[Occam]

Syrah,

Typically, an inverter based VFD (variable frequency drive) generates a horrid noisy pwm voltage waveform but a sorta sinosoidal current draw... To deal with the harmonics, noise and high current pulses in this topology either  dc reactors (inductors) are incorporated between the rectifiers and storage capacitors within the unit, or ac reactors are placed before, externally on the AC line.
The later might work quite well, with the addition of appropriate X2 caps in minimizing its noise. All the X2 caps, save for some from Mallory, are rated for 240vac.

Dunno if you're powering your AC unit(s) with single phase (single ended or split) or 3 phase, but HVAC often uses a 3 phase line reactor, wired appropriately for all of them. That being said, I'm not familiar enough with the requisite calculations for sizing the reactors, or familiar enough with motor drive noise that I can advise you.
I assume your AV system is powered via a dedicated circuit(s) and the air conditioning unit was installed properly....

You really need a electrician, EE or PE familiar with (typically) HVAC, from a noise sensitive environment like hospitals or labs, if you want to deal with that noise at its source.
There might well be such filters available from electrical supply houses, but I'm not familiar with them.

[Speedskater]

Does the Air Conditioner make noise continuously (when it's operating) or just when it starts and stops?

It's always best to deal with noise at it's source.

[Syrah]

Thanks guys.  Boy I'm learning a lot about electricity these days.  I've got a new sparky and it sounds like I need him back.  He explained to me that (contrary to what I thought) my AC is not two 120v lines out of phase, but just a center tapped transformer giving me two 120v lines.

When he wired up my stereo room, he used thick four wire cable and gave me a dual receptacle where I could use either or both poles.

So here's the funny thing.  Normally, one pole is quieter than the other.  Except when this air conditioner compressor goes on, then the (normally quiet) pole becomes the super loud one.  The (normally slightly louder one) does not seem to be effected by the air conditioner compressor.

All slightly strange since the ac uses 240v, so it's using both poles.

So for now, I'm using the slightly louder pole since my X caps and P10 seem to quiet that slight hum down - although still not dead quiet.

Maybe what I'll do over the weekend is switch the connections on the ac unit.  That way I might end up with one dead quiet pole, and one extremely loud pole - as an interim measure.

Then I'll speak to my sparky about what to do to get rid of the noise completely.  Maybe I need a Y cap from the noisy pole to ground at the ac unit, combined with a Felix type filter.  But that's one I'm not going to take on by myself.  Since my X cap stack filter after my P10 seems to be useless against this ac noise, I'm guessing it common mode noise.

Another conceivable issue is that the ac unit is wired with only 3 wire cable.  I think that's all I had at the time.  I'm wondering if the guy who installed it did it properly.  Another question for my sparky.

Thanks.

[Folsom]

I don't know if I'd recommend a Y cap, at least from hot to safety (ground).

This might be a stretch, but if you could draw the electrical layout...

[mjosef]

Thanks Paul.  I live on a Caribbean island though, so it's tough getting things I don't already have, even something as simple as a breadboard.  I have to be like the Professor on Gilligan's Island.

Robinson Crusoe would suggest something like this, in the absence of breadboards:



A few short pieces of 12 g. solid core wire for the rails. Finish up with wrapping the whole shebang with rubber splicing tape to prevent fireworks.

[Syrah]

Nice work.

Why are Y caps not a good idea?  I thought they were good for common mode noise and if they Y rated they're safe.

[Folsom]

When they go from hot to safety ground they tend to have an adverse effect in some equipment because safety ground typically has a connection to circuit. In balanced units it's a non-issue, so in some systems it's fine.

[Syrah]

Thanks.  The Dodd amps do not connect to ground at all.  So it sounds like this might be worth a go.  What would good value Y caps be?  I read somewhere that they are typically lower uF than X caps because of the relevant noise frequency of CM noise.

[Syrah]

And would it make sense to put them in place of the CMC on the Felix board?  So instead of the CMC I would put Neutral to Ground caps (value(s) TBD) and Hot to Ground caps (value(s) TBD)?

