Tube DIY Asylum Do It Yourself (DIY) paradise for tube and SET project builders.

---

[ Follow Ups ] [ Post Followup ] Thread: [ Display  All  Email ] [ Tube DIY Asylum ]

---

[ Alert Moderator ]

---

RE: Modeling the Bugle 45 PSU, any comments?

Posted by mqracing (M) on June 29, 2009 at 16:46:46

In Reply to: RE: Modeling the Bugle 45 PSU, any comments? posted by drlowmu on June 29, 2009 at 13:03:18:

Hi Jeff:

you wrote;

::::I had a friend in Venice CA measure it for me and I do recall it was astronomically high in DCR.

In any case, its a SICK part to use in any DHT amp, because it suffers from too high a DCR. It instantly guarantees a non dynamic amplifier.

No wonder people objected to my posts on DCR when I said "20 Ohms or less" is needed for the high voltage secondary for "real quality"....255 Ohms is TWELVE TIMES too high from my listening evaluations over the years.:::::

************************************************************************

this goes back to what I have written and illustrated in the past... that if you evaluate a component based on one parameter alone (in isolation) your asking for either trouble, foolishness, or mischief.

It is not the DCR of the copper windings in isolation that is the overarching singular "quality factor" of any transformer--- and in fact designing for low dcr in isolation is simple as pie--- though dangerous in terms of spec'ing a transformer on this basis alone.

For starters... and just starters at that.... without also accounting for the core losses--- copper losses in isolation really doesn't tell us much---- more important (though not singularly even when combined) is what are the TOTAL copper and core losses.

Remember you must energize the iron circuit in a transformer--- so you surely want the iron circuit to be well behaved and easy to energize. Additionally core losses are going to introduce distortions in the waveform... so having a good low loss and low distortion core is perhaps even moreso important than emphasizing simply low copper losses. Copper losses (in and of themselves) are pure and simple "losses" but copper losses do not (in and of themselves) introduce any distortion of the waveform passing through it.

In the case your looking at--- the efficiency of the copper circuit introduces a bit less than 3% voltage drop if run at full rated current.
so that the voltage drop in absolute terms is approx 20 volts at it's full current rating.

What your missing critically is the many other performance properties of the subject power transformer. How hot does it run? What is the voltage regulation from no load to full load? How hard is the core run? What is the core distortion of the iron path? Is the core run so hard that it suffers mechanically induced noise (i.e., that it becomes noisy). How large of an external magnetic field does the power trans throw off?

As I've said before... if we are willing to give up a host of performance factors as indicated above and design SIMPLY for low dcr... then transformer design becomes trivially easy.

the real challenge is in balancing BOTH copper and core losses... while running the core at reasonably low flux density levels and maintaining good waveshape fidelity (i.e., keeping distortion low).... and maintaining good voltage regulation from no load to full load. If you expand your areas of "performance parameters" to include all of or some of these additional factors... then you quickly learn that "ramping up" one parameter at the expense of all the others is not smart nor is it the paradigm of "high performance" design.

The same trans you bemoan and belittle... has been topside of an amplifier that over a period of more than ten years of constant production has itself won numerous "great reviews", "best of shows", "best products", "one of the top ten amplifiers ever built", and on and on.

Why? At least in part because when evaluated from a fuller range of performance parameters--- it is in fact--- a pretty darn good power transformer.

The irony to me is that if I built power transformers solely to a low dcr number... these are the designs that I would not ever want to use on any amp of my own? Why? The quickest way to get a low dcr is to run your design ALL the way up to the knee of the BH curve. This will reduce the number of turns that you need on the primary.... and thus by proportion the number of turns needed on the seccndary... and allow you to use a larger wire size--- if you reduce the number of turns and increase the wire size---yep... your DCR will go down. The price you'll pay for this is a much greater likelihood that the magnetic circuit has been comprimised at the expense of the copper circuit--- so that a healthy balance of losses btwn the copper and magnetic circuits becomes quite elusive.

And interestingly--- even the measurement of the copper losses themselves are done at the most amateur level--- the dcr (almost always) given is the "cold" DCR--- it is not the DCR of the winding while it is "working" and delivering it full rated current. Some of these low (when measured cold) dcr components if measured for DCR at the design operating points will tend to have much greater temperature rises... and hence their dcr will "grow" on you quite a bit. This is especially the case if you minimized the core cross sectional area so that you could keep the MLT of the copper circuit smaller. So that now the core runs so hot--- let's say it has a 65 degree C temp rise instead of let us say a 30 degree C temp rise--- that the copper circuit (which is a good thermal conductor) picks up and tries to dissapate this heat coming from a too small, too hard driven iron circuit.

If your doing a lopsided design for just low DCR then predictably there will be a good number of cases in which the core's losses will, in fact, be the dominant loss factor... and these core losses will probably be much, much greater in absolute magnitude than the copper losses. Sometimes by a few hundred percent. And remember... what we really want is a good balance of copper and core losses... so that our total losses are not too high. One parameter done at the expense of all the others is not "optimal", "progress", "high performance" or anything at all to write home about.

I have previously illustrated similar points re: "low dcr" in previous posts--- wherein I showed several different choke designs... one of them having record setting low dcr's, but, that had much greater overall losses than a competing design wherein the copper and core losses were much moreso in balance with one another and in absolute terms it was the units with the greater DCR's that actually had the lowest overall losses.

In my view... reducing the "goodness" of a complex product--- one that has many different performance parameters that ought to be optimized--- to one singular parameter whose main attribute appears to be the ease with which it can be meausured... is sophmoric beyond belief.

msl

Builder of MagneQuest & Peerless transformers since 1989

This post is made possible by the generous support of people like you and our sponsors:

WEET Music Caps

---

[ Top of Page ] [ Contact Us ] [ Support/Wish List ] [ Copyright © 2025, Audio Asylum, all rights reserved ]

---

[ General ] [ Speakers ] [ Tubes ] [ Vinyl ] [ Digital ] [ Hi-Rez ] [ Video Asylum ] [ Cables ] [ Tweaks/DIY ] [ Music ] [ Films ]

---