Jump to content

VSD harmonics


Alan H

Recommended Posts

Hi people

I am about to install a new site. Currently installed is the utility TX is at 1 MVA to a 900kW genset (burning waste gas). No other site loads (yet).

New on site will be:

1 x 250kW VSD (compressor),

2 x 300 kW VSD (cent pumps),

The existing 1 x 900 kW Genset,

30 kW of other bits, so I have a relatively high VSD contribution to the planned site load.

 

My concerns are:

1) Starting the last motor (assuming genset not on).

2) The harmonics from the VSDs hitting the genset and the utility transformer (5% THD permitted)

 

My questions

1) The 1 MW TX was installed a year ago with the genset (no site load at the time!) Now with 3 good sized VSDs, am I pushing it to stay at 1 MW for the utility TX.

2) Are there any reliable "rule of thumb" methods to confirm the TX needed size (I have done many calcs, but the results vary)

3) Should I be considering harmonic filters and if so, what are the key factors to watch for.

4) Are the hamonics going to upset the genset? (genset not capable of islanding !!)

 

Thanks for any help

Link to comment
Share on other sites

Hello Alan

 

Welcome to the forum.

 

Yes, you will have a high level of harmonics and this could cause issues with the supply loading.

 

I assume that the VSDs are all standard 6 pulse input drives and will therefore have a harmonic current causing at least 30% THD current. The effective load of these drives will be higher than the quoted KW due to the harmonics and I usually rate the supply to the drive at around 130% of the drives ratings. This includes the transformer, so I would tend to be calculating on the VSD load being equivilent to around 1.105MVA which does not give you any room for further load on the transformer and I would expect that you could easlity have issues if you try to run the drives at full load on the genset.

 

You could add some filtering to reduce the harmonics and therefore the transfomer loading, but these filters are very expensive if they do any good. I am not sure how the cost of the filters will compare with increasing the TX rating.

 

VSDs do not have a high current draw, so starting the last motor will not be a problem. The start current on a VSD will be no more than 120% of the rating of the drive, and if you slow down the rate of acceleration, the start current will not exceed the Full Load rating of the motor.

 

If the 5% THD that you refer to is voltage at the point of copmmon coupling, then you would need to know the supply impedance to calculate this.

 

Best regards,

Link to comment
Share on other sites

Dear Alan_H,

for your case, I would recommend an harmonic filter "passive type - non-tuned". There is a leader manufacturer in Canada.

The application of an harmonic filter should seems expensive at first sight, but is very useful to reduce active current asked to the generator, this will reduce the working temperature of windings and at the end you will get some more power from existing generator.

Please note that you have to check the capacitive reactance of this filter, that must be low if you want to use the generator also at low load (that means also low speed for VSD).

 

Regards

Mario

Mario Maggi - Italy - http://www.evlist.it - https://www.axu.it

Link to comment
Share on other sites

 

3) Should I be considering harmonic filters and if so, what are the key factors to watch for.

 

 

Hello,

 

One area where harmonic filtering will be a benefit is on ground-fault relaying. Due to the system capacitance to earth and the nature of harmonics, harmonic currents tend to end up on the ground conductor and can cause nuisance tripping of ground-fault relays.

 

One way to mitigate the effects of harmonic related nuisance tripping is to install a ground-fault relay that employs harmonic filtering. The relay would be installed on the line side of the VSD and the filter would be set to the fundamental frequency.

 

Regards,

Don Selkirk

Startco Engineering Ltd.

Link to comment
Share on other sites

  • 2 weeks later...

Hello Alan

 

Welcome to the forum.

 

Yes, you will have a high level of harmonics and this could cause issues with the supply loading.

 

I assume that the VSDs are all standard 6 pulse input drives and will therefore have a harmonic current causing at least 30% THD current. The effective load of these drives will be higher than the quoted KW due to the harmonics and I usually rate the supply to the drive at around 130% of the drives ratings. This includes the transformer, so I would tend to be calculating on the VSD load being equivilent to around 1.105MVA which does not give you any room for further load on the transformer and I would expect that you could easlity have issues if you try to run the drives at full load on the genset.

 

You could add some filtering to reduce the harmonics and therefore the transfomer loading, but these filters are very expensive if they do any good. I am not sure how the cost of the filters will compare with increasing the TX rating.

 

VSDs do not have a high current draw, so starting the last motor will not be a problem. The start current on a VSD will be no more than 120% of the rating of the drive, and if you slow down the rate of acceleration, the start current will not exceed the Full Load rating of the motor.

 

If the 5% THD that you refer to is voltage at the point of copmmon coupling, then you would need to know the supply impedance to calculate this.

 

Best regards,

 

Link to comment
Share on other sites

Hi Guys

Thankyou for your advice so far.

Mark.

Yes, it is 6 pulse.

I have since concluded (more rule of thumb than calculation) that if the VSD contribution to the load is above say 30%, I probably have a problem. For a $60k drive, the harmonic filters would be about $25 k and about 250 kg !!.

We are now looking seriously at ABB and AB active front end VSDs as well, but somewhat more costly.

Also looked at 12 pulse, which kills the 5th and 7th, but then the 11th & 13th then exceed what the lines company will take.

 

Mario.

Yes, the generator vendor has now indicated that heating may be an issue, thus passive filters are a definite possibility. Also should expect some additional heating in other 3 phase motors on site.

We really want to export as much power as we can, as the gas is free, so extra heating is not welcome

 

Just of interest, I think I had a similar problem last year at a remote site.

