HUM in Tube Amp (I Tried My Best)
Apr 7, 2015 at 2:13 PM Post #16 of 55
So even though all my grounds sort of "feed" in to the main grounding star it is causing ground loops? I thought a ground loop occurred only when the same component had 2 paths to ground, which it doesn't seem like. How should I re-ground it to fix the issue?
 
I 100% would love to build a killer amp, but I have to solve the electrical flaws before I case it up obviously.
 
Apr 7, 2015 at 2:29 PM Post #17 of 55
  So even though all my grounds sort of "feed" in to the main grounding star it is causing ground loops? I thought a ground loop occurred only when the same component had 2 paths to ground, which it doesn't seem like. How should I re-ground it to fix the issue?

 
Without doing a major chassis rebuild, my suggestion is to run a ground wire from each component that needs a ground to a central ground. Quick and easy way to do this is the following:

To further explain what's going on, I found some pictograms. Just ignore all of the amateur radio stuff and imagine each of your components. Shamelessly stolen from http://kc.flexradio.com/knowledgebasearticle50426.aspx
 

 
That is bad. What's happening is the PC is getting a different ground reference than the Transceiver, and the rest of the "chain" of equipment.
 
A single point ground, like this:

 
Will reduce the relative ground reference between each component, which is what causes your hum and "ground float."
 
Hopefully my unrelated examples will help you understand what's going on. What I'd do is put one of those grounding bars (found at your local home depot, lowes, or electrical supply store) and connect earth to that, then run a wire from each component to that bar.
 
 
If you were to build a new chassis, or place everything on a plate of steel or aluminium (I don't suggest Al, because of corrosion and the galvanic effect), you would just tie everything directly to the chassis which is connected to ground. You would then have the shortest possible ground path internally, which would reduce all of the harmonic issues (which you probably won't have with an Audio Spectrum amplifier anyway, but good practices remain.)
 
Hopefully everything made sense, if not I'll do my best to answer your questions.
 
Apr 7, 2015 at 2:39 PM Post #18 of 55
The top example looks like a single bus wire to connect all components, which is what others recommended to me lol. Running a wire from each grounded component to one central point seems odd, since so many people write about star and single bus wire grounding.

I have aluminum plates that I can use to cut the chassis
 
Apr 7, 2015 at 3:06 PM Post #19 of 55
If you were to build a new chassis, or place everything on a plate of steel or aluminium (I don't suggest Al, because of corrosion and the galvanic effect), you would just tie everything directly to the chassis which is connected to ground. You would then have the shortest possible ground path internally, which would reduce all of the harmonic issues (which you probably won't have with an Audio Spectrum amplifier anyway, but good practices remain.)


There is no easier way to build an amp that hums than to tie all the ground points to the case. Do not do this. (It is common to do this in guitar amps because a little hum is more acceptable there. A headphone amp is a different sort of animal.)

Do, however, take the other advice of a) not touching a live circuit, and b) using a star ground of some sort.
 
Apr 7, 2015 at 3:10 PM Post #20 of 55
The top example looks like a single bus wire to connect all components, which is what others recommended to me lol. Running a wire from each grounded component to one central point seems odd, since so many people write about star and single bus wire grounding.


The difference is the distance and the amount of current. A bus wire that is a few inches long, made from thick copper, and used in a circuit drawing a few mA's is not the same as several components spaced several feet apart along a copper wire each drawing several amps.


I have aluminum plates that I can use to cut the chassis


That's what you should be using.
 
Apr 7, 2015 at 3:53 PM Post #21 of 55
There is no easier way to build an amp that hums than to tie all the ground points to the case. Do not do this. (It is common to do this in guitar amps because a little hum is more acceptable there. A headphone amp is a different sort of animal.)


Fair enough. Most of my experience is with tube RF amplifiers. I've yet to build a headphone amp.
 
Apr 7, 2015 at 4:01 PM Post #22 of 55
There is no easier way to build an amp that hums than to tie all the ground points to the case. Do not do this. (It is common to do this in guitar amps because a little hum is more acceptable there. A headphone amp is a different sort of animal.)

Do, however, take the other advice of a) not touching a live circuit, and b) using a star ground of some sort.

You mean by tie all ground points to different parts of the case? Or that and not have a ground return from the case to ground.
 
Apr 12, 2015 at 5:19 PM Post #24 of 55
So, after the advice given, I redid some of my ground wires, Shortened them up, and moved my PSU about 5 feet away from the amp. Without input tube, there is still hum. The voltage drop across my long umbilical is only 1.6 mV, so I don't think the issue is with a ground loop. I think it is weird that with no signal, there is still hum.
 
Apr 12, 2015 at 5:53 PM Post #25 of 55
  So, after the advice given, I redid some of my ground wires, Shortened them up, and moved my PSU about 5 feet away from the amp. Without input tube, there is still hum. The voltage drop across my long umbilical is only 1.6 mV, so I don't think the issue is with a ground loop. I think it is weird that with no signal, there is still hum.

Hrm.
 
 
Is the output transformer grounded on both sides? I'm not sure I'd do that.
 
How much DC do you have on the output?
 
Apr 12, 2015 at 6:01 PM Post #26 of 55
Hrm.


Is the output transformer grounded on both sides? I'm not sure I'd do that.

How much DC do you have on the output?


The common taps meet at the Sleeve of the 1/4" jack. A wire then runs from that to ground.

Isn't there supposed to be 0V DV at any headphone output?
 
Apr 12, 2015 at 6:12 PM Post #27 of 55
Yeah, any speaker (or headphone) output should have 0V DC. Your design looks like it's possible for there to be DC on the output, which is why I asked.
 
Apr 12, 2015 at 6:15 PM Post #28 of 55
How should I measure?

I would think since the transformer only outputs AC, it would be 0DC, but I'll check
 
Apr 12, 2015 at 6:40 PM Post #29 of 55
Yeah, any speaker (or headphone) output should have 0V DC. Your design looks like it's possible for there to be DC on the output, which is why I asked.


Also, with b+ disconnected, there is 0 hum.
 
Apr 12, 2015 at 7:22 PM Post #30 of 55
Another note, it is more of a mix of noises, maybe a buzz? Its not straight hum though, which I think eliminates Magnetic coupling and bad filtering, but not sure.
 

Users who are viewing this thread

Back
Top