• 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

Help —SX-838 disaster

Looks like you do have blown transistors in the left channel power amp output section.
Unless you have a broken power supply wire to the AWH-031 (pin 1).
 
merlynski said:
Looks like you do have blown transistors in the left channel power amp output section.
Unless you have a broken power supply wire to the AWH-031 (pin 1).
Click to expand...
.
I will double check the power supply wire just to verify, thank you.
I pulled out the -031 board and rechecked all voltages on the remaining boards. Everything is in range and all pin to pin wires are correct. I do have some minor AC voltage running throughout the circuits so I may have some leaky capacitors somewhere, but yes, my main problem is with the -031 board. I plan to test each component (left channel) for expected voltage to see if I can find my fault(s). I think the transistor(s) blew because some other component(s) are bad. I tested the transistors before installing so they should have been functioning. The foil side is in rough shape with some of the foil separating/missing at the pins. When I soldered, I added connecting wire where needed, so even though I checked all these trails carefully, I haven’t fully ruled out a non connection or a short. (The other -031 board I bought had some significant surgery in the past and is in worse shape than this original one).

one other thing, as you suggested, I disconnected the output transistors and in order to access the foil side of the -031, I also removed it from the heat sink. There are two components attached to the heat sink which have a symbol on the schematics of 5 triangles. From reading other posts, these components are no longer accessible and folks on this platform discussed alternatives. The consensus opinion was three 4004 diodes in series, so that is what I have connected on the -031. I do have two of the original parts but they are fragile so I have them in a safe place for now.
 
Those thermal diodes that are so fragile control the bias, so must be in the circuit when testing it. If the bias control components (D2, VR2, D4 left channel) are open or missing the output and driver transistors will saturate and flow full current at power on, possibly destroying them, and other components. If the substitute diodes have too high a pn junction drop the same thing happens. If they are not thermally bonded (mounted) to the heatsink the output transistor circuit will go into thermal runaway as they warm up.
 
Would these diodes need to be ‘thermally mounted’ if I do not have output transistors connected? Also, do you agree with the conclusion that these three diodes is series are a substitute for the original thermal diodes?
 
For the purposes of brief testing, the thermal compensation diode array can remain hanging free in space, but connected in the circuit. Howeverm once the unit is assembled they need to be mounted to the same heat sink used by the output devices in order to be effective.
 
dustmaker said:
Would these diodes need to be ‘thermally mounted’ if I do not have output transistors connected? Also, do you agree with the conclusion that these three diodes is series are a substitute for the original thermal diodes?
Click to expand...

If you do not have the output transistors installed you cannot set the bias current, as there is no bias current. One you have installed the output transistor you must have a method to control bias as the output transistors heat up. A pn junction voltage (Base-Emitter) decreases as temperature rises, so for the same drive applied to the BE junction the collector current increases, increasing the transistor's power dissipation, it heats up more . . . The current increases as the transistor goes into thermal runaway and destroys itself, and the components around it. The bias circuit provides negative feedback by sensing the temperature rise of the output transistors through the physical/thermal connection of the heatsink and reducing the drive to the transistors, controlling the current through them.

In order for a 'substitute' to work correctly its thermal response curve has to track pretty closely to that of the output transistors BE junction.

Here is a link to an AK thread about this: https://audiokarma.org/forums/index.php?threads/replacing-the-stv-3h-and-4h-diodes.279530/
 
I need help. I finally got around to reconnecting the -031circuit. I turned the unit on and within 10 seconds I had smoke. I shut off it off and flipped the 031 over and I see that resistors R31 and R33 are cooked. The diode array on that same channel is also toast as I don’t read anything with the DMM and it has completely ripped away from the foil. So I don’t know if the diode array breaking from the foil caused the resistors to cook or something else. I need your advice before I start swapping out parts
 
Were you on a DBT?
That happens when the bias control circuit is open: D2, VR3, D4, or an open trace. This makes the driver transistors turn full on and applies the max +/- power supply voltage across their emitter resistors, R31 & R33, and lets out the magic smoke. You can test D2 with this circuit:

STV_test.jpg

Pick a value for R1 of about 3.3kΩ, not critical, and you can use one 9V battery. You should get about 2.4 vdc if it is good, maybe a bit less. It will be the battery voltage if it is open.

It can also happen if a driver transistor is shorted E to C.
 
Last edited:
merlynski said:
Were you on a DBT?
That happens when the bias control circuit is open: D2, VR3, D4, or an open trace. This makes the driver transistors turn full on and applies the max +/- power supply voltage across their emitter resistors, R31 & R33, and lets out the magic smoke. You can test D2 with this circuit:

View attachment 1901972

Pick a value for R1 of about 3.3kΩ, not critical, and you can use one 9V battery. You should get about 2.4 vdc if it is good, maybe a bit less. It will be the battery voltage if it is open.

