Reducing hum suggestions?

[Mike_J]

I am getting back to the AB763 – one channel build. Since I will be using one less 12A_7 tube than the PT heater supply is rated for was thinking about a way to use the mA. Decided to try to get the amp as quiet as possible.

First, going to install a Humdinger pot. Looks like 500R trimmers are available at Mouser. Any preference as to which trimmer works best? Not sure if single-turn or multi-turn would be best?

Second, going to elevate the heater supply to a DC voltage. Can I run this voltage off the bias tap or is it necessary to take it of the B+ tap? Any suggestions would be appreciated.

Third, going to switch the first preamp tube to DC. Will take the mA that would have been used by the unused tube to run DC to this tube. Have numerous schematics showing how to do this. The Valve Wizard's suggestion looks pretty simple. Could put it on a small board. Any suggestions would be appreciated.

Thanks
Mike

[HotBluePlates]

I am getting back to the AB763 – one channel build. ... Decided to try to get the amp as quiet as possible. ...

I've found if you execute an AB763 build correctly, as Fender designed it (or as Hoffman re-arranged it), there is no hum to get rid of.

D.C. heat is probably not needed unless the amp has massive gain after the input stage (like a 4-5 gain stage metal amp).

Elevated heater voltage is fine, but only helps if you happen to be using marginal, leaky tubes.

I don't know how many amps you've built. If you've built a few, you probably have some without all these features which still don't exhibit objectionable hum/noise. If you've played any vintage Fender amps, you probably know most don't hum, some may/may not have background hiss. You can do the belt & suspenders approach, but it's added work/cost and may be a solution in search of a problem...

[sluckey]

Humdinger... You need a 100Ω or 200Ω or 500Ω wire wound pot, not a trimmer pot. Two 100Ω 1/2W 1% resistors would probably be just as effective as a correctly adjusted pot.

Don't mess with the bias circuit to elevate your heaters. Bias circuit is too critical. Too much stuff waiting to burn up if the bias fails. Use a voltage divider at one of the downstream B+ nodes. You need two resistors and a filter cap.

I don't think dc filaments for the preamp tube will be much of a benefit in an AB763 amp.

[Mike_J]

I am getting back to the AB763 – one channel build. ... Decided to try to get the amp as quiet as possible. ...

I've found if you execute an AB763 build correctly, as Fender designed it (or as Hoffman re-arranged it), there is no hum to get rid of.

D.C. heat is probably not needed unless the amp has massive gain after the input stage (like a 4-5 gain stage metal amp).

Elevated heater voltage is fine, but only helps if you happen to be using marginal, leaky tubes.

I don't know how many amps you've built. If you've built a few, you probably have some without all these features which still don't exhibit objectionable hum/noise. If you've played any vintage Fender amps, you probably know most don't hum, some may/may not have background hiss. You can do the belt & suspenders approach, but it's added work/cost and may be a solution in search of a problem...

Most of my amps don't make much noise. However, some do and I would like to experiment with ways to reduce hum and hiss. It has become my arch enemy that I am out to destroy.


As far as hiss goes I know it is effective to use large (3 watt) metal film resistors on the plates and cathodes of the preamp tubes. Reduces a lot of objectionable hiss without a noticeable change in tone to my ears versus carbon comps.


My Fender amp builds don't have significant hum until you throw a lot of pedals at them. Have the available heater current to try these hum reducers. I am sure I have a special knack for finding marginal, leaky tubes. Will probably build the amp without the hum mods but layout the amp with room to install them if deemed necessary. Want to know what is needed so I can try these if it may help.


Am going to go through all of my amps after this build and try to make each one better. Just seems as though these things would be more weapons in my war against amp noise.


Thanks
Mike

[Mike_J]

Humdinger... You need a 100Ω or 200Ω or 500Ω wire wound pot, not a trimmer pot. Two 100Ω 1/2W 1% resistors would probably be just as effective as a correctly adjusted pot.

