Traction motor commutation

[Skogsgurra]

I am looking at a possible conversion of a diesel-electric locomotive so that it can be run off the catenaria instead of from the usual DC generator.

The railway grid in question is a 16.7 Hz grid and the ripple in the rectified voltage will cause a lot more armature current ripple (33.3 Hz) than the existing DC motors were built for. At least, that is what I think.

I haven't been very successful in searching for information on this problem. Anyone out there with first-hand knowledge? Especially practical data on allowed ripple percentage, allowed current A/s, needed inductance and other such things for DC traction motors at around 1000 kW.


Thanks a lot for your inputs.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[waross]

An anecdote that may be helpful.
A friend lives in a small city where freight trains are often stopped.
These are very long trains and the location is at the eastern edge of the Rocky mountains. The majority of the freight is Canadian grain destined for the seaport of Vancouver. There is an issue of residential foundations cracking in the city. It is attributed to the cogging of the locomotive drives as the trains start.
These are AC motors supplied by PWM drives.
Yes, I know you are working with DC motors, but it may be well to consider whether a ripple on the DC will cause cogging.
My thought is that 16.7 Hz is fast enough that any cogging will be negligible.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Skogsgurra]

Thanks a lot Bill! That is the kind of anecdote that makes one think. I can surely imagine that those motors, that usually aren't even PWM fed, but with a very rudimentary GTO switching with a primitive full block modulation* (if you can call that modulation) and goes through the whole spectrum from zero Hz to perhaps a few hundred Hz, can act as giant vibrators and crack just about anything that resonates when the frequency sweeps across the spectrum during acceleration. Torque produced is usually two or three times rated torque in that phase of acceleration. So lots of things can happen.

My concerns are not about the community, though. It is enough to have the commutator to think about...


*Full block switching is used because the switched voltage is usually in the 1500 - 2000 V region in those drives and switching losses are quite high.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.