Is it Safe to Operate a VFD Motor Above Nominal Speed?

[russ_watters]

As many of you know, I'm an HVAC (mechanical) engineer. I do a lot of fan, motor and variable frequency drive specification. As you go from fan to motor to drive, the device goes from purely mechanical, to purely electrical, and my level of knowledge/understanding drops off...

Many building owners, electricians and maintentance guys are still skittish about VFDs, for issues such as minimum and maximum drive frequency. I'm less so, but there are a lot of minutae about what they do and what side effects they have that I am not clear on. Most vendors of both say they are safe to operate from close to zero up to 120hz. But there is a catch: They have to be compatible with the fan at the right speed and power output.

If the fan is belt drive, life is relatively easy because you can specify sheaves (pulleys) to make the motor run near nominal speed to get the fan to run near nominal speed. But what about direct drive?

I have a direct drive fan specified to operate at 2350 rpm and 11.1 horsepower. I've selected a 15hp motor, but at what rpm; 1800 or 3600?

Ordinarily, 11.1/15=0.74 is a nice comfortable 26% safety factor. But if I select the 3600 rpm motor it will run at 39hz. VFD output voltage is a linear function of frequency (up to nominal), so instead of 480V it will run at 312V, which also means - assuming consistent power factor and efficiency - the motor is now only capable of 9.8 horsepower. Oops.

So I've selected an 1800 rpm motor, which will then run at 78hz. My understanding of the standards (is thin) is that as long as the motor is rated properly for VFD operation it will be fine, but I'm not really 100% sure of that. A few potential issues:

1. Running at higher rpm should mean more friction and shorten motor life, right?
2. Electrical signal issues; harmonics, spikes, etc?

I'm not one of those who equates operation of a motor at above 60hz as "overspeeding", but at the same time, I'm not totally clear on what issues may occur with operation above nominal speed. Anyone able to help clarify?

 

[anorlunda]

http://www.csemag.com/single-articl...nd-vfds/bb818a2058bbc1a611cf663490f8c4f5.html

Says

. VFDs can be used not only to slow down a motor, but also to drive it beyond its rated speed (overspeed) by increasing the VFD output frequency above 60 Hz. Some HVAC equipment manufacturers overspeed a motor to allow for fan and motor combinations that really shouldn't have been allowed in the first place. While I have seen a letter from a motor manufacturer that grudgingly stated that it was probably OK to a certain extent with a particular piece of equipment, recognize that consistent overspeeding of a motor will reduce its useful service life. As the design engineer, you will need to evaluate the load profile with the amount of time expected at the high end of the speed range and determine if there is a cost benefit that offsets the potential reduced reliability. Also note that available motor torque can fall off quickly beyond 60 Hz because you cannot maintain a consistent V/Hz ratio. This is because your drive output voltage cannot exceed your fixed input voltage.

Engineers always give and get the same answer, "It depends."

 

[Averagesupernova]

I would think modern VFD and VFD compatible motors should handle harmonics fairly well as they have evolved together. It is intuitive and of course correct to assume that bearing wear and etc. will be expedited with an increase in speed. That is not to say that a VFD compatible motor with its horsepower rating at 1800 RPM does not have bearings that will hold up well at 3600 RPM. In fact that may be one of the requirements to rate a motor for VFD. I would want to know the horsepower required to move the fan at 1800 RPM. From a bit of quick googling that I did I would say that the horsepower is linear from the minimum to the rated speed (1800 in your case). After that the horsepower is constant. In other words, increase the speed a bit past rated speed and you lose torque. .
-
https://www.google.com/search?q=VFD...hUDQCYKHZCGBv4Q_AUICCgD#imgrc=TQvO9OA4q1fJkM:

 

[jim hardy]

Wow what an interesting concept. I never encountered VFD's in the power plant , but now they're in washing machines and airconditioners.

Here's an interesting paper by Siemens that gives a broad-brush introduction to the concepts, probably a good place to start building vocabulary.
http://www.industry.usa.siemens.com...fication/documents/nema-application-guide.pdf

Interestingly it even mentions "bearing currents" .
Before VFD's they only bothered machines bigger than about 400hp. They pit ball bearing races with regular rings that look like a raccoon tail. They're mysterious because if you don't take the machine apart at first signs of bearing noise instead let it run , the bearing fails catastrophically destroying the evidence and the bearing manufacturer gets blamed..
Sleeve bearings they pit randomly and the lubricant gets blamed..

Now i'll lay awake trying to figure out speed-torque curve at elevated frequency.
At normal frequency speed-torque curve is proportional to square of voltage. At elevated frequency, same %slip is more hz so rotor reactance is higher ... wow.
Hysteresis that heats the iron is in proportion to frequency but so is cooling airflow from the motor's fan,,,

Thanks i think !

old jim

 

[jim hardy]

here's a more practical article
https://deltaautomation.wordpress.c...ds-than-their-nameplate-speed-can-it-be-done/
i've only skimmed it
looks like they assume you don't raise voltage as you overspeed so use the torque = volts/hz squared relation.
That looks to me so straightforward it's just too good to be true.

first approximation 18hp X (60hz/78hz)^2 becomes 10.6 hp

incorporating both torque and rpm in approximation
18hp at 1800 rpm = 52,5 ft lbs
rpm in proportion to frequency
1800 rpm X 78/60 = 2340 rpm
torque in proportion to (volts/hz)^2, same voltage at 60 hz and 78 hz ?
52.5 ft lbs X (60hz/78hz)^2 = 31.07 ft lbs at 2340 rpm = 13.8 hp

it's operating at lower flux so is capable of less torque. Check current to make sure you are still within rated running amps, not into starting amps region of slip..
Stator iron should be quite happy at lower flux.
i don't know about rotor amps , I'll have to think on it a few days.
I'd wager @Babadag knows the equation for rotor current by heart .
old jim

 

[anorlunda]

@jim hardy , check out that link in #2, it has lots of interesting stuff on the pitfalls of VFD.

Bearing currents? Funny that you should mention that because I just saw this video recently. If the video's thermal explanation is correct, no wonder that lifetime is reduced.

 

[jim hardy]

Thanks anorlunda, i like Yoon's concise style . He covers a lot of ground in that one page.

Did you check that electromagnetic explanation for Latheman's ball bearing motor ?
http://www.physics.princeton.edu/~mcdonald/examples/motor.pdf

Bearing currents (called shaft currents in my day) were attributed to imperfect, slightly asymmetric flux distribution in the machine resulting in some flux linking the shaft causing current to circulate longways through the rotor and frame via the bearings. I noticed it on the 400 kva generators that fed our control rod drive system.

i think high frequency components in applied voltage aggravate it. My machines fed a thyristor rectifier so had high frequency components in load current.

Insulating one of the bearings is the classic fix.
Some plants installed ceramic bearings that are nonconductive. Some plants reported no trouble. So i believe there's a bit of old fashioned luck involved.

old jim

 

[anorlunda]

Did you check that electromagnetic explanation for Latheman's ball bearing motor ?
http://www.physics.princeton.edu/~mcdonald/examples/motor.pdf

Holy mackerel