SX-950 Protection Issue

krs5264

New Member
Hi there,

I recapped a 950 about 2 years ago and it was working great until recently, and now it won't come out of protection.

Below are my measured voltages for the protection and power supply boards.
Pins 1,3,5,6,7,8 on the protection board seem suspicious so I measured Q4 and Q5, and the collector on Q4 was reading 17.38 instead of the specified -0.6.

It's been a while since I worked through a schematic, anyone have any suggestions for what went wrong or what to measure next? Thanks

Protection
1 = -0.01
2 = -13.87
3 = -0.01
4 = 0
5 = - 0.02
6 = 0
7 = 0
8 = -0.02
9 = 20.74
10 = 34.42
11 = 34.42
12 = 8.09 AC

Power Supply
1 = 0
2 = 6.16
3 = 6.13
4 = 13.83
5 = 13.83
6 = 50.4
7 = 50.4
8 = 34.38
9 = 34.40
10 = 34.40
11 = 26.52
12 = 0
13 = -13.87
14 = -53.6
15 = -19.9
16 = 37.8 AC
17= -49.9
18 = 50.0
19 = 37.8 AC
20 = 0
21 = 0
22 = 8.10 AC
23 = 8.09 AC
24 = 8.24 AC
25 = 48.6 AC
26 = 48.6 AC
27 = 20.3 AC
28 = 20.3 AC
 
The emitter on Q4 should be -0.6V not the collector. Your readings of both boards do not look bad. Check your DC offset and idle current adjustments.950.jpg
 
The voltages on protection pins 10 & 11 are a tad low but should be more than enough to pull in the relay. The relay voltage drop should be 24V so with pin 10/11 at 34.42V, pin 9 should be roughly 10V. When in the protection mode, no current flows through the relay coil and pin 9 would measure the same as pin 10/11 but you are seeing 20.74V. So, it appears that it isn't in protection but not enough current is being drawn through the relay to activate it. That would suggest that either Q7 is weak or it's not being driven into saturation by Q6. I'd take a closer look at those two devices.

Les
 
Last edited:
The voltages on protection pins 10 & 11 are a tad low but should be more than enough to pull in the relay. The relay voltage drop should be 24V so with pin 10/11 at 34.42V, pin 9 should be roughly 10V. When in the protection mode, no current flows through the relay coil and pin 9 would mesure the same as pin 10/11 but you are seeing 20.74V. So, it appears that it isn't in protection but not enough current is being drawn through the relay to activate it. That would suggest that either Q7 is weak or it's not being driven into saturation by Q6. I'd take a closer look at those two devices.

Les

Not bad Les. :D
Also check the 220 ohm 1 watt R28 as well. It would raise the emitter voltage of Q7, upsetting the careful biasing of the q6 stage.

You could AS A TEST connect the Q7 collector and emitter together and see if the relay pulls in.
If the relay pulls in, look at the transistor. If the relay DOES NOT pull in, look at the resistor and the relay coil itself.
 
Alright first I checked DC offset, it was 4mv for left and 9mV for the right channel, not that far off so I didn't mess with it since there are bigger issues.

Then I measured Q6 and Q7

Q6
B = -1.51
E = 0.004
C = 17.47

Q7
B = 17.34
E = 10.88
C = 21.01

I did connect the collector and emitter for Q7 and the relay did not pull in. Visual inspection of the R28 didn't reveal anything, and probing the solder pads didn't register a resistance on my meter. I'm guessing Q6 is the problem with the low emitter value?
 
This information points towards R28 being bad IMO. Q7 Vbe is roughly 7 volts which indicates that there is only a small amount of emitter current. Vbe should be closer to 0.7V in order for Q7 to be saturated. In addition, the result of the test suggested by Mark further points towards R28.

Les.
 
This information points towards R28 being bad IMO. Q7 Vbe is roughly 7 volts which indicates that there is only a small amount of emitter current. Vbe should be closer to 0.7V in order for Q7 to be saturated. In addition, the result of the test suggested by Mark further points towards R28.

Les.

Agree, R28 MUST be checked. Shorting Q7 collector to emitter takes ALL of the transistors out of the "equation", leaving ONLY the relay coil and R28.
There SHOULD have been "resistance" registering between Q7 emitter to ground. 220 ohms worth. That is the ONLY valid resistance reading to do here.

as for Q6, it is an NPN transistor, and it's normal state is turned OFF, high impedance,
allowing the timing capacitor to charge to where Q7 turns on and pulls in the relay..

Thus the base voltage of -1.51v with an emitter voltage of +0.004 is (probably) ok.
WHEN the protect is SUPPOSED to open up the relay contacts, Q6 conducts continuously, keeping the timing cap discharged.
 
Alright I pulled R28, the board behind it and the surrounding resistors looked a little crispy, it read 210 ohms, but since it's out, I'll go ahead and replace it
 

Attachments

  • DSC_0365.JPG
    DSC_0365.JPG
    78.2 KB · Views: 45
  • DSC_0366.JPG
    DSC_0366.JPG
    55.5 KB · Views: 44
show picture of other side of board, also I never remember specifying a glass diode for pins 9 to 10. I believe I have been specifying robust 1n4004's in that position...

When replacement resistor is in, repeat the test where q7 collector and q7 emitter are tied together to make the relay pull in.
 
Since R28 tests good out of circuit, I wonder if it's perhaps just a faulty solder connection. :idea:
 
Also check that the incorrect diode across pins 9 and 10 hasn't turned into a 'die-ristor'. (dropping voltage in both directions)
 
I did re-solder the R28 pads when it came under question prior to removal, and there were no noticeable differences in voltages so I don't think that was an issue.

Hmm not sure where I got the part number for that diode... Well I'll get an order in for that resistor and the *correct* flyback diode and update once they're installed. I appreciate all the help.
 

Attachments

  • DSC_0367.JPG
    DSC_0367.JPG
    81.1 KB · Views: 21
I would touch up the pin solders from 1 to 12, I had an issue similar to this awhile back and it was a cold solder on one of those pins.

Also that flyback diode might be backwards, looks like the stripe is in the wrong direction, listen to mark and use a 1N4004 there, the striped end cathode should be attached to pin 10 and the anode attached to pin 9
 
disconnect that diode, reinstall the resistor, resolder pins 1 to 12 and try the q7 collector to emitter connect test again to get the relay to pull in..
something has been mighty fishy in there.
 
I don't recall that resistor heating to the point of fading the value stamp. As if something besides the coil is drawing current. But it's not like I've seen hundreds of examples.
 
I don't recall that resistor heating to the point of fading the value stamp. As if something besides the coil is drawing current. But it's not like I've seen hundreds of examples.

#1 - the writing on the resistor can be lost many ways including cleaning AND removal (desoldering heat)
IF the resistor got too much current, it could have been from a backwards diode in parallel with the coil...
and the diode itself would take damage to where it would present REALLY weird characteristics.
A diode that small COULD have that happen, a 1n4004 in backwards would take the overload and shrug it off.
 
Update:

I replaced the resistor with a new one, pulled the incorrect, improperly installed diode and low and behold the relay activated. I then added a 1n4004, in the correct direction... I measured all the pins again and the only appreciable change is on pin 9 on the protection board which is now 8.21 V

I'm amazed it lasted over 2 years with the wrong diode in the wrong direction until serious problems manifested.

Seems like all is well, thanks again for the help!
 
That's one for the memoirs....

That pin 9 voltage is correct and what I expected to see, all's well that ends well.
 
Back
Top Bottom