In a recent long and wandering thread, (phase converter questions 8/24) it
was written:

"A pair of good posts. It really is refreshing to see
some solid input on 3 phase phase conversions which is both
soundly based and backed up by direct measurement. I hope it
will dispel some of the myths on minimum sizes of idlers and
the practical usefulness of precision "tuning"."

This, of course, refers mainly to the excellent report from Jerry Martes and
tests made with his dynamometer. Perhaps, I know a bit more than others
about Jerry and his dynamometer. I have knowledge of some of the design
particulars of Jerry's dyno and I know that it is a first class machine both
with regards to the way it loads and the way it measures HP. Therefore, I
would be the first to say Jerry's test were soundly based and backed up by
direct measurement. Jerry's tests proved that well designed 3-phase motors
have a built in power reserve. The amount of "reserve" varies among
manufacturers, and probably the quality of their products; it is known as
"service factor" and would account for a 3-phase motor running on
single-phase being able to deliver full rated 3-phase power out for a
limited time. Jerry explained it very well.

To those which may be "new" to rotary phase converters, I would caution not
to be mislead by erroneous conclusions that may be drawn from the RCM
thread, above. Jerry reported the load vs. current characteristics of
3-phase motors quite accurately and showed they can be driven beyond full
rated output for short periods, whether running on single phase or 3-phase.
Note that Jerry did not recommend this practice - he only reported on it.

Having a bit of experience with rotary phase converters, frankly I cannot
say where the "1.5 X" minimum size of idler came from. Perhaps Fitch threw
it out several years ago when researching the subject, I just don't know. I
do know that to successfully start a 3-phase load from an idler motor,
whether "balanced" or not the idler must have a certain minimum size in
order to take the load. How large, I'm not sure and have never experimented
in this area. Perhaps, for starting duty only, the 1.5 figure is a bit high
in some cases. I do know, though, that 1.5 is a good "platform" to work
from toward voltage balancing a rotary and load system. The case for
balancing can be made from a starting current standpoint. As various
respondents have reported, no ordinary residential circuit breaker could
stand the surge of starting a 15 or 20 HP non-balanced motor on single phase
current.

Precision "tuning", like magic, lies in the mind of the beholder. How much
precision lies somewhere between none and much too much. Myths? I think
not. No, a proper RPC system is merely an example of a serious metalworking
hobbyist adhering to good engineering procedure.

Bob Swinney

In article , Robert Swinney says...

Having a bit of experience with rotary phase converters, frankly I cannot
say where the "1.5 X" minimum size of idler came from. Perhaps Fitch threw
it out several years ago when researching the subject, I just don't know.

I may be guilty here - if not of originating, then at least
propogating the factor. Seems like anyone who ever tried to
build a converter with a motor that's 1X never gets it to work,
over the years of observing. They need to go bigger - the 1.5
factor works well if you do the 'capacitor thing.'

I probably first saw the number from Fitch Williams. Likewise
the old saw that the 3600 rpm motors don't make good converters.

Jim


--
==================================================
please reply to:
JRR(zero) at pkmfgvm4 (dot) vnet (dot) ibm (dot) com
==================================================

"jim rozen" wrote in message
...
In article , Robert Swinney says...

Having a bit of experience with rotary phase converters, frankly I cannot
say where the "1.5 X" minimum size of idler came from. Perhaps Fitch
threw
it out several years ago when researching the subject, I just don't know.

I may be guilty here - if not of originating, then at least
propogating the factor. Seems like anyone who ever tried to
build a converter with a motor that's 1X never gets it to work,
over the years of observing. They need to go bigger - the 1.5
factor works well if you do the 'capacitor thing.'

I probably first saw the number from Fitch Williams. Likewise
the old saw that the 3600 rpm motors don't make good converters.

Jim


Jim

I suspect there is alot of good guidance to be obtained from most "old
wives tales". Anything written by Fitch Williams or Don Foreman is
valuable.
I havent been able to determine why the 3600 RPM idler would be
undesireable. I sure dont have data 'either way', 1800 or 3600 being
better for idlers.
The 1 1/2 time tool motor seems like a good starting place for identifying
an idler size. I was actually surprized to learn that idlers as small as
1/10th the tool motor's HP could spin up the 3 phase tool motor. The "size
factor" is very dependent on the amount of load on the tool while
'spinning-up'. I have noticed that a Big idler and Tuning (ala Fitch)
really helps get a heavily loaded 3 phase tool spin-up. The Big idler and
the *proper* tuning* both really help smooth out the 3 phase tool motor's
pulsating when the tool is heavily loaded.

What really surprizes me is that the 3 phase tool motor's performance
doesnt seem to care if it has an idler or not whenever the tool is loaded to
less than about 1/2 its name plate max rating. I may have missed something
in my testing and thinking. But, I do have confidance in the findings
since Don Foreman has been kind and patient enough to guide me thru all the
aspects of this project I couldnt understand without his help. I suspect he
would have picked up on any serious errors I made and corrected my approach
and faulty conclusions.

It sure pleases me to read that Pentigrid and Bob Swinney appreciate the
efforts Don and I have made to get some data on how RPCs perform for RCM
type considerations.

Jerry

Yeah, Jim. It's funny how things get lifted from context only to be shoved
back down as gospel. Thanx for the confirmation.

Bob Swinney
"jim rozen" wrote in message
...
In article , Robert Swinney says...

Having a bit of experience with rotary phase converters, frankly I cannot
say where the "1.5 X" minimum size of idler came from. Perhaps Fitch
threw
it out several years ago when researching the subject, I just don't know.

I may be guilty here - if not of originating, then at least
propogating the factor. Seems like anyone who ever tried to
build a converter with a motor that's 1X never gets it to work,
over the years of observing. They need to go bigger - the 1.5
factor works well if you do the 'capacitor thing.'

I probably first saw the number from Fitch Williams. Likewise
the old saw that the 3600 rpm motors don't make good converters.

Jim


--
==================================================
please reply to:
JRR(zero) at pkmfgvm4 (dot) vnet (dot) ibm (dot) com
==================================================

I probably first saw the number from Fitch Williams. Likewise
the old saw that the 3600 rpm motors don't make good converters.

I looked at some of Fitch's old posts. He contents that any good
motor will make a good RPC. The original theory was that a 3600
RPM motor will have more mass and hold more energy for that peak.
Fitch debunks that theory with his own theory. His theory is that
the idler motor is a rotary transformer not a flywheel. If the motor
is big enough and has enough copper and iron to make the transformation
the motor RPM doesn't matter.

I have been following this thread and experimenting a bit on my own.
I now believe that a RPC isn't doing much until the load motor has
a significant load. Both motors are pretty much running on single
phase. At some point, the load motor starts to slow down(ie slip)
and the RPC motor is still running at full speed and starts suppling
some power to the load motor on the third leg.
Maybe my theory is all wet, but I have tried connecting up three
different motors without balancing caps and I see almost no current
in the third leg when the motors are unloaded.

chuck

On 30 Aug 2004 08:26:51 -0700, jim rozen
wrote:

I probably first saw the number from Fitch Williams. Likewise
the old saw that the 3600 rpm motors don't make good converters.

Perhaps a 3600 RPM motor is less desirable because it's harder to
start, having to build much more angular momentum to reach
operating speed.

On Mon, 30 Aug 2004 13:48:01 -0500, Don Foreman
wrote:

On 30 Aug 2004 08:26:51 -0700, jim rozen
wrote:

I probably first saw the number from Fitch Williams. Likewise
the old saw that the 3600 rpm motors don't make good converters.

Perhaps a 3600 RPM motor is less desirable because it's harder to
start, having to build much more angular momentum to reach
operating speed.

