A Thought Experiment

[Andy™]

so look at the control gear and see if its wired star or delta...

or just go buy a new one, less hassle

 

[Sidewinder]

The control gear is DOL.

The machine is obsolete so you can't buy a new one, plus there is no rating plate on the motor to tell you what to buy, if you go to a motor supplier.

OK, I have sketched a first thought for a test circuit for a Delta winding.

I think I have an answer, but I'm not going to give it out in one hit, as I think it defeats the object of the reason I started the thread, to get people THINKING about the theory and the underpinning knowledge, which is why we do college etc. in the first place, so we can learn this stuff, to enable us to solve difficult problems from first principles.

Whiteboard-2014-03-17-21-39-15.pdf

 

[NozSpark]

Could you test the load to see how much torque is needed to rotate it???

 

[Andy™]

just connect it up any way. 50/50 chance. if it all goes wrong, go here and buy the last item to repair it. then reconnect the other way

 

[Doc Hudson]

Deke,

I don't think resistance will tell us, each of the windings are equal resistance.

Letting the ends be as you have labelled them a, b, & c.

a to b will be two windings, b to c will be two windings, c to a will be two windings, if star.

a to b will be one winding, b to c will be one winding, and c to a will be one winding, if delta.

Trouble is we don't know what the resistance of one winding is, so we also don't know what the resistance of two windings is either.

Not blindingly obvious really.

Hence why I posted.

I'll have a look at what technician has posted later if I get time.

I think we need to stick known currents & voltages in and see what we get as an initial guess.

As you say in Star the resistance between a-b, or b-c, or c-a will be 2x the resistance of a single winding. But if Delta wouldn't the resistance between a-b, or b-c or c-a  be One winding in parallel with Two windings? i.e. resistance measured = 0.66 of the resistance of a single winding. Not sure that helps resolve the issue though.

Doc H.

 

[Doc Hudson]

Measure the resistance between any pair of windings call that value 'A'.

Short out one pair of terminals (which will either be one or two windings), then measure the resistance between the unshorted terminal and the two shorted terminals.

If Star isn't the new resistance going to be 'A' + (2x'A' in parallel) i.e. 1.5x'A'

If Delta isn't the new resistance going to be 2x'A' in parallel as the third winding is shorted out ie. 0.5x'A'

So if your new reading is greater than original reading it is Star if less than it is Delta.

Or am I missing something? 

Doc H.

 

[Technician]

I have a theory, and I'm going to go through what technician has posted next.

Sidewinder; I will have to post the sketches and yes,

Kirchoff's laws underpin this lot.

It is a good many years ago but I knew the guys who were

REALLY up against it. 

The problem was a tank of liquid methane and the submerged

pump single phased and blew the HRC fuses.

The procedure in this event was the use of another submersible

pump, followed by a warm-up, an inerting procedure with dry nitrogen,

followed by a gas free and certification prior to entry.

AND then, only then, the repair.

Followed by....inerting,  cooldown and recommissioning.

The guys simply HAD to prove that it was the motor and

nothing else.  The approach I outlined was how they did it,

and when the motor was isolated, one set of windings was indeed

down, and the analysis proved which one it was and it was

found to be so.

 

[apprentice87]

Hi there,

Say all the windings had a resistance of 10 Ohms.

Now taking delta first. measure the resistance between two of the wires, it will be 6.6667 Ohms.  If you then linked two of the wires and then measured the resistance from the linked wire to the third one, you would have a reading of 5 Ohms.

Now divide the 6.6667 by 5. You will end up with the ratio between the two of 1.333

Now do exactly the same, but for the windings in star.

The resistance between two wires will be 20 Ohms. The resistance between two linked wires and the third one will be 15 Ohms. Divide 20 by 15 and you will get the ratio between the two, it is exactly the same at 1.3333 ******!!!

What you have discovered though, is that whichever way you measure it, the motor in star will have readings exactly 3 times higher than the delta one.

So, look at motors with the same shaft and keyway size, and same foot dimensions. Compare your readings with one of the "new" ones. If your readings are roughly the sameish, your motor is wired the same. If yours are about THREE times higher or lower, your motor is wired the "other" way.

Only thing i can think of..

john..

 

[apprentice87]

Hi Canoe,

.76 being the reciprocal of the 1.333 i gave above, so no, as you say, it will not work!!!!!!! ****** and double ******!!

john...

 

[kerching]

Well, a short while ago there was a bloke on here who had done some panel building....had it on his profile so it must be true.

Cannot remember his name though as he has not posted for a few weeks!

I will warrant and wager a slap up feast at Mrs Miggins Pie Emporium that he will know the answer

If only I could remember his name...........

 

[Technician]

I have had a look at this all ends up....I have even tried to

examine it using a "black box" technique.

Consider this;  If one conductor is disconnected in delta

only one set of windings is energised.   If one conductor

is disconnected in star two sets of windings will be energised.

