I could try moving it away from the tone circuit, but I don't think it'll do much good. That wire is connected directly to ground.

The fact that it has very close to zero voltage on it is not the point. It has current pulses going through it and they can couple to the tone caps like a transformer winding. Moving it probably won't fix anything, but good wire routing can't hurt anything.

 

The fact that it has very close to zero voltage on it is not the point. It has current pulses going through it and they can couple to the tone caps like a transformer winding. Moving it probably won't fix anything, but good wire routing can't hurt anything.

Ok, that makes sense. I'll reroute it over towards the chassis wall when I add the standby switch.

I just finished taking some SPL measurements with all my house lights off and the volume knob at 0, 2, and 5. The readings were about 50, 54, and 61 dB respectively. I also looked at the spectrum and their are strong returns at 120 Hz pointing to rectifier ripple? There's also a softer spike at 60 Hz, but I'm not sure if that's a limitation of the iPhone's microphone, filament bleed through the 12AX7, or I have some intermodulation distortion going on.

Thanks for helping me make sense of all this. The mod is scheduled for tonight!

 

One of my personal tricks is to place 1000 pf across each rectifier diode. In your case, at least 600 volts rating, ceramic or metal film. I'd love to see if that affects the frequency analyzer results or if it was just a tiny bump on the wave that I found with my scope and it really doesn't make any difference.

 

Putting 0.3 VAC on the cathode would likely be to cancel power supply hum. The challenge is not to minimize AC ripple on the power supply outputs, but to get the amount of ripple that the 0.3 VAC is supposed cancel.

The effect is to that similar to when using a humbucking transformer but cheaper.

 

Putting 0.3 VAC on the cathode would likely be to cancel power supply hum. The challenge is not to minimize AC ripple on the power supply outputs, but to get the amount of ripple that the 0.3 VAC is supposed cancel.

Wow! Very interesting. That makes a lot of sense. I'm not sure how well the cancelation is working but thanks for that info.

 

I have my doubts, for this particular case.
If the amplitude of the hum goes up with the volume control, what is turning up the filament bias on the output tube cathode?
Nothing.
The cathode can not track the volume control.
Therefore, I say you have a different problem.
Besides, how you gonna adjust the filament voltage that was applied to V2 cathode?
There's no adjuster.

 

I have my doubts, for this particular case.
If the amplitude of the hum goes up with the volume control, what is turning up the filament bias on the output tube cathode?
Nothing.
The cathode can not track the volume control

Another great point! Thanks for all the input. I obviously came to the right forum.

 

Another great point! Thanks for all the input. I obviously came to the right forum.

You did.

 

Well, adding a choke and tripling the filter capacitance was a complete bust. The noise is still there and as loud as before.

I just finished looking at another review of this kit and another builder complained bitterly about "120 Hz" noise. Moving the treble and bass controls have zero effect on the hum if that helps any. Maybe I need to try another preamp tube (different brand)?

 

I found an explanation of the noise inherent in this kit from a guy who builds guitar amps. I'm still trying to make sense of it, but maybe some of you see a solution here? Thanks.

This MOD 102 design is described as “single ended” because it uses a single output tube. This is a good configuration for a low-power and low-cost design, but a single ended design can be prone to hum compared with a push-pull or differential pair design where common-mode noise such as hum is cancelled.

The source of hum is typically the heater’s AC voltage, and there are generally three ways to reduce it. The first is to use a center tapped heater winding on the transformer, which is not used on the MOD 102, presumably for cost reasons.

The second method is to create an artificial center tap by connecting matched resistors from each lead of the heater winding to ground. This arrangement is commonly seen in many Fender amps, such as a Tweed Deluxe. It adds the cost of two resistors and a higher current draw from the heater winding, so possibly a larger and more expensive transformer. We’re guessing, but presumably this is why it was not used in the MOD 102.

The third method is called “DC elevation” and the MOD 102 uses a variation on it by connecting one side of the heater winding to the cathode of the V2 power tube, which is biased to about 5.5 Volts DC. This floats the 6.3 volt AC heater voltage on all three tube filaments (two in the dual 12AX7 and one in the EL84) on top of the 5.5 volts DC. This approach minimizes the hum induced from the heater to the cathode because the induced current becomes close to a constant value when the voltage is above a threshold. In other words, although there is still leakage current, it is constant and therefore the noise voltage appearing at the cathode becomes DC, which is inaudible. From our testing of the amp, discussed below, the approach is effective as the amp is very quiet.

​

 

I designed and built an amplifier very much like this with a 12AX7 input and a 6V6 single ended output and it is quiet as a mouse. (I stole most of the concepts from the Supro or the Fender Champ.) Maybe that's because I used DC for the filaments?

Rectify the filament winding, add a capacitor and an LM7806 voltage regulator, and send that to the filaments...if the transformer can handle the added power requirement.

and, no, which type of tube can't cause hum. Any individual tube can be bad, but not the type of tube.

I would suggest removing the cathode connection and see if that helps.

 

Thanks. I was thinking aren't tube filaments just as home with DC as they are with AC? After all, they were powered by the "A" battery in the old days, correct?

I'll pull up the details on the power x-former and check it ratings.

 

The virtual center tap might work.
http://www.valvewizard.co.uk/heater.html

 

I looked at the schematic again and see that the V2 cathode is putting +5.5 volts DC on ALL the filaments at ALL places on every filament because there is no other path to ground in the filament string. Suddenly I understand this. The filament transformer allegedly can't be injecting hum in the V2 cathode, but we already ruled that out because the hum responds to the volume control. The peak of 6.3 volts AC is 8.8 volts, and this isn't centered at all, so I expect some of the filament voltage to exceed the 5.5V bias. Thus, DC could make a difference.

Here's a chart of derating factors for rectifying a transformer winding.

 

Thanks everybody. Lots of great ideas and I'm learning a lot about tube amps, which was the primary reason for building this kit in the first place. I believe we're approaching the point where it would be easier to simply build another amp using the tips and knowledge gleaned so far rather than make any more mods to this one - unless they were very simple. My next step in to simply build a cabinet for this amp to make it safer for storage and moving around.

I'll continue some small tinkering with this kit, but now I'm moving towards another project, an overdrive pedal (mainly to use with this amp).

Thanks again.

 

It's been a pleasure working with you.

 

Yes, #12, it is as you said.

 

The virtual center tap is only 2 resistors, you might try it.

 

The virtual center tap is only 2 resistors, you might try it.

That sounds simple enough. So the only issue is making sure the power transformer filament windings can supply the necessary current to the two tubes (3 filaments) AND whatever current flows through the two resistors? Could I then remove the connection from the filaments to the cathode on V2?

 

If you make a virtual center for the filament winding you must disconnect the connection to V2 cathode. The, "usual" value is a pair of 100 ohm resistors.