I've been looking and can't find it but I read the math which shows
that all things being equal the 1800 rpm motor is better. I was
surprised when I read it a few years ago and I'm no motor guy so I
could be totally wrong. All that being said, I've got three store
bought RPCs. Two spin at 1725 and the other at 3450. The fast one came
without enough capacitance to balance the load correctly when the load
was close to the maximum rated load. I added capacitors using info
gained from several posters to this group. The other two are really
well balanced machines and the voltage on all legs on the largest is
always within 6% when the load is between 1 and 33 hp. I have not
measured the current to see how well it is balanced at all possible
loads. The manuals for my machines all only talk about the voltage in
the supply.
ERS

On 30 Aug 2004 08:26:51 -0700, jim rozen
wrote:

I probably first saw the number from Fitch Williams. Likewise
the old saw that the 3600 rpm motors don't make good converters.

Perhaps a 3600 RPM motor is less desirable because it's harder to
start, having to build much more angular momentum to reach
operating speed.

I've been looking and can't find it but I read the math which shows
that all things being equal the 1800 rpm motor is better. I was
surprised when I read it a few years ago and I'm no motor guy so I
could be totally wrong. All that being said, I've got three store
bought RPCs. Two spin at 1725 and the other at 3450. The fast one came
without enough capacitance to balance the load correctly when the load
was close to the maximum rated load. I added capacitors using info
gained from several posters to this group. The other two are really
well balanced machines and the voltage on all legs on the largest is
always within 6% when the load is between 1 and 33 hp. I have not
measured the current to see how well it is balanced at all possible
loads. The manuals for my machines all only talk about the voltage in
the supply.
ERS







FWIW years ago (before they made RPC's too) Phase-A-Matic had instructions on
how you could make a rotary phase converter starting with one of their static
units and buying a used 3 phase motor. Quoting them: " The idler motor should
be at least 50% larger than the largest motor that you want to run to
accomodate the higher starting current. A good quality 3600 R.P.M. Y wound 3
phase 220-V motor is the best choice. 1800 R.P.M. motors can be used on
applications not heavily loaded."
David Lindquist

With some trepidation, I have to say that I don't think it's "just a
rotary transformer". Energy must be stored and released somewhere to
create a third phase, because a third phase provides power when there
is none available from the mains (during zero crossings).

One way to check this (and perhaps prove me wrong) would be to make
measurements of phase currents (with no "tuning" capacitors) with
various loads with and without additional mass (flywheel) on the rotor
of the idler motor.

It would also be interesting to observe rotor speed on the idler as
load is increased on the driven motor. If it decreases, that would
suggest that some exchange between mechanical power and electric
power is taking place in the idler, with the process of generating
the third phase creating countertorque that slows the idler rotor.

Fitch is *very* busy designing his new house at the moment, but he'll
be done with that eventually and have time again for thinking about
other things.

On 30 Aug 2004 18:34:58 GMT, (Charles A.
Sherwood) wrote:

I probably first saw the number from Fitch Williams. Likewise
the old saw that the 3600 rpm motors don't make good converters.

I looked at some of Fitch's old posts. He contents that any good
motor will make a good RPC. The original theory was that a 3600
RPM motor will have more mass and hold more energy for that peak.
Fitch debunks that theory with his own theory. His theory is that
the idler motor is a rotary transformer not a flywheel. If the motor
is big enough and has enough copper and iron to make the transformation
the motor RPM doesn't matter.

I have been following this thread and experimenting a bit on my own.
I now believe that a RPC isn't doing much until the load motor has
a significant load. Both motors are pretty much running on single
phase. At some point, the load motor starts to slow down(ie slip)
and the RPC motor is still running at full speed and starts suppling
some power to the load motor on the third leg.
Maybe my theory is all wet, but I have tried connecting up three
different motors without balancing caps and I see almost no current
in the third leg when the motors are unloaded.

chuck

In article , Don Foreman says...

With some trepidation, I have to say that I don't think it's "just a
rotary transformer". Energy must be stored and released somewhere to
create a third phase,

Sure it does - it takes energy to provide the rotating B field
that spins with the rotor's squirrel cage. The energy is in
the kinetic energy of the rotor. That seems to be pretty consistent
with the 'rotary transformer' approach to converter operation.

Jim


--
==================================================
please reply to:
JRR(zero) at pkmfgvm4 (dot) vnet (dot) ibm (dot) com
==================================================

On 30 Aug 2004 18:34:58 GMT, (Charles A. Sherwood) wrote:
I probably first saw the number from Fitch Williams. Likewise
the old saw that the 3600 rpm motors don't make good converters.

I looked at some of Fitch's old posts. He contents that any good
motor will make a good RPC. The original theory was that a 3600
RPM motor will have more mass and hold more energy for that peak.
Fitch debunks that theory with his own theory. His theory is that
the idler motor is a rotary transformer not a flywheel. If the motor
is big enough and has enough copper and iron to make the transformation
the motor RPM doesn't matter.

Agreed. But there's also the idea that a 1800 RPM motor will have lower
windage losses than a same size 3600 RPM motor. Windage loss, and
bearing friction loss, are the two major non-electric loss mechanisms for
the idler. The electric loss mechanisms are winding resistance and flux
leakage (same as for a transformer).

Fitch demonstrated that a flywheel doesn't help. In fact you want the
lowest rotating mass that permits low winding resistance and flux
leakage. That's because you *want* the idler to respond as rapidly as
possible, ie draw more current from the 1 ph line, when the load varies.
It is that current which creates the rotor flux which allows the idler to
work as a RPC.

I have been following this thread and experimenting a bit on my own.
I now believe that a RPC isn't doing much until the load motor has
a significant load. Both motors are pretty much running on single
phase. At some point, the load motor starts to slow down(ie slip)
and the RPC motor is still running at full speed and starts suppling
some power to the load motor on the third leg.
Maybe my theory is all wet, but I have tried connecting up three
different motors without balancing caps and I see almost no current
in the third leg when the motors are unloaded.

Your theory is not all wet. Assume the idler and load motor are identical.
When both motors are running unloaded, with the same slip, they have
the same back EMF on their wild legs, so there is no voltage difference
to cause current to flow in the wild leg. When the load motor is loaded,
its slip increases, its back EMF falls, and a potential difference is created
between the idler wild leg voltage and the load motor wild leg voltage
which then forces current to flow between the idler and the load motor.

Obviously, wild leg current reaches a maximum as load on the load
motor reaches maximum. In a well tuned system (ie a system with
good voltage balance), this current should be the same as the currents
in the other two legs (within a few percent anyway) at full rated load.

Note this is at the *load* motor's rated load. If you were running a 1/2 hp
load motor off of a 20 hp idler, max wild leg current would occur when
the load motor is mechanically loaded to 1/2 hp, the idler hp rating is
irrelevant as long as it is significantly larger than the load motor's hp.

Of course during load motor starting, slip starts out infinite, ie the
back EMF of the load motor is zero, so maximum current flows
between the idler and the load motor. This is exactly what you want
for starting a 3 ph motor.

Again, in a reasonably well tuned system, the currents on all 3 legs of
the load motor should be equal during starting. But if you don't require
maximum starting torque, the idler's voltage imbalance can be rather
large and the motor will still start, though not as rapidly. That's why folks
can get away with using an untuned converter.

In summary, during starting, the load motor back EMF is small, so
significant current flows on the wild leg, properly starting the motor.
While running at speed with no load, wild leg current will be small,
for identical idler and load motors it will be zero. As load increases,
wild leg current increases, until at the limit when the load motor is
running at full rated load, it is drawing full rated current on the
wild leg, same as it is on the other two legs.

Gary

Don sez: "Perhaps a 3600 RPM motor is less desirable because it's harder
to start, having to build much more angular momentum to reach operating
speed."

