VFD Whine

Not me the motor. I bought a used Millport mill. I never heard the spindle run because the vari drive had a problem. I rebuilt the vari drive and hooked up an AC Tech SM230S to drive the motor. While rotating the motor I get a high frequency whine not very loud but detectable. I was wondering if this is something you get with a VFD or is it a bearing in the motor complaining?

Thanks,
neilca

neilca said:

...hooked up an AC Tech SM230S to drive the motor. While rotating the motor I get a high frequency whine not very loud but detectable.

Click to expand...

Hi Neil,

That's coming from the VFD, and it's normal. If you change the operating frequency of the drive, the pitch of the whine will change. (Not really advisable, but a way to test the analysis.)

What you're hearing is the electrical signals inside the VFD interacting with the mechanical assembly and creating the same effect as a speaker.

I have two AC Tech drives and they both do the same thing. The same phenomenon has been reported for other drives.

- Leigh

What you are hearing is the whine from the carrier which is normal and by design.

Feel free to raise the carrier frequency so that the whine is no longer objectionable.

I have not heard that raising the carrier frequency is inadvisable but on the contrary is advisable to lower objectionable noise. Looking at it another way, if it was "bad" for some reason, then why did the manufacturer make it adjustable?

The reason it is set so low is mostly due to European EMI standards. Europe has radio stations that operate at VFD carrier frequencies so reception of those radio stations would be a challenge near an operating VFD. As a result it is illegal to raise the carrier in EU.

Leigh,

Since you have these drives maybe you can help me with the operation. I have been able to run the motor by turning on the VFD and adjusting the frequency manually on the VFD with the 1 an 11 terminals connected. According to the manual I should be able to reverse by connecting 11 and 13A, but when I do this nothing happens. I would eventually like to connect 1 and 11 via a drum switch and have the drive run at 60Hz then switch the 1 and 13A to reverse.

I was able to input the password and get to the programming section but really wasn't sure of what values to change. I did change the carrier frequency to one step higher (8 kHz) because the manual said it was ok. Two steps higher (10 kHz) and you have to derate the drive.

I have chosen not to use a potentiameter to adjust speed since my mill can do the speed adjustments.

Thanks,
neilca

You've allready discovered the down side.

The VFDs will run hotter with a higher carrier frequency, which is why they get de-rated. There's actually sort of a curve where the carrier frequency makes the drive run MUCH hotter after a certain frequency. That's why they say to de-rate after 10kc. Below 10kc it doesn't make a big difference in operating temp, so it's allowed to be adjusted however. If it gets to be a hot day, you may have to lower the carrier frequency and live with the whine if it gives you an overtemp error.

Pierce Butler said:

I have not heard that raising the carrier frequency is inadvisable but on the contrary is advisable to lower objectionable noise. Looking at it another way, if it was "bad" for some reason, then why did the manufacturer make it adjustable?

Click to expand...

Increasing the operating frequency of any AC power-generating device will increase the internal losses and raise component temperatures. This is just a fact of life. The frequency is adjustable so you can avoid resonances which may develop in a given installation.

Pierce Butler said:

Europe has radio stations that operate at VFD carrier frequencies so reception of those radio stations would be a challenge near an operating VFD.

Click to expand...

Radio stations operating at 10 kiloHertz? The only things in that range that I know of are WWVL, which transmitted an unmodulated standard frequency on 20 kHz (discontinued July 1972), and some submarine data systems used by the military.

The lowest frequency listed on the Wikipedia VLF page is 11.905 kHz, for a Russian navigation system.

The modulation bandwidth available at carrier frequencies below 100 kHz is inadequate to transmit amplitude-modulated (A3) voice in any intelligible form. VFDs don't operate in that frequency range. Radio stations operate at 500 kHz and above, although there may be a few in the 400 kHz range. I know of no voice broadcasts below 400 kHz.

On the AC Tech SCF-series drives, the factory default is 6 kHz, and the highest available frequency is 10 kHz. For their SMV series, the default is 4 kHz for the NEMA 4X drives and 6 kHz for the NEMA 1 drives. The maximum on both types is 10 kHz.

The difference is that the NEMA 1 drives are open and have better cooling than the 4X drives. This is a direct example of the fact that the heat generated increases with increasing frequency.

- Leigh

Yep, my VFD on the drill press had a moderately annoying whine. Reading
the manual, I found it was on "medium" frequency. Decided to try the "low"
setting. Arrgh. Whine got five times worse.

So I set it to the "high" setting. Whine got a factor of two or three better than
the med. setting.

For those of you discussing PWM schemes, remember that a sharp rising square
wavefront has fourier components about ten times the the inverse of the
risetime. You can't just say, the carrier is at 10 kHz so that's that.

If have a microsecond risetime on a 10 Khz PWM signal, it'll have a considerable
amount of energy up to 10 MHz. Because they probably don't alter the
risetimes on those VFDs when the mod frequency is changed, I find it hard
to believe that anyone is restricting the carrier frequency based on RFI issues.

It may be that the EU restricts the risetime first, and once you do that, it
places an upper limit on the carrier frequency.

