Chasing hum in a new tube amp

[bwarden]

OK, took the plunge, and built my first DIY tube amp, a direct coupled Darling. Some trouble at first (bad soldering), but got it working with a little help (thanks, Bob).

Sounds great, except for the hum. I searched here and elsewhere, but I can't find any reference to hum the way mine presents.

It is variable with the volume pot, but in what I guess is an unusual way. The hum starts at about 5-10% of full volume, peaks at about 30%, falls off to about 45-50%, and is dead silent above that. So if I play it full on, no hum. Seems equal in both channels (maybe slightly louder right than left).

I don't know how to describe the frequency, except to say its coming out of the mid, not the woofer.

Can anyone help a fella out? Thanks in advance...


Bill

[pmillett]

It could be several things...

Most likely may be noise being capacitively coupled into the input wiring between the volume control and the first tube's grid. When the volume control is mid-span, the source impedance driving the input is at it's highest, so the wiring is more susceptable to picking up noise. When it's all the way down, it's almost grounded; all the way up, it's connected to the source, which (probably) has a low source resistance.

You usually see this if the input pot is 100k or higher (not if it's 10k, for example).

The coupling can be inside the tube itself (from the filament). In this case usually elevating the filament to some DC voltage (50V or so?) usually cures it.

I've also seen this behaviour if the input tube is oscillating (which sometimes manifests itself as a hum sound). Same idea... when there's a high impedance to ground on the grid, it is more likely to oscillate. A grid stopper resistor in series right at the pin usually cures this.

Another possible cure is to ground the pot case - sometimes that helps too.

Pete

[Shoog]

I've also seen this behaviour if the input tube is oscillating (which sometimes manifests itself as a hum sound). Same idea... when there's a high impedance to ground on the grid, it is more likely to oscillate. A grid stopper resistor in series right at the pin usually cures this.

I have also found this. If the anode load resistors are not straight onto the valve base pin (ie flying leads) , this can also produce oscillation. A small value resistor on the pin will usually solve this. When I had this issue with a 5687 it manifested as a hum coupled off the heaters.

Shoog

[FastEddy]

I don't see how you guys can diagnose this problem without looking at a schematic or a picture ... pmillett and Shoog suggest adding parts !! ?? !! (Note that the Darling already has some resistors in front of the grids ( http://www.diyparadise.com/Darling.html )

In the absence a picture or diagram of the circuit, I would have guessed the need to fix any and all cold solder joints as bwarden already admits to several = " ... Some trouble at first (bad soldering) ...".

pmillett: " ... Another possible cure is to ground the pot case - sometimes that helps too. ..."

... but I would make sure all the solder joints are good before adding any parts or "grounding" any cases ...

[Shoog]

I have made no definate advise, simply my experience of what has caused a similar issue for me in the past. These are basic tools in the debugging armoury and can very easily be tried and rejected without causing to much hassle.

He said he had checked for cold solders so I had to assume this was not the issue.

Calm down.

Shoog

[pmillett]

Originally posted by FastEddy
In the absence a picture or diagram of the circuit, I would have guessed the need to fix any and all cold solder joints as bwarden already admits to several = " ... Some trouble at first (bad soldering) ...".

Can you propose a scenario where a bad solder joint would cause the problem reported? Keep in mind that the amp is perfectly functional, and sounds good, except for the odd hum only at mid-volume?

I couldn't think of one. That's why I suggested other things to look at.

Pete

[bwarden]

Soldering was one heater connection, and a binding post. Continuity of everything checks point to point, no visible cold solder joints.

Schematic attached.

[bwarden]

And the power supply ...

[Tom Bavis]

I'm with Pete... the symptom points to ONE spot... the wiper of the volume pot. The case of the pot should ALWAYS be grounded - it's a shield - so if it's not mounted to grounded metal, take care of that.

[atmasphere]

I've seen the complaint often enough! Ground the body of the volume control, install shielded wiring from the input connector to the control and from the control to the input of the amplifier itself and *definitely* install the grid stop resistors as previously advised. Make sure that they are as close to the tube socket as possible.