Studer A800 audio noise from PSU

[Pat Maki]

Greetings good people at GrouDIY. Getting a ~100hz hum off the repro head when recording (electrical hum). It's on channels 17-24 only, so logic tells me it's caused by something that effects an entire row of channels. After testing with 24/15v cables in the VU section, I tried swapping audio power supplies (the 3 biggies at the back bottom) and that made it clear the issue is with the third audio power supply. I'd previously updated all the elcaps and rifas and double checked all solder joints, so it's not a cap, or poor connection issue. Wondering if it might be a bad rectifier. Have a scope, but not sure what I should measure to trace the issue. Any guidance by the experts here would be greatly appreciated.

 

[juanito2008]

Check with your oscilloscope the +/-15V rails.
And verify the Molex that come from the power supply.

 

[guy_4]

Probably a bad rectifier bridge.

 

[Pat Maki]

Check with your oscilloscope the +/-15V rails.
And verify the Molex that come from the power supply.

Thanks @juanito2008. Was able to get PSU #2 and #3 (GR 70/71) both plugged in fully and with lids taken off to run some comparison tests. Your idea on the +/-15V seems to be on to something. I found that the red wire that goes from the +15V stabilizer board to one of the 22,000uf elcaps is showing ~23V instead of 15V! Also the thin red wire next to it that goes to the power supply PCB is reading at 23V as well. No bueno. Next, I need to get in there to see if I have something wired incorrectly, or if it's a faulty component, or something else. Thanks for the tip!

 

[Brian Roth]

The schematic you posted shows 23V at the positive end of C2, a 22,000 cap (although it shows the wire as orange). That would be correct. The regulator needs a voltage input higher than the desired 15V in order to function.

However, the voltage between terminals 4 (or 6...same thing) and terminals 5 (or 2) should be 15. Another place to measure is across D2 since it connects to terminals 4/6 and 5/2.

If there is indeed 23V there, the regulator isn't doing it's job. A likely culprit would be a collector to emitter short on Q1 which can be checked with an Ohmmeter with the power off and maybe a minute wait after power down to allow the voltages to bleed off.

Bri

 

[zamproject]

As Bri say, 23V at regulator card input from bridge/big C is fine.
But It's confusing because you say it's a red wire there, shemo call for orange wire ?!?
Are you sure you don't wire the thing upside down ?

 

[Walrus]

Orange and Yellow feed the +15V regulator card, and Red and Brown feed the -15V regulator card.
The Output of the +15 regulator card is Red and should be +15V when measured against the Yellow wire.

I always write the terminal numbers on the cards with a Sharpie, whenever I first have to open up a supply, saves time constantly
looking back at the manual!

 

[Pat Maki]

The schematic you posted shows 23V at the positive end of C2, a 22,000 cap (although it shows the wire as orange). That would be correct. The regulator needs a voltage input higher than the desired 15V in order to function.

However, the voltage between terminals 4 (or 6...same thing) and terminals 5 (or 2) should be 15. Another place to measure is across D2 since it connects to terminals 4/6 and 5/2.

If there is indeed 23V there, the regulator isn't doing it's job. A likely culprit would be a collector to emitter short on Q1 which can be checked with an Ohmmeter with the power off and maybe a minute wait after power down to allow the voltages to bleed off.

Bri

Thanks @Brian Roth and @zamproject. I was thrown by the wire color coding in the schemos as well so I double checked and on all 3 of the audio PSUs, the wiring is different than the schemo. Ie., on this bridge, the schemo shows Red/Brn coming in and Yel/Orn coming out, BUT on each of my PSU's, all three bridges have same wire colors coming out as coming in. So in this case, instead of Yel/Orn, I have Red/Brn coming off the bridge and feeding the Big elcap/stabilizer pcb. Not sure if this was a factory variance or if it was changed by a previous owner, but tend to think the former, since when I bought the unit, it was all original caps throughout. ANYWAYS....so I pulled the board with Q1. To test it for shorting, I set the DMM to resistance, put - probe on the base and the + probe on the emitter....reading was 5 ohms. Then reversed the probes and got the same reading. According to the internet, there should be a high resistance on one of the readings and low resistance the other way. same resistance both ways indicates a short. If that methodology is correct, then indeed, Q1 has a short. If there's a better way test Q1, I'd be interested in your thoughts. Thx!

