[Amps] FW: Power factor correction on tube amps
Jim Thomson
jim.thom at telus.net
Sat Mar 9 08:36:09 EST 2013
Date: Thu, 07 Mar 2013 18:49:27 -0500
From: KA4INM <ka4inm at gmail.com>
To: amps <amps at contesting.com>
Subject: Re: [Amps] FW: Power factor correction on tube amps
On 03/07/13 18:07, Matt KM5VI wrote:
> Help me out here guys...
> I thought that conventional power supplies (transformer-bridge-filter) were
> relatively clean on reflected harmonics?
No, the transformer is all inductive and the diodes conduct for most
of the sine wave, but not all, at that transition the current changes
abruptly causing distortion.
(The capacitors are polar and not able to compensate for the
inductance in the transformer.)
--
Ron KA4INM
## with a resonant choke set up, like used on the henry amps, PF is typ .9
## On a high C filter, you not only have the ‘normal’ pf, but you also have loads
of odd harmonics up to the 9th. So if you just measure AC line volts X AC line current,
you will get VA power. Trouble is the load is capacitive, by virtue of the high C filter.
## One of the hams ( forget his call) used a HP PF analyzer to look at the odd harmonics, and how
much current was in each harmonic, it was just wicked. When added to the fundamental, they
would be through the roof. The only reason they are not is... some of the harmonics cancel, or
partially cancel each other....or are out of phase with the fundamental.
## I have tried off and on for years now to try and reduce the PF in high C, HV supplies. It’s extremely
tough to pull it off. Even then, it will only partially work for a 100% dead cxr. It won’t work on SSB, since
the load is varying up-down like a yo yo. You either have to cancel out individual odd harmonics...and / or
change the phase of the odd harmonics so they cancel with each other..and /or the fundamental.
It’s the 3rd and 5th that are the biggest offenders. You need equal amplitude + opposite phase to cancel stuff out.
## In the end, I gave up and through in the towel. I went through this with the late Peter Dahl, and I should have taken his
advice, don’t even try it, it’s an exercise in futility with any high C, HV supply.
## Typ values of PF are like .57 to .85 It’s all over the map really. It doesn’t always get worse with bigger caps either.
Depends on xfmr used. The high Z types, originally intended for choke input supplies, with high dc sec resistance don’t
fare too well. The low Z types do a lot better..and the lower the dc resistance, the better, but not always. There are other variables
at play here, including Z from pri of xmfr back to the source, pri to sec C, whether the xfmr will even pass the odd harmonics, or
alter their phase etc.
## If you want unity-perfect PF use switching supplies that have pf correction built into them.
Jim VE7RF
More information about the Amps
mailing list