Jim

Actually, there may be radio signals on 10kc!

From what I understand, submarines use a very low frequency to communicate while they're submerged, pretty sure they still use it. I dunno what the frequency is, but it may well be 10kc. Actually, most of the human voice range is under 500hz. It doesn't take a real high carrier frequency to transmit an intelligible human voice. Granted, the higher the carrier, the better it sounds, but it CAN be done with 10kc.

neilca said:

Since you have these drives maybe you can help me with the operation.

Click to expand...

Hi Neil,

I don't have the SM-series drives, nor a manual for same. I have the SCF- and SMV-series drives. There's a definite "family resemblance" to the programming of these two, and I think it will apply equally to yours.

neilca said:

According to the manual I should be able to reverse by connecting 11 and 13A, but when I do this nothing happens. I would eventually like to connect 1 and 11 via a drum switch and have the drive run at 60Hz then switch the 1 and 13A to reverse.

Click to expand...

There should be two parameters which control this function.

1) Terminal 13A can be used for several things. The desired function must be programmed specifically.

2) Motor reverse must be enabled. The default is disabled, so the motor won't run in reverse no matter what you do, until you enable the function. That's another programming parameter, distinct from the terminal 13A parameter.

Review the wiring diagrams (around the middle of the manual probably). Decide which of these best matches your installation and duplicate that wiring exactly. There should be notes regarding the correct programming for each diagram.

- Leigh

Human voice inteligebility needs to have 200 to 3000 kc for good rendition. That was
the old western electric standard for phone work.

ELF submarine communcatinos happen at much lower frequencies than 10 khz.

Jim

As Leigh mentions, there is some programming required to activate reverse:

P112 (Rotation) must be set to 1 (Forward and Reverse)

P121 (TB-13A Function) must be set to either 10 (Reverse Rotation), 12 (Start Reverse) or 14 (Run Reverse) depending on which type of Start/Stop control scheme you are using.

Look at the note that follows the description of parameters 121-123, as there are example wiring diagrams for the different control schemes.

John

Re: Whiney drives

First off, the whine is in both the drive and the motor. The motor whine is usually a lot worse, but there is usually more going on and it's harder to notice.

The reason why the carrier frequencies are adjustable is... because they can be.

When PWM VFDs first came out, the carrier frequency was fixed by the manufacturer. Every time the transistor is switched on, there are what are called "switching losses"; heat rejected into the heat sink. The higher the switching rate or "carrier frequency", the more heat losses there are in the devices. So early drives, those based on Darlington transistors, used relatively slow carrier frequencies because Darlingtons can get hot really fast. On top of that, there is a "natural" resonance that works well for motors at about 1 - 1.2kHz, during which the switching losses in the motor are minimal. Any higher than that and you start getting the molecules in the motor windings a little more excited, meaning you are increasing the heat in the motor for a given load. Mind you, having a VFD usually means you are running it at less than full load, so that isn't always a problem, but in the business of using VFDs simply as phase converters, it can be. So with higher carrier frequencies, the drive AND the motor heats up. Therefore, the lower the cf the better in the long run.

But...
Carrier frequencies below 10kHz create that whine in the motor which is annoyingly audible to younger people (I no longer hear them above about 5kHz). Most of the time, a motor or the machine it is on was making more noise so nobody cared. But when VFDs started getting applied to HVAC motors, especially in-line vane axial fans where the motor was inside of the air duct, the whine was getting transmitted into nice quiet office spaces. people started complaining and claiming headaches etc. So in the early 80s, Toshiba came up with a high carrier frequency "Quiet Drive" technology and started kicking ass in the HVAC market. People didn't realize at first that the higher cf was damaging the motors faster, but once they did, it became another problem on top of the solution to the previous problem (as most of these things go). So competitors started making the cf adjustable so that IF you wanted to use the higher frequencies, you could, but they now explain the ramifications to you. By the way in case you didn't catch that subtle issue, the whine is ALWAYS there, its' just that above 10kHz, most humans can no longer hear it. Your dog probably can though...

So bottom line: if you are not fully loading your motor AND your drive is rated for it, no problem increasing the cf to get the whine out of your hearing range. If you are running your motor close to the edge, expect it to die an earlier death with a higher cf. But keep in mind that the term "earlier" might mean 10 years at high duty cycles instead of 25 years, so it you are not using it all the time, you may never know the difference.

It Works!

I talked to the tech guru at AC Tech and he walked me through programming the VFD. Real nice guy who actually knew his business (unlike the Hewland Packard rep I get on the phone). I did exactly what he said and IT WORKED. Technology is great when it actually works.

I played with the carrier frequency parameters (1) 4kHz, (2) 6 kHZ, (3) 8 kHZ and (4) 10 kHz. The whine went away on the 10 kHz setting. I chose to use the 6 kHz setting because the whole machine noise changed when this was used. The audible noise from my mill seemed to be a lower frequency than with the other settings. Kinda weird. I figured from this there is some kind of resonance frequency at play. Bottom line the mill sounded happier than with the other settings.

Hope everyone has a good high tech day,
neilca