Running fan motors at over 50Hz?

[gremlin_95]

Hello all,

Bit of a strange one this as I'm probably missing something extremely obvious. Unfortunately, having been trained to be mechanically biased, I still have quite a bit to learn about electronics etc.

At work, we are currently procuring some large air handling units which are to be equipped with 2 equally sized Plug Fans working together to provide over 20m3/s of air. The fans do not run at a fixed speed and will have frequency inverters to change the speed as and when required (a minimum speed of 30% is required).

We've had a few proposals and most consist of 2no. 18.5kW motors for each fan, however, the frequency for the motor driving the fan at its 'duty point' is in the region of 66Hz. Now, our consultant has rejected these proposals by simply stating 'frequency too high, maximum is

50Hz'...and no supporting information.

The air handling unit manufacturers only build the units in their casing with all the other bits such as heaters etc. and having enquired with them, their response is that the motors work best between 40Hz and 70Hz AND that their current proposal offers the best specific fan power (watts/litres second of air). They also state that larger fans will stall if you try to slow them down to 30% which seems logical to me. As seems to be the case with all of these projects, there is very limited space for bigger fans.

Can anyone shed any light on why running at a high frequency is such a problem? I was under the impression that it isn't a problem - as long as the full load current rating of the motor isn't exceeded?

[Bob Minchin]

Three phase induction motors run from appropriate inverters (ie those with sensorless flux control) will provide excellent torque characteristics down to 5-10 Hz and I regularly run such motors up to

100Hz ie twice plated speed. Quite often motor cooling can be too low at for long duration running at low speed but given you are running fans the air being drawn through the fans should flow over the motor and so this should not be an issue.

[Syd Rumpo]

Your 'consultant' is a no-nothing cowboy, lose him. HTH.

Cheers

[John Rumm]

I don't claim any particular expertise in this area, but I was generally of the understanding that a higher frequency will increase the speed, increase the cooling, possibly lower the power factor, and will not result in a current draw increase. Basically it ought not be a problem - but you may not get the full rather power output from it.

On something like a fan you may find the 20% shift in speed will increase the loading quite a bit.

Going for lower frequencies can be more of an issue as you can push the current draw over the max full load amperage, while the cooling will fall (although less of an issue with a motor driving a fan if it itself is in the air stream).

[Fredxxx]

I guess your consultant is worried about running motors above their design speed. I think I would ask him why he has rejected these proposals!

One thing that worries me. The low speed torque characteristics of motors when run at a corresponding frequency are remarkably good with the right choice/match of inverter. The load from running a fan at a low speed will of course be much lower than full speed. Therefore, unless your consultant can explain himself, I would get a second opinion as I can't think of a valid reason for suggesting a motor would stall at low speed (assuming it is run at low frequency).

He could be suggesting the fan bladed might be in stall at low speeds, but I really can't see that happening. In fact efficiency is more likely to drop off at high speeds.

BICBW

[Thomas Prufer]

Either the consultant has forgotten the frequency inverter, and is thinking that this will run straight off mains. (This would speak against his/her qualification, or thoroughness...)

OTOH, there may be some obscure reason for not running the motors at over 50 Hz, like "EMC". (I would be surprised if this were the case.)

This surely is the problem and responsibility of the air handling unit manufacturers. They *would* notice if they ran their motors hot, or over-current, these fans run for long times...

I'd ask the consultant to clarify.

(If the answer is "mains in the UK is 50 Hz, not 60", don't pay for the consultation -- because a random collection of people in a newsgroup can do better for free...)

Thomas Prufer

[Brian-Gaff]

Also depends on the waveform of the frequency. I can well recall many years ago, a wizzard plan to run induction motor control using a frequency adjustment on a fairly high power inverrtor either cooked the transistors if in sine wave mode, or the motor if in what the manufacturer called raw mode, presumably a rounded of square wave. Nothing as high powered as yours though I have to say.

