SA-7100 Ripple Current measurements-effects on protection board?

[Gobstopper]

Hi Everyone, I've been wrestling with a SA-7100 that has a horrid buzzing noise, for a couple of years now. I replaced electrolytics(including filter caps) and transistors using guides on this forum to no avail. Last night I used an android app to measure sound frequency, and found that when the pre and main amps are connected there are 60 hz and 120hz peaks, with many higher harmonics. With the pre/main jumpers disconnected, the 60hz is gone and I just have 120Hz plus harmonics. I'm pretty sure I have at least 2 different sets of problems. In any event, I borrowed an oscilloscope and started looking at the output voltages from the power supply board, thinking the 120hz noise would logically point there. Outputs at pins 6,7,12 and 13 were at the appropriate DC voltage and showed barely any ripple current. However, pin 11, which feeds the protection board, has a conspicuously larger ripple voltage(DC was okay). Looking at the circuit diagram(and trying to apply my rusty circuit theory) it seems that the pin 11 path,in contrast to paths for pins 12 and 13, doesn't have any smoothing capacitors in it, in which case the ripple current wouldn't be unexpected.Is this correct? I wanted to check with more experienced and knowledgeable folks before discarding the finding. Out of curiosity, could excess ripple into the sa-7100 protection board cause buzzing in the audio path?
Any other things I should check while I still have the oscilloscope?

 

[markthefixer]

D1,D2 are S1b01-02 diodes, 200v 1a
mouser: 863-1N4004G @ $0.21 ea

D4-D7 are DS-118a diodes, 200v 3 amp diodes. With 0.01uf snubber capacitors.
mouser: 863-1N5404G @ $0.41 ea

Check the diodes out of circuit.

I trust the 2sa726 and 2sc1451 transistors have been replaced long ago.

My bet is abused D4-D7 and/or a bad snubber cap across one or more of D4-D7.

Any buzz in pin 11 would also be in pins 12 & 13. The smoothing isn't that miraculous.

You can measure the AC buzz on the DC circuit by using a blocking capacitor in series with the DMM probe. 0.1uf 150v metal film cap.

 

[Gobstopper]

Hi Mark, thanks so much for your reply, it's great to see you are still here on the boards helping everyone out after all these years. I replaced all the diodes on the power supply during my rebuild
as well as the 2sa726 and 2sc1451 transistors. (http://audiokarma.org/forums/index....erence-in-voltages-at-l-and-r-outputs.768115/)
I'll pull the diodes and check out of circuit just in case. I'll also try the blocking cap + DMM instead of the oscilloscope and see how the ripple measures compared. Finally, if it were to be a bad subber cap(those are the big ceramic disks, no?), without a dedicated ESR device, what's the best way to check?

 

[Gobstopper]

I measured ripple current with a blocking capacitor and DMM. AC voltages were 0.006V on pin 11, 0.008V on pin 12 and 0.005V on Pin 13. Pin 7 and Pin 6 both had .028 V AC. What is confusing me is that on the scope AC voltage at pins 12 and 13 show almost a flat line, while pin 11 has a very wide band(with analog scope set to 5V per division, band is almost 4 divisions from top to bottom) with not much obvious structure(see attached photo,edit: AC readings @ Pins 6 and 7 also appear as flat line). Any ideas about what is going on? Scope is to set to read AC. I attached a shot of AC voltage downstream of D1 and D2 for reference.

 

[llwhtt]

The voltage at Pin 11 is straight from the regulator, Q1 being the pass element, and should be clean so I wouldn't call C4/5 smoothing caps more like decoupling capacitors. If you notice there are two grounds on the power supply board, Pins 10 and 14, are both OK? What does the voltage on the collector of Q1 look like? The regulator appears to be working as your DC voltages seem to be OK, I'm thinking the ground at Pin 10 is missing.

Craig

 

[petehall347]

just for fun try using rca cables to jump pre main instead of the jumpers . guess it wont do anything but will maybe rule something out .

 

[Gobstopper]

The voltage at Pin 11 is straight from the regulator, Q1 being the pass element, and should be clean so I wouldn't call C4/5 smoothing caps more like decoupling capacitors. If you notice there are two grounds on the power supply board, Pins 10 and 14, are both OK? What does the voltage on the collector of Q1 look like? The regulator appears to be working as your DC voltages seem to be OK, I'm thinking the ground at Pin 10 is missing.

Craig

I tested the voltage at the Q1 collector:
DC (DMM)
to chassis: 48.5V DC
to Pin 14: 46.8V DC
to Pin 10: 48.2V DC
For AC, used scope and compared between using Pin 14 or Pin 10 as ground. Pin 14 as ground and the waveform looked fine. With Pin 10 as ground, the width of the waveform(thickness of trace) seemed to fluctuate pseudo-perdiodically over time. If I touched the chassis with my hand(while ground at Pin 10), then the thickness of the trace got noticeably thicker(see attached photos). As for checking grounds, visually it looks okay(pin 10 wire wrapped to same post as where blue wire from transformer connects) and Pin 14 grounds to a different, but neighboring set of grounds, and no difference in voltage between Pin 14 to Ground and Pin 10 to Ground on DMM, between Pin 10 and Pin 14, a -0.2mV DC difference(edit: tested again and no difference).
For kicks, I only connected one lead of oscilloscope to chassis (or either of the two grounds) and left the other lead floating, see a mostly sinusoidal ~ 60Hz waveform with +5V to -5V peak-to-peak. Is this normal? Any other checks for proper grounding?

I also replaced the jumpers with RCA cables, but buzzing still there.