This means that in delta the motor will hardly turn.  In star,

on no-load, the motor may try to turn, one way or the other.

 

[Evans Electric]

I will be interested in the solution to this ...I have no idea TBH  although I remember it cropping up some time ago ,  somewhere.

I 'd probably take it round to the rewinders and ask them .......they know strange things you'd never dream of .  

One thing I'd add to the thread is there was a job in a large boiler room once ,  a row of 6 or 7   3 ph pumps around  1/2 HP  I think .   All existing , all connected   in Star .     New pipework etc was added at the end of the row and a new pump, identical  but configured in Delta .   I looked at the other 7 motors , all Star  so altered the new one to Star .  

Bad decision ....heating guys run it up and a winding failed .    I never found out why 7 pumps were Star and that one was Delta.

 

[Technician]

I think I have the answer.....using a "black box" approach.

Just examining my reasoning.

 

[Sidewinder]

...  

Bad decision ....heating guys run it up and a winding failed .    I never found out why 7 pumps were Star and that one was Delta.

Deke,

I've been there, cost me a motor, breakdown, customer screaming, got new motor, looked @ old one, star, changed links on new one to star, job's good un.

Machine away.

Less than 24 hours motor burned out!

I did not look at the rating plate, 230/400 vs 400/750 or summat, can't remember now.

Canoe will I suspect, I'd have to look up the reasoning, and theory etc.

Canoe, can you explain this one, whilst I try to catch up on the ideas on the theory of identifying the windings, I genuinely have not had enough time to think about this for long, and I don't have a definitive answer, there is one, and, I "think" at the moment that the key is Kirchoff.

Next is to write out & try to draw out the suggestions by Technician, to get my head around these.

Remembering it is the exception that disproves the rule.

I realise that this is totally theoretical, but, I feel it is a useful thing to be able to know how to do to be honest, plus, this could then be extrapolated out to PM servo windings, as in BLDC motors is I believe another thing they are called, but, never got my head around that, because I was trained at a manufacturer, I really only learnt their terminology and techniques etc.  :coat

 

[Sidewinder]

I think I have the answer.....using a "black box" approach.

Just examining my reasoning.

I "think" it "has" to be a black box approach Technician, as we don't know if it is Y or D, and there is no way to tell, apart from the meter readings...

curses  upon theses storemen who cannot maintain the records system....look at the trouble they have caused

You told the store man that he was delinquent in keeping records, and now you also have a black eye to go with the motor that you can't identify the winding configuration of!

 

[steptoe]

I "think" it "has" to be a black box approach Technician, as we don't know if it is Y or D, and there is no way to tell, apart from the meter readings...

You told the store man that he was delinquent in keeping records, and now you also have a black eye to go with the other one your wife gave you

:slap

 

[Evans Electric]

In a star connected generator   the phase voltage would be  Vp

                                                                                               Sq, root 3

In mesh  the line current  =  sq,root3  X phase current.  

I don't think it helps.

 

[kerching]

I find that storemen are very quick on their feet when they choose to be, BUT they have to be in their own mysterious "racked out and sorted alphabetically and numerically storage facility" and wearing the compulsory 'brown warehouse cloak/coat of invisibility'. Along with the 'clipboard of cussedness' which they are given when they go through the "5 different coloured pens in the top pocket ceremony'

Remove these symbols of office and they could not catch a cold, never mind give out a black eye. Probably the worst you would receive would be a withering look

Just saying

 

[poni]

Why does it matter?, it's a replacement, it's DOL Star or Delta, just stick it in and let it run

 

[Sidewinder]

I have had an idea.

Arrange to pass a small direct current through

the motor windings using a small power source,

like a battery.

Measure said current.  Do this three ways.

Record results.  They should be the same.

Now, let one terminal be A the next B and the next C.

The battery and its ammeter are connected to A & B.

Measure the voltage between terminal A and C and

then B & C.

----------------------------------------------------------------------

If this measured voltage is half the applied voltage,

i.e. V (A/C) and V (B/C)  is half voltage A/B then

chances are this is a delta connected machine.

-----------------------------------------------------------------

Now, connect another ammeter so that it is teminated

at winding C at one side and to A on the other.

This puts the windings at A and C in parallel.  This

ammeter should show half the reading of the other one

which is measuring total current.

If you do this on a delta connected motor you will short

the energy source and this ammeter will go off scale.

Hope this is clear but I am open to suggestions & questions.

I have sketched this out on a cigar box and it looks OK.

I have split the post into 3 sections, and tried to sketch out the options, my comments are on the sketch, I don't see how this can work, you are suggesting connecting an ammeter in parallel with one of the windings, thus it will merely act as a shunt.

On my diagrams I have used a delta looking triangle, as we know we have 3 connections it is just the inside of these we are not sure about, this is the reason the connecting wires are shown mid side of the triangles.

Got a little computing issues, I'll upload my sketches for my thoughts very soon!

Sorry,

Document-2014-03-18-21-43-24.pdf