[Folsom]

The filter would be significantly weaker without the CMC. The CMC influences the capacitor behavior positively.

You can certainly try the Y capacitors, see how they work for you. But just to note if your other equipment in the system uses safety ground the potential is still there, since your amps need a source.

[Syrah]

Thanks.  The rest of the system is not on the P10, and it's on a different circuit, so hopefully the path to ground is fine - but pretty easy to try it out.  What values are typically good for audiophile purposes, 0.01uF and lower?

[Folsom]

You could do a couple at different values like the Felix has.

If you're 100% that the poles (line and common) will never get switched, the ground to neutral has the sky as the limit. You can actually use massive values of non Y rated caps. I use a 200, 10, and .1. However I never deploy this unless I know the circumstances of the poles because the values are very high and would be dangerous in the wrong setting. (I may come up with a workaround for safety on this, but haven't found a suitable one yet)

[Syrah]

Getting back to the snubber idea.  I'm wondering if I can't give this a shot myself.

Here are a couple of picks from inside the Dodds.  It looks like those blue wires coming out of the transformer are the secondaries?  Then it looks like they're both soldered to those round copper circles.

Assuming that's the case, and I could snap some more photos of the wiring to be sure, the snubber (330 ohm carbon resistor and 0.022uf orange drop cap) would be wired in between both copper circles?

Thanks.

[randytsuch]

The red and yellow wires, that go to the high voltage cap banks, are likely to be the high voltage secondaries.  They generate the positive and negative high voltage needed for tubes.

The rings look to be grounds to me, the single point ground.

I would also guess there is a 2nd transformer, to the right of the high voltage one.  Guessing that the 2nd one is for the heater power.  Is the power input on the top left part of the picture, that got cut off?

Do you have a meter to measure stuff?

It looks like there are resistors on the lower high voltage cap bank, so I'm pretty sure there is another set on the top bank
These resistors drain the high voltage after you turn off power.  So, a few minutes after you turn off power, there should be no high voltage, and the amp would be safe to measure.

Just to be safe, I would very carefully measure the voltage across the caps after a few minutes after turning off.   Make sure it is 0, or some low voltage.

Randy

[Syrah]

Thanks Randy.

I'll snap some more pictures of the back tonight.

I know from having changed that internal fuse you see, that there is still a good chunk of voltage after it's unplugged.  I learned that the hard way!

So it might be that the snubber goes across those red and yellow connections, subject to further investigation.

[randytsuch]

On the bottom row of caps, third from left, there are two "things" soldered to it.  I thought they were bleed resistors.  Are they fuses?

If you don't have bleed resistors, I would add them.  Makes it safer to work on, guess you learned the hard way.  High voltages can be lethal, so be very careful.

Randy

[Syrah]

They're not fuses.  I was referring to that single fuse on the circuit board.  I now only change it with gloves on.

I will double check if they are resistors on that cap.

Thanks.

[Syrah]

Those two black wires are the mains coming in to the transformer.  Next to them are the green, red, and orange wires coming out of the same side of the transformer.

In the other photo you see the red wires going into the first caps on the one side of the amp.  The similar cap on the other side of the amp is connected to yellow wires which seem to go to the IC.

I can try to trace those orange wires, it looks like one of them goes to another cap at the end of the row of caps that the red one is plugged in to.









If that is a resistor on the caps, it’s a funny looking one.

That other transformer in the middle must be an output transformer.

What do you think?

Could it be that red and orange coming out of the transformer are my DC + and - ?

[randytsuch]

I think you're right, the other transformer is likely an output transformer.

The black things on the caps are diodes, they are rectifying the voltages to generate a DC voltage.

There are multiple outputs from the power transformer for a tube amp.
Need heater power, B+ (high voltage), and maybe another high voltage.  In thinking about it, not sure if there is a negative high voltage or not, there may be two different high DC voltages generated.
Hard to tell by looking at pics exactly how things are connected.

One thing you could do is get the part number of the transformer.  I think it starts with 17 but the rest is hidden under wires.  With the part number, you can find the transformer data sheet, and that will tell you what the different wires are.

Randy