diesel generator for motor 1

[godel123]

hi

a 50 KW asynchronous motor needs a 7 to 10 times that power to start(direct start) .

if we need to use a diesel generator we will need to 450 to 600 KVA !!

 

[zdas04]

I probably should have left the room when invited, but I felt like I was on the verge of getting something from the discussion.

How does that rule of thumb work when there are multiple motors? For example, could I run five 10 kW motors off a 65 kW genset if all the motors had VFD and there was no way to start more than one at a time (all are manual start, one guy has to start them all, and they are 1/4 mile apart).

David

 

[waross]

You can stay Scotty. Under Frequency Roll Off and volt/hertz limiter are similar enough to not matter. On the small machines the UFRO gets a lot more use than the volt/hertz limiter on the big ones.
Wayne, Under Funded, Really Optimistic really gives Under Frequency Roll Off a run for it's money.
You are always welcome David. That is an interesting suggestion.
The quick answer is yes it will work. A more accurate answer may be, I don't know.
If I may share an anecdote that may have some relevance;
I was on a project years ago when forbidden frequencies for pump drives were not well understood. We had some 60 HP centrifugal pumps with VFDs under PID control. At one particular speed setting, the pressure pulse reflections from the first 90, about 50 feet away coincided with the impellor vane period and the pressure pulses started to reinforce. The end result was a piece of casting about the size of my hand being blown out of the pump body. Twice. Lesson learned about forbidden frequencies.
What has that to do with generators? There may be a combinations of motor speeds and or loadings that causes the non-linear current draw of VFDs to reinforce with a bad interaction with the AVR.
I don't know, but were I faced with the installation that you suggest, I would tread cautiously.
Good question!!

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

 

[LionelHutz]

David - You'd likely be getting close to 60kW of load on the generator assuming all motors running at full load and I would be worried about the harmonics of the VFD rectifier causing issues with the generator more than the generator being capable of starting the motors. There is also the issue of the VFD harmonics possibly messing up the regulator unless the regulator has the appropriate filtering in it. You should be able to set the VFD so that any single motor would not draw much over 10kW as it starts.

 

[zdas04]

I have to use 10 kW motors because that is the smallest motor the manufacturer will put on this application. Actual load is closer to 3 kW. I think that in real life it is going to be a lightly loaded system, but it would be irresponsible to do the analysis based on that (I have a meeting with a EE next month and I'll drop it all in his lap, I'm just trying to understand enough to recognize smoke being blown up my pant legs).

Bill, that "forbidden frequency" discussion is new to me. Has it been discussed anywhere you could point me to? I have a bunch of folks that want to use centrifugal pumps for an application that I think should be PD (hugely variable head and capacity requirements, order-of-magnitude changes happen in under a second and can happen dozens of times in a minute), if I start chasing huge changes with a centrifugal pump I find myself in surge pretty early, but I didn't consider that there were frequencies that I had to avoid as well.

David

 

[waross]

Hi David. It has to do with the speed of the pump which is closely related to frequency.
As you know, as each impeller vane passes the outlet it generates a pulse or pressure peak. The timing of these pulses will depend on the number of vanes on the impeller and the speed of rotation. These pressure pulses will be reflected back from the first sharp bend in the piping. When the transit time of the pulses is equal to or a small multiple of the vane timing the pressure pulses will reinforce and dangerous peak pressures may build up in the pump housing and the discharge piping. Problems are easy to anticipate knowing the speed of a pressure pulse in the fluid in question. Look for VFDs that have provision for a forbidden or do not run parameter. The VFD will hold just below the forbidden speed until the control parameter increases enough that the speed setting will jump to a speed just above the forbidden speed.
Jeff (jraef) can add a lot more here.
As for the generator size;
If the 10 kW motor sizing is power in rather than mechanical power out, and the starts are staggered you should be fine.
The issue with generator sizing has to do with the interaction between the Automatic Voltage Regulator on the gen-set and the non-linear current demands of the VFD front ends.
I don't know but Jeff may be able to add comments here; should the loading of several motors result in similar wave forms adding there may be stability issues with the AVR.
I would add a disclaimer to the contract mentioning possible stability issues and holding the designers responsible for the overall performance of the system.
Please share with us, as much as you are able, the results of this project.
Hi Lionel. Your concern is my concern also. Not capacity or overloading but possible interaction with the AVR and the resulting instability.

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

 

[zdas04]

Thank you. Maybe this will help. It is a downhole application with probably 80 stages of centrifugal pump. I will introduce this "forbidden frequency" into the discussion and then duck. My client REALLY wants to use centrifugal pumps (there is a manufacturer of this kind of pump in South Africa and he's trying to minimize his imports as much as possible). I was concerned about the applicability of them even before I heard of this twist.

David

 

[waross]

It has to do with the pressure variations as the impellor vanes pass the discharge outlet.
For downhole, I am thinking that the pressure pulse from one stage (the last) won't be as severe as from a single stage pump.
Keep us posted.

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

 

[williamjohn]

We had a 1200 hp motor, huge inertia and 17 second run up time at 4.5 full load current. No problems starting on two generators each 2 MW, 2.5 MVA (total 5 MVA) supplying the power. I tried splitting the board and starting on one generator but the generator tripped on overload after 15 seconds by which time the motor was almost up to speed. Pretty sure that, if necessary, the motor could have been started on one generator with the overload setting increased. The above comments about power are relevant. The stating current is low power factor except for the last one or two seconds and the generator inertia should see it through this.

 

[itsmoked]

Look for VFDs that have provision for a forbidden or do not run parameter.

I don't think I've seen a VFD that doesn't offer this lately.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[itsmoked]

Oh, and to continue with the resonances, VFD users may need to avoid all sorts. Fan blade/housings/ducts, machine resonance frequencies, and even regions where a system just doesn't perform well.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[waross]

Thanks for the update Keith. I haven't been around VFD apps for a few years.
Zdas04; Another issue with down hole applications is long leads and the need for sine wave filters and/or inverter rated motors.

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

 

[zdas04]

The science of very long downhole cables is pretty mature. The industry has had endless problems associated with cables over the last 50-60 years, but the number of cable-related pump issues seems to have fallen off dramatically over the last decade or so. Pump manufacturers look like they've solved the worst of the recurring issues.

David

 

[waross]

Inverters introduce a whole new set of issues and problems. Non VFD pump cables do not have to withstand high voltage high frequency pulses. The voltage may increase with cable length.

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

 

[KllrWolf]

We see it all the time on submersible motors. VFD's and long cable lengths are not the best of buddies. It is not that the cables get damaged, but more of what happens to the motors. If a VFD is to be used on a submersible or any motor with long leads, an output is filter is definitly required. The filter used will depend on many factors, but a true sine wave filter is the best option, with a dV/dt an ok second.