Some years ago, I wrote an essay entitled “Using One 115V Fan On Two Line Voltages.” Its essence was that a fan motor rated for one line voltage could be switched to running at twice that line voltage as in 120V/240V service by having a power transformer’s primary windings serve as an autotransformer.

The gist of that essay is in these first two illustrations:

What I omitted was a look at how much of the fan’s power would need to be delivered via the magnetics of transformer T1. A look at the following SPICE model addresses that issue.

My version of SPICE does not include an autotransformer. Therefore, I had to devise a model using two 1:1 turns ratio transformers. The left side, the yellow shaded windings, is where the autotransformer properties are exploited. The coupling between the two right side windings allows this model to function in step-up or step-down service as in the following two illustrations: 

Putting this to work on a one-hundred watt load (an arbitrary wattage choice made here just for purposes of illustration) in step-down fashion, we can examine the power transfer through the magnetics by examining voltage and current in those two right-side coils as follows:

Half of the load power is carried to that load via the magnetics.

I had once worked out this conclusion algebraically, but those notes are long gone. Still, it felt good to see that result confirmed in this way.

John Dunn is an electronics consultant, and a graduate of The Polytechnic Institute of Brooklyn (BSEE) and of New York University (MSEE).

This is an example of complete misuse of SPICE! Some “good old academic knowledge” hopefully learned in power electronics class will suffice to give this answer. This is not a high-speed switching nor RF problem; it is 50 or 60 Hz “perfect” magnetics plus series resistance analysis (it is even simpler than that, one need only treat the transformer as a black box with a max rating & losses at that load).

Simply looking at the winding currents gives the same answer, while sitting at a contractor engineering review, ON-THE-SPOT (Can you wait while I whip out my SPICE PC & run this problem? If you needed to do that, it would never occur to you to do it.)

Been There Done That (contractor claimed there was no benefit in efficiency for this very setup, (versus full isolation transformer with switched primaries). Contractor should have known better, as presenter was engineer in their an in-house transformer shop. SPICE didn’t exist then!

“Buck-boost” transformers are routinely used because of this (way larger than 50%) advantage when the buck or boost factor is low. Power engineers (& BSEE degree’d engineers) should have this as common knowledge.

This analysis is applicable for any setup involving isolated power conversion (such as stacking DC/DC converter’s output upon the input).

Misuse? I never thought anyone would actually be offended by the use of SPICE to illustrate a concept.

Of course, autotransformation is not limited to line frequency applications. Tapped “loopstick” coils wound with Litz wire over adjustable ferrite cores are also autotransformers as applied to RF purposes. The first essay was supposed to have pointed out the general applicability.

The fan service at line frequency actually came about when I designed the Bertan High Voltage HVL-25 electronic load. The transformer supplier to whom I defined that usage was rather taken aback. He’d never seen that done despite many years in the transformer business and even though the idea wasn’t anything original nor was it something new to the industry.

However, from feedback that I’ve gotten via various LinkedIn groups, there were/are other folks out there to whom this idea was unfamiliar. If trying to edify others somehow causes offense to you, I’m afraid there is no remedy.

In the manual for a SPICE-based simulator I found: Know what to expect from simulation before running it! Simulation is no substitute for thinking. So the best way to handle such a subject is to first show the math and then the simulation. I am sure, that John knew exactly what to expect, but omitting the math may lead some beginners in the wrong direction (simulate before thinking!). I am sure, there are other interesting points in this application to point out: I expect the fan supplier to have only an approval of the fan for 110V applications, because that is the rated voltage. But using it this way now a higher voltage occurs in the system against ground potential, so the fan must be rated for this.

I didn’t have any access to SPICE when I was doing the design at Bertan High Voltage of which this was a part. The governing equations were in my lab notebook at the time, but that analysis is long since lost. The company, Bertan High Voltage, was eventually acquired by Del and then by Spellman and so far as I understand, all of my notebooks and other documentation at Bertan were discarded. (Sigh)

What I was getting at is that SPICE makes it easy to demonstrate the underlying concept.

Also, by keeping the fan and its associated transformer winding tied one-end-to-neutral, there is no stress ever applied to the fan that exceeds any of the fan’s ratings or published capabilities

You must Sign in or Register to post a comment.