Brian

[harry]

All induction/PM motor speed is frequency related. The problem is when you run them at above 50Hz they might easily overload when driving a fan. To double the speed of a fan takes four times the power. So even a small "overspeed" may overload the motor. The motors are likely sized to run at 50Hz max. Only a small overspeed might well cause a significant overload.

Added to this, many electronic speed controls don't produce a perfect sine wave. This means a motor has to be derated as there are increased magnetic losses leading to possible overheating with non-sine waves.

[Krankie]

Granted it is many years since I earned my crust in the fan industry. In my day fan power varied as the cube of fan speed, i.e. doubling the fan speed multiplied fan power eight fold.

[newshound]

You also risk the flow becoming more turbulent, getting resonances in the blades or other parts of the structure, etc.

I would certainly begin to worry about cooling, even in an airstream, at much lower speeds than normal. Not too difficult to monitor with an IR thermometer, though.

[gremlin_95]

I have requested some further information from him, I will most certainly share his response!

As it happens, on the site I was on today, there were a few fans running via a frequency inverter at 60Hz without any issues, these particular fans were serving a kitchen extract hood and were 'bifurcated' so the motors were not in the air stream.

I have also asked a commissioning engineer for his opinion since they are the ones who set up the inverters etc. for these sort of jobs.

[gremlin_95]

Very true, these people are employed directly by the client to oversee the project. However, I can think of many cases were they do more harm than good.

Indeed, we are seeking a more substantial response from them.

We shall see :)

[Johnny B Good]

As others have pointed out, there might be a cooling issue at very low speeds but in this case, the torque demanded by the fan will drop considerably compared to the max torque at full speed (which should be comfortably within the motor's maximum torque rating anyway - current demand correlates very closely to torque loading).

Presumably the variable frequency inverter is a purpose made speed controller in which case the output voltage will also vary in direct proportion to the frequency. Provided that the torque stays below the motor's maximum limit when run at the highest speed setting (highest frequency and applied voltage from the speed controller) there shouldn't be any problem assuming you don't take the upper frequency/voltage to an extreme, risking insulation failure of the motor windings.

I'd get the contractor to confirm that the 'inverter' is actually a speed controller with the characteristics I've just described. If the output voltage varies in direct proportion to the frequency then you're "good to go".

If, when you appraise your 'consultant' of the speed controller, he still rejects the specification with no further justification as to this decision, sack him and get a qualified consultant who *does* understand the basics of frequency/voltage converter speed control.

[The Natural Philosopher]

Why? at lower speeds the implication is the current will be lower, and even with the weird duty cyucles imposed by whatver chopping system is used to create the frequency, heating will be lower.

This seems to me to be analagous to the brushless model aircraft fdan motors, just done bigger.

An inceased in heating for less poewr input implies that efficioency is going down faster than power input is! A very strange idea.

The heating effect on the motior is worst at full throttle, full load, and reduces as the load or RPM reduce. The controllers however tend to get hottest at a little below full throttle, for reasons that are very subtle, to do with inducatnce and actual waveforms.

Neither is the cooling effect of airflow much reduced at low airspeeds. It tends to be logarithmic*. A little airflow does a lot, a lot of airflow does a little more. That is the general experience from fan blown heatsinking data.

*May be square root..

[harry]

I thought it was doubling the volume shifted quadrupled power needed? But if you're right, it makes things even worse.

[dennis]

The one in my fan oven has two fans on the shaft, one hot one in the oven and one outside with the motor to cool it. It is a pretty common way to do it but it does mean less cooling if you slow it down. Its why upmarket power tools have separate cooling fans.

[The Natural Philosopher]

Sometimes you do talk total unadulterated wombat turds, Dennis.

Three lies and a non sequitur and a couple of logical fallacies thrown in for good measure.

[dennis]

You're crazy, care to state where I lied?

[The Natural Philosopher]

"Its why upmarket power tools have separate cooling fans. "

[dennis]

So show me all such ones that don't do that as you have obviously seen every power tool that exists.

What about the other two lies you claimed I made?

You are crazy there is no doubt about that.