That would have to be true. In my limited experience, low-power
installations such as we encounter in home-shop applications, 3600 RPM idler
motors seem to work as well as the more-common 1800 RPM types. I have seen
some 7-1/2 HP, 3600 RPM idlers that appear "normal". They are sorta like
ugly babies - if they are otherwise able to kick and scream, their mothers
don't notice the difference.

As an aside to all the great points set forth in recent threads, I want to
acknowledge a key point that Don Foreman set me straight on. I sent Don a
copy of a small, unpublished "paper" I had written on RPC's. In there, were
comments on the obvious (obvious to me anyway) point that an idler and its
load motor are in parallel. BAM! Don hit me right between the eyes with
"dead squirrel" reasoning which proved they were *not* in parallel. Don's
reasoning was eloquent in its simplicity and enhanced my understanding of
RPC current flow by a quantum leap!

Previous to Don's proofing of my little paper, I could not reconcile 3rd leg
current measurements against the notion that the 3rd legs were in parallel.
Duh! Of course, they can't be in parallel. Thanks Don.

Current flow in a RPC is quite complex and does not easily yield itself to
calculation. The stated goal of "balancing" the 3 legs of a RPC is to
achieve reasonably similar *voltages* across all 3 legs, or phases.
Interestingly, from measurements of loaded and unloaded voltage balanced RPC
systems, it can be seen that the *current* through the 3rd leg is always
maximum; or of an amount sufficient to create the desired 3rd leg voltage.
That is to say if the 3rd leg voltage is made to equal the voltages in the
other 2 legs, then the 3rd leg current has to be manipulated in such a way
as to make that happen.

Third leg *current* flows in a tortuous path indeed. It might be stated
that in a voltage balanced RPC, 3rd leg current is made to flow in a manner
that emulates a parallel connection of idler and load. Measurements taken
at various nodes show current tends to flow in opposite directions through
the 3rd leg - but, of course, the resulting aggregate current is what
establishes 3rd leg voltage. In other words, idler and load are made to
*behave* as if they are in parallel in a voltage balanced RPC. Thanks again
to Don Foreman for showing that the idler and load in a RPC are not in
parallel.

Bob Swinney


"Don Foreman" wrote in message
...
On 30 Aug 2004 08:26:51 -0700, jim rozen
wrote:

I probably first saw the number from Fitch Williams. Likewise
the old saw that the 3600 rpm motors don't make good converters.

In article , Robert Swinney says...

load motor are in parallel. BAM! Don hit me right between the eyes with
"dead squirrel" reasoning which proved they were *not* in parallel.

Um, maybe in some non-intuitive theoretical treatment - but from
a topological sense the motor windings *are* in a parallel circuit,
at least for delta wired motors. Granted the incoming line is
across two of the paralleled windings of course.

Where's the dead squirrel?

Jim


--
==================================================
please reply to:
JRR(zero) at pkmfgvm4 (dot) vnet (dot) ibm (dot) com
==================================================

In article , Gary Coffman says...

In summary, during starting, the load motor back EMF is small, so
significant current flows on the wild leg, properly starting the motor.
While running at speed with no load, wild leg current will be small,

Actually, for the driven motor running at speed with no load, *all*
the currents will be nearly zero. Granted an amp-clamp will show
some larger value, but once again that is reactive current.

Measuring the in-phase currents on all three legs of the driven
motor, with no mechanical load, will probably yeild nearly
equal values. SWAG.

Jim


--
==================================================
please reply to:
JRR(zero) at pkmfgvm4 (dot) vnet (dot) ibm (dot) com
==================================================

I like your explanation of induced EMF in the load's wild leg
varying with load motor slip speed, Gary. That clearly and concisely
shows how the load motor draws additional third-phase power under
load.

I'm sure not inclined to dispute Fitch's observations from data. .
I will, however, try to understand what that data tells us.

He may have found that *additional* flywheel makes no observable
difference. I'd bet that he did his experiment with an idler at
least 1.5 times as large (HP) as the load motor. I rather doubt that
he built an induction motor with a rotor of negligable mass.

The mere fact that an induction motor draws many times rated load
current to reach speed in a period long compared to 1 cycle suggests
that the kinetic energy in the rotor is considerably greater than the
integral of rated line power over one cycle or fraction of a cycle.
Variation in rotor speed (kinetic energy) over 1/60th of a cycle would
therefore already be difficult to observe. The effect of adding more
mass would then be even more difficult to observe. One would need a
high resolution (fraction of RPM) speed sensor with a very high
sample rate to observe the ebb and flow of kinetic energy over each
cycle.

Some clarification of the term "rotary transformer" might be helpful.
Simply stating that it is a rotary transformer doesn't clear anything
up for me! What exactly is a rotary transformer? If the rotating
mass is irrelevant, then how is rotation relevant?

If the wild leg is generating power during periods when little or no
power is available from the mains each cycle, where does that power
come from if not from kinetic energy stored in the idler rotor? If
that *is* where it comes from, then I contend that idler motor rotor
mass is indeed relevant, though I could sure see how adding more
mass may not make any noticable difference.





On Wed, 01 Sep 2004 01:41:14 -0400, Gary Coffman
wrote:

On 30 Aug 2004 18:34:58 GMT, (Charles A. Sherwood) wrote:
I probably first saw the number from Fitch Williams. Likewise
the old saw that the 3600 rpm motors don't make good converters.

I looked at some of Fitch's old posts. He contents that any good
motor will make a good RPC. The original theory was that a 3600
RPM motor will have more mass and hold more energy for that peak.
Fitch debunks that theory with his own theory. His theory is that
the idler motor is a rotary transformer not a flywheel. If the motor
is big enough and has enough copper and iron to make the transformation
the motor RPM doesn't matter.

Agreed. But there's also the idea that a 1800 RPM motor will have lower
windage losses than a same size 3600 RPM motor. Windage loss, and
bearing friction loss, are the two major non-electric loss mechanisms for
the idler. The electric loss mechanisms are winding resistance and flux
leakage (same as for a transformer).

Fitch demonstrated that a flywheel doesn't help. In fact you want the
lowest rotating mass that permits low winding resistance and flux
leakage. That's because you *want* the idler to respond as rapidly as
possible, ie draw more current from the 1 ph line, when the load varies.
It is that current which creates the rotor flux which allows the idler to
work as a RPC.

I have been following this thread and experimenting a bit on my own.
I now believe that a RPC isn't doing much until the load motor has
a significant load. Both motors are pretty much running on single
phase. At some point, the load motor starts to slow down(ie slip)
and the RPC motor is still running at full speed and starts suppling
some power to the load motor on the third leg.
Maybe my theory is all wet, but I have tried connecting up three
different motors without balancing caps and I see almost no current
in the third leg when the motors are unloaded.

Your theory is not all wet. Assume the idler and load motor are identical.
When both motors are running unloaded, with the same slip, they have
the same back EMF on their wild legs, so there is no voltage difference
to cause current to flow in the wild leg. When the load motor is loaded,
its slip increases, its back EMF falls, and a potential difference is created
between the idler wild leg voltage and the load motor wild leg voltage
which then forces current to flow between the idler and the load motor.

Obviously, wild leg current reaches a maximum as load on the load
motor reaches maximum. In a well tuned system (ie a system with
good voltage balance), this current should be the same as the currents
in the other two legs (within a few percent anyway) at full rated load.

Note this is at the *load* motor's rated load. If you were running a 1/2 hp
load motor off of a 20 hp idler, max wild leg current would occur when
the load motor is mechanically loaded to 1/2 hp, the idler hp rating is
irrelevant as long as it is significantly larger than the load motor's hp.

