• 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

Need help getting rid of hum

Replace C51 and C 57. They don't have anything to do with the problem but if they are out that far, it's a problem in that particular area. Do you have a schematic of the SDS Board and it's pinouts.

Answers in RED!
thornev said:
Larry - I agree 100% about the SDS board. I'd be VERY willing to dump it (for the exact same reasons you cite) and replace it with a more accessible, revamped power supply though I would need help with the parts and installation. I've seen plenty of revamped power supplies online so I know what they look like, just not part specs. Can you refer me to something? Only thing I can suggest is to find another 500c WITHOUT a SDS Board, and copy the power supply part for part. It's mainly a Hayseed cap kit, a couple of Diodes (1n5006), and some resistors. If you go with the Hayseed kit, for the BIAS Voltage caps, use the Nichicon caps.

As far as the current, the measurements I posted above were after I resoldered the big resistor. I checked the IBAM board and still have the same issue - cannot get them below 37mA and V11 is all the way down on a 24-turn pot ! Your Bias voltage is too low. Get it bumped up to at least -19 to -22v (If it has a Silicon Diode Bridge rectifier, you'll need to add a resistor to match the voltage drop of a selenium.)

You say my output tubes are starved. What is it for which they're starved? I thought that since I was dialed in at 33mA and voltage to the tubes was 398v that I was spot on with everything about the output tubes. Actually I was worried that 438v out of pin 3 on the tubes was too much. I know that -19 is greater than -24 so I don't understand why you say it is "too low." I was saying V14, 15, 16, & 17 heaters are starved. The BIAS circuit is too HOT. And the V14 thru 17 filaments are in this circuit. Which is why your filament voltages are LOW. The closer to ZERO VOLTS the hotter the bias. -19v is nominal, better is -21v to -24volts.

The big question... How do I remediate to solve the "starving" issue? I've read about this issue and remember that I may have to add a resistor somewhere? I think at this point I'm going to move the IBAM power lead back to V14, see if that fixes all this. These numbers are drastically different from measurements I took on May 11. Something went wrong. Or the May 11 measurements were on a 120vac line. Could that difference in AC voltage (110 vs 120) account for the variations in measurements? Normally the 500-C gets powered from a power strip whose voltage I have set at 110vac. Once we get the voltages straightened out, variac the unit and adjust it so the V14 thru 17 string is getting an average of 12.6V per tube +/- 5%. or 12.0v to 13.0V (rounded off to Whole Volts.) Note that variac voltage and run it at that.

Could it be the metal film 10 ohm 1/4w 1% resistors I put at all pin 5 of output tubes? I assumed metal film would be OK. I read up and believe metal film is a better choice for a few reasons. No it ain't that.

Here are all the voltages coming off the SDS board (see attached parts image):

A = 335 vdc (spec = 430. this worries me) goes to left PI pot, center terminal. This one(430v) should go to the OUTPUT TRANSFORMER PRIMARY. You've got A & C reversed???
B = 356v (spec = 375. under spec) goes to daisy-chained pin 1 of output tubes. Presuming you have 100Ω resistors from pin 1 to 4, this would be your screen voltage.
C = 397v (spec = 350. over spec) goes to t-strip terminal where R143 non-tube side is connected.
D = 299v (spec = 295) goes to orange wire to MPX power supply. This one is ok
E = 208v (spec = 215) goes to purple wire where R22 and R19 meet which connect to V17. This one ok.
filament = -17v (that's worse than -19 I measured before, eh? no spec for this part) Look at the DIODE Bridge on the schematic. -22 is original spec.
filament winding @ 8.2 ohm 5w resistor = -6.76v
D1 = 396v at stripe side; 197v other side. 197v is the R/Y transformer lead.
D2 = 197v at stripe side; 0v other side.
C14 = 197v in; 0v out
R2 = 183v on one side; 190v other side.
R3 = 190v on one side; 197v other side.

V14 pin 4 = -17v This voltage is too low (Heater / BIAS circuit) should be at least -22vdc spec.)
V17 pin 4 = -17v This voltage is too low (Heater / BIAS circuit) should be at least -22vdc spec.)


output tubes:
pin 1 = 356v
pin 3 = 390v Plates
pin 4 = 354v If pin 1-4-8 connected they will be predominately the same. (Screens)
pin 5 = 373v - 381v Your IBAM ISN'T WORKING OR you missed the Decimal point in the meter reading. Should be MV
pin 6 = -16v Way too low (towards ZERO VOLTS) Should be around -18 to -20v depending on the tube. If they are Tungsols figure -22 to -24 to get in 75% range.
Click to expand...

Give serious consideration to taking it over to Mike @ Hudson Valley HIFI. He's a member here and does good work. He should be able to alleviate the problems with the SDS board.

