why vary pitch in a generator 7

[eleclew]

What is the affect of varying the "pitch" in the windings of a generator - is it to reduce eddy current losses and harmonic currents - we have a generator with a 7/9 pitch (so we're told).

 

[rbulsara]

1. Cost or economics, certain pitch uses less copper.
2. Reposes to harmonics content varies. (Caterpillar has a good data sheet on it LEKX3115, ask your local rep).
3. To match existing units, if they were to run in parallel.
4. It also affects impedance, a certain value is important.





Rafiq Bulsara

http://www.srengineersct.com

 

[electricpete]

Short pitch coils (as opposed to full pitch) have a dual benefit of reducing copper requirements and improving harmonic performance. There was a good disussion of harmonic published b Caterpillar entitled "Generator Winding Pitch and Harmonics" by Caterpillar.

=====================================
(2B)+(2B)' ?

 

[rbulsara]

Hate typos. Reposes=Response. In item 4, "if" certain value is important....

Rafiq Bulsara

http://www.srengineersct.com

 

[catserveng]

Here is a link to a condensed version of the CAT info referenced above used by dealers for sales info.

http://www.toromontcat.com/powersystems/pdf/ConsultantsCorner/Generator Windings Pitch.pdf

I have a pdf version of the complete article on my compute at home but am traveling now, so I'll try to post it when I ge back next month.

Or you contact your local CAT dealers engine sales department.

Hope that helps,

Mike L.

 

[electricpete]

A brief analysis of 7/9 pitch effect on harmonics using the harmonic pitch factor formula found in almost any electric machinery book:

Kp = sin{nu * (y/yp) * (Pi/2)}
where y is coil span, yp is full-pitch coil span, and nu is the order of the harmonic.

For y/yp = 7/9, we calculate:
nu Kp(nu)
1 0.940
3 -0.500
5 -0.174
7 0.766
9 -1.000
11 0.766
13 -0.174
15 -0.500
17 0.940
19 -0.940

This does a good job of cutting the 5th spatial harmonic of flux. As you probably know, even harmonics aren't of concern in symmetric machines and triplen harmonic currents are often limited if zero-sequence path is restricted, so 5th can end up being the next logical target to limit with pitch factor as above.

=====================================
(2B)+(2B)' ?

 

[zlatkodo]

This is a well-known rough estimate of the influence of pitch value on the higher harmonics:
- a/ a full pitch will have no damping effect on any harmonic frequency.
- b/ a 2/3 pitch will eliminate the third harmonic and subsequent triplens i.e.: 9th, 15th, 21st, 27th, etc ,
- c/ a 4/5 pitch will eliminate the 5th harmonic.
- d/ a 6/7 pitch will eliminate the 7th harmonic.
-e/ a 5/6 pitch will:
1. minimize the 5th harmonic, but not eliminate it as will a 4/5 pitch.
2. minimize the 7th harmonic, but not eliminate it as will a 6/7 pitch.
- f/ it should be noted that only a full pitch gives maximum power from the core.
Reduced, 2 / 3 pitch , for example, gives approximately 15% less power compared to full pitch ( for same core).
Zlatkodo

 

[electricpete]

All good points.
As a re-inforcement, they can be shown from the equation Kp = sin{nu * (y/yp) * (Pi/2)}:
* a is shown by setting y/yp=1 which leaves Kp = sin{nu * (Pi/2)}=1 for nu odd (which are the only nu of interest)
* b/c/d are a result of the fact that Kp=sin{nu * (y/yp) * (Pi/2)} is 0 iff {nu*(y/yp)} is an even integer. (for the harmonic corresponding to full pole pitch, nu/yp=1 and Kp is 0 when y is an even integer).
* e was shown in my previous post. (note 5th is reduced much more than 7th)
* f is shown by considering nu=1

Again that is not to detract from those points in any way. I just think they stick in our minds more if walk thru the logic.


=====================================
(2B)+(2B)' ?

