Sansui QRX7001 repair

[par4ken]

Why Wima brand film caps? Why not Kemet films, or any other brand?

I only mention Wima because I know that they will fit. By all means use something else if you can fit them in. Look for something like 5mm lead spacing (pitch). I've also used axial lead capacitors by mounting them on end.

 

[rwwear]

Electrolytics don't last 50 years.

The large caps can and often do.

 

[rwwear]

I have replaced the electrolytic caps on the F-2428 Filter and power supply board. It feeds 25V to the 2088 QS phase control board. Both QRX Restore and 4channelsound stress the importance of 25v. Before the rebuild I read 26v with the 2088 board in place. Checking the output on pin 5, I have 29v with the pot turned all the way down. This is with no boards in place. Is this normal? Will it settle down with the 2088 board in place? I am reluctant to put the 2088 on with 29v present. Did the new caps change the voltage?

It could be reading high because when it was designed to operate on 117 volts instead of the average 128 volts we have now. You may need to change some of the unreliable transistors and check the resistors for drift. You may also need to change the pots.

 

[rwwear]

The schematic came with the unit and looks to be of the same vintage.

When I rebuilt a CA2000 I found a few discrepancies on the schematic. It's generally a good idea to go by the factory install. And take lots of pics.

 

[SPXER]

A basic question. Is there an easy way to locate the solder points on the underside of the board after finding the component on the top side?

 

[rwwear]

It's hard sometimes but shining a flashlight on the component from the top helps.

 

[mandrix]

A basic question. Is there an easy way to locate the solder points on the underside of the board after finding the component on the top side?

Continuity, meter?

 

[SPXER]

Lest anyone think I have been goofing off, I made these calipers to find the solder joints under the board. They are actually helpful.

 

[SPXER]

Also made this signal tracer.

 

[SPXER]

Got the WIMA caps. $50 worth.

 

[Sonik Wiz]

It's hard sometimes but shining a flashlight on the component from the top helps.

Good point. Done that many times myself.

 

[rwwear]

Thanks for that, mandix.
The cap I removed (C16) was installed with the neg going to the transistor. Opposite what the board shows as + to the transistor. I think it is a misprint on the board. Also there are two TR09s printed on the board and no TR07. Another misprint. I take the upper line to be + and the lower to be neg. as it is marked ground. I would assume (and we know what that means) the positive of the cap goes to the upper line and the negative to the transistor (closer to the ground line). I think I should put it back as I took it out.

Edit: found C12 is put in opposite of the board marking.

View attachment 98642

How is C15 installed?

 

[SPXER]

How is C15 installed?

The schematic shows C14 (same as C13, C15, C16) pos to the top rail, neg to the transistor.

 

[rwwear]

I would go by the way the factory installed them. Unless there has been a service alert.

 

[SPXER]

Back to the weak output problem. I replaced C12, C24, C40, and C52 with the WIMA caps hoping to cure the weak output. No luck. Using the signal tracer, comparing a good circuit to a bad, I find roughly the same signal (compared to good circuit) through C12, C24, and C40. Not the same at C52. makes me suspect around TR12. How do I check the function of TR12 in place? Or does one just replace it and see what that does?

 

[barfle]

Back to the weak output problem. I replaced C12, C24, C40, and C52 with the WIMA caps hoping to cure the weak output. No luck. Using the signal tracer, comparing a good circuit to a bad, I find roughly the same signal (compared to good circuit) through C12, C24, and C40. Not the same at C52. makes me suspect around TR12. How do I check the function of TR12 in place? Or does one just replace it and see what that does?

View attachment 99055

There aren’t any balues on the resistors or capacitors, so it’s impossible to tell you how large the signals you see should be, but I’ll tell you in general terms what ought to be happening around TR12.

TR12 is a gain stage. The gain there is set by the ratio of the resistors R76 and R80. The idea is that the current through both resistors should be the same, so the voltages across them, from Ohm’s law, should be the product of the current through the resistor and the value of the resistor. E=IR. The capacitor C48 is used to increase the AC (signal) gain of the circuit by effectively reducing the AC resistance on the emitter. That resistance will change dependent on the frequency, but from my days as a circuit designer, the value of C48 should be quite high so as to minimize its reactive value and maximize the AC gain of the circuit.

I would compare the signals you see at the emitter and collector of TR12. The emitter should be just about nothing, the collector should have a pretty large signsl on it, out of phase with the base.