 

[llwhtt]

There should be a ground terminal on your scope, ground it to the chassis of the Pigoneer.. The ground should be a smooth trace no ripple.

Craig

 

[markthefixer]

OK, I haven't read the rest of the thread, this is an addendum to my earlier post.

I should have been more specific: the DMM should be connected ACROSS the TWO connections each of each individual main capacitor.
There will be a DC reading for EACH big cap
There will ALSO be an AC reading for each big cap where the DC present on that cap is blocked by the blocking capacitor.

Again, these readings are made ACROSS each individual capacitor, AT the base of that capacitor.

 

[Gobstopper]

Thanks everyone. I don't think the sinusoidal waveform is that big of a deal, since I wasn't measuring while grounded, but I'm still concerned about readings at the Q1 collector while grounded at Pin 10, and still not sure why pins 12 and 13 look okay on scope, but pin 11 looks so messy.

Mark, when you say "big caps" are you referring to the flat ceramic disks in parallel with the big diodes(D4-D7) on the rectifier bridge, or the BIG BIG filtering caps? Assuming the ceramics, am I correct in understanding that I measure DC voltage with the DMM by attaching one probe to each lead of the capacitor?
Same process for AC, but with the blocking capacitor in series between one lead of capacitor probe, and other probe attaching directly to other lead of same capacitor?

 

[markthefixer]

BIG BIG 6800uf 25v chassis mounted.

 

[Gobstopper]

BIG BIG Caps ( replaced the old ones as part of my rebuild last year with 8200uf 50V)
DC: 27.7 V (both)
AC 0.027 V (both)

 

[Gobstopper]

I'm assuming the above measurements are okay?

 

[markthefixer]

just got back on... yes - 27mv of ripple, same on both caps.

note - just figured out you have new caps and diodes (1n5404?? 3 amps) in there.

 

[Gobstopper]

Yes, replaced main caps, and the caps, transistors and diodes on the power supply. Given that main caps look okay, where should I check next?
edit for clarification: followed a previous thread and replaced e-caps and most transistors on all boards.

 

[markthefixer]

Back to basics: speakers connected and power connected, NO OTHER CABLES.then
AC volts at speakers at full volume.

Then try with plug reversed in the power socket.

Then DC voltage at pin 11 and if it isn't between 33.25 and 36.75v (i.e. WORKING!!) , the ecb voltages of q1 and q2 AT THE TRANSISTOR LEADS!!!

We will concentrate on getting the 35v correct.

Note Q2, the 2sc945 replacements have TWO pinout flavors. the ksc945C has a center collector lead, making then eCb. The other is ksc945 with an eBc pinout.

 

[Gobstopper]

Got to this a bit late at night, so only managed one orientation of plug for speaker test(buzz/hum is too loud for neighbors). I measured AC at speakers, where wires connect (older speakers, where connection is basically a flat head screw that bare wire is wrapped around, then with screwdriver tighten down)

L: 0.1-0.13 VAC
R: .07 VAC
I'll repeat tomorrow with both orientations, but in past has not made a difference in buzz. Granted, I've done most of the testing with headphones.

Pin 11 Measurements:
37.5-39V DC depending where exactly I ground to chassis. Also, if I touch bare chassis with my hand voltage starts fluctuating all over the place

DC Voltage measurements
Q1: (512-KSC2073TU) E: 37.66 C:45.2 B: 35.4
Q2: (512-KSC2383YTA,ECB pinout) E: 12.6 C:35.2 B: 13.6

 

[Gobstopper]

Okay, started earlier today and was able to do the speaker AC voltage measurements
Plug orientation 1: Left = 0.01 V AC
Right = .023V AC

Plug orientation 2: Left = 0.14 V AC
Right = 0.125 V AC

Not sure why values changed for orientation 1 between last night and today. Orientation of plug had no effect on sound. Again, I'm measuring at the metal screw terminals on speakers, with unit at full volume.

 

[markthefixer]

I want you to put direct shorts across your RCA inputs from the ground connection on the jack to the center conductor - perhaps making up a few sets of shorted plugs for the inputs (tuner, aux 1, aux 2, phono 1, phono 2) to see if there are any changes with different shorted inputs.

Touching the chassis and making such dramatic changes (37.5 to 39v) should not be possible.

There's something basic (and wrong) at work here.'

The unit does NOT have "death caps" or safety resistors (2.2 megohm)

It shows a grounded shell transformer with NO powerline neutral reference locations.

Normally the chassis ground is connected to the unit's neutral power plug prong with a 2.2 megohm resistor which is used to steer the power transformer's capacitive leakage current in a predictable direction.

0.1v of ac hum at the speakers is surprsing.

what do you mean by plug orientations 1 and 2?

Turn off and disconnecte the unit from the power line, and with the plug out in midair and the power switch turned on, take an ohmmeter reading between the metal chassis and the power plug pin(s - two readings) - it will probably be millions of ohms, but that's ok.

 

[Gobstopper]

Thanks Mark. "Orientation 1" and "Orientation 2" referred to the the power plug orientation relative to the outlet. I am out of town and away from the unit this week for the Thanksgiving break, but will try the shorting plugs and the resistance measurements to chassis and power plug pin(s) as soon as I get back. Is idea to get a few RCA cables, snip ends, and then connect ground cable to the center pin cable for each plug? Also, as for touching the chassis, it didn't just oscillate between 37 and 39 Vdc, it swung wildly such that I couldn't get a steady reading and seemed to swing down to zero and then back up again. Happy Thanksgiving to everyone!