Larry
 
Honestly Thorne, I think you have problems in the power supply (either unintentional due to mods, or otherwise). I'm not that good with Solid state unless I have a schematic, and then I have problems with Transistors and power flow. The board layout confuses me. It looks like to me that a couple of your power points on the board (a & c) are possibly reversed, causing a lot of problems. Let's see if someone else picks up on it and can run with it. I've hit my limit. Sorry Man.
 
Thank you for all your responses, Larry. You've always been my wing man. Yeah, something has gone wrong in the last week and I can't explain what could have happened. I'll figure it out or get it figured out and post here. Thorne
 
Well, Larry, I finally figured out what causes the drastic change in voltages. It was clipping the ground wire at the YEL/BLK V14 pin 4 that I did to reduce hum. Just for kicks I reconnected it and all the voltages were back to "normal." Also the IBAM adjustments are again operative. With the reconnected wire all the output tubes mV are around 0.25 which means I need to adjust them upward back to where they were. Can you explain this electronically? I thought the "double ground" was a bad thing, but it seems the SDS board needs it that way. I get -22mV for the bias current now.

Unfortunately, the hum is back. I moved the IBAM power lead to V17 pin 4 and that decreased the hum a little. Want to find another place where I can get -21 vdc and try there.

And you were right - I did have A and C reversed on the SDS board ! Very perceptive !

EDIT: The numbers now:

SDS A = 352mV
SDS B = 374
SDS C = 409 (spec = 430 so I'm concerned. this is middle terminal on left PI pot)
SDS D = 313
SDS E = 216
filament = -21vdc
V8 p1 = 374v
V8 p3 = 406
V8 p4 = 374
V8 p6 = -19.6 vdc

And I've got the current dialed in at 34mA.
 
Last edited:
I find it hard to believe that SDS or the installer used this one lead's shield as a ground. If you can, Run a new wire from the grounding point to a chassis ground very close to the board. And disconnect the shield wire(leaving the end at the tube connected) from the board. This will give a chassis ground reference AND should bring the hum back down to the level it was y'day., negating a "ground loop".
 
Last edited:
No. That lead isn't the SDS board's main ground. There is a thick green lead that goes to the ground by can cap C92 (which is deactivated). It's marked as GND in the schematic (in post 77).

So this new ground... Do not connect it to the same ground point as what the SDS GND uses? The new ground lead must be as near as possible to the SDS board?

Re "leaving the end at the tube connected"... I cannot disconnect wherever that lead goes because it goes underneath the SDS board. I can't get the SDS board out without cutting a bunch of other wires. That's another one of the things I dislike about the SDS board - it's a PITA to get it out and see what's underneath.
 
Last edited:
NO. Don't add another ground to the SDS Board, you just exacerbate the groundloop issue. But DO CHECK for continuity between the SDS board ground point and chassis. It should be 0.0Ω or a decimal below 1Ω. If NOT, take the end of the Big green ground wire connected to the chassis and clean the end, and the chassis point. USE a new bolt, STAR WASHER and a nut (find a convenient chassis hole close to the SDS board). Any other grounds are going to cause a ground loop, and problems. I think what's happening is the Green wire is loose or corroded somehow, and they either got lazy and connected that shield wire up or it was connected when the SDS board was installed. Which is wrong on both counts. Once you get the Green Wire cleaned up, resoldered and showing good continuity, and the voltages are good, disconnect the shield wire on the heaters and recheck.
 
I'm pretty sure I did test for continuity between the SDS board ground point and chassis. Moving where it's connected to chassis to a different location closer to the board is worth a try. I'm wondering if there are additional SDS board grounds I can't see underneath the board. Geez I hate that board.

What do you think about this idea... Since that V14 YEL/BLK lead has a ground to the SDS board, what if I disconnect that green wire marked GND on the SDS board (i.e. disconnect from chassis)? That would leave the one ground for the board near V14 which should also solve the issue? The YEL/BLK ground lead is only like 20AWG while the green GND is like 14AWG. Does that matter? I believe length of ground lead doesn't matter. Maybe not a good idea since SDS board probably relies on its GND terminal to have a real ground. Instead maybe connect the YEL/BLK ground to SDS board green wire GND and disconnect green wire from chassis ground? Maybe not because I don't know where the YEL/BLK ground goes under the SDS board and that might screw up the voltages again. Sigh.

I emailed HV HiFi. Waiting for response. I hope you don't mind that I name-dropped you. Maybe that was a bad thing? :D (kidding)
 
Last edited:
I had the same problem, with a SS magnavox Imperial.... sounded like a guitar amp without anything plugged in ... just a power hum ... not loud, the music would cover it up when it was playing... I played with grounds , added more grounds , reflow solder joints ... nothing worked, pulled it and put on the bench so I could trace with my scope ... turned out to be a leaky preamp transistor ... changed it and power hum dissapeared...
 
