| whitelabrat |
I've read bits and pieces throughout this forum on troubleshooting power related hum, but nothing that gives a real strategy for going through things.
My dilemma. I've just about completed a very sweet tube preamp (12AX7, 12AU7), but there is a faint 60Hz buzz that could possibly induce madness.
I've grounded the chassis and twisted various power, input, output interconnects with some improvement. Carefully touching the insulation of the input wires makes the buzzing worse. So I tried wrapping the inputs with foil which didn't help any. The strange parts what after turning on an input source the buzzing decreased (but not 100%). It must be RF related.
So this leads me to believe that perhaps some shielding or filtering of the power coming off the power transformer is needed. I've got two 6.3Vac, a 300-0-300Vac, a 15Vac and, 12.6Vac leads coming from the transformer. |
|
|
| cbutterworth |
Whitelabrat,
What I did that really reduced the buzz was to install a ground-lift resistor network. I take the star-earth connection and wire it to 20 ohm resistor, that sits across a diode-bridge and cap filter. Supposedly, the cap filters any noise, and the dioide bridge provides a safe path to earht should the resistor fail, as diodes fail shorted and can tolerate high voltage.
When I first set this up, I used a 10 ohm resistor (5W, I think). The hum level dropped significantly. Then I upped it to 20 ohms. Now the hum is barely audible, although noticeable during extremely low volume music. It sounds almost like a gently idling car parked 30 yards away. Of course, I also sometimes confuse it with the quiet hum emanating from the fridge in the kitchen (the curse of open-plan living).
I am also wondering if the small amount of hum may be tube induced, as it seems to get quieter as the amp warms up. Although I am going to reflow all my solder joints just in case.
If you want more info on the ground-lift circuit, let me know, and I'll find it for you.
Charlie |
|
|
| gofar99 |
Hi, Many things may be going on here. What is the schematic? Is it a commercial kit or completely diy? What is hooked up on the input and output sides? My best guess with out more info is a ground loop or need for better (or more) shielding on signal conductors.
Good listening
gofar99 |
|
|
| whitelabrat |
It is a commercial Lite Audio LS7b kit with an R-core transformer. I've got the transformer and PCB mounted on standoffs inside a recycled chassis. I'll have to shoot a photo and post it. Everything looks good. Solder points are clean and solid. Good parts appear to be used.
The inputs are simple RCA jacks wired to the PCB using solid core copper wire. No potentiometer or anything. I figured I'd put my stepped attenuator in it's own chassis in case I'd like to use it in a passive configuration. Likewise with the channel selector. One note about the RCA jacks; They are isolated on a PCB material and their ground does not touch the chassis at all.
I've have the outputs wired to two sets of RCA jacks. A these aren't twisted, but they are very short.
For testing I'm using a cheap DVD player connected to the preamps's inputs. No attenuation. The output go to inexpensive computer speakers. |
|
|
| gofar99 |
Hi again, I searched for your preamp and apparently the only site is in a language that my server won't translate for me. A photo would help. Schematic too. From your description it sounds like a quality product, so I would concentrate on the ground loop angle. One thought comes to mind. With no pot, presumably on the input (although some have them on the output side), is there any fixed resistor from the input grid to the signal ground. If is floating, it would be rather prone to hum pick up when nothing is plugged into it.
Without the schematic and photo I'm just guessing. There could be a ground loop through the power connections and if you use three prong power cords between the preamp and power amp. I had one like that and it drove me nuts until I found it. If the power transformer isn't gounded, it can radiate a strong ac field and be picked up by the circuitry. I usually remove the paint from one foot of each transformer and attach a ground wire to the chasis ground point (not the signal ground). This helps quite a bit. Hope these musings help.
Good listening
gofar99 |
|
|
| whitelabrat |
Ok. I've attached a photo of things so far.
The ground is soldered to the power supply and attached to the chassis at one of the standoffs of the power supply. I've checked it with a multimeter to be sure everything is good.