Don't mess with the bias circuit to elevate your heaters. Bias circuit is too critical. Too much stuff waiting to burn up if the bias fails. Use a voltage divider at one of the downstream B+ nodes. You need two resistors and a filter cap.

I don't think dc filaments for the preamp tube will be much of a benefit in an AB763 amp.

I have another source from the HT line that I will use. Saw a thread somewhere where the two resistors and filter caps were used.


Attached a schematic for the hum mods. As soon as I find the thread for the voltage supply will include that in the schematic.


Thanks
Mike

[mresistor]

Mike, I recently built a Single Channel Deluxe Reverb on a turret board and it is as quiet as a church mouse. I used Magnetic Components iron and mallory caps. Payed attention to lead dress and Hoffmans grounding scheme. It's in a very tight for space chassis. I also used flameproof metal oxide resistors in the power supply. Noise can also come from tubes.

[Mike_J]

Mike, I recently built a Single Channel Deluxe Reverb on a turret board and it is as quiet as a church mouse. I used Magnetic Components iron and mallory caps. Payed attention to lead dress and Hoffmans grounding scheme. It's in a very tight for space chassis. I also used flameproof metal oxide resistors in the power supply. Noise can also come from tubes.

Thank you for the reply mresistor. My plan is to go through each of my amps and make each as good, and quiet as possible. My other AB763 which is a two channel is very quiet. Will probably try using these techniques on each amp that has enough room for the circuitry just to maximize noise protection. Costs less than $15.00 per amp for the components. Given the amount of time spent on each build it is worth it to me to do the little bit extra.


Have attached the schematic for how I see the components.


Thanks
Mike

[Mike_J]

Here is the layout diagram. Please comment if you see anything wrong or that can be done better.


Thanks
Mike

[trobbins]

The PI 22k tail resistor could be changed to 10k+12k, and a high value RC filter off say the 10k (~+35VDC) can provide a simple DC bypassed voltage for heater elevation.  The RC could be something like a 220k and 10uF 50V, where the 220k will isolate the cap from PI interaction.

A pot can be more useful than fixed resistors if you have some capacitance leakage to input valve grid which needs an offset humdinger to null that out.

[Mike_J]

The PI 22k tail resistor could be changed to 10k+12k, and a high value RC filter off say the 10k (~+35VDC) can provide a simple DC bypassed voltage for heater elevation.  The RC could be something like a 220k and 10uF 50V, where the 220k will isolate the cap from PI interaction.

A pot can be more useful than fixed resistors if you have some capacitance leakage to input valve grid which needs an offset humdinger to null that out.

Trobbins, thank you for your response. Unfortunately, what you are describing is over my head. If you could markup the schematic to show what you mean it might make more sense. By saying a pot is more useful than fixed resistors are you referring to the Humdinger pot?


Thanks
Mike

[trobbins]

Yes a humdinger pot allows a particular type of hum to be alleviated more than just using two resistors of the same value, or a heater CT.  Sometimes the heater voltage from one side of the heater couples over to the input stage grid more than the other end of the heater - remember that the heater ends are at opposite phase to each other if you were sitting in the middle (CT or fixed humdinger point).  For example one of the heater wires could be innocently placed closer to the grid pin than the other heater wire.  The capacitance between the nearby heater wire and the grid pin can add some hum - the level depends on the type of cathode circuit used.  Some types of valve also have wiring from one heater terminal placed closer to the grid than the other heater terminal.

I can upload a quick sketch of the elevated voltage circuit in a few hours.  It just may be an easier way to get a DC voltage of about 30-40V, which is the sort of level that would likely be appropriate.

As you are keen to chase down the last uV of hum, you may find something useful in:
http://dalmura.com.au/projects/Hum%20article.pdf

Ciao, Tim

[kagliostro]

Many Thanks for sharing Tim

Franco

[Mike_J]

As you are keen to chase down the last uV of hum, you may find something useful in:
http://dalmura.com.au/projects/Hum%20article.pdf

Ciao, Tim

Tim


Excellent information in your article. Will have to read it a number of times to absorb it. Appreciate your help and advice.


Thanks
Mike