Tektronix high voltage transformer ferrite flux balancer explained

[oz2cpu]

starting at 23 mins in this video


I have not seen any other ever digging deep into this,

[TomKatt]

The ingenuity involved in creating these things is amazing.  I often wonder how they knew to even approach the problems they found like this.  Such an elegant solution.

I've read that we are losing the knowledge of transformer design because all the knowledge is just in the heads of the guys who developed solutions like this.  And as they retire and leave the workforce, their knowledge goes with them.  And given the magic involved, I can begin to appreciate how that could be true.

Thanks for sharing!

[oz2cpu]

I think it is also funny to think about the voltage pr vinding, see the diode filament run at 1.25V and is using one side branch ferrite one winding to get this voltage,
the center part contain double that flux, so here we got 2.5V pr winding.

[NiHaoMike]

I've read that we are losing the knowledge of transformer design because all the knowledge is just in the heads of the guys who developed solutions like this.  And as they retire and leave the workforce, their knowledge goes with them.  And given the magic involved, I can begin to appreciate how that could be true.

Small high voltage, high frequency transformers (where "high voltage" in this context is >1kV) are indeed a shrinking niche now. CRTs were pretty much phased out decades ago, fluorescent and neon lighting have largely been replaced with LEDs, and even gasoline engines are declining. In warmer climates, gas heating is being increasingly replaced with heat pumps.

At least many of the remaining consumer uses for such transformers, most significantly electrostatic air filters and ionizers, are unlikely to be replaced by low voltage alternatives anytime soon.

[Conrad Hoffman]

Easy low voltage to high voltage- piezoelectric transformers!

[moffy]

Easy low voltage to high voltage- piezoelectric transformers!

Yeah, Jim Williams did an app note with one, but when I searched they were hard to find, wish they were more readily available.

[moffy]

What a beautifully simple way to balance flux, the effective voltages cancel one another if the flux is the same, and any flux imbalance gets compensated by the current flow, subtracting from one side and adding to the other, it wouldn't be perfect but good enough.

[NiHaoMike]

Yeah, Jim Williams did an app note with one, but when I searched they were hard to find, wish they were more readily available.

It doesn't help at all that they came right when CCFL backlights were being replaced with LEDs. That was the exact market where the advantage of small size mattered.

[mawyatt]

What a beautifully simple way to balance flux, the effective voltages cancel one another if the flux is the same, and any flux imbalance gets compensated by the current flow, subtracting from one side and adding to the other, it wouldn't be perfect but good enough.

Thanks to Thomas for posting this!!

Agree, another one of Tektronix's absolutely brilliant concepts

Reminds of the Ross "T" coils and Ft doubler

Best

[oz2cpu]

what if we try this balancer concept and put it on wide band transformers, like audio hifi transformers,
they are used for signal isolation, but also very much in tube amplifiers, signal and power size, so it should be easy to add this winding
and measure how it performs before vs after..
plenty of active tube amplifier tech people out here who love cool ideas to try out ?

[moffy]

what if we try this balancer concept and put it on wide band transformers, like audio hifi transformers,
they are used for signal isolation, but also very much in tube amplifiers, signal and power size, so it should be easy to add this winding
and measure how it performs before vs after..
plenty of active tube amplifier tech people out here who love cool ideas to try out ?

I think it might only be applicable when there are windings on each leg of the transformer, E core, I assume Tektronix did that for voltage isolation purposes. With audio isolation transformers the windings tend to be on top of one another to reduce leakage inductance so the flux will balance naturally.

[oz2cpu]

I am sure tektronix did it so you can have an uneven number of rectifier diodes on the ferrite side parts,
and their peak currents and usage current dont create an uncontrolled voltage at the filaments,
remember the center flux is fully feedback regulated, due to sec winding voltage, NOT the sides individually,
in some of the different scopes, you find 3-5 diodes.
so they desperately needed this balancer winding.

[ferdieCX]

An incredible elegant solution.
It makes me think about, how many times today a "brute force solution" (read brute number of transistors in a chip) is used, instead of basic knowledge in physics, etc

[Jay_Diddy_B]

Hi,

The flux balancer is a 'figure-8' of wire wound around the center of the E core. The result is that equal flux is in the two outside limbs of the core. Since the flux on the outside limbs is 1/2 the flux in the center limb the volts per turn is also half the center limb.

The technique of flux splitting allows for fractional turns in power transformers. The Tektronix implementation is 1/2. Other fractions are possible.

Jay_Diddy_B

[oz2cpu]

oh yes, other fractions are possible, wow that is wild, i had not figured that it,
but yes, if that twisted figure 8 winding, had another ratio, you could make the flux stronger one side,
if you needed this for some type of crasy feature,