Of course during load motor starting, slip starts out infinite, ie the
back EMF of the load motor is zero, so maximum current flows
between the idler and the load motor. This is exactly what you want
for starting a 3 ph motor.

Again, in a reasonably well tuned system, the currents on all 3 legs of
the load motor should be equal during starting. But if you don't require
maximum starting torque, the idler's voltage imbalance can be rather
large and the motor will still start, though not as rapidly. That's why folks
can get away with using an untuned converter.

In summary, during starting, the load motor back EMF is small, so
significant current flows on the wild leg, properly starting the motor.
While running at speed with no load, wild leg current will be small,
for identical idler and load motors it will be zero. As load increases,
wild leg current increases, until at the limit when the load motor is
running at full rated load, it is drawing full rated current on the
wild leg, same as it is on the other two legs.

Gary

The B field does not come from rotation. It takes current to produce
a B field. Since there are no permanent magnets present, rotation
alone cannot produce a B field.

Given that there is a B field, it takes energy to keep it rotating if
it induces voltage (hence current) in a winding somewhere that
therefore produces a counter field and counter torque.

Calling it a "rotary transformer" doesn't clear anything up. What is a
rotary transformer and how does it work?

Transformers work by inducing voltage from a varying B field.
Induction motors (and generators) work with a rotating B field of
constant magnitude, the rotation causing the field linking a
stationary coil to vary with time. One method of analyzing
single-phase motors and unbalanced polyphase motors is by the method
of symmetrical components, which is a collection of constant B fields
rotating in one direction or the other but all at the same speed,
whose resultant sums to the actual field.

A complete electrical understanding requires a description, by
whatever method, of how the field (or contrived fields as in
symmetrical component analysis) vary with time at the location of
each winding and in the rotor. Once this is known, the exchange of
electrical power and mechanical power (instantaneous torque * speed)
can be described by the interactions of the fields with currents.

This has been done. Fitch found it and sent it to me some years ago.
It is a mathematical nightmare!





On 31 Aug 2004 21:55:20 -0700, jim rozen
wrote:

In article , Don Foreman says...


Sure it does - it takes energy to provide the rotating B field
that spins with the rotor's squirrel cage. The energy is in
the kinetic energy of the rotor. That seems to be pretty consistent
with the 'rotary transformer' approach to converter operation.

Jim

In article , Don Foreman says...


The B field does not come from rotation. It takes current to produce
a B field. Since there are no permanent magnets present, rotation
alone cannot produce a B field.

My exact comment was "rotating B field."

Given that there is a B field, it takes energy to keep it rotating if
it induces voltage (hence current) in a winding somewhere that
therefore produces a counter field and counter torque.

Exactly.

Calling it a "rotary transformer" doesn't clear anything up. What is a
rotary transformer and how does it work?

A rotary transformer could be thought of as a permanent magnet
that rotates inside the three windings on the stator, which
are wound to produce the correctly phased voltages when the
*rotating* B field sweeps over them.

You might say that this is nothing more than a permanent magnet
three phase generator and you would be right.

Now the PM rotor has to be replaced with a 'squirrel cage'
that can be found in any common induction motor. The currents
that flow in the buss bars of this device make it behave
like the PM version mentioned above.

So one simple explaination is that the primary is the circuit
inside the squirrel cage in the rotor, which generates a B
field. The B field is in motion with respect to the stator
windings, so even though its *magnititude* is constant, there
is a voltage impressed on the secondary (stator windings) because
of the relative motion.

This story is somewhat incomplete because something has to
be exciting the rotor's current, or it would rapidly decay
away and the rotor would come to a standstill. I like to think
that the power is coupled from the one excited winding in the
stator, to the rotating field that the rotor generates, to
the other two windings in the stator.

So granted it's a bit of a convolution, but I like to think
of the primary as the one driven winding, and the secondary as
the combination of all three windings. The *rotating* B
field of the armature's conductive buss bars is the item that
couples them together.

How do you model one of the old Alexanderson Alternators? There
you have a set of driven coils coupled to another set of
coils that see only the B field of the first coils. The field
is chopped with a variable reluctance (iron vanes on a rotor
inerposed between the two) so that the secondary current is
modulated at the chopping frequency. That's another kind of
'rotary' transformer, albeit a much simpler one to comprehend.
Even though the excitation is dc, the field is still varying and
it works as a transformer.

Transformers don't have to have alternating current flowing in
the primary winding to create the time-varying B field seen
by the secondary. In most cases of course you *do* need to
have the secondary see a time-varying field for the voltage
to be induced (faraday's law) but the mathmatical difference
between, say, a field generated by alternating current in a
stationary coil, versus a field generated by direct current in
a winding that is in motion relative to the secondary,
is nearly inconsequential when analyzing the transformer action.

All that matters is that the field seen by the stator windings
changes over time. It does not matter *how* the change is
effected. The term 'transformer' is typically reserved for
a device where all the windings are mechanically fixed relative
to each other. So 'rotary transformer' is a fairly novel and
unconventional term I admit. It does a pretty good job of
describing the energy flow in a rotary converter though - inductive
coupling of two different circuits via a time-varying B field.

Transformers work by inducing voltage from a varying B field.
Induction motors (and generators) work with a rotating B field of
constant magnitude, the rotation causing the field linking a
stationary coil to vary with time. One method of analyzing
single-phase motors and unbalanced polyphase motors is by the method
of symmetrical components, which is a collection of constant B fields
rotating in one direction or the other but all at the same speed,
whose resultant sums to the actual field.

A complete electrical understanding requires a description, by
whatever method, of how the field (or contrived fields as in
symmetrical component analysis) vary with time at the location of
each winding and in the rotor. Once this is known, the exchange of
electrical power and mechanical power (instantaneous torque * speed)
can be described by the interactions of the fields with currents.

This has been done. Fitch found it and sent it to me some years ago.
It is a mathematical nightmare!

I am sure that if Fitch took the time to do a complete analysis, it
is a) correct and b) extremely intricate. Even though there is a
great deal of hand-waving in the 'rotary transformer' explaination,
it still provides a good seat-of-the pants understanding for those
who either cannot understand the math of a full treatment, or who
don't want to take the time to do so. As they say, it will be left
as an exercise to the reader to figure out which camp I reside in!

:^)

Jim


--
==================================================
please reply to:
JRR(zero) at pkmfgvm4 (dot) vnet (dot) ibm (dot) com
==================================================

"Don Foreman" wrote in message
...
I like your explanation of induced EMF in the load's wild leg
varying with load motor slip speed, Gary. That clearly and concisely
shows how the load motor draws additional third-phase power under
load.

I'm sure not inclined to dispute Fitch's observations from data. .
I will, however, try to understand what that data tells us.

He may have found that *additional* flywheel makes no observable
difference. I'd bet that he did his experiment with an idler at
least 1.5 times as large (HP) as the load motor. I rather doubt that
he built an induction motor with a rotor of negligable mass.

The mere fact that an induction motor draws many times rated load
current to reach speed in a period long compared to 1 cycle suggests
that the kinetic energy in the rotor is considerably greater than the
integral of rated line power over one cycle or fraction of a cycle.
Variation in rotor speed (kinetic energy) over 1/60th of a cycle would
therefore already be difficult to observe. The effect of adding more
mass would then be even more difficult to observe. One would need a
high resolution (fraction of RPM) speed sensor with a very high
sample rate to observe the ebb and flow of kinetic energy over each
cycle.

Some clarification of the term "rotary transformer" might be helpful.
Simply stating that it is a rotary transformer doesn't clear anything
up for me! What exactly is a rotary transformer? If the rotating
mass is irrelevant, then how is rotation relevant?

If the wild leg is generating power during periods when little or no
power is available from the mains each cycle, where does that power
come from if not from kinetic energy stored in the idler rotor? If
that *is* where it comes from, then I contend that idler motor rotor
mass is indeed relevant, though I could sure see how adding more
mass may not make any noticable difference.


"..the induction motor may be considered as a transformer whose secondary
winding is permitted to rotate. Consequently, much of the analysis applied
to the transformer may be modified and utilized in the analysis of the
induction motor."

From my old "Energy Conversion" text book. Yes, the math becomes
involved-much more than I can recollect after so many years!

Jim, more concerned with a dead squirrel, sez:

"Um, maybe in some non-intuitive theoretical treatment - but from a
topological sense the motor windings *are* in a parallel circuit, at least
for delta wired motors. Granted the incoming line is across two of the
paralleled windings of course."

Nope! Draw out a capacitor-assisted rotary phase converter...not just 2
ordinary three-phase motors in parallel. Granted, two phases are across the
line, and as such are in parallel. For the other "phase" assume a direction
of current flow from the 1st motor's "Y"center point through the end of the
3rd leg. Then look at that same point on the other motor and 'viola', the
difference will strike you -- dead squirrel between-the-eyes fashion!

Jim, you are right that two 3-phase motors are in parallel such as in your
own non-balanced RPC. But when you connect them as a capacitor-assisted
RPC, the convoluted 3rd leg current is forced to emulate an ordinary
parallel connection. This is more obvious if you consider that each set of
windings in a 3-phase motor acts as both consumer and generator. The
generator aspect is augmented by capacitor action to creat an aggregate
current that does, in fact, flow in the same direction as if the motors were
truly in parallel.

Bob Swinney



"jim rozen" wrote in message
...
In article , Robert Swinney says...

load motor are in parallel. BAM! Don hit me right between the eyes with
"dead squirrel" reasoning which proved they were *not* in parallel.


Where's the dead squirrel?

Jim


--
==================================================
please reply to:
JRR(zero) at pkmfgvm4 (dot) vnet (dot) ibm (dot) com
==================================================

On Tue, 31 Aug 2004 22:58:21 -0500, Don Foreman
wrote:



Fitch is *very* busy designing his new house at the moment, but he'll
be done with that eventually and have time again for thinking about
other things.

Glad to hear that and all the best wishes to him, since I took the
jump to retired life, I still haven't been able to determine how I
EVER had time to work for a living!
Gerry :-)}
London, Canada

On Wed, 01 Sep 2004 11:19:05 -0500, Don Foreman wrote:
I like your explanation of induced EMF in the load's wild leg
varying with load motor slip speed, Gary. That clearly and concisely
shows how the load motor draws additional third-phase power under
load.

Thanks. If you look a bit closer, you can see that the back EMF
determines the load current drawn by the windings of any motor,
not just the wild leg of a 3 ph motor running off a converter.

I'm sure not inclined to dispute Fitch's observations from data. .
I will, however, try to understand what that data tells us.

He may have found that *additional* flywheel makes no observable
difference. I'd bet that he did his experiment with an idler at
least 1.5 times as large (HP) as the load motor. I rather doubt that
he built an induction motor with a rotor of negligable mass.

No, he didn't, but he did put a *heavy* flywheel on it, and noted no
improvement. In fact, the converter can't respond as quickly to changes
in load with a heavy flywheel attached. (And we want it to, since the
majority of the energy in, and all of it that is passing through, the
system is electrical, not mechanical.)

The mere fact that an induction motor draws many times rated load
current to reach speed in a period long compared to 1 cycle suggests
that the kinetic energy in the rotor is considerably greater than the
integral of rated line power over one cycle or fraction of a cycle.

True, but the lighter the rotor, the less power needed to change its
speed (and hence its slip). Since that slip is mechanically coupled for
all bars of the squirrel cage, more power can be immediately drawn
from the primary feed if the rotor mass is low. Unless we can draw
this power, we can't pass it along to the load.

In electrical terms, a lower mass rotor lowers the source impedance
of the RPC, and a lower impedance source can provide a stiffer output,
ie less sag under changing load.

Variation in rotor speed (kinetic energy) over 1/60th of a cycle would
therefore already be difficult to observe. The effect of adding more
mass would then be even more difficult to observe. One would need a
high resolution (fraction of RPM) speed sensor with a very high
sample rate to observe the ebb and flow of kinetic energy over each
cycle.

Or simply watch the voltage and current on a dual channel scope,
then, if you have a good scope, integrate the two to get instantaneous
real and reactive power.

Some clarification of the term "rotary transformer" might be helpful.
Simply stating that it is a rotary transformer doesn't clear anything
up for me! What exactly is a rotary transformer? If the rotating
mass is irrelevant, then how is rotation relevant?

Well, you have 3 sets of stator coils, and one rotating squirrel cage
made up of a number of bars. The latter are all energized in parallel,
so the electrical phase is the same on all of them at any instant. But
their positions aren't the same, and the whole thing is turning. This
creates a mechanically rotating B field which sweeps past the fixed
position stator coils inducing a back EMF in them (all of them, including
the ones hooked to the wild leg, ie the "transformer" secondary).

*Some* of the stator coils (primary) are also energized by line current
from utility power. This is electrically time varying, producing its own
electrically rotating B vector. This vector is what induces currents in
the squirrel cage in the first place to allow it to produce its own rotating
B field.

The rotor is just an intermediate between primary and secondary which
has the interesting time varying property of mechanically displaced
(phase shifted) synchronization with the primary field.

If the wild leg is generating power during periods when little or no
power is available from the mains each cycle, where does that power
come from if not from kinetic energy stored in the idler rotor? If
that *is* where it comes from, then I contend that idler motor rotor
mass is indeed relevant, though I could sure see how adding more
mass may not make any noticable difference.

It is important to realize that a RPC (or more conventional transformer)
has mostly *reactive* currents circulating in it. These are inductively
reactive, so the current lags voltage. (If the coils were lossless, the
lag would be exactly 90 degrees.) These reactances consume no power,
but they do store considerable energy in their magnetic fields. It is this
energy which is transferred from input to output of any transformer,
whether rotary or not, as the fields rise and collapse. Negligible
mechanical energy is exchanged.

What's different about a rotary transformer is that the coils are in
different spatial relationships at different points in time. This means
that the geometry of the stators and rotor are such that inducing and
induced currents are of differing phase (time). So even though the
primary is going through a voltage zero, the secondary is seeing a
rising voltage induced by the collapsing rotor field, and vice versa.

In other words, the B field induced into the rotor doesn't instantly
die when the primary goes through a zero. That field is still collapsing
since it is inductively lagging, as it mechanically approaches a
secondary coil. The collapsing field induces a current in the secondary
which lags the current in the primary by the *mechanical* phase (time)
difference between stator locations.

A non-rotating transformer can't do this phase shift, but a rotary
transformer can, and does. That's how it is able to produce 3 ph
from 1 ph. But all the energy being transferred is *electromagnetic*,
the mechanical rotation is just there to provide the phase shift.

The ratio of real power to reactive (imaginary) power in the system
varies with load. But the system stored energy is constant (at least
until you overload it enough to drive it into saturation). In the steady
state, ie after sync speed is achieved in the idler, energy out equals
energy in less system losses. Same as for any transformer.

Gary

On 1 Sep 2004 08:44:35 -0700, jim rozen wrote:
In article , Robert Swinney says...

load motor are in parallel. BAM! Don hit me right between the eyes with
"dead squirrel" reasoning which proved they were *not* in parallel.

Um, maybe in some non-intuitive theoretical treatment - but from
a topological sense the motor windings *are* in a parallel circuit,
at least for delta wired motors. Granted the incoming line is
across two of the paralleled windings of course.

Where's the dead squirrel?

Topologically, the two windings are in parallel as viewed from the
outside. But from the *point of view of the current*, they are in series.
In other words, current has to flow through one then the other to
close the loop. That's a series circuit.

(Note that you always use the POV of the current when determining
whether a circuit is series or parallel. Nothing else makes any electrical
sense.)

The wild leg winding in the RPC can be pictured as a battery, and the
winding in the load motor can be pictured as a light bulb. The bulb won't
burn unless you have a closed loop for the current, and you do. It is the
*other* leg connecting the *other* ends of the windings.

Of course in reality the "battery" is producing a time varying voltage,
and the "bulb" actually represents the mechanical output of the load
motor. But none of that changes the fact that the two coils are in
series from the POV of the current.

Gary

On Wed, 1 Sep 2004 13:14:42 -0500, "Robert Swinney" wrote:
Jim, more concerned with a dead squirrel, sez:

"Um, maybe in some non-intuitive theoretical treatment - but from a
topological sense the motor windings *are* in a parallel circuit, at least
for delta wired motors. Granted the incoming line is across two of the
paralleled windings of course."

Nope! Draw out a capacitor-assisted rotary phase converter...not just 2
ordinary three-phase motors in parallel. Granted, two phases are across the
line, and as such are in parallel. For the other "phase" assume a direction
of current flow from the 1st motor's "Y"center point through the end of the
3rd leg. Then look at that same point on the other motor and 'viola', the
difference will strike you -- dead squirrel between-the-eyes fashion!

Jim, you are right that two 3-phase motors are in parallel such as in your
own non-balanced RPC. But when you connect them as a capacitor-assisted
RPC, the convoluted 3rd leg current is forced to emulate an ordinary
parallel connection. This is more obvious if you consider that each set of
windings in a 3-phase motor acts as both consumer and generator. The
generator aspect is augmented by capacitor action to creat an aggregate
current that does, in fact, flow in the same direction as if the motors were
truly in parallel.

Capacitors aren't required for this to be true. The RPC is mechanically
unloaded, the load motor isn't. So the back EMF on the wild leg winding
of the RPC will be greater than the back EMF of the load, thus you have
a potential difference, and current will be forced to flow from the higher
potential to the lower potential. No tricks with capacitors are required
for this to happen.

Capacitors serve a different purpose entirely in a tuned RPC. They're
there simply to equalize the phase to phase voltages at the RPC. They
do that by forming LC phase shifters which offset some of the inductive
lag with capacitive lead so that you get 120+120+120=360 instead of
167+135+58=360, which you might see in a really badly tuned converter.

Gary

The capacitors serve as a source of reactive power.

In article , Gary Coffman says...

Capacitors aren't required for this to be true. The RPC is mechanically
unloaded, the load motor isn't. So the back EMF on the wild leg winding
of the RPC will be greater than the back EMF of the load, thus you have
a potential difference, and current will be forced to flow from the higher
potential to the lower potential. No tricks with capacitors are required
for this to happen.

OK but from a circuitry standpoint the two windings on the two motors
(delta configured both) are in parallel. Yes there is current that
flows in the circuit of course.

Jim


--
==================================================
please reply to:
JRR(zero) at pkmfgvm4 (dot) vnet (dot) ibm (dot) com
==================================================

In article , Gary Coffman says...

Topologically, the two windings are in parallel as viewed from the
outside. But from the *point of view of the current*, they are in series.

Yep. Consider a "parallel" tuned rf tank circuit. A capacitor
and coil connected to each other. They call it a parallel
circuit but of course the internally circulating currents are what
give it the interesting properties.

We have a nomenclature impass here. :^)

Jim


--
==================================================
please reply to:
JRR(zero) at pkmfgvm4 (dot) vnet (dot) ibm (dot) com
==================================================

In article , Gary Coffman says...

*Some* of the stator coils (primary) are also energized by line current
from utility power. This is electrically time varying, producing its own
electrically rotating B vector. This vector is what induces currents in
the squirrel cage in the first place to allow it to produce its own rotating
B field.

Obviously the term 'transformer' is most often used to describe
stationary devices with no mechanical moving parts. But thinking
of a rotary converter as the primary, excited winding coupled
to the three output windings via the rotating B field of the
armature makes it look a great deal like a transformer. The
time-varying B field in this case is coming from a mechanically
moving object rather than a stationary winding with a time
varying current in it.

The fly in the ointment is that the primary winding and the
secondary windings share a common element. This makes it even
tougher to wrap ones mind around.

Jim


--
==================================================
please reply to:
JRR(zero) at pkmfgvm4 (dot) vnet (dot) ibm (dot) com
==================================================

In article ,
jim rozen wrote:
:In article , Gary Coffman says...
:
:Capacitors aren't required for this to be true. The RPC is mechanically
:unloaded, the load motor isn't. So the back EMF on the wild leg winding
:of the RPC will be greater than the back EMF of the load, thus you have
:a potential difference, and current will be forced to flow from the higher
:potential to the lower potential. No tricks with capacitors are required
:for this to happen.
:
:OK but from a circuitry standpoint the two windings on the two motors
delta configured both) are in parallel. Yes there is current that
:flows in the circuit of course.

Topologically in parallel, yes. Consider that when you connect a light
bulb across a battery, they too are topologically in parallel, but that
isn't a particularly useful way to look at the circuit.

--
Bob Nichols AT comcast.net I am "rnichols42"

On Tue, 31 Aug 2004 22:58:21 -0500, Don Foreman
wrote:

With some trepidation, I have to say that I don't think it's "just a
rotary transformer". Energy must be stored and released somewhere to
create a third phase, because a third phase provides power when there
is none available from the mains (during zero crossings).

One way to check this (and perhaps prove me wrong) would be to make
measurements of phase currents (with no "tuning" capacitors) with
various loads with and without additional mass (flywheel) on the rotor
of the idler motor.

It would also be interesting to observe rotor speed on the idler as
load is increased on the driven motor. If it decreases, that would
suggest that some exchange between mechanical power and electric
power is taking place in the idler, with the process of generating
the third phase creating countertorque that slows the idler rotor.

Fitch is *very* busy designing his new house at the moment, but he'll
be done with that eventually and have time again for thinking about
other things.





A slightly different way of looking at this vexed
problem may help.

It doesn't directly clarify the operation of a real
real rotary phase converter with all its losses and second
order effects but it does at least give a reasonably
convincing (to me anyway) insight into the way it works.

Consider a 2 pole 3 phase wound LOSSLESS squirrel
cage motor supplied with single phase power to one winding.
There is no mechanical load so there are no losses of any
kind. Once the rotor is spun up to operating speed, the
stable operating condition is with the rotor spinning at 2
pole synchronous speed.

This necessarily means that the rotor is
diametrically magnetised as a single N-S magnet as a result
of the induced current that is circulating in in the
superconducting rotor bars. Because these are perfect
conductors this current CANNOT decay so that the rotor
behaves exactly as a rotating permanent magnet.

The rotating magnetic field produced by this magnet
induces equal voltages into each of the three phase windings
and it is this rotating field that produces the true
balanced three phase output voltage pattern.

In the case of the energised winding, in this
completely lossles system, the induced voltage (the back
EMF) has risen to be both equal to, and in phase with the
supply voltage so that no current is drawm from the supply.

If a load current is taken from either or both of the
unenergised windings the current in the windings produces a
magnetic field that tries to slow down the rotor. It doesn't
succeed because the motor is operating in a synchronous
mode. What happens is that there is a slight change in the
angular position of the rotor so that the voltage maximum of
the induced back EMF is no longer exactly coincident with
the supply voltage maximum. This shift causes just enough
supply current to flow to cancel the drag forces generated
by the load currents.

In this lossless motor there are no voltage changes
resulting from voltage drops in the winding resistance so
the output remains at three voltages all equal to the supply
voltage and 120 deg apart.

Exactly the same correction mechanism applies if a
mechanical load tries to slow down the rotor. This means a
lossless motor delivers balanced three phase out independent
of variations in both mechanical and electrical loading.

It's interesting to note that this explanation does
not require any direct transformer action between the
windings, all the energy transfer is via the rotor.

A factor,so far ignored, is that the torque generated
by input supply currents fluctuates between zero and a
maximum value twice per cycle of the input waveform. Because
of this there must be enough mechanical inertia in the rotor
to continue the rotation through the low current parts of
the cycle. This is only matters if the inertia is so low
that the input torque fluctuations produce significant
instantaneous rotary velocity changes at twice supply
frequency. In practice the rotor inertia of the commonly
used motors so large that this is not a problem. Any
variation large enough to matter would show itself as a
modulation distortion of the phantom phase voltage(s)
waveform at twice supply frequency.

Jim

In article ,
says...

It's interesting to note that this explanation does
not require any direct transformer action between the
windings, all the energy transfer is via the rotor.

I think that both gary and I would be the first to
agree that without the intermediary of the rotating
B field caused by the squirrel cage in the armature,
the converter does not work.

Hence the name *rotary* transformer. Something has
to be spinning for it to work. Nobody is suggesting that
the stator windings on their own will function as a
transformer to generate three phase power.

Jim


--
==================================================
please reply to:
JRR(zero) at pkmfgvm4 (dot) vnet (dot) ibm (dot) com
==================================================

In article , Robert Nichols says...

Topologically in parallel, yes. Consider that when you connect a light
bulb across a battery, they too are topologically in parallel, but that
isn't a particularly useful way to look at the circuit.

No, they're in series.

The example I gave before was what is classically known as
a parallel tuned tank circuit. The internal current of
the two devices are flowing as though they are in series,
but the combination, as connected to the outside world,
are in parallel.

Your light bulb example doesn't work, because there is
noplace for current to flow besides in the single loop
battery - load.

In the "parallel" tank circuit, as in the rotary converter,
each node in the circuit has more than one other connection.

Jim


--
==================================================
please reply to:
JRR(zero) at pkmfgvm4 (dot) vnet (dot) ibm (dot) com
==================================================

On 1 Sep 2004 11:03:20 -0700, jim rozen
wrote:

(snip)

So granted it's a bit of a convolution, but I like to think
of the primary as the one driven winding, and the secondary as
the combination of all three windings. The *rotating* B
field of the armature's conductive buss bars is the item that
couples them together.

Thanks! I certainly agree that it is "a bit of a convolution", but I
see nothing wrong with it -- and now I know what you mean by "rotary
transformer" !

All that matters is that the field seen by the stator windings
changes over time. It does not matter *how* the change is
effected. The term 'transformer' is typically reserved for
a device where all the windings are mechanically fixed relative
to each other. So 'rotary transformer' is a fairly novel and
unconventional term I admit.

Yes. My problem was making sense of an explanation using that term.

I am sure that if Fitch took the time to do a complete analysis, it
is a) correct and b) extremely intricate. Even though there is a
great deal of hand-waving in the 'rotary transformer' explaination,
it still provides a good seat-of-the pants understanding for those
who either cannot understand the math of a full treatment, or who
don't want to take the time to do so.

Well....maybe your pants, cuz you already had the term in your pocket!
The math is easier for me to understand, but whatever works!

On Thu, 02 Sep 2004 04:04:47 -0400, Gary Coffman
wrote:

On Wed, 01 Sep 2004 11:19:05 -0500, Don Foreman wrote:
I like your explanation of induced EMF in the load's wild leg
varying with load motor slip speed, Gary. That clearly and concisely
shows how the load motor draws additional third-phase power under
load.

Thanks. If you look a bit closer, you can see that the back EMF
determines the load current drawn by the windings of any motor,
not just the wild leg of a 3 ph motor running off a converter.

Yup.

I'm sure not inclined to dispute Fitch's observations from data. .
I will, however, try to understand what that data tells us.

He may have found that *additional* flywheel makes no observable
difference. I'd bet that he did his experiment with an idler at
least 1.5 times as large (HP) as the load motor. I rather doubt that
he built an induction motor with a rotor of negligable mass.

No, he didn't, but he did put a *heavy* flywheel on it, and noted no
improvement. In fact, the converter can't respond as quickly to changes
in load with a heavy flywheel attached. (And we want it to, since the
majority of the energy in, and all of it that is passing through, the
system is electrical, not mechanical.)

OK with the first sentence. Did he observe and report what you
assert in the second sentence? You note that he noted no
improvement. Did he note any degradation?

The mere fact that an induction motor draws many times rated load
current to reach speed in a period long compared to 1 cycle suggests
that the kinetic energy in the rotor is considerably greater than the
integral of rated line power over one cycle or fraction of a cycle.

True, but the lighter the rotor, the less power needed to change its
speed (and hence its slip). Since that slip is mechanically coupled for
all bars of the squirrel cage, more power can be immediately drawn
from the primary feed if the rotor mass is low. Unless we can draw
this power, we can't pass it along to the load.

In electrical terms, a lower mass rotor lowers the source impedance
of the RPC, and a lower impedance source can provide a stiffer output,
ie less sag under changing load.

Variation in rotor speed (kinetic energy) over 1/60th of a cycle would
therefore already be difficult to observe. The effect of adding more
mass would then be even more difficult to observe. One would need a
high resolution (fraction of RPM) speed sensor with a very high
sample rate to observe the ebb and flow of kinetic energy over each
cycle.

Or simply watch the voltage and current on a dual channel scope,
then, if you have a good scope, integrate the two to get instantaneous
real and reactive power.

Some clarification of the term "rotary transformer" might be helpful.
Simply stating that it is a rotary transformer doesn't clear anything
up for me! What exactly is a rotary transformer? If the rotating
mass is irrelevant, then how is rotation relevant?

Well, you have 3 sets of stator coils, and one rotating squirrel cage
made up of a number of bars. The latter are all energized in parallel,
so the electrical phase is the same on all of them at any instant. But
their positions aren't the same, and the whole thing is turning. This
creates a mechanically rotating B field which sweeps past the fixed
position stator coils inducing a back EMF in them (all of them, including
the ones hooked to the wild leg, ie the "transformer" secondary).

*Some* of the stator coils (primary) are also energized by line current
from utility power. This is electrically time varying, producing its own
electrically rotating B vector. This vector is what induces currents in
the squirrel cage in the first place to allow it to produce its own rotating
B field.

The rotor is just an intermediate between primary and secondary which
has the interesting time varying property of mechanically displaced
(phase shifted) synchronization with the primary field.

If the wild leg is generating power during periods when little or no
power is available from the mains each cycle, where does that power
come from if not from kinetic energy stored in the idler rotor? If
that *is* where it comes from, then I contend that idler motor rotor
mass is indeed relevant, though I could sure see how adding more
mass may not make any noticable difference.

It is important to realize that a RPC (or more conventional transformer)
has mostly *reactive* currents circulating in it. These are inductively
reactive, so the current lags voltage. (If the coils were lossless, the
lag would be exactly 90 degrees.) These reactances consume no power,
but they do store considerable energy in their magnetic fields. It is this
energy which is transferred from input to output of any transformer,
whether rotary or not, as the fields rise and collapse. Negligible
mechanical energy is exchanged.

A conventional transformer may have relatively very little reactive
current when operating at full load. In fact, this is a necessity in
high-current high-freq switchmode power xfmrs and great pains are
taken to achieve very tight coupling so as to minimize stored energy.

