RCA BA-2C Buzzing Issue

[Cranehazard]

By choke I mean an LC filter in the power supply. An RC filter can do the same but it takes a big resistor and capacitor to filter out all the hum and the voltage drops too much. This has been my experience. Until the tube hiss is masking the hum, it's not quiet enough for me. Then the step up transformer on the input is louder than the tube hiss and your in business!

 

[mattrad]

Awesome, thanks for the tip! I'll look into LC filtering.

 

[dthodoris]

Big thanks to everyone for helping out.

Got the capacitors soldered in- experimented with a few values and ended up with 100nf as it was the quietest. Still a slight PSU hum left over that I need to figure out, interesting that I could cancel the noise completely by moving the caps around on gator clips. But noise is much more manageable now!

Reading that some different diodes may further help me out... does anyone have experience with UF4007s as opposed to 1N4007s in rectifiers?

Hi, it seems to me that by moving the caps around you essensially have a noise source that you move around.This noise from the leads couples into a sensitive part of the circuit.Now depending on the exact position you move the caps/leads into, this coupling can be in or out of phase with the original noise you began with.Unfortunately this cancellation is almost always dependent on the amp settings (gain volume or even eq) So you might have complete cancellation at low volume settings but at high volumes you might have more noise than you originally had.

 

[MisterCMRR]

The fact that moving the capacitors (and clip leads I assume) allowed a total cancellation of the buzz at one position, tells me this is capacitive coupling (through the air). If the power transformer secondary is center-tapped, the two diodes and capacitors have anti-phase voltage on them and would allow for the electric fields to cancel at some pick-up point in the signal path. I'd keep all that wiring, diodes, and caps as far as possible from the signal path. You might try some shielding as well. A simple way to test is to start with a square of aluminum foil, fold one corner so you can attach a clip lead without ripping the foil - and then cover the foil on both sides with plastic household tape (so it won't short out to terminals) and then ground the foil via the clip lead to chassis. Move the sheet carefully inside the chassis wiring. If the shield is between the power supply and the signal path, It may almost completely stop the buzz coupling. Experimenting with the position may suggest a good location for a little shield made of thin aluminum and screwed to the chassis (possibly with little notches at the bottom for wires to pass under).

 

[mattrad]

Very nifty trick with the tin foil on a clip lead- i'll give that a shot and see if it helps! Thanks again for all the help.

 

[mattrad]

Still trying to hunt down the remaining 120 buzz, tried moving around a shield (used a soda can) on a clip lead to ground. No results there, couldn't get any change in the buzz no matter where I put it. There is a fairly loud buzz emitting from the PT (or the diodes) when the unit is on. Much louder than I would expect- you can hear it across the room. Had to make sure the buzz wasn't just my test microphone picking up the actual buzz coming from the unit! Tried some UF4007s, no luck there. Wondering if anyone has any further ideas. Thanks!

 

[mjrippe]

Is your power transformer rated for more than the total current draw of the preamp, or just barely enough? A stressed transformer will make noise and heat (and eventually smoke). Or the laminations could be loose, but if it is a new tranny that is unlikely.

 

[MisterCMRR]

This is sounding like the power transformer is running in magnetic saturation! Contrary to popular belief, saturation (and the resulting lamination buzz and huge radiated magnetic field) is NOT caused by over-loading it's output, but by over-voltage on its primary side. Is your AC line voltage over 120 V? A common move to save money on power transformers is to operate them very close to magnetic saturation - sometimes a primary voltage increase to just 123 or 125 will put them into saturation (and they'll get hot even if there's no load on the secondary at all!). Overheating due to too much loading of a transformer secondary is caused only by too much load current and the DC resistances of the windings. It amazes many people when I explain that the strength of the magnetic field in a power transformer is the same, whether it has no load or a full load (it actually decreases very slightly under heavy loading). The field strength is determined only by the primary voltage and frequency. Saturation is also commonly caused by operating a 60 Hz transformer on 50 Hz!

When a power transformer saturates magnetically, the magnetic field literally squirts out of the transformer and it has a very "pulsey" character that often magnetically induces a "buzz" voltage in every wire near it. A good way to test for saturation is to power the unit from a Variac. If everything calms down if you reduce line voltage to say, 105 VAC, then you've got a saturation issue! I hope this little mini-tutorial helps!

 

[mjrippe]

Thank you Bill, clearly I was not aware of that. Funny, I was watching a video by Photonicinduction on YouTube last night (yes, he is back!) and he talked about core saturation as well. Would a few shorted primary windings cause this issue as well, or am I thinking backwards?

The OP mentions using a Hammond transformer which is a well established brand as I'm sure you know. Could still be the issue though.

 

[mattrad]

Thanks everyone! So i checked out the wall voltage- 119V at the socket. I have it wired up to the 125V primary. Also tried the 117V primary option, still buzzes. The buzz surges loudly right when the unit is powered on, then quiets down for a few seconds before it returns to a steady buzz. Don't have access to a Variac currently but I could get my hands on one. I do have another PT of the same model installed in a different unit, maybe it's worth it to swap them out if a defective PT is a possibility.

 

[MisterCMRR]

Hammond's are well-designed and 119 V to a 125 V winding rules out saturation. Even a single shorted turn will generally cause serious overheating. But the behavior of buzz louder at power-up suggests that it's the current in the energy-storage capacitor (the large on charged directly from the rectifiers) that's somehow injecting current into the ground system. Perhaps I should look more closely at the photos of the wiring that you posted earlier.

Getting back to your original post: "It is amplified with the volume pot, and when I short the grid of the first gain stage to ground it goes away completely." Is this still true? Could you post a complete, as-built, schematic of the unit? Some of the clues are a bit puzzling.