I would first try replacing C48, if you haven’t already.

 

[SPXER]

There aren’t any balues on the resistors or capacitors, so it’s impossible to tell you how large the signals you see should be, but I’ll tell you in general terms what ought to be happening around TR12.

TR12 is a gain stage. The gain there is set by the ratio of the resistors R76 and R80. The idea is that the current through both resistors should be the same, so the voltages across them, from Ohm’s law, should be the product of the current through the resistor and the value of the resistor. E=IR. The capacitor C48 is used to increase the AC (signal) gain of the circuit by effectively reducing the AC resistance on the emitter. That resistance will change dependent on the frequency, but from my days as a circuit designer, the value of C48 should be quite high so as to minimize its reactive value and maximize the AC gain of the circuit.

I would compare the signals you see at the emitter and collector of TR12. The emitter should be just about nothing, the collector should have a pretty large signsl on it, out of phase with the base.

I would first try replacing C48, if you haven’t already.

Thanks barfle. This is the kind of education I need. I have changed C48. The circuit worked before I changed all the electrolytic caps. So, I have to conclude the issue is something I introduced. I will change C48 again.
C48=100uf
R76=3.9K
R80=820
According to the schematic the voltage should be 30v on the high side of R76. 17.5v at the collector, and 25v (Edit: 2.5v) at the emitter. (I have not measured.)

Service manual available here- https://ia802301.us.archive.org/24/items/manual_QRX7001_SM_SANSUI/QRX7001_SM_SANSUI_text.pdf

 

[SPXER]

I replaced C48 and found a broken trace, so I bent the lead over to R80 to repair the break. That did restore the expected voltages (17.5) at the collector of TR12 and 2.5 at the emitter. (I thought the print said 25v, but I read 2.5v.)
But now there is no signal at C52 at all. Also, touching the signal tracer to the base of TR12 trips the protection relay.
The signal at C24 is still good.

Edit: more numbers. 15.3v drop across R76, 3v drop across R80,
14.5v drop across TR12 C to E.

Edit: found another broken trace and that fixed the other bad channel I hope. I think changing the signal caps twice was more than the board could take or maybe I was in too much of a hurry.

 

[barfle]

I replaced C48 and found a broken trace, so I bent the lead over to R80 to repair the break. That did restore the expected voltages (17.5) at the collector of TR12 and 2.5 at the emitter. (I thought the print said 25v, but I read 2.5v.)
But now there is no signal at C52 at all. Also, touching the signal tracer to the base of TR12 trips the protection relay.
The signal at C24 is still good.

Edit: more numbers. 15.3v drop across R76, 3v drop across R80,
14.5v drop across TR12 C to E.

Edit: found another broken trace and that fixed the other bad channel I hope. I think changing the signal caps twice was more than the board could take or maybe I was in too much of a hurry.

Lifting traces is, unfortunately, part of the service tech’s experience path. I think a lot of manufacturers didn’t plan for much repair even in the days when through-hole components (as compared to surface-mount) was the norm.

Those DC voltages seem about right for those resistor values. Your circuit tracer probably put a significant base current through TR12, causing the protection relay to trip.

Which side of C52 are you looking at? Do all three components at that node (transistor, resistor, capacitor) read exactly the same? If there’s an AC signal on the base of TR12, you should have about the same signal in the emitter (with about a 0.6VDC lower reading), and a signifigantly larger copy of that signal (polarity reversed) on the collector, as well as on both sides of C52.

It’s possible that the output side of C52 is loaded heavily. That isn’t shown on the schematic, but of course it goes somewhere, and that might be shorted.

I have a device that I’m troubleshooting myself that has a -12VDC power supply issue. One of the boards it feeds is presenting a heavy load (short circuit) to that supply, and tracking that down is taking a lot of time I don’t have just now. It’s not an important part of my system (a slow-scan TV converter that I have drawings and manual for), so I’m not frantic about it just yet. I have another project that works fine while it’s spread out on the bench, but when I put it together, it acts very badly.

Old electronics certainly have their gremlins.

 

[Owen Smith]

Edit: found another broken trace and that fixed the other bad channel I hope. I think changing the signal caps twice was more than the board could take or maybe I was in too much of a hurry.

What temperature are you using on your soldering iron? The more heat you use the higher risk of lifting tracks.