balance as well.
Better check the outputs for condition. You can check them in situation. Look for a short between the three leads of each output.
Make sure the volume is at minimum while on the DBT.
 
Last edited:
transistors replaced:
q5,6 2sc1451 --- KSC3503DSTU
q1,2,3,4 2sa726's --- ksa992FBU
Q9, 10: 2SB528 --- A940
Q7, 8: 2SD358 --- C2073

Will check as much as I can and report back. Assuming the DBT kept anything from frying? Thank you...
 
Back
Top Bottom