 

[mattrad]

Yes, it is still amplified with the volume pot. When the grid is shorted to ground the buzz disappears. To be clear, the buzz surges audibly from the power supply upon startup, but doesnt come through the headphones until the tube warms up and starts passing signal. There IS however, a more hum-like noise that surges through the headphones upon startup mimicking the PT noise, but once the hum settles down its fairly negligible. I drew up a schematic posted below, as well as the original RCA BA-2C schematic. This only includes one channel as the 2nd channel isn't wired up to the power supply yet.

The differences:

1. SS Rectifier instead of 6X5
2. Anode choke + capacitor instead of gapped OT
3. Voltage drop resistors (R8 and R9 on original schem.) reduced by roughly 1/2, in preparation for 2 channel unit from one power supply
4. R11 and C8 removed - used for tone shaping as I understand.
5. Output attenuator added

 

[tony hunt]

Could you draw you grounding layout too?
On the photos it is really difficult to follow and seems like a mix of star–grounding along with several connections to the steel chassis.

 

[mattrad]

Yes the wiring is a bit of a mess in those pictures! The grounding layout is essentially as it is in the schematic- all preamp grounds are tied together at one point and sent to the filter cap grounds, which all bolt to the chassis at only one point- on the negative end of the first filter cap. The OT primary ground is included in this star ground, unlike in the schematic, as well as the high voltage CT and the heater virtual CT. XLR pin 1s are connected to chassis at the jack.

 

[tony hunt]

I have sketched of what I understood above and what I see in the photos - please check:
a) in the photos it seems the power transformer CT has a bolted connection to the steel chassis on the PT tag-strip (between the PT and the filters caps) right?
b) there is another bolted connection to the steel chassis at the negative of the first reservoir cap & the 470k bleed-resistor?
c) shown dotted: a long meandering loop of black wire connecting the above chassis connections a) and b) ?
I cannot really see where the OT is connected, so I showed it at the last filter cap.


Quote from above post: "3. Voltage drop resistors (R8 and R9 on original schem.) reduced by roughly 1/2, in preparation for 2 channel unit from one power supply"
Halving the values seems excessive. The second half of the circuit will only draw a few mA, I guess 6-8mA. You have thrown away I guess somewhere around 25dB ripple rejection. That may be a component of either the buzz or the background hum.

Primarily I still think it is a layout issue. A steel chassis requires more careful attention to layout than non magnetic metal enclosures. Please check the sketch.

(Edit on 22. July 2021: sketch of grounding edited to reflect post #58: removal of chassis connection directly at PT-CT)

 

[Cranehazard]

Star grounding hasn't worked for me with mic preamps. I create a continuous ground plane and use the shortest possible paths to ground, preferably 1/16 inch. I also completely ignore the difference between chassis ground and signal ground. This has allowed me to get my tube preamps as quiet as I need. As quiet as my solid state stuff maybe even better because of the step up transformers.

 

[mjrippe]

Back to the transformer possibility, check this thread:
https://groupdiy.com/threads/false-center-tap-blown-why.78188/post-999620

 

[mattrad]

The high voltage CT is connected to a floating lug on the PT tag strip that is then connected to the filter cap chassis ground via the meandering loop- just one point bolted to chassis. Not the most elegant layout but i cut the CT wire short for other purposes then had to work with it! OT is connected to the same star point. Ill do the rounds experimenting with grounding schemes again. Thanks for the heads up about the dropper resistors, ill return them to stock value and see if it has any affect on the buzz. EDIT: Changed them to stock, no affect.

Also, interesting to read in the other thread about a defective Hammond PT mechanically buzzing- swapping the PT might be worth it just to rule it out.

 

[mattrad]

Also, was just experimenting with different cap values across the HT winding (in addition to the caps in parallel to the diodes.) This knocks the buzz down further and it seems the higher I go in capacitance the less the obtrusive the it is (the capacitance forces the buzz into a lower frequency as i understand it.) Interestingly the mechanical buzz from the PT changes frequency as well when I connect/disconnect the capacitor. My question is, is there a limit as to how high can you go with capacitance in this situation before problems arise?

 

[tony hunt]

The high voltage CT is connected to a floating lug on the PT tag strip that is then connected to the filter cap chassis ground via the meandering loop- just one point bolted to chassis. Not the most elegant layout but i cut the CT wire short for other purposes then had to work with it! OT is connected to the same star point. Ill do the rounds experimenting with grounding schemes again. Thanks for the heads up about the dropper resistors, ill return them to stock value and see if it has any affect on the buzz. EDIT: Changed them to stock, no affect.

Also, interesting to read in the other thread about a defective Hammond PT mechanically buzzing- swapping the PT might be worth it just to rule it out.

Good that the high voltage CT is on a floating lug. I have corrected the sketch in post '55 to reflect that.

I understand the limitation with the short CT wire. But the solution, via the meandering loop, may indeed be the cause of the 120Hz buzz. The connection between the negative of the first reservoir capacitor and the CT should be kept as short as possible. That meandering loop is like a stinky sewer spewing 120Hz anywhere it gets a chance.

If I get time, I intend to make you a layout suggestion. If I get held up, I think you can turn your tag-board with the reservoir und filter caps around and get real close to the CT.

Due to the steel chassis, I would mount the PT with non-magnetic / non-ferrous bolts, insulating washers and make an additional safety ground to the PT. I do this to prevent any PT core flux from flowing into the chassis.

I am still pondering your last statement about the "the mechanical buzz from the PT changes frequency as well when I connect/disconnect the capacitor."
Is this the first time that you mentioned a mechanical buzz?
That 200 series Hammond PTs does not have as much iron as the 300 series. Even though the company has a good rep I would not rule out an issue. At this stage, check that the laminations are not loose and the the bolts are tight.