How to calculate time for a vertical rotor to coast down - for mechanical enggs

[khader316]

Dear Friends,

This is my first time posting something on eng-tips website.

I need to know how to calculate time for a electric generator rotor to coast down from full speed to zero. I was trying to us (Torque = inertia x acceleration) correlation but this would give me an incorrect result as the torque is not constant. I know the bearing losses at speed but as the rotor slows down, the bearing losses will also reduce. There must some way of calculating the time for rotor coasting.

Any advice will be greatly appreciated.

-Thanks
Khader

 

[Skogsgurra]

It may be that the bearing losses decrease when speed gets lower. But the friction may remain constant or even increase.

Windage is usually dominant at higher speeds and decreases when speed goes down.
You will never get a correct result from such a calculation. Bearing friction is never given as an exact number but more often as a minimum/maximum interval or a typical value. It also changes with temperature..

What is the purpose of such a calculation?

Gunnar Englund

http://www.gke.org

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

 

[khader316]

Thank you Skogsgurra.

The purpose is to know how long the battery should last to operate the DC motors that supplies the cooling oil to the bearings.

 

[Skogsgurra]

I understand. I was involved in a similar problem with gas turbines. Very critical application - make them last five times longer than your most pessimistic assumption. Batteries age.

Gunnar Englund

http://www.gke.org

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

 

[jraef]

Might it be worth it to monitor the speed to make sure? Or is this for pre-design?

"Will work for (the memory of) salami"

 

[khader316]

Skogsgurra,

What do you recommend to calculate the coastdown time?

 

[waross]

Have you investigated the possibility of using the batteries to also inject direct current into the windings to reduce the coast-down time?
You may be able to shorten the coast-down time enough that net battery amp/minutes is reduced.
Do you need bearing cooling at near zero speed or is there a speed below which the bearings do not need to be cooled?

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

 

[LionelHutz]

You'll just have to make a best guess at the torque. I'd use full speed friction torque x speed^2 curve to represent the loss torque.

 

[Skogsgurra]

My recommendation is to use only windage and pretend that the bearings are completely frictionless. That way, you will get a worst case stopping time. Then double (at least) that when selecting battery capacity.

Gunnar Englund

http://www.gke.org

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

 

[mikekilroy]

asked before, is this for a machine already built or being designed?

You say "I know the bearing losses at speed:" how? Does this mean you know the no load torque required to rotate it at that speed?

If so, I add my agreement to Gunnar answer and use THAT value for calculatio, then at least double the batt ah answer. This is how we do it, and yes, we use the formula you mentioned in this form T=Jw/t or

t_in_sec = J_in_#-ft-sec^2 * (rpm_max)/(9.55*T_in_#-ft)

http://www.KilroyWasHere<dot>com