Cascading Transformers, a bad idea?

[mac]

I have a HV tube preamp in mind and some 220vac to 12vac dicroic lamps transformers.
I want near 300vdc for this project.

If I tie together both secondaries I have a 220vac-->12acv-->220vac transformer, right?
That is near 300vdc, and the 12acv can be used for the heaters.

The problem is hum, even though I have 100 ohm resistors to gnd as in many tube amp designs. Leaving heaters floating is even worse.

When I use and external source for the heaters, not floating, hum is reduced a lot to useful levels. So it is not the preamp wiring what is causing hum.
(And now than I'm typing I'm thinking that I did not try to reverse the way I tied the secondaries)

The question is: two transformers --> a lot of magnetic field?

mac

[R.G.]

The problem is hum, even though I have 100 ohm resistors to gnd as in many tube amp designs. Leaving heaters floating is even worse.

When I use and external source for the heaters, not floating, hum is reduced a lot to useful levels. So it is not the preamp wiring what is causing hum.
(And now than I'm typing I'm thinking that I did not try to reverse the way I tied the secondaries)

The question is: two transformers --> a lot of magnetic field?

Leaving tube heaters floating will always give you hum.

Two transformers do approximately double the magnetic field leakage. However, that is less than a perceived doubling of the hum loudness.

I suggest a test. Magnetic field falls off with the square of distance from the source. Make up a test rig with some wire to move the transformers a couple of meters away from the working circuits. If this reduces hum a lot, then yes, a big portion of your hum is from M-field. If it does not, then it was not the magnetic field, but something else.

[merlinb]

Magnetic field falls off with the square of distance from the source.

Isn't magnetic field strength inversely proportional to the distance (not the square), according to Ampere's law?

[R.G.]

Isn't magnetic field strength inversely proportional to the distance (not the square), according to Ampere's law?

http://www.solarviews.com/eng/edu/invsquar.htm
http://www.emfs.info/Sources+of+EMFs/how+fields+fall+with+distance.htm
Seems opinions vary. A static field varies as 1/r, but alternating is 1/r² or 1/r³. Depends on the geometry and source.

In actual fact, since the core of a transformer is not a point-source radiator or a solenoid, but instead has M-field concentrations at core nonlinearities, it will fall off as a complex function of the geometry until you're far enough away for it to be considered a point source.

In any case, a couple of meters should tell the tale.

[mac]

RG,
It was not only the distance.

I forgot to say that this transformers were a few mm from each other, glued in a kind of silicon base.
I broke the glue, separated and rotated from each other, and hum decreased even more.

In actual fact, since the core of a transformer is not a point-source radiator or a solenoid, but instead has M-field concentrations at core nonlinearities, it will fall off as a complex function of the geometry until you're far enough away for it to be considered a point source.

This is also applied in Quantum Mechanics when you deal with scattered waves. At the detector, far from the interfering source,  the scattered wave tends to a free wave, and the scattering effect is only seen at the detector as a phase shift. f(x) --> f(x+phase)

mac

[Rob Strand]

The problem is hum, even though I have 100 ohm resistors to gnd as in many tube amp designs. Leaving heaters floating is even worse.

I broke the glue, separated and rotated from each other, and hum decreased even more.

These facts should be a hint.

The first suggest the heaters are not balanced relative to ground.   You could also try adjusting the values of one of the 100R resistors up or down.  Some of the bassman amps had a hum balance pot.   A friend of mine had humm problems one of those and I adjusted the pot and the hum was gone.

The second suggests there is also some sort of field coupling.   I'd try reversing the windings and moving the transformers relative to each other.

Good transformers had copper shorting loops to stop stray fields - possible lighting transformers have higher leakage fields.

Series transformers: 220-> 12V -> 12V ->  usually less than 220V maybe down 20%.

Incidentally field depends on geometry (AC and DC doesn't matter).
- field radially out from a wire is 1/r (hence Ampere)
- axial field of a solenoid is 1/r^2
- axial field from a loop is 1/r^3
(- IIRC off axis field from a loop is also 1/r^3 but lower, probably cos or sin of angle)

[mac]

The first suggest the heaters are not balanced relative to ground.   You could also try adjusting the values of one of the 100R resistors up or down.  Some of the bassman amps had a hum balance pot.   A friend of mine had humm problems one of those and I adjusted the pot and the hum was gone.

Thanks for the tip.

*************
BTW, I'm using 12vac for the heaters. But I could rectify, then down to 12vdc o 6.3vdc using a LM317.
DC would make any difference?

mac

[R.G.]

BTW, I'm using 12vac for the heaters. But I could rectify, then down to 12vdc o 6.3vdc using a LM317.
DC would make any difference?

Hum comes from many places. If the imbalance to ground is part of where the hum is coming from, DC will help. If it's the stray fields, it won't.

[Minion]

My latest amp build is a 2 channel and I am useing a really small chassis so my power transformers (3 sperate transformers in the same chassis) are really close to the preamp circuit (within 3 inches) and I don"t have any humm problem , I have a piece of aluminum sheet between the curcuit and the transformers and PSU"s and I have my AC heaters elevated to 25v ......
The amp sounds really good and uses 4 12ax7 and uses relays and footswitch for the chanel switching and indicator lights to show what channel is selected , it has an FX loop and a 50w LM3886 power amp , all using home etched PCB"s ......

Cheers

[Rob Strand]

If you are using half wave rectification on the high voltage side you could also try a full-wave.  Maybe the asymmetrical loading on the transformer is modulating the heater voltage in an asymmetrical way.

I think you definitely have two sources of humm - your initial experiments show that.

[mac]

I found the problems.
One was the relative position of the T's.
And both 12vac wires were a bit long and were forming a closed path, or loop. I put the 12vac outputs face-to-face, twisted wires and problem solved.

RG, Rob, thanks.

mac

[Rob Strand]

Hey, cool.  Maybe there's bigger lesson to be learned here - twist the heater wires all the way back to the transformer!

[mac]

Hey, cool.  Maybe there's bigger lesson to be learned here - twist the heater wires all the way back to the transformer!

For me the big lesson is:
when you can't find electronics parts your imagination is pushed to the limit which is cool, but improvising has side effects. On the other hand you learn a lot! 

mac