Compressor On Vfd

[martin]

Hi Marke,

I have a compressor system made in USA. The cooling fan motor is DOL and is tripping the overload. It is rated at 60Hz but running at 50Hz. I know that cooling will be affected, but the current will increase also?...by what percentage?

The compressor is also a 60Hz motor...what will the current increase by in terms of percentage?

The drive is 50/60Hz, 380-480Vac....no problem.....the site was measured at 500Vac at the input.

The drive couldn't start, it is 75kW and very, very marginally rated to the compressor motor at 100HP.

After 5 minutes of trying to get the compressor running, there was a loud sound from the VFD; I suspect the IGBT's.

To make matters worse, this particular VFD has a program for inputting the rating of the compressor motor; and what they inputted was nonsense.

There was no autotuning carried out as they are in speed control with no feeback devices such as Tacho etc...another problem

Another problem is that they told the drive the motor weight was 550Kg when it was 550lb....about half in kg...

Now, if you tell the drive nonsense (or nothing for autotuning) you will get out nonsense!

They also selected a brand of motor in the drive setup that was clearly incorrect; hence I think the motor characteristics (resistance, inductance etc..) that was automatically load when it selected the wrong motor made thngs worse when the autotuning was not carried out.

Now I know you can run 60Hz motors at 50Hz and they will get hotter with higher current and less internal cooling and the lifetime will be affected.

Please comment on the above if possible with your thoughts (and Raef). I know it is a mess but your input would be valuable.

Regards

Martin

 

Hi Marke,

Martin again; forgot to mention the utility is 50Hz region, but compressor was made by a 60Hz region; not sure that was clear!

Regards

Martin

[marke]

Hello Martin

 

If you operate a motor at 50 Hz and it is designed for 60Hz, there are a number of potential issues.

1. The motor will run slower. i.e. at 5/6 speed.
2. The motor output power will be reduced. i.e. at 5/6 power.
3. The voltage applied to the motor must be changed so that the flux in the iron is correct. i.e. 5/6 voltage If you do not keep the V/Hz ratio constant, then you will either have too much flux or too little flux. To little flux will reduce the torque capacity of the motor. Too much flux will cause the motor to draw too much current and result in excess iron loss and motor failure.
4. If the motor is a single phase motor, then the capacitors may need to be altered.
5. If the motor is a single phase motor, then there may be a problem with the centrifical switch not operating. - designed to switch out the start winding at close to full speed.

[martin]

Hi Marke,

the site gave me more information. It is actually a 120HP motor that they are using at 100HP. They have autotuned with the correct procedure with 120HP selected as the choice so the motor winding info should be correct. The settings on the drive appear correct and make sense when looking at the circuit drawing.

When the drive failed after 5 minutes running, there was a sound from the drive followed by an E on the display, meaning that the processor was in its booting phase....if it was stuck in E for a long time it would appear that they loast the firmware and hence the parameter set. I can confirm however, that this happened after the sound.

The sound could be a number of things...namely the the AFE or inverter sections....or the IGD (semiconductor on the DC bus)

As they tell me the caps are not missappen or leaking and none of the indicators have "popped" I should think these are OK.

This leaves the motor...I will megger the cabling and do an measurement of the windings with a milliohmmeter....look for signs of visible damage to try to find the cause. I was thinking of cooling and the heat factor but at this point it is mute until I see the situation on Monday.

Regards

Martin

[marke]

Hello Martin

 

If the drive failed after 5 minutes, I would not expect that to be due to a cooling problem. The thermal inertia of the system would allow operation for longer than 5 minutes.

I would also be surprised if the firmware was lost, but there could easily be problems in the control card after an explosion in the drive.

 

If is quite possible that the drive failure was just a premature failure and nothing to do with settings or installation. The probability of a failure occuring are highest at turn on and reduce as the run hours are increased, and then towards the end of the life of the equipment, they increase again. We call this the bath tub effect.

A graph of the probability of failure against time looks like a bath tub.

 

Best regards,

[martin]

Hi Marke,

Some updates!

Been to site. Its a 460Vac system and the transformer secondary to the drive was measured at 512Vac no load. This voltage is actaully still in spec....a bit high but we can use. Unfortunately the secondary on the transformer is star connected and they didn't connect the star point to ground. I think therefore the input phase to phase can go as high as X 2 right..??...I looked at the power card and sure enough there was a flashover at the input of the control electronics.

The power components of IGBT's, thyristors etc...were OK....he DC link capacitors were fine....it would appear the control input voltage flashed over therefore killing the brain of the drive and therefore stopping the drive from working.

The customer also said he saw sparking on the motor before this failure happened. The motor has been meggered and milliohmmeter tested and looks fine.

The star point has been ground....and the voltages look far more stable.

We will buy some new cards and try try again!!

Thanks for your help on this and Happy New Year of the Rat!

Martin

[marke]

Hello Martin

 

If the secondary of the transformer is floating, then any high voltage transients on the primary can be capacitively coupled to the secondary, resulting in high voltage transients on the secondary relative to earth. This can damage the drive inputs and outputs.

 

Best regards,