Would the polarity of the plug matter? |
|
|
| frank754 |
I found that based on the preamp I built, if you run the AC into the transformer as usual, the polarity shouldn't matter, as long as everything is on the "other side" of the power transformer (filament supplies, HV,etc.) and you don't treat the incoming AC end any differently, one side or other. You should (if you can) use a polarized plug and have the switch and fuse on the hot side of the AC (unless of course you are in Europe or using 240 where they are both hot.
On most amps it won't matter, but for my pre-amp, I found I could only maximize hum reduction with a 3-prong wire, with the wall ground (3rd wire) grounded to the chassis. And again, I didn't need a 3-wire cord on any other more powerful amps. |
|
|
| gofar99 |
Whitelabrat,
Comments by frank 754 parallel my experiences. Although I have found that if more than one piece of equipment in my system has a third wire ground I get a ground loop. My choice is to have such a ground on the preamp only.
I just looked at your photo and genrally everything looks pretty good. The board seems OK. I have concerns about the proximity of the transformer to what appears to be the output tube. There also seems to be a lot of stray wires in that area. General practice is to get any wires that have ac on them away from signal paths and components. Also they are usually run as close to the metal chasis as possible. I can't tell from the photo how or if the signal and chasis grounds are connected. My overall sense is that the case is too small. If the power transformer was fully shielded it might not be as much a problem. There are a number of sources for a switching mode power supply that might solve your problems if the trannie truns out to be the source. Most such supplies are around $50US.
good listening
gofar99 |
|
|
| whitelabrat |
The board and transformer are sitting on roughly 1 inch standoffs. Perhaps things should be closer to the chassis? I've tried a different set of tubes with no improvement in hum/buzz. I do have a lot of industrial lighting in my listening area which may be contributing to the noise. The transformer should be as quiet as a toroidal, and much better than an EI type.
This thread seems to be on it:
http://www.diyaudio.com/forums/show...497#post1145497
I'm curious about cbutterworth's ground circuit.
In any case, at least I'm not alone! I thought I was doomed. Thanks for all the replies. |
|
|
| cbutterworth |
The circuit that I use for the ground can be found at:
http://sound.westhost.com/earthing.htm
Rod gives a good explanation of this ground-lift circuitry and its safety advantages. Anyway, I am using a 20 ohm resistor and will swap it out for 30 ohms (maybe even 40). Either way, there certainly seem to be advantages to using some sort of connection to earth rather than leaving the entire circuitry floating.
I ordered parts from Mouser and can the whole circuit costs only a few dollars.
Regards,
Charlie |
|
|
| whitelabrat |
| One more thought I just had. What about putting ferrite blocks on the transformer outputs? Correct me if I'm mistaken, but these are also referred to as chokes? |
|
|
| Zibi |
| Only for high frequencies. |
|
|
| Zibi |
"The inputs are simple RCA jacks wired to the PCB using solid core copper wire. No potentiometer or anything."
No any resistor between input tube grid and ground? |
|
|
| whitelabrat |
| Yes and no. I keep a stepped attenuator external to the preamp. There isn't much room in the chassis for it otherwise. |
|
|
| Jaap |
| The transformer could be too close or in the wrong plane to the circuit. If possible try an outboard powersupply or outboard transformer to check this option. Perhaps it is possible to have a piece of shielding material between the circuit and the transformer, or tilt it in another direction ? |
|
|
| whitelabrat |
The buzzing is almost inaudible since I've gotten everything plugged in. It's power strip seems to have some filtering that may have helped.
I get a very nice sound from the preamp now but the S/N is pretty high. Anyone have suggestions for quieting things down? Replacing cheapo capacitors? Better resistors? |
|
|
| whitelabrat |
I've tried attaching a 20 ohm resistor to the ground from the transformer. The ground from the wall is attached directly to the chassis. No improvement. I even tried a bypass around the 20 ohm resistor to compare the results. None.
The most noticeable change in hum occurs when unplugging the source from the inputs. Things go from small hum, to BIG HUM!
