View Full Version : Phase converter idiosyncrasies

04-25-2015, 12:28 PM
Hi All,

Time for another display of ignorance of an electrical nature.

A customer brought me an older officially built 20-30hp Roto Phase phase converter complaining of not starting reliably, tripping breakers etc.

Opened up the large junction/capacitor box, found the usual giant mouse nest complete with chewed insulation, urinary corrosion, and pre cooked carcasses. Same scenario in the rotary portion, with the addition of a side course of fried snake to compliment the mouse carcasses.

Anyhow, remove the old capacitors, clean it up as best as I can and ship it off to the local rewind shop for a dip and bake, as I see some loose windings and weed seeds that have been left by the previous inhabitants.

Get it back and proceed to reassemble the rotary portion when I notice what looks like some deliberate modifications to the shorting bars on the rotor. The bars on each end each have 2 cuts, one end looking like clean, deliberate cuts, the other end looking something like the result of a hammer and punch or chisel. The cuts are spaced about 90 from each other on each end and about 90 from each other end to end.

I had read in the past about a "high resistance rotor" design (one descripiton http://www.gwm4-3phase.com/uands/high1.htm) for the rotating component of a rotary phase converter. Apparently this modification allows the initial starting of the idler motor to be accomplished with just the run capacitors that would normally be connected for balancing the output of the 3 legs of the converter, rather than needing a separate array of start capacitors that need to be controlled/disconnected once the idler motor reaches operating speed and that is the way this particular unit operates. This resembles that description, but I was wondering if anyone else had seen this type of converter before.

But the real question I have is about the current draw of a layout like this. The converter seems to start ok with just the run capacitors hooked up. It was/is a little noisy when running so I played with the amount of capacitance connected as the new Chinese capacitors may have different values/characteristics than the ones that they replaced. In doing so, I noticed the the pitch of the noise coming from the converter seemed to vary at a regular beat. The clamp meter confirms this. The single phase current draw varies from approximately 24 amps to 32 amps at a regular frequency of about 5 seconds. Other than this anomaly, the unit seems to run just fine, no heating or wanting to vibrate off the bench.

If this is in fact a "high resistance rotor" modification, would this account for the cyclical current draw characteristic? Or is there some other plausible rationale.

What say ye great mavens of mechanical marvels??





04-25-2015, 02:21 PM
Cutting the rotor bars drastically lowers the initial current and even the running current.
In the early days before cast in rotor bars, the bars were copper with silver soldered copper end rings, when a motor would overheat, it would often fling the solder off and the result was the motor would draw extremely low current and refuse to turn.
Re-soldering the end rings usually solved the problem.
Cutting the bars is not a great idea IMO.

J Tiers
04-25-2015, 08:18 PM
Cutting the end shorting bars in 4 places is not really a part of a "high resistance rotor".

What probably IS part of that is the fact that the bars appear to have been turned DOWN from the OD. THAT will raise the resistance and confer the benefits (and problems) of the high resistance rotor.

The cutting of the end bars appears to create 4 groups which may be intended to improve performance. I believe that to have been done by some later Bubba, who thought he could create "poles" in that way (it doesn't work that way). The beat between the rotational frequency of these and the 60Hz MAY be causing the cyclical fluctuation of current, since as one of the cuts passes under the stator pole, it changes the distribution of rotor current, and probably not in a good way.

A high resistance rotor has the benefits described, but it also may be less effective as an idler due to the "pole" imposed on it by induced rotor current being dissipated faster as I^2-R loss. The rotor current is more "in phase" when the resistance is higher, I am not sure if that gives a benefit or not as an idler, without considerably more thought.

The better way to make the rotor is likely to combine the two, as is done with my Arco Roto-Phase unit. It has two sets of rotor bars. One set of high resistance bars is brazed copper, and on the surface. The other is cast aluminum, and buried.

The surface high resistance bars give good starting, when the slip frequency is high. The buried bars have too much inductance to respond.

After the motor is up to speed, the slip frequency drops, and the high inductance low resistance bars carry most of the current This produces the rotor field. The high resistance of the surface bars causes them to have little current, and small contribution.

The reason the cuts do not produce poles, is that the rotor is not synchronous, and so the cuts may be between poles, or directly ON poles, at a regular beat frequency rate.