 

[electricpete]

* e [5/6] was shown in my previous post. (note 5th is reduced much more than 7th)

Correction - my post above addressed 7/9. For 5/6, both 5th and 7th harmonics are reduced (both have Kp=0..25) as zlatkodo said.

=====================================
(2B)+(2B)' ?

 

[wolf39]

Returning from a trip, I came across this interesting post.

The first and only reason a hydro generator designer has in mind to select a certain pitch is to reduce harmonics in the sine waveform voltage. Standard wye-connected stator windings take care (remove) harmonics of 3, 9, 15, etc. For a reduction of the 5th and 7th harmonics a pitch in the region of 5/6 is the most effective compromise.

It is true that a certain saving of winding copper can be achieved by selecting a substantial pitch. However, this is only a welcome side-effect, not a primary factor for a good generator design. But, selecting a high pitch has the disadvantage that the number of slots or number of coil turns has to be increased for the following reason:

A full pitch does utilize the magnetic flux of a generator by 100% whereas the flux utilization of a 5/6 pitch drops to sin (90° * 5/6) = sin 75° = 0.966. In other words: For a given voltage the number of slots or number of coil turns have to be increased by 3.4% if a pitch of 5/6 has been selected.

A 2/3 pitch is unheard of in the hydro generator field as the flux utilization would drop to sin (90° * 2/3) = sin 60° = 0,866, i.e. the stator slot number or number of coil turns have to be increased by 13.4%. It would be interesting to hear from CAT in which cases such a pitch is considered.

Hydro generators can run parallel with no problems at all regardless of different pitches. Here, too, I'd like to know the background of concern.

The influence of a pitch variation on impedance values is negligible. High pitch stator windings may have lower stator leakage reactance figures because of shorter end windings. However, this is counterbalanced by the increase of slots or number of coil turns.

Wolf

http://www.hydropower-consult.com

 

[electricpete]

It would be interesting to hear from CAT in which cases such a pitch is considered.

From the article cited above you can gather it was a competitor of Caterpillar that used 2/3 pitch. The application is relatively small generators where I gather that the need for efficient utilization of materials is tempered by a goal to produce a single machine which could adapt to multiple applications. The ability to connect the winding in delta helped this goal. In delta 2/3 pitch was useful to reduce circulating 3rd harmonics in the delta.

The benefit ifrom copper reduction from full pitch coil is generally more pronounced for 2-pole machines with large pitch and large endwindings than for slow hydro machines with relatively smaller pitch and smaller endwindings.

Look at the example pitch cited 5/6, reduction in fundamental = 1-0.966 =3.4% so we’ll say for simplicity increase in number coils is 3.4%.
What is reduction in copper per coil? Reduction in copper per coil = (1-5/6)*X where X is something like ratio of endwinding length (excluding knuckle) to total coil length. I think 0.5 or more is typical value of X for a 2-pole machine (lower for low-speed machines with short spans). Using 0.5 Reduction in copper = (1-5/6)*0.5 = 8.3% which more than compensates for the 3.4% inrease in coils.

This is one of reasons 2-pole tend toward lower pitch ratios like 4/5 (along with constructability, coolability, and strength of the endwinding area).

For hydro it may be tend to be much less a consideration as you pointed out.


=====================================
(2B)+(2B)' ?

 

[electricpete]

I'm guessing maybe X=0.25 might be representative for hydro (?). If you plugged X=0.25 into above example it would roughly be a wash (reduction in copper from coil span is negated by increase in coils to counter reduced fundamental).

=====================================
(2B)+(2B)' ?

 

[wolf39]

Pete:

As I said: The designer would like to go for a full pitch to get the most economical design but harmonics dictate him to select a suitable pitch. In your calculation you save 8.3% copper weight for a 5/6 pitch. The cost to manufacture and install 3.4% more coils (labor, insulation, copper) may well outweigh the pitch related copper savings.

All posts sofar are referring to multiple coil windings, i.e. lap windings. For large low speed hydro generators with Roebel bars the standard type of windings are wave windings. For such windings there is no pitch-related reduction of end winding dimensions and copper savings.

It would be interesting to hear to what type of generator eleclew is referring to.

Wolf

http://www.hydropower-consult.com

 

[electricpete]

Wolf,
A quote from "Design of rotating electrical machines" by Pyrhonen, Jokinen, Hrabovcova ISBN 978-0-470-69516-6 (

Because of short pitching, the coil end has become shorter, and the copper consumption is thus reduced. On the other hand, the flux linking the coil decreases somewhat because of short pitching, and therefore the number of coil turns at the same voltage has to be higher than for a full-pitch winding. The short pitching of the coil end is of more significance than the increased number of coil turns, and as a result the consumption of coil material decreases.

I would say what he wrote in bol is a generalization. You and I could compute more exactly how the balance shifts for a specific machine. But my basic point: certainly the savings in copper deserves to be mentioned and considered among the advantages of fractional pitch coil.I'm not disagreeing with anything you said as long as you're not disagreeing with anything I said

=====================================
(2B)+(2B)' ?

 

[wolf39]

As said before, for the most efficient design the generator designer would select a full-pitch stator winding. In order to reduce harmonics from the sine waveform voltage a certain pitch has to be selected and a pitch of in the region of 5/6 is the most effective as the 5th and 7th harmonics are suppressed equally well. However, for a design with 3 slots per pole and phase, the pitch selected can only be 7/9 or 8/9, both not very close to the 5/6 figure. In such case the designer would select 8/9 for the more efficient design.

Because of the above, the question of "varying the pitch" cannot be answered satisfactorily. A designer always would go for 5/6 or slightly above and wouldn't fiddle around with other figures.

In order to find out how much the pitch can influence a design, I designed two hydro generators with the following main data:

40 MVA 13.8 kV p.f. 0.9 50 Hz 14 poles Xd 1.00 p.U.

The first design had 168 slots and a full pitch of 12/12, the second design had 174 slots and a pitch of 10/12 3/7. The result didn't surprise me. The copper saving of the second design was a mere 105 kilograms, equivalent to a saving of US$ 1000 at todays copper prices. Nevertheless, this is a certain copper saving, as pointed out by others already. However, the loss situation was less favourable. The armature losses of the second design were 8 kW lower, as could be expected. In contrast, however, the stator core losses went up 8 kW compared with the first design. At first glance one could say that this is a break-even in the loss situation. With a loss evaluation of 9000 US$ per kW core loss and US$ 6000 per kW armature loss (these are reasonable figures as I recall a loss evaluation for a recent hydro project in the U.S. of about US$ 17500 per kW), the second design would end up with a disadvantage of US$ 23000 compared with the full pitch design. Again, a full pitch design has to be ruled out because of the harmonics. But in general it can be said that the shorter the pitch, the less economical the overall design will be.

BTW, the reactances were not affected considerably as the synchronous reactance can be adjusted by varying the air gap and the subtransient reactance of the second design was only 0.6% less (18.7% versus 19.3%) compared with the full-pitch design.

As one can see, an at first glance easy looking question can turn out to be quite complicated to answer.

Pete: The copper reduction per coil for the second hydro generator design is (1 - 5/6) * 0.42, but you have to add the copper weight of 6 coils. I guess that for large 60 Hz 2-pole machines the X-figure will be in the region of 0.25 or even below. Such machines have core length's of about 8 metres and rotor diameters of about 1.1 metres.

Wolf

http://www.hydropower-consult.com

 

[electricpete]

The first design had 168 slots and a full pitch of 12/12, the second design had 174 slots and a pitch of 10/12 3/7. The result didn't surprise me. The copper saving of the second design was a mere 105 kilograms, equivalent to a saving of US$ 1000 at todays copper prices. Nevertheless, this is a certain copper saving, as pointed out by others already. However, the loss situation was less favourable. The armature losses of the second design were 8 kW lower, as could be expected. In contrast, however, the stator core losses went up 8 kW compared with the first design.

If I read the bolded portion literally, it suggest the second design = fractional-pitch coil design had higher core losses than the first design = full pitch.

But I think you meant to say it the other way around, correct?
(i.e. full pitch core should have higher core losses due to the higher harmonic content).
I’ll assume that’s what you meant.

So, to my way of thinking, what you have shown fully supports exactly what we said. Reducing coil pitch gave copper savings even for slow speed hydro design. It was very simple math above that showed the copper savings tend to be even more pronounced for fast-speed machines than slow speed since the endwinding is a larger fraction of the coil.

That there are also benefits from reduced core loss for fractional pitch coil is not a contradiction either because we said that reduced harmonics are also a benefit of fractional pitch coil. So again, no disagreement from me and no conradiction that I see.

Pete: The copper reduction per coil for the second hydro generator design is (1 - 5/6) * 0.42, but you have to add the copper weight of 6 coils.

That is why in my 17 Jan 11 10:36 post I compared 8.3% to 3.4%.... the 8.3% represented copper reduction per coil and the 3.4% represents increase in number of coils for that example.

Nothing you said surprises me and I have no disagreement with anything you said other than to the extent you imply that coil length is irrelevant. If my mention of “full pitch coil” in my very first post is what’s bothering you, then I withdraw it…I was just trying to describe in a few words the effect of change of pitch and never intended to suggest that a full pitch coil would be a viable design.


=====================================
(2B)+(2B)' ?

 

[electricpete]

I guess that for large 60 Hz 2-pole machines the X-figure will be in the region of 0.25 or even below.

That part was a point that I did not correctly describe, good point. Looking among a variety of machines I think faster machines tend to have higher ratio "X" than slower. But does not apply for large turbo machines.

=====================================
(2B)+(2B)' ?

 

[wolf39]

Pete:

The fractional pitch coil design had iron losses 8 kW HIGHER than the first design. More slots lead to more core teeth and higher flux densities in this region. In addition, losses in pressure plates and pressure fingers increased and the iron losses in the pole shoe surfaces went up because the air gap had to be decreased in order to keep the synchronous reactance Xd = 1.00 p.u. constant (otherwise the excitation losses would have gone up). The 5th and 7th harmonics don't play a substantial role iron loss-wise.

All the discussion started with the following question: "What is the affect of varying the "pitch" in the windings of a generator". The contents of some of the replies were beside the point and were more or less based on hearsay, not on personal experience. I was trying to explain - obviously without success - to the forum that a designer doesn't play around with different pitch figures. It's your way of thinking that what I have shown fully supports exactly what you and others said. I fully disagree in this point.

Please explain to me the logic of your 17 Jan 11 12:59 statement: "I'm not disagreeing with anything you said as long as you're not disagreeing with anything I said". Imagine if one of us is wrong.


Wolf

http://www.hydropower-consult.com

 

[electricpete]

Please explain to me the logic of your 17 Jan 11 12:59 statement: "I'm not disagreeing with anything you said as long as you're not disagreeing with anything I said".

I was explaining where I was coming from while trying to figure out where you were coming from. The contradiction was that I kept on hearing a tone suggesting I was wrong (for example when 2 in a row posts starts with "As I said....)" and yet I honestly didn't see any contradiction between what you said and what I said. So I was explaining where I was coming from. I wasn't disagreeing with anything you said except the tone implying that I was wrong about something which I never could pin down. I hope that explains the logic of the statement you quoted... if you thought it was an insult than I believe you misunderstood it. However I'm sorry for whatever it is that I said that you may have found offensive in some way.


=====================================
(2B)+(2B)' ?

 

[edison123]

pete

"This is one of reasons 2-pole tend toward lower pitch ratios like 4/5 (along with constructability, coolability, and strength of the endwinding area)."

I have done quite a few 2 pole machines and they all were 2/3rd pitch.

As Wolf said, it's usually 5/6th in lap wound hydro machines to reduce 5th & 7th harmonics.

In diesels, all kinds of pitches from 2/3 to 4/5, 5/6 to 3/4 are used.



Muthu

http://www.edison.co.in