So. Either my tranformer isn't up to snuff, or something isn't happy on the circuit board.
Excuse me while I go nuts! :whazzat:
Hi-Rez photo: http://mysite.verizon.net/oldamps/ls-7b-smal.jpeg |
|
|
| salas |
You are experiencing a ground loop. Where is that pcb's ground to chassis, at what point? Obviously the circuit looks for signal ground and finds it wrongly through the return side of the input interconnects.
Are the RCAs insulated to chassis? |
|
|
| whitelabrat |
There's only one place where the board is grounded. It's the 0 where the 300-0-300 connects to the board. All the inputs and outputs are insulated from the chassis which is grounded and I verified this with my multimeter. I also poked around the board where other connections are made to the transformer. Nothing going to ground.
The circuit board is also isolated from the chassis and I again verified this with a multimeter. Something could be bad on the board.
I guess my next question regarding troubleshooting would be, how do I test the power transformer and supply for trouble? I'm also thinking the problem may be on the input side of the circuit? The buzz comes from both channels though. Anyone suggest any good books that may help me to have a clue?
I'll try to post a recording of the buzzing. It's a very sharp buzz rather than a mellow hum. Not that it makes a difference. |
|
|
| salas |
| So you have a PSU ground and no signal ground to chassis. Please get crocodile clips and a 10 Ohm resistor. Attach one clip to an input RCA ground side and through the 10R attach around to chassis and see what happens. Do L&R channels test zero R when you probe L & R input RCA ground sides? |
|
|
| jrevillug |
| quote: | Originally posted by frank754
(unless of course you are in Europe or using 240 where they are both hot.
[/B] |
Here in Europe, despite using 240V, only one of our lines is hot- we don't have the + and - 120v and ground thing going on that you have in the states, IIRC. |
|
|
| Colt45 |
| quote: | Originally posted by jrevillug
Here in Europe, despite using 240V, only one of our lines is hot- we don't have the + and - 120v and ground thing going on that you have in the states, IIRC. |
ding ding |
|
|
| whitelabrat |
Once again thanks everyone for their help!
I double checked the polarity going into the power transformer. Oll Korrect there.
I then connected a 20ohm resistor to the input ground and then to the chassis. Most of the buzzing disappeared. Just a very low warm hum remained. |
|
|
| whitelabrat |
This is an interesting discussion on ground loops.
http://www.dself.dsl.pipex.com/ampi...ps/grndloop.htm
I suppose I should hunt for a voltage drop somewhere in the circuit? Should I ground the inputs to the earth via resistor? I'm not sure about the latter, but I think I need to check the voltage in the circuit?
In the circuit, could a bad capacitor cause this sort of problem? |
|
|
| EC8010 |
| quote: | Originally posted by whitelabrat
It's a very sharp buzz rather than a mellow hum. Not that it makes a difference. |
Oh yes it does. A sharp buzz suggests problems with reservoir capacitor current pulses producing voltages where they shouldn't. There should be one loop connecting the reservoir capacitor to the rectifier and the load should be connected across the capacitor - never anywhere inside the loop. |
|
|
| salas |
| quote: | Originally posted by whitelabrat
Once again thanks everyone for their help!
I double checked the polarity going into the power transformer. Oll Korrect there.
I then connected a 20ohm resistor to the input ground and then to the chassis. Most of the buzzing disappeared. Just a very low warm hum remained. |
Way to go! Ground loop Zzzz nailed. Now hunt around that PSU ground related hum. Is there any connection to chassis from negative post of 1st smoothing PSU cap? |
|
|
| whitelabrat |
| Further tests show that connecting a source and attaching the input ground to the chassis cuts the buzz and provides a good clean signal. The multimeter shows 13Vac moving over the ground. I couldn't get a current reading with my meter. Very good! |
|
|
| salas |
| See what is best. Test 0R, 3R3, 6R8, 10R also. Maybe the 20 Ohm you used already proves best, maybe there is better. |
|
|
|
