For alternators, what are practical design aspect ratios (lngth/dia)?

[mdx]

The alternator is a fairly simple maching with a stator and rotor being the two primary components. For different applications it may be desireable to either minimize the diameter or minimize the length. Hoover Dam, for example, seems to have short length generators with the diameter being approximately twice the length, whereas many gen-sets (engine generator assemblies) have alternators with the length being about two times the diameter. So, without compromising function in any significant way what is a practical range for the aspect ratio (length/diameter) for an AC generator?

 

[ScottyUK]

You are comparing very different machines: high speed turbo-alternators with two poles have long and relatively slender rotors which are machined from a single forging. A 'typical' ratio might be 8:1 for length : diameter. A low speed salient pole machine with perhaps 30 poles might have a ratio of 1:6 for length : diameter. Turbo machines are limited by mechanical stresses on the rotor and the ability to produce such a massive forging of uniform quality; and above a certain size they are usually constructed as 4-pole machines running at half the speed of a 2-pole type. Salient pole machines have much lower stresses and normally use a built up method of construction, and can thus be constructed with very large pole counts and in very large diameters for low speed applications.


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

If we learn from our mistakes I'm getting a great education!

 

[Hoxton]

The basic formula is P (power in VA) = D2 x L x N

Where :

D2 is diameter of core (not the frame) squared
L is length of core
N is speed

So this drives towards a short fat machine rather than a long thin one. The real limits are material stress (hence turbogenerators tend to be long and thin because of their high speed), and winding cost, the stator endwinding having a high labour element because of all those connections made manually.

 

[ScottyUK]

Hi Hoxton,

What are the other units - metres, inches, radians/second, rpm? I tried a couple of units I'm familiar with and couldn't get sensible results.


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

If we learn from our mistakes I'm getting a great education!

 

[starkopete]

Hi,
The formula that I have seen concerning diameter and length is attached. It is a little different than the one mentioned above. I have used it for estimating the HP of DC machines and found it to be pretty close.

 

[ScottyUK]

If the formula result was in kVA and the diameter and length are in metres and speed is rpm then Hoxton's formula is not far off for a 100 - 200MW machine. Note that 'core diameter' is the stator core - I initially used the rotor diameter which really screwed things up.


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

If we learn from our mistakes I'm getting a great education!

 

[Hoxton]

Hi

Al long as you are consistent, it does not matter, the result in this simple case is a factor, so you can check if your designer is in the right ball park.

When i was very very young and started life in the commercial department at Bxxxh, all our price submissions had to have the d2l space filled in. One of our directors was from Gxx in rugby and alleged that they always checked this.

In those days, it was inches and rpm

so a 70:84 3000rpm was 12.34 (we divided by 100,000 to get a sensible number). So if the design came to 20, say, questions were asked.

Actually, thinking about it, the N may have been poles, not speed, but the principle is the same.

 

[mdx]

mdx here. Thanx for all your responses.

The crux of the issue here may be the limit on diameter is due to the stresses induced by the rotation. How can I estimate the practical limit on diameter for a generator in the 1-10MW range for common RPMs such as 900, 1000, 1200, 1800? (I expect lower speed allows for a larger diameter) Also consider that the generator is mounted vertically (not horizontal axis).

Is there a reference source that contains the answer for generators in this range.

 

[electricpete]

In your original post you mentioned a big difference in aspect ratio between Hoover dam and steam turbines which are big difference in speed and as you mentioned slow speed machines tend to have large D/L... fast speed machines tend to have small D/L.

You'll notice as you change from slower speed applications to larger speed applications, the design will change from salient pole (high D/L) to smooth rotor (low D/L), and I think you are right that a big driving factor in that change has to do with the stresses and the costs of manufacturing a machine to cope with those stresses.

But stress is not the only factor and probably not the determining factor once you are comparing machines on the high end of your speed scale that all have cylindrical rotors. I say that because of course there are machines that are far larger ratings at 1800 and 3600 rpm and they find a way to cope with those stresses.

Some of the other factors may include:
* endwinding construction... high speed machines .. especially 2-poles tend to have large endwindings due to large span. To minimize wasted material in this area and also make the design easier to build may push toward lower D/L.
* number of slots and poles. There can be lower limits on slots per pole (or limited choices of slots per pole in the lower range) which become a problem for very low speed machines and drive the machine towards larger number of slots which may increase D/L

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

 

[wolf39]

mdx:

You have to make up your mind. As others pointed out already, you can't compare turbo generator type units (cylindrical rotors) with salient pole machines. Assuming that you have turbo generators in mind, the only practical pole numbers for three-phase units are either 2 or 4. However, single-phase turbo units usually have 6 poles.

In practice the designer is selecting the largest possible diameter (limiting factor is the permissible mechanical rotor stress). The core length then depends on the generator output.

Wolf

http://www.hydropower-consult.com

 

[electricpete]

In practice the designer is selecting the largest possible diameter (limiting factor is the permissible mechanical rotor stress). The core length then depends on the generator output.

Are you saying that a 1000 MW 2-pole generator has the same diameter as a 10 MW 2-pole generator, but 100 times the length?


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

 

[mdx]

OK. With these last two responses its getting clearer.
Yes, it makes sense that for a more expensive construction the rotor can be made stronger to allow for larger diameters at a given operating speed.

wolf39 - I don't know the fundamental difference between the "turbo generator type units (cylindrical rotors)" and the "salient pole machines", but I'd like to know what large D/L can practically be for generator applications where the need is for high efficiency electricity generation, 1-10MW and constant speed.

If there's a manufacturers published list for these machines with specs that may help.

Thanks again.

 

[wolf39]

Pete:
You are right, I should have explained all this in more detail.

I was thinking of large generators, i.e the 500 MW plus region. Such units are mechanically designed to maximum possible diameters, otherwise the stator cores would become too long.

Depending on the rotor material selected, permissible diameters do vary, of course. It makes no sense to design a 10 MW rotor like a 1000 MW rotor. In the latter case the designer has to select materials with higher mechanical stress limits (i.e. very expensive). On the other hand, high loss evaluation may dictate to select smaller diameters for less windage losses in the case of low output machines.

Other variables are cooling matters (10 MW units are air cooled, 1000 MW units H2/water cooled) and loss evaluation requirements. Air cooled generators have to be designed with far lower utilization factors (Esson' factors) than water cooled units.


mdx:
Look into the internet and find out what the rotor of a turbogenerator looks like (cylindrical rotor) in comparison with the rotor of a typical hydro generator which field poles are mounted to a rotor rim.

Wolf

http://www.hydropower-consult.com

 

[electricpete]

That makes sense Wolf - thanks.

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