Thanks for chiming in, Elton. Can you tell me how you used the scope to track down the bad part? I tried doing that on my 500-C but couldn't figure out where in the circuitry I could use the scope. Are there other concerns besides high voltage that could damage the scope?

The hardest part for me was identifying which circuits hold the audio signal so I know I should get a nice sine wave. I tried all the pins on all the tubes except where 400v was exceeded as that's close to the limit of my scope (Rigol DS1054z).

If you could write the step by step process you used, I'd greatly appreciate it. Thorne
 
Last edited:
No DON'T disconnect the GRN GND. You need the gauge on that line so it will handle the load without overheating. (Same principle as using 8-3 on an oven or dryer, replace it with 12-3 and you'll have a FIRE due to increased resistance and inability to handle current, wire insulation melts and catches fire! ). Go ahead and check again the GRN wire and post results, with and without the Yel/Blk shield wire connected to the board. Numbers, please. No problem name dropping. Mike knows who I am.
 
Tomorrow I should be getting replacements for C51 and C57. I'll replace them and then post your requested results. Were you looking for voltages and/or currents and anything else?
 
Post continuity results of grn gnd wire, with and without yel/bulk shield wire connected. POWER OFF.
 
Continuity results of grn gnd wire, with and without yel/bulk shield wire connected. POWER OFF. ELECTRIC UNPLUGGED. Green ground wire soldered to big glob of solder at C92, black MM lead connected to chassis ground elsewhere; red MM lead connected to "big glob of solder":

1. With yel/blk ground lead CONNECTED -> continuity exists, 0 ohms.
2. With yel/blk ground lead DISCONNECTED -> continuity exists, 0 ohms.

I also discovered that plugging in AC electric in one direction reduced hum compared to reversing the AC plug. But it's not enough of a reduction to declare "problem solved."
 
Continuity from one end of grn wire at board to chassis? You want to test thru the wire.
 
I emailed the designer of the SDS board (he has a web site). He was unable to offer any assistance except that he's willing to help if I send him pictures.

I cannot get to the underside of the board where any exposure to the lead's metal is accessible. On top the lead's insulation is snug up against the board.

EDIT: Should I make a small incision in the green GND wire at the board so I can measure as close to the board as possible with one lead and the other lead at the chassis end of the green wire?
 
Last edited:
Take a straight pin (raid the wife's sewing kit) and pierce the insulation getting the pin into the wire. Do in such a way as to NOT get the pin near anything powered. Touch your probe to the wire at that point, with the other probe on the chassis somewhere other than the "Big blob of solder".
 
OK... pin in green GND wire about 1/4" inch from SDS board. Common MM lead attached to pin. + lead from MM touched several different non-solder-glob chassis locations both near and far from SDS board:

1. yel/blk ground lead attached, continuity exists, 1 ohm.
2. yel/blk ground lead disconnected, continuity exists, 1 ohm.

I believe this test proves that the green SDS board GND wire is indeed a ground for the board, right? Unfortunately the yel/blk ground lead must also remain connected so that the voltages around the chassis don't go way off the farm which they did when I measured with that lead disconnected.

I ordered new C51 and C57 caps, but they're too large. Reordered smaller, should arrive tomorrow. Thorne

EDIT: If I'm going to keep this SDS board, maybe it's time I do what I have to in order to get pictures of the underside of the board and the chassis. Larry, do you think it's necessary to do that in order to move forward? It's not a trivial effort because I'll have to cut several wires in order to get the board free. And even then, I don't know how much play there is in the wires underneath so I can move around underneath.

EDIT2: I found a problem. Apparently when I resoldered the yel/blk ground lead to a ground point, I must have melted the insulation around the + lead because I now have continuity between ground and the + lead of the yel/blk wire (13 ohms). But what's weird is that I opened up the insulation and I don't see where the 2 wires are touching. That means V14 pin 4 (which is a heater) has been grounded. I'm lucky it didn't cause a catastrophic short. Fixing now. Will report back here. Maybe this issue explains why my voltages were all over the place with the ground lead disconnected.

My analysis when comparing SDS to stock... SDS board C10 and C11 are equal to C97A and C97B in stock and serve as the bias filter caps (filament which is "B-", right?). This filament is carried in the YEL/BLK wire to V14, V15, V16 and V17 pins 4 and 5 which are heaters in a 12AX7. But I also see on the 500-C schematic that those pins on the 4 tubes are grounded. Does this make sense?
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom