Question on phase angles

I ran across an interesting claim from a manufacturer of rotary phase converters. He says that you cannot use an ordinary three phase motor
to build a phase converter and get true 120 degree phase timing. You have to use a "specially modified" motor, or the phase angles will be off. I don't quite understand this, because a standard three phase motor has its windings physically located 120 degrees apart. Am I missing something here, or is this just a bunch of sales hype? Here's a link to the article in question:
http://www.deselectric.com/panels
--
JR

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
He's probably referring to the fact that the motor has "slip" so its not turning at a synchronous speed, but the phase relationship should remain correct.

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
| He's probably referring to the fact that the motor has "slip" so its not | turning at a synchronous speed, but the phase relationship should remain | correct.
The high leg voltage coming back out of the motor would not always have the same phase or voltage. Think of it as an unregulated generator. The slip will change under loading conditions, such as when starting another motor, thus drawing current and power from the idler motor.
| |> |> I ran across an interesting claim from a manufacturer of rotary phase |> converters. He says that you cannot use an ordinary three phase motor |> to build a phase converter and get true 120 degree phase timing. You |> have to use a "specially modified" motor, or the phase angles will be |> off. I don't quite understand this, because a standard three phase |> motor has its windings physically located 120 degrees apart. Am I |> missing something here, or is this just a bunch of sales hype? Here's a |> link to the article in question: |> |> http://www.deselectric.com/panels |> |> |> -- |> JR | |
--
-----------------------------------------------------------------------------
| Phil Howard KA9WGN | http://linuxhomepage.com/ http://ham.org/ |
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 4 May 2004 13:50:21 GMT, snipped-for-privacy@ipal.net said:

I've found that the generated leg will show a voltage around 10% lower than the single phase line voltage being used to run the idler motor, as measured to either of the other two legs, but this can be corrected with running capacitors between the generated leg and each of the other two legs. I know that the actual voltages on the idler motor will vary somewhat with the physical load on the driven motor, but as far as the phase angles being 120 degrees apart, wouldn't that remain constant because the windings are still physically located 120 degrees apart? If not, what sort of alterations to the idler motor would correct that?

--
JR

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
| On 4 May 2004 13:50:21 GMT, snipped-for-privacy@ipal.net said: |
|> |> | He's probably referring to the fact that the motor has "slip" so its not |> | turning at a synchronous speed, but the phase relationship should remain |> | correct. |> |> The high leg voltage coming back out of the motor would not always have |> the same phase or voltage. Think of it as an unregulated generator. |> The slip will change under loading conditions, such as when starting |> another motor, thus drawing current and power from the idler motor. | | I've found that the generated leg will show a voltage around 10% lower | than the single phase line voltage being used to run the idler motor, as | measured to either of the other two legs, but this can be corrected with | running capacitors between the generated leg and each of the other two | legs. I know that the actual voltages on the idler motor will vary | somewhat with the physical load on the driven motor, but as far as the | phase angles being 120 degrees apart, wouldn't that remain constant | because the windings are still physically located 120 degrees apart? If | not, what sort of alterations to the idler motor would correct that?
The problem is you are taking 1 line from the idler motor and the other 2 lines from the main supply. The phase angle difference is between these. If you were getting all 3 lines from the idler motor, then you might have phase jitter, but it would all be in a consistent 120 degree configuration. But to do that, you need a motor generator set.
--
-----------------------------------------------------------------------------
| Phil Howard KA9WGN | http://linuxhomepage.com/ http://ham.org/ |
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
in article snipped-for-privacy@news.alt.net, JR at snipped-for-privacy@hotmail.corn wrote on 5/4/04 12:14 AM:

While I consider myself rather savvy on electrical engineering topics, I am not always familiar with gargon such as *CNC*. The site above does not help. It does not define many of the terms it uses. Its English is atrocious and stands in the way of understanding the points it is trying to explain.
Bill
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
<snip>

am
help.
On must remember that when selling any brand of snake oil, hard data is your enemy.
Disclaimer: I have not evaluated referenced product and therefore cannot definatively state that it works any better or worse than any other product on the market.
Charles Perry P.E.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Tue, 04 May 2004 17:08:11 GMT, Repeating Rifle said:

A CNC-rated converter is supposed to provide near-perfect 3-phase power for machinery that has sensitive electronic control circuits for servo motors, etc. I'm looking for a panel that will be suitable for this purpose. The guy from DES Electric claims that you have to use a "special motor" to make a true CNC converter, and that using an ordinary 3-phase motor and balancing the voltages on all three phases is not enough, because the phase angles won't be 120 degrees apart. One of his competitors claims that's not true, because the windings in a three phase motor are already 120 degrees apart, and that's the way they should be. I'm inclined to believe the second guy, because I can't imagine that anyone would make a "special motor" with the windings arranged any differently, and even if they did, why would you position them at any angle other than 120 degrees if you wanted the output to be at 120 degrees? I'm just wondering if there's something I'm missing here, like maybe there's some other modification to the idler motor other than changing the orientation of the windings.
--
JR

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
in article snipped-for-privacy@news.alt.net, JR at snipped-for-privacy@hotmail.corn wrote on 5/4/04 10:25 AM:

If this was supposed to clarify the question, it did not! Let me just ask the question: What is a true CNC converter? Please avoid arty-farty descriptions.
Bill
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Wed, 05 May 2004 00:29:50 GMT, Repeating Rifle said:

I'm not sure what the "arty-farty" part of that was. A "CNC converter" is simply a rotary phase converter that supplies well-balanced three phase power from a single phase source, suitable for running CNC machinery. For purposes of this discussion, let's just say that CNC machinery is any machinery containing sensitive electronic control circuits that might not operate properly if the manufactured three phase power was not well-ballanced (nearly identical voltage readings between any two phases), and the phase angles being 120 degrees apart. It is DES's contention that a normal three phase motor cannot be used to produce clean three phase power with 120 degree phase angles, and that he uses some "special motor" to overcome this. My questions are: (1), is there any reason why a three phase motor wouldn't produce 120 degree phase angles on the output when used to make a rotary phase converter, and (2), if there is, what modifications would be necessary to make it do so?
--
JR

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
| I'm not sure what the "arty-farty" part of that was. A "CNC converter" | is simply a rotary phase converter that supplies well-balanced three | phase power from a single phase source, suitable for running CNC | machinery. For purposes of this discussion, let's just say that CNC | machinery is any machinery containing sensitive electronic control | circuits that might not operate properly if the manufactured three phase | power was not well-ballanced (nearly identical voltage readings between | any two phases), and the phase angles being 120 degrees apart. It is | DES's contention that a normal three phase motor cannot be used to | produce clean three phase power with 120 degree phase angles, and that | he uses some "special motor" to overcome this. My questions are: (1), | is there any reason why a three phase motor wouldn't produce 120 degree | phase angles on the output when used to make a rotary phase converter, | and (2), if there is, what modifications would be necessary to make it | do so?
I think the problem here is that whoever is promoting the "CNC converter" is dwelling too much on the existance of the problem, and not enough on just how his particular product is able to effectively solve the problem better than something else (at least in the same price range).
So the question is "what (the hell is) a CNC converter?" ... and this is asked NOT in terms of what it is FOR, but rather, how it accomplishes its goal. So answering "it is a phase converter for CNC machinery" is not the answer being sought. What is being sought is how this converter is different from other converters that are considered unsuitable for CNC machinery.
Sales people tend to give that former (wrong) answer. If you ask a sales person "how does it work", they tend to think you asked "how WELL does it work" and answer "It works very well".
Engineers don't give tiddly about sales drool. They want to know what makes it tick, and whether it actually solves the problem at hand, and fits into other issues like reliability, maintainability, etc.
Smart managers (yes, there are some around) listen to their engineers.
Dumb managers (all the rest) listen to the sales drool.
Now to your questions:
| is there any reason why a three phase motor wouldn't produce 120 degree | phase angles on the output when used to make a rotary phase converter,
A three phase motor BY ITSELF, which is designed with 120 degree angled armatures, will produce consistent angle differences of 120 degrees between the phases. But the "BY ITSELF" is the issue here. If you are running a syncronous single phase motor driving a three phase generator then you'll get some nice three phase output if you don't load it so hard as to drag the mover motor out of sync. But consider what happens if you run this genset with an inductive squirrel cage motor. You will get 120 degrees between the 3 lines from the generator, but you won't get a consistent frequency. It will be lower that your mains frequency and will change with load.
Now, if you are taking just ONE line of power from that generator, and comparing that to 2 lines from the mains (opposite poles of single phase) you'll see either frequency variation or phase variation depending on the design and load. Frequency variation makes things rather useless. Phase variation within a certain window can be used. Just what that window is depends on the application.
| and (2), if there is, what modifications would be necessary to make it | do so?
It's a matter of degree (I love that pun). How perfect do you want? Even the utility won't be perfect because of factors (ooh, another pun) like the loads from other customers. But it is usually close enough for everything but the most picky equipment.
An idler motor will give you a consistent frequency, but the phase will vary depending on how much the load causes the idler rotor to lag. The specifics I can't tell you, but I've read of cases where it lagged as much as 30 degrees under normal loads. Your mileage may vary.
There are probably a number of things to do to adjust that phase or reduce the variation in it. A larger idler motor would help. Or maybe a large inertia on that motor.
A system that senses the phase error and adjusts things to correct it would be the way to go. There are probably a number of things that can be adjusted, from applied voltage on the idler, to adding varying amount of the single phase power to compensate. Mixing voltages from single phse doesn't get you anywhere because it's all one dimensional. But once you have something different than 180 degrees, you can mix varying quantities of voltage to get any phase angle you want (within practical limits).
What does the "CNC converter" actually do? That I don't know. And that is the whole of the technical questions. No engineer can advise you on whether it is suitable for your needs without first knowing what that means. And unfortunately, when marketing names a problem based on what problem (market) it is intended for, rather than on what kind of design it has, then it leaves other engineers in the dark.
--
-----------------------------------------------------------------------------
| Phil Howard KA9WGN | http://linuxhomepage.com/ http://ham.org/ |
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:>

remove the urine to answer

The use of a 3 phase induction motor with two terminals connected to a single phase supply and all 3 legs used to provide 3 phase to another load ( appreciably smaller than the rating of the 3 phase machine ) will and does work quite well. The phase relations will be correct although there may be some voltage magnitude unbalance where capacitors can help. It seems to require delta or ungrounded Y. Frequency stability is not a problem .The slip doesn't matter as, at any slip the rotor field will be at synchronous speed with respect to the stator and this "synchronous speed" is determined by the primary supply.
--
Don Kelly
snipped-for-privacy@peeshaw.ca
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Thu, 06 May 2004 02:58:09 GMT, Don Kelly said:

I understand the principle behind single-phasing a three phase motor, and then tapping power from all three windings to get a three phase output, and I understand how the balance between phases on the output can be corrected with running capacitors. I'm assuming that this balance cannot be kept constant under varying loads, but I'm not understanding how that throws the 120 degree timing off, or how DES proposes to correct it with the following statement:
"The motor being used for the converter must have either the windings converted for the phase angle differences,,, or the rotor must be modified for this to happen."
Is this just a bunch of sales hype, or does somebody understand what he's talking about when he mentions "converting" the windings for the phase angle differences, or "modifying" the rotor?
--
JR

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
JR wrote:

[snip]
I vote for sales hype. Since the phase lag and voltage drop on the generated leg are a function of load, I don't see how they can "convert" the windings to correct for these effects. Unless they have found a way to re-wind a motor while its running.
As far as "modifying" the rotor, its not likely. Typical rotors produces a field that is fixed with respect to the rotor (synchronous) or has a constant magnitude and rotates around the rotor at the slip frequency (induction). As the field cuts each stator winding, it generates the same induced voltage in each (assuming equal turns and geometry). The only way to produce compensation in individual phases from the field would be to modulate the field flux magnitude and phase position. Its been done, but its not easy.
My guess is that their rotary converters are simple three-phase induction motor types in combination with either switched capacitor or some similar kind of dynamic compensation to correct the generated phase.
P.S. I wonder how much I'd trust their sales info. since they can't even spell "OSCILLASCOPES" correctly.
--
Paul Hovnanian mailto: snipped-for-privacy@Hovnanian.com
note to spammers: a Washington State resident
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:
| I vote for sales hype. Since the phase lag and voltage drop on the | generated leg are a function of load, I don't see how they can "convert" | the windings to correct for these effects. Unless they have found a way | to re-wind a motor while its running.
[snip]
| My guess is that their rotary converters are simple three-phase | induction motor types in combination with either switched capacitor or | some similar kind of dynamic compensation to correct the generated | phase.
One way I could see doing this, if the speed of correction is not an issue, is with a couple of motor driven variacs. One would add some voltage from the supply phase to bring the lagging (or leading if you use a dual phase variac) teaser leg back to 90 degrees. The second variac would correct the teaser leg voltage.
--
-----------------------------------------------------------------------------
| Phil Howard KA9WGN | http://linuxhomepage.com/ http://ham.org/ |
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
| The use of a 3 phase induction motor with two terminals connected to a | single phase supply and all 3 legs used to provide 3 phase to another load | ( appreciably smaller than the rating of the 3 phase machine ) will and does | work quite well. The phase relations will be correct although there may be | some voltage magnitude unbalance where capacitors can help. It seems to | require delta or ungrounded Y. Frequency stability is not a problem .The | slip doesn't matter as, at any slip the rotor field will be at synchronous | speed with respect to the stator and this "synchronous speed" is determined | by the primary supply.
The rotor field will be synchronous, but lagged a number of degrees that depends on load. The connections that are also feed from the single phase supply will have the non-lagging supply phase are part of its total vector while the 3rd leg will not. The ratings are probably the issue as to how many degrees of difference between induced and supply voltage there will be.
The connection will be a T, so the loads cannot assume grounded wye. But if there is a need for genuine wye, a delta-wye transformer could then be added to get a separately derived wye system.
--
-----------------------------------------------------------------------------
| Phil Howard KA9WGN | http://linuxhomepage.com/ http://ham.org/ |
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

load
does
be
.The
synchronous
determined
Be careful here. The rotor field will only lag if the *shaft* is loaded, *not* the 3-phase output. If the motor is run unloaded so the only mechanical power developed is to overcome bearing friction and windage, the phase angle difference would be miniscule. The output of the third phase *can* lag if the electric load is high compared to the machine's rating. But that's a different phenomenon, the inductive voltage drop in the windings. This is one reason why the electric load must be kept relatively small compared to the machine's rating. Some of this can be compensated with capacitors, but unless the load is constant, it's hard to strike a balance of capacitance between no-load and full-load. And the power factor of the load itself can be an issue.

be.
This can easily be verified by simply measuring the voltage from third-phase to one incoming line, and comparing to the voltage from the third-phase to the other incoming line. With a delta output, that's the only real difference one will see. And if the rotor shaft is not loaded, and the electrical load is kept to a fraction of the machine's current ratings, this voltage difference will be quite small.
This type of setup has been used many times for powering three-phase motors where no three-phase service is available. IIRC, the OP was concerned about 'sensitive electronics'. For most electronics that I've seen, the three-phase is simply rectified/filtered for DC, so a slight phase shift wouldn't matter at all (unless the rectifier loading is at the extreme limits). VFD's 'convert' the 3-phase to variable DC with thyristors and the gating of those is tied to the incoming phase, so they wouldn't care much about a slight phase-phase voltage imbalance.
For some types of servo-mechanisms that use line AC to feed the reference windings and electronics to control a second AC, a slight phase shift *could* cause the servo to not null-out properly and stop moving when the electronics wanted it to. But such high-gain servo systems could probably adapt for the slight shift with little noticable effects.
If the requirements are *really* critical, probably couldn't use utility three-phase either and would need to generate three-phase from an M-G or inverter anyway.

if
added
Correct.
daestrom
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:
| Be careful here. The rotor field will only lag if the *shaft* is loaded, | *not* the 3-phase output. If the motor is run unloaded so the only
You're saying that no matter how much power is drawn from the generated teaser phase, this won't put drag on the motor?
| mechanical power developed is to overcome bearing friction and windage, the | phase angle difference would be miniscule. The output of the third phase | *can* lag if the electric load is high compared to the machine's rating.
How much for how much?
| But that's a different phenomenon, the inductive voltage drop in the | windings. This is one reason why the electric load must be kept relatively | small compared to the machine's rating. Some of this can be compensated | with capacitors, but unless the load is constant, it's hard to strike a | balance of capacitance between no-load and full-load. And the power factor | of the load itself can be an issue.
So the converter (the machine generating the extra leg) needs to have a rating way over the driven motor? Well, then it would seem to me that the costs would now much much higher. Why not just have a motor-generator set at the load rating instead of a motor alone at a much higher rating?
| This type of setup has been used many times for powering three-phase motors
Or in the case of motors that can run on single phase once started, the third phase can be disconnected once the motor is up to speed, and used to start another motor.
| If the requirements are *really* critical, probably couldn't use utility | three-phase either and would need to generate three-phase from an M-G or | inverter anyway.
That's what I was thinking. I don't get three phase at home, so I don't have anything I can hook up and test right now.
--
-----------------------------------------------------------------------------
| Phil Howard KA9WGN | http://linuxhomepage.com/ http://ham.org/ |
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 7 May 2004 01:48:53 GMT, snipped-for-privacy@ipal.net said:

It's typical, when using a rotary phase converter to power CNC-type machinery, to use an idler motor at least twice the horsepower rating of the motor in the machinery. This is mainly because this type of machinery often has to reverse rotation instantly, or undergo abrupt changes in the mechanical load.

What I'm getting out of this thread so far, is if AB is the single phase input, and you're able to tune the phase angles under no-load conditions using running capacitors so that the sines are 120 degrees apart, that condition will only hold true in the unloaded condition. Under a mechanical load on the driven motor (not the idler) the timing will shift, so that BC is either more or less than 120 degrees, and CA will be comprised of 360 degrees, minus AB and BC. Is this correct? If so, it still sounds like nobody in here is aware of any magical way to dynamically correct that by "having the windings converted for the phase angle differences" or "modifying the rotor" as DES is claiming. This is what I mainly wanted to know, because I can't think of any possible way to make a mechanical alteration to a three phase motor that would allow it to dynamically correct a varying condition... correct?
--
JR

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
| It's typical, when using a rotary phase converter to power CNC-type | machinery, to use an idler motor at least twice the horsepower rating of | the motor in the machinery. This is mainly because this type of | machinery often has to reverse rotation instantly, or undergo abrupt | changes in the mechanical load.
The more horsepower in the idler (I'd call it a converter), the less the phase angle should lag for some given load.
| What I'm getting out of this thread so far, is if AB is the single phase | input, and you're able to tune the phase angles under no-load conditions | using running capacitors so that the sines are 120 degrees apart, that | condition will only hold true in the unloaded condition. Under a | mechanical load on the driven motor (not the idler) the timing will | shift, so that BC is either more or less than 120 degrees, and CA will | be comprised of 360 degrees, minus AB and BC. Is this correct? If so, | it still sounds like nobody in here is aware of any magical way to | dynamically correct that by "having the windings converted for the phase | angle differences" or "modifying the rotor" as DES is claiming. This is | what I mainly wanted to know, because I can't think of any possible way | to make a mechanical alteration to a three phase motor that would allow | it to dynamically correct a varying condition... correct?
I'm unaware of any magic. It can be done, but it requires adapting to the particular load. The way I would think about correcting it would be at the electrical level. But I could imagine correcting it by shifting the 2 armatures that are connected to the new phase by some number of degrees. But that's a whole new motor design I wouldn't get into. If I wanted to correct this mechanically, I'd have a detection control circuit drive a couple variac motors to change compensation voltages to get the new leg back in the correct position relative to the other two.
But this gave me a thought. If you can access the discrete leads from the windings (6 leads for 3 windings, 12 leads for 6 windings, etc), you could separate the windings so that ONE winding is run from the single phase supply, and the other 2 windings are wires with only the new phase end in common, forming an OPEN DELTA. A capacitor starter of the right values could power that open delta to bring the idler up to speed, then switch out to just single phase power to tap power from the open delta in three phases which should hold a good 120 degree relationship. The voltage might drop a lot under load, and the whole thing could slow down or speed up as load changes (effectively rotating the phases around relative to the intended frequency). But I think the phase _relationships_ should hold fairly tight with this. The motor would need to have a lot more of a rating than the load, not just because it is an idler, but also because an open delta has has to power 3 phases from just 2 windings alone.
--
-----------------------------------------------------------------------------
| Phil Howard KA9WGN | http://linuxhomepage.com/ http://ham.org/ |
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.