Natural gas recip compressor with electric 110kW motor, soft starter, fed from a 350 kW hired genset.

I had the most unbelievable problems trying to start the compressor, because the genset kept tripping with a voltage or excitation problem even though the current was well below the 350kW limit.

I had to get a monster 500 kW set in before startup became reliable.

My suspicion is that the soft starter caused harminics on startup which confused the AVR. In fact I suspect the genset was big enough, but the AVR was just intolerent of harmonics.

 

Cheers

Link to comment
Share on other sites

Dear Alan_H

 

My suspicion is that the soft starter caused harmonics on startup which confused the AVR. In fact I suspect the genset was big enough, but the AVR was just intolerent of harmonics.

 

I agree with you. Generally speaking, some manufactures simply don't take into consideration the possible presence of harmonics.

 

Regards

Mario

 

Mario Maggi - Italy - http://www.evlist.it - https://www.axu.it

Link to comment
Share on other sites

Hi Alan

 

Your experience with the generator is not uncommon and can be due to a number of things.

 

1. AVR. The AVR (Automatic Voltage Regulator) comes in a number of forms, but the most common is a single phase half wave monitoring regulator. The problems with this form of AVR stem from the fact that they are peak reading and effectively sample only once per cycle. The single sample, peak reading architecture results in sensitivity to waveform distortions (harmonics) and a slow response time. Transient response tends to be poor and subject to overshoot and instability. This AVR is good for static loads and will track small variations without problems, but sudden variations in load can result in voltage variations.

At the other end of the scale there is the three phase averaging AVR which sample six times faster and is not as sensitive to waveform distortions and is very resposive to transient load changes.

 

The voltage control on the generator system is a closed loop system and the loop stability is dependent on the loop characteristics.

 

2. Soft starter. The soft starter reduces the voltage applied to the motor by phase controlled Thyristors. This results in current harmonics which will cause voltage distortion due to the finite altenator impedance. The single phase half wave AVR can be affected by the hamonics.

The soft starter can either be an open loop soft starter with a ramping voltage only, or closed loop where there is a feedback system controlling the current or torque.

If the starter employs an internal feed back and the generator loop is slow responding, the two feedback systems can begin to chase each other causing "system" instability. This can result in current surging and voltage swing and the generator will often shut down on overcurrent or over voltage. Some soft starters are more prone to this type of interaction than others. I experienced this with one particular brand of soft starter, but when we set the start voltage above 50%, the stability was much better. At 30% start voltage, the instability was bad and the genset shut down. At 50% start voltage, the system was sufficiently stable to start the motor without the generator shutting down. Another brand on the same generator works very well.

 

3. Excitation. The standard economic genset uses self excitation where the excitation energy is taken from the output voltage. When the generator becomes overloaded, or the voltage sags due to a slow AVR, the reduciong voltage causes a reducing excitation which further reduces the voltage etc - Result, the voltage collapses for small overloads or transients. The alternaive excitation can be either external, or derived from a permanent magnet generator. This will allow a much higher overload with out the voltage collapsing. PMG excitation is recommended for applications requiring a transient overload performance.

 

So, at the end of the day, the system is a little more complex than just applying a motor and soft starter to a generator set. A properly engineered system will yield good results, a poorly engineered system will result in major performance issues.

Your problems could have been due to a half wave peak reading AVR, or weak self excitation, or the feedback system with in the soft starter interacting with the genereator feedback loop, or a combination of all of the above.

As a matter of interest, what was the brand of the soft starter and how was it set up.

 

Best regards,

Link to comment
Share on other sites

Hi Mark

Many thanks for your rather comprehansive reply.

 

This is one of those "Oh no. You've done WHAT ?" jobs.

The Soft Starter was free issued with the compressor for me to install on a site (55 kW mains supply).

The 350 kW genset was therefore got in from one of the local hire shops.

 

Motor = Toshiba 110kW Type CD280, 4 pole Ex'd', FLC = 193 amps

Soft Start = Toshiba TMS7, 250% start current (else compressor won't start).

Other load = 55 kW motor on a hydraulic power pack, DOL. This is used to run a Weatherford well lift pump (shaft all the way down to 1700 metres below ground, vertical recip, about 2 metres motion).

 

The 350 kW genset was then upped to 500 kW, and this allowed the compressor to be started reliably.

The diesel costs are out of this world, by the way.

 

Cheers

Link to comment
Share on other sites

Hello Alan

 

That soft starter will not be an issue with gensets, however there can be an issue if the control voltage drops too low. The control voltage is rated at + 10% -15%. If the control voltage drops below this, you can have issues with SCRs not being triggered on reliably and this can cause over currents and trips.

 

As you were well inside the generator ratings, I would suspect that it had a half wave peak reading AVR that went unstable.

 

Have a good day,

Best regards,

Link to comment
Share on other sites

Hello Alan,

 

Although the manufacturer of the TMS7 (not Toshiba) specifies a control voltage range of +10% / -15%, the products have in fact proven reasonably stable outside of this range. Our own studies suggest one should expect to experience the problematic behaviour outlined by Marke when the control supply drops to 25 - 30% below nominal.

 

It may also benefit you to know that there has been a recent change to the instantaneous trip point for the lower threshold of the out of frequency range range protection on this starter. It has been reduced from 45Hz to 36Hz thereby assisting to provide more reliable operation in generator supplied installations.

 

If you are experiencing under-frequency trips at any of your sites, you may wish to contact Toshiba and request a control module with the latest firware. I suspect your Toshiba contacts may not be aware