A fully loaded induction motor has power factor considerably below
0.5, which says that power converted from electrical to mechanical is
more than power being stored in the magnetic fields.

I'm not saying you're wrong about RPC's, mind you. I don't know. It
would be interesting to measure the inductances of a typical 3phase
motor to see if the stored energy is consistent with your assertion.

Your assertion *is* consistent with (but not proven by) the
observation that idlers should be at least 1.5 times the rating of the
load, since induction motors run at very low power factors at light
load.

What's different about a rotary transformer is that the coils are in
different spatial relationships at different points in time. This means
that the geometry of the stators and rotor are such that inducing and
induced currents are of differing phase (time). So even though the
primary is going through a voltage zero, the secondary is seeing a
rising voltage induced by the collapsing rotor field, and vice versa.

In other words, the B field induced into the rotor doesn't instantly
die when the primary goes through a zero. That field is still collapsing
since it is inductively lagging, as it mechanically approaches a
secondary coil. The collapsing field induces a current in the secondary
which lags the current in the primary by the *mechanical* phase (time)
difference between stator locations.

A non-rotating transformer can't do this phase shift, but a rotary
transformer can, and does. That's how it is able to produce 3 ph
from 1 ph. But all the energy being transferred is *electromagnetic*,
the mechanical rotation is just there to provide the phase shift.

The ratio of real power to reactive (imaginary) power in the system
varies with load. But the system stored energy is constant (at least
until you overload it enough to drive it into saturation). In the steady
state, ie after sync speed is achieved in the idler, energy out equals
energy in less system losses. Same as for any transformer.

All true ... and all based on your initial thesis that all (or nearly
all) energy storage is in magnetic fields. None of the above
argument addresses that question but builds on an assumed answer to
it.

Again, I'm not saying your wrong.


Gary

On 2 Sep 2004 08:14:10 -0700, jim rozen
wrote:

We have a nomenclature impass here. :^)

We certainly have a nomenclature disparity! I'll agree with Gary
that the wild legs are in series as regards current flow, and further
argue that they are therefore also in series topologically. I sent
Bob Swinney a 3D CAD model to support my wild-legs-in- series
topology assertion without regard to current flow.

As Clinton would have said, "define parallel....." G

On Thu, 02 Sep 2004 23:50:16 -0500, Don Foreman wrote:
On 2 Sep 2004 08:14:10 -0700, jim rozen
wrote:

We have a nomenclature impass here. :^)

We certainly have a nomenclature disparity! I'll agree with Gary
that the wild legs are in series as regards current flow, and further
argue that they are therefore also in series topologically. I sent
Bob Swinney a 3D CAD model to support my wild-legs-in- series
topology assertion without regard to current flow.

As Clinton would have said, "define parallel....." G

Yeah. Draw a tank circuit like this:


x--------------x
| |
| )
= )
| )
| |
x--------------x

and it looks like a parallel circuit.

But draw it like this:

x---------||-------))))---------x
| |
x-------------------------------x

and it looks like a series circuit.

But electrically they are the same circuit.

Gary

In article , Don Foreman says...

No, he didn't, but he did put a *heavy* flywheel on it, and noted no
improvement. In fact, the converter can't respond as quickly to changes
in load with a heavy flywheel attached. (And we want it to, since the
majority of the energy in, and all of it that is passing through, the
system is electrical, not mechanical.) [gary]

OK with the first sentence. Did he observe and report what you
assert in the second sentence? You note that he noted no
improvement. Did he note any degradation?

I think the experiment of interest would be to somehow
create a rotor with a much lighter than normal mass, and
see how *that* behaves.

It is pretty apparent that the ratio of energies is what's
being discussed the the kinetic energy of the rotor vs
some other energy stored in a magnetic sense. Possible the
moment of the squirrel cage field in the applied field of
the excited stator winding. That's won't be exactly right
but it would give an upper bound on the stored magnetic
energy.

Jim


--
==================================================
please reply to:
JRR(zero) at pkmfgvm4 (dot) vnet (dot) ibm (dot) com
==================================================

I passed circuits 1 in college; a little help would be appreciated.

Given the following assumptions:

50 hours per year run time on home phase converter
220v 1/2 hp 1 phase motor owned w/ extra caps
reasonable wiring skills
$.00013/kwh

If one buys a reasonably priced used 3 phase motor to make a phase
converter for a bridgeport or lathe in the home shop

How much more expensive would it be to run it by using a 220v 1 phase
motor belt driving the 3 phase motor as a generator? This as compared
to using the 3 phase motor as a rotary phase converter.

A tank circuit is usually a parallel circuit because the inductor and
cap are both in shunt from driving point to AC ground. A "series"
resonant circuit is driven on one end and grounded (or connected to a
load) on the other end. See modifications to your drawings below.
The O is a voltage or current source.





On Fri, 03 Sep 2004 03:44:13 -0400, Gary Coffman
wrote:



Yeah. Draw a tank circuit like this:


---- x--------------x
| | |
| | )
O = )
| | )
| | |
|---- x--------------x

and it looks like a parallel circuit.

But draw it like this:

x---------||-------))))---------x
O
x-------------------------------x

and it looks like a series circuit.

But electrically they are the same circuit.

Gary

In article , Don Foreman says...

A tank circuit is usually a parallel circuit because the inductor and
cap are both in shunt from driving point to AC ground.

Right. Topologically they are in "parallel" when viewed from
the outside world. As a combination their properties have
specific, well understood behavior when connected to an outside
circuit.

But those properties exist because of the circulating currents
of the series combination. The currents that circulate internally
are what make the pair of practical interest. Indeed you don't
need to connect them to any external circuit, simply bringing
a grid dip meter (did I just date myself?) near will show what
is going on internal to the pair.

The terms "series" and "parallel" work great when demonstrating
light bulbs and batteries in a grade school text, but the descriptions
get kind of more complicated with things like rotary phase converters.
I think the only way to do this rigorously is the way it's already
been done, writing out the loop and mesh equations and solving
them explicitly.

Jim


--
==================================================
please reply to:
JRR(zero) at pkmfgvm4 (dot) vnet (dot) ibm (dot) com
==================================================

In article ,
jim rozen wrote:
:In article , Robert Nichols says...
:
:Topologically in parallel, yes. Consider that when you connect a light
:bulb across a battery, they too are topologically in parallel, but that
:isn't a particularly useful way to look at the circuit.
:
:No, they're in series.
:
:The example I gave before was what is classically known as
:a parallel tuned tank circuit. The internal current of
:the two devices are flowing as though they are in series,
:but the combination, as connected to the outside world,
:are in parallel.
:
:Your light bulb example doesn't work, because there is
:noplace for current to flow besides in the single loop
:battery - load.

OK, I over simplified. Here's a circuit:

----------
+--------| Device 1 |--------+
| ---------- |
| |
| |
| ---------- |
+--------| Device 2 |--------+
| ---------- |
| |
| |
| ---------- |
+--------| Device 3 |--------+
----------

Without knowing anything about the devices, tell me what's in series and
what's in parallel. If I tell you that device 2 is a battery and the
other two devices are lamps, does this change your answer? How about if
I now substitute a battery charger for device 1? My point (and I'm in
agreement with Jim's tank circuit example here) is that topology alone
can't always tell you whether you have a series connection or a parallel
connection. Until you pick a point of view it may be impossible to make
the distinction.

Even for the fairly trivial case where device 2 is a battery and the
other devices are lamps, there can be times when you need to treat the
battery and one of the lamps as parallel elements, e.g., to answer the
question, "If the battery has a finite internal resistance, what is the
equivalent source resistance seen by the lamp in position 3?"

--
Bob Nichols AT comcast.net I am "rnichols42"