 

[Walrus]

You are measuring the 4.7Ohm resistor R3 that is connected between base and emitter!
Is there any low reading between emitter collector?
If you really had 23 Volts coming out of the regulator card, then the protection circuit would have shut down the PSU.

Edit: Are all the screws in the transfomer terminal block tight?

 

[Pat Maki]

You are measuring the 4.7Ohm resistor R3 that is connected between base and emitter!
Is there any low reading between emitter collector?
If you really had 23 Volts coming out of the regulator card, then the protection circuit would have shut down the PSU.

Edit: Are all the screws in the transfomer terminal block tight?

Thanks @Walrus. Took Q1 out of the circuit and redid the reading. With + probe on Emitter and - probe on Base, got a 19.33 ohm reading. Then switching probes, got a reading that cycled back and forth between 3.5 ohm and 0. Does this reading sound ok?

 

[Brian Roth]

Measure resistance between collector and emitter (out of circuit).

Bri

 

[Pat Maki]

Measure resistance between collector and emitter (out of circuit).

Bri

Thank-you for the guidance @Brian Roth. With - probe on emitter, got .5 ohm and 0 ohm when switching probes. Also put the DMM in diode mode, put the - probe on the base and both the collector and emitter read .48V. Thoughts?

 

[Brian Roth]

A good transistor should read infinity ohms (or perhaps a very high resistance) between collector and emitter when out of circuit, regardless of which meter probe is on which transistor pin.

Bri

 

[Pat Maki]

A good transistor should read infinity ohms (or perhaps a very high resistance) between collector and emitter when out of circuit, regardless of which meter probe is on which transistor pin.

Bri

ok, so this sounds like a bad one then. What would be considered a high resistance in this example? 10 ohm? 100? more?

 

[Brian Roth]

Something like 100's of Kohms or in the mega ohms. In theory (and typically) infinite Ohms each direction out of circuit. Awhile back, I ran across a power transistor that raised my suspicion when testing out of circuit and I saw something like 500K Ohms collector-emitter. But, a brand new one I had in stock (and it was legit...from Mouser) read the same.

Bri

 

[Pat Maki]

Something like 100's of Kohms or in the mega ohms. In theory (and typically) infinite Ohms each direction out of circuit. Awhile back, I ran across a power transistor that raised my suspicion when testing out of circuit and I saw something like 500K Ohms collector-emitter. But, a brand new one I had in stock (and it was legit...from Mouser) read the same.

Bri

Ah, ok, so closer to infinity than I was thinking! So, once I got a few of the 2N5884's on order, I swapped the +15v board from PSU 70 with 71 (the one with issues) just to do a swap test. Sure enough the 23V reading I was getting went to +15V as it should. So I'm pretty confident that our diagnosis will get 'er back online soon. Thanks again to @Brian Roth @Walrus @juanito2008 and @guy_4 for your inputs and expertise. Very much appreciated!!!

 

[Brian Roth]

Don't run that machine with the errant PSU connected. You run the risk of frying opamps due to the overvolt condition. Also, buy the replacement transistor from a totally legit source...Mouser, Digikey, etc and NOT from Ebay, Amazon, etc.

Bri

 

[Pat Maki]

Don't run that machine with the errant PSU connected. You run the risk of frying opamps due to the overvolt condition. Also, buy the replacement transistor from a totally legit source...Mouser, Digikey, etc and NOT from Ebay, Amazon, etc.

Bri

Thanks, well noted. Struggled to find one since they're obsolete on all the major distributor sites. Not sure what to do other than roll the dice.

 

[Brian Roth]

Future Electronics and Rochester Electronics are not "primary" suppliers, but they often have NOS for obsolete parts. I've bought some odds and ends from them over the years and they seemed OK. Can't 100% say they don't sell counterfeits. Rochester seems to specialize in NOS.

https://www.futureelectronics.com/p...s--bipolar-transistors/2n5884g-onsemi-1429786

https://www.rocelec.com/part/01t4w00000NZ6jyAAD-2N5884G

The picture at Rochester is wonky but the specs seem correct. I also don't know what MOQ both companies currently have.

Bri

 

[Brian Roth]

BTW, Jameco is another source I use for NOS. In the past few years I've bought a "smallish" inventory of 74xx TTL logic chips from them for maintaining some Ampex ATR-100 and MCI JH-110 recorders. Alas, Jameco didn't have the 2N5884.

Bri