Overspeed with a VFD 4

[MKMason]

Hopefully this is an easy question. Which motor is capable of going overspeed, induction or synchronous? How? What would be the maximum increase allowable (%)? Does service factor figure into the equation, ie, does a motor (power) SF of 1.15 limit the speed also to 1.15 x synchronous speed, or some other factor limit?

 

[ozmosis]

MKMason
Firstly, it would be useful to know what your application is. Do you know the load (torque) characteristics?
Secondly, why do you want to over speed?
This is a good starting point.

 

[MKMason]

Oh dear, it seems I've over-simplified the question. The driven equipment is a high flowrate slurry pump, ~500 HP, and the client is attempting to vary the developed pump head to match an increasing downstream restriction that there is no other way around. One could overspeed to raise the pressure head and still control the flowrate. If you could see the (confidential) process diagram, you would agree with the design. However, I need to go from 50% of rated head to somewhat over the rated head to overcome the slowly increasing downstream restriction. I need to know what type of motor will get me there, assuming I'll be using a VFD either way.

 

[LionelHutz]

If you're picking the motor then pick a motor with a lower rated voltage then the line power and VFD rated voltage. That would allow you to maintain the V/Hz ratio above line frequency which means you continue to get a constant torque and linear HP beyond the motor rated frequency. For example, a 380V motor on a 600V source allows you to run to 158% of the motor speed and get 158% or the motor nameplate HP.

As for motor type. I would expect most any induction motor to be capable of some over speed, but I wouldn't expect the same from a synchronous motor. Most induction motors seem capable of about 150% over speed but it's always best to check with the manufacturer.

 

[MKMason]

Thank you Lionel. You are obviously not in the USA. Will this idea you have also work for 480V on a 4160V system? It's a much bigger delta-V. I calculated my required overspeed, and it will likely be 125%.

 

[John2025]

I know this idea has been explained in other posts, but I'm still struggling to understand. I get the constant V/Hz and torque plus higher RPM, thereby higher HP, but doesn't the extra power equate to extra heat and wouldn't that cause the motor to overheat? Maybe I'm just a pessimist, but it seems too good to be true.

 

[KllrWolf]

Yes, running a motor over sync speed does create more issues. Many times you could also be running into a critical frequency not too far above the sync speed, in addition to greater heat and other losses.

 

[waross]

At the current of a loaded motor is closely proportional to the torque produced. The greatest heat is I[sup]2[/sup]R losses in the windings.
By maintaining the V/Hz ratio and supplying an over voltage at over frequency we get more HP due to the greater speed and voltage at the same current.
HP is dependent on current and voltage.
My first choice would be a couple of small dry type transformers in open-delta auto-transformer boost configuration.
My table shows a typical current for a 500 Hp motor at 460 Volts as 590 Amps.
If you use a 25% boost you need to add 120 volts. Your transformer size will be 590 Amps x 120 Volts. or 70.8 KVA. A pair of 480:120, 100 KVA transformers will do the job nicely. The transformers must be rated for boost operation.
My second choice, if this is a new installation and a transformer must be purchased, consider a 600 Volt secondary. 600 Volts is a common Canadian voltage and most manufacturers will be able to supply a 600 Volt transformer.
Another third option is to have the motor reconnected in delta. That will let you go to 173% of base speed.
Another option may be to oversize the motor. A 600 HP motor with a 1.15 service factor will be the equivalent of 690 HP.

What is the maximum speed and the HP required at the maximum speed?

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

 

[John2025]

Bill,
Sorry if I appear dense (heck, maybe I am). If you're saying heat is from I2R and the current stays the same, then it follows that there won't be much, if any, extra heat loss even up to say twice the nameplate HP? Why aren't motors just built this way in the first place? Is it just the fundamental limit of number of poles and 50/60 HZ?

 

[macmckim]

MKMason

Based upon your comments:
You have a 500HP, 4kV induction motor driving a slurry pump.

Going to assume motor is a 1780RPM (4pole mtr) to a gearbox and then to the pump.
You want to drive the motor from the 50% to 125%. (885 to 2225 RPM)

Typically slurry pumps (centrifugal type) are considered variable torque applications.
So between 885 to 1780RPM a typical 4P-500HP motor can operate without any issues.

From 1781 to 2225 RPM, the motor become Constant HP unit, meaning HP stays the
same through the overspeed, but torque decreases as the speed increases.
IE: at 1780 RPM motor will produce approx. 1474 lb.-ft. of torque.
At 2225 RPM motor will product approx.. 1179 lb.-ft. of torque.

You will then need to talk to your slurry pump mfg. and find out what is the motor torque
requirement at 2225RPM. Also ask pump mfg. about operating the pump at the higher
speeds. If there is a gearbox in between pump and motor, same question to the gearbox mfg.

This is just the starting point. There will be many other issues to consider.

Good luck
Mac

 

[MKMason]

WOW. I'm sure getting a lot of great advice here. I don't think the client will like the transformer idea, and has already stated that 600V is not available. It needs to be a VFD, they say. I TOTALLY forgot about the critical speed issue. That could be the achiles heel. Thanks for the reminder. It has no gearbox, and will likely run at 1750 rpm. The pump curves are running up from about 450 HP to 600+ HP depending on flowrate. I was hoping to limit the HP to 500, but I can see now that won't be possible. I will need to talk this over with the pump supplier concerning the torque -- constant or not? What am I after besides more pump head and flow? Only the process engineer can tell me.
To help clarify my inquiring mind, just what features of how an induction motor is designed allows it to go overspeed, while a synchronous motor cannot?
Again, thanks to all of you for this phenomenal thread.