It can also happen if a driver transistor is shorted E to C.
Click to expand...
I don’t know what DBT means.

Thanks for the above circuit. I’ll test D2 and advise
 
merlynski said:
DBT - dim bulb tester, protects from releasing magic smoke:
Dim Bulb Tester diagram pictorial... please feel free to copy and post.
I didn't know about it either, 'till I came to Audiokarma :).
Click to expand...
I finally made a dim bulb tester. Originally, it would light up when I came near it. I’d back up and it would dim, I’d get close and it light up. The closer I’d get, the brighter the light......I realized I didn’t create a dim bulb tester, I’d created a dim bulb DETECTOR instead.

First things:
1) I have not replaced the smoked resistors or Transistors, etc
2). The Original diode array I used I can’t find so I can’t test it. I probably threw it out so I made a new one. I am using 1N4004 diodes in this arrays. Not sure where I found the reference for those, but that’s what I’m using to test. I do have 1N4148s if those are superior for this preliminary testing.
3) The red power wire was loose. I reattached it

when I power on the 838 using the DBT, the bulb (60w) dims, not a lot, but it does dim. I don’t have any additional smoke but I did shut every thing off after about 8 seconds.

I do have a variac if it is useful in running down my issues

What is the best way to proceed at this point?
 
Last edited:
Dim Bulb Detector . . . |:rflmao:|

The DBT bulb should go down to orange or even brown.
Are the output transistors uninstalled? Uninstall if not already out.
Make sure the VR3 and VR4 are turned down to 0Ω. Check in circuit with ohmmeter (power off).
Retest on DBT.
Start troubleshooting!
Input set to Aux, no signal input, Volume to minimum. No speakers or loads attached.
Power up on dbt and check power supply voltages:
AWR-055, power supply,
Pin 21, +8.5vdc:
Pin 17, +51vdc:
Pin 14, +13vdc:
Pin 13, +35.5vdc:
Pin 7, -45vdc:
Pin 8, -13vdc:
Pin 4, +44vdc:
Pin 5, -44vdc:
Pin 22, 7.5vAC:
Some of these may be a bit low because of the dbt, just get the baseline.

AWH-031, power amp,
Pin 5, 0vdc:
Pin 20, 0vdc:
Pin 1, +44vdc:
Pin 2, -44vdc:
Pin 24, -44vdc:
Pin 25, +44vdc:

AWM-025, Protection,
Pin 11, +35.5vdc:
Pin 10, +35.5vdc:
Pin 2, -13vdc:
Pin 9, ≈+9vdc*:
Pin 12, 7.5vAC:
* may be ≈+35.5vdc
If these are OK, post values and continue.


Did you look over the threads from the links in posts 47 and 48?
pa_200 used 3 1N4148 diodes in his project to replace the STV-4.
Put replacement for STV in place if not done already.
Replace any other smoked transistors (except outputs).
Replace the damaged resistors.
 
Last edited:
The DBT bulb should go down to orange or even brown. After changing resistors and bad transistors, it did get dimmer
Are the output transistors uninstalled? Uninstall if not already out. Yes
Make sure the VR3 and VR4 are turned down to 0Ω. Check in circuit with ohmmeter (power off). Turned counterclockwise to stop. DVM showed 0
Retest on DBT.
Start troubleshooting!
Input set to Aux, no signal input, Volume to minimum. No speakers or loads attached. Done
Power up on dbt and check power supply voltages:

I tested everything using DBT andallvoltages were in line, except a bit low. I disconnected DBT and the following are current voltages
AWR-055, power supply,
Pin 21, +8.5vdc: 7.5
Pin 17, +51vdc: 51.9
Pin 14, +13vdc: 12.3
Pin 13, +35.5vdc: 32.8
Pin 7, -45vdc: -45.8
Pin 8, -13vdc: -13.2
Pin 4, +44vdc: 44.5
Pin 5, -44vdc: -44.5
Pin 22, 7.5vAC: 8.0vAC
Some of these may be a bit low because of the dbt, just get the baseline.

AWH-031, power amp,
Pin 5, 0vdc: -0.15
Pin 20, 0vdc: -0.5
Pin 1, +44vdc: 44.5
Pin 2, -44vdc: -44.5
Pin 24, -44vdc: -44.5
Pin 25, +44vdc: 44.5

AWM-025, Protection,
Pin 11, +35.5vdc: 32.8
Pin 10, +35.5vdc: 32.8
Pin 2, -13vdc: -13.3
Pin 9, ≈+9vdc*: 7.83
Pin 12, 7.5vAC: 8.0vAC
* may be ≈+35.5vdc
If these are OK, post values and continue.