 

[jraef]

If we ASSume that the motor size was chosen based on full speed of the motor connected to that pump running at that speed, increasing the speed, on a CENTRIFUGAL pump will increase the POWER required by that pump by the cube of the speed change. Increasing the speed and voltage (as suggested) also increases the output mechanical power of the motor, but not at the same rate. So at 125% speed, that same pump will require 1.25[sup]3[/sup] power, or 1.953X the HP. If you increase the motor terminal voltage appropriately to match the new speed, your 500HP motor is now capable of delivering 625HP at the shaft, but your PUMP will now require 976HP at that speed.

Oops...

"Will work for (the memory of) salami"

 

[jraef]

[sub]Meant to add this... ohhh for an edit function....[/sub]

The affinity laws work in all directions, not just for reduction of speed.

"Will work for (the memory of) salami"

 

[KllrWolf]

An induction motor speed is based on the frequency of the power, thus a 2 pole motor will run at 3600 RPM at no load and 60 Hz source (3000 at 50 Hz). At load the RPM actually drops as the load increases. A synchronous motor is designed to run at a specific speed that stays at that speed even under an increasing load.

 

[waross]

The issue with synchronous motors has to do with the field more than the motor. A synchronous motor may be started on a VFD to reduce starting current because a synchronous motor starts as an induction motor. When the motor is close to synchronous speed the field comes on and the motor locks onto the frequency. Also, synchronous motors are often used for power factor correction. That may become complicated if the frequency is varied very much.
You have noticed that induction motors always run a little below synchronous speed. The speed difference, translated to frequency is called the "Slip frequency". The slip frequency is what induces the currents in the rotor to magnetize the rotor.
For a motor rated at 1760 RPM the slip is 40 RPM. 40 RPM/1800 RPM is a ratio of 1/45. 60 Hz x 1/45 = 1.33 Hz. The slip frequency stays much the same regardless of motor speed.
Another option may be to connect the motor for 230 Volts. Now you can go to 200% speed and 200% HP. You may want to change the bearings out for the bearings that would be used on a 3600 RPM motor.
Take this back to the pump group. For the price of a 500 HP motor you can give them 1000 HP at 200% speed. (You will have to use a 1000 HP VFD)
You can have the impeller trimmed to bring the head at 200% speed into your desired range.
Are there any winders following this thread? Are there any issues rewinding a large 3600 RPM motor for 1800 RPM? That would take care of the speed issues.
The pump load drops off so much with reduced speed that cooling fan efficiency should not be an issue.
Anecdote: We had a submersible sewage sump pump that was grossly overpowered. When it started the pressure forced a gravity connection apart and there was stuff everywhere. The mechanical folks decided to trim the impeller to reduce the head. They calculated the amount to take off the impeller and sent it to the shop. The instructions got mixed and the impeller was trimmed on the face instead of the diameter. It worked until a solid or gas bubble went through and then it would cavitate and stop pumping totally. Glad to see the end of that job and the accusations that there was something wrong electrically when the pump stopped pumping.



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

 

[KllrWolf]

Rewinding a 2 pole to a 4 pole is an issue. For one, the same horsepower requires a larger frame. In the fluid filled motors I deal with, a 75 Hp 2 pole is an 8" motor and a 4 pole is a 10". Second is the torque, a 4 pole creates about double the torque at the same Hp. Depending on the lamination design used in the rotor and stator, the performance would change.

 

[LionelHutz]

No, you can't run a 480V motor on a 4160V drive. You likely can run a 2400V or 3300V motor on a 4160V drive. You have to pick the drive to match the full-load current of the motor.

You really need to define the speed and power needed and then use that to figure out the motor HP and rpm required at 60Hz and the over-voltage ratio required.

For example - 750HP @ 2200rpm, assuming a 1750rpm motor will be used. 2200rpm/1750rpm = 1.26x over speed. 750hp/1.26 = 595hp. 4160V/1.26 = 3300V. So, you need a 600hp, 60Hz, 3300V motor to achieve 750hp @ 2200rpm when using a 4160V drive. Make sure the drive current rating matches the FLA of this 600hp motor.

 

[jraef]

Lol, I feel as though I'm watching a debate on the upholstery in the lifeboats of the Titanic when the fact that there aren't enough boats is immaterial...

Whether or not you successfully get to the higher speed with that motor is probably still going to leave you about 300HP short of what the pump is going to demand at that speed.

"Will work for (the memory of) salami"

 

[MKMason]

Yes, jraef is correct. The pump speed N1/N2 CUBED is = the power ratio (HP1/HP2), I could easily jump from 300 HP to 2000 HP going from 1800 rpm to 3600 rpm.
Sooooo... I will use an 1800 rpm motor with the VFD, run normally at 25% less, cut it down 50% more when I need that lower pump head (pressure) and forget about overspeed. It was a bad idea, sounds like. I'll be content with the torque I get, and the process engineers will just have to get off their ivory pedestals and learn how to use valves.

 

[LionelHutz]

LOL, well I think it's funny someone is so worried about the motor being overloaded when it doesn't even appear that MKMason has the motor yet....

On the flip side, if the flow and head require 976HP then you're going to need a motor capable of 976HP, even if you change the pump so it's giving the flow and head at 1800rpm.