Did you look over the threads from the links in posts 47 and 48?
pa_200 used 3 1N4148 diodes in his project to replace the STV-4. Done - 3 1N4148
Put replacement for STV in place if not done already.
Replace any other smoked transistors (except outputs). Done
Replace the damaged resistors. Q5, Q7, Q9. Q1&Q3 seemed to be ok
 
Last edited:
dustmaker said:
The DBT bulb should go down to orange or even brown. After changing resistors and bad transistors, it did get dimmer
Are the output transistors uninstalled? Uninstall if not already out. Yes
Make sure the VR3 and VR4 are turned down to 0Ω. Check in circuit with ohmmeter (power off). Turned counterclockwise to stop. DVM showed 0
Retest on DBT.
Start troubleshooting!
Input set to Aux, no signal input, Volume to minimum. No speakers or loads attached. Done
Power up on dbt and check power supply voltages:

I tested everything using DBT andallvoltages were in line, except a bit low. I disconnected DBT and the following are current voltages
AWR-055, power supply,
Pin 21, +8.5vdc: 7.5
Pin 17, +51vdc: 51.9
Pin 14, +13vdc: 12.3
Pin 13, +35.5vdc: 32.8
Pin 7, -45vdc: -45.8
Pin 8, -13vdc: -13.2
Pin 4, +44vdc: 44.5
Pin 5, -44vdc: -44.5
Pin 22, 7.5vAC: 8.0vAC
Some of these may be a bit low because of the dbt, just get the baseline.

AWH-031, power amp,
Pin 5, 0vdc: -0.15
Pin 20, 0vdc: -0.5
Pin 1, +44vdc: 44.5
Pin 2, -44vdc: -44.5
Pin 24, -44vdc: -44.5
Pin 25, +44vdc: 44.5

AWM-025, Protection,
Pin 11, +35.5vdc: 32.8
Pin 10, +35.5vdc: 32.8
Pin 2, -13vdc: -13.3
Pin 9, ≈+9vdc*: 7.83
Pin 12, 7.5vAC: 8.0vAC
* may be ≈+35.5vdc
If these are OK, post values and continue.


Did you look over the threads from the links in posts 47 and 48?
pa_200 used 3 1N4148 diodes in his project to replace the STV-4. Done - 3 1N4148
Put replacement for STV in place if not done already.
Replace any other smoked transistors (except outputs). Done
Replace the damaged resistors. Q5, Q7, Q9. Q1&Q3 seemed to be ok
Click to expand...

observations:
1). NO SMOKE
2) relay takes about 2 seconds to engage. Prior to all this, it would engage in about 1 second. Pretty sure I heard it click off and back on after about a minute
 
dustmaker said:
AWH-031, power amp,
Pin 5, 0vdc: -0.15
Pin 20, 0vdc: -0.5
Click to expand...
If these readings are volts rather than millivolts, you have a dc offset issue.
Still on DBT? Good, if so, the other voltages look OK.
DC Offset can be adjusted on DBT.
Clip black to ground, red to pin 5, adjust VR2 to get 0.0vdc +/- 50mVdc.
Next, red to pin 20, adjust VR1 to get 0.0vdc +/- 50mVdc.
If it will stabilize less than +/- 20mVdc that is great. Some slow drift is normal.
 
merlynski said:
If these readings are volts rather than millivolts, you have a dc offset issue.
Still on DBT? Good, if so, the other voltages look OK.
DC Offset can be adjusted on DBT.
Clip black to ground, red to pin 5, adjust VR2 to get 0.0vdc +/- 50mVdc.
Next, red to pin 20, adjust VR1 to get 0.0vdc +/- 50mVdc.
If it will stabilize less than +/- 20mVdc that is great. Some slow drift is normal.
Click to expand...

I will double check...I assumed volts, but I didn’t look for milivolts

these readings are NOT on DBT. My readings, while on DBT were consistent with your posted voltages...so I went ahead and did readings while not on DBT.

A clarification, I turned it on again, and the relay takes almost 5 seconds to engage.

I will go back on DBT until you say OK to go live.....i will get back to you on whether these are volts or mili-volts before I proceed
 
dustmaker said:
I will double check...I assumed volts, but I didn’t look for milivolts

these readings are NOT on DBT. My readings, while on DBT were consistent with your posted voltages...so I went ahead and did readings while not on DBT.

A clarification, I turned it on again, and the relay takes almost 5 seconds to engage.

I will go back on DBT until you say OK to go live.....i will get back to you on whether these are volts or mili-volts before I proceed
Click to expand...
Were the output transistors installed when you were of the DBT?
 
You must log in or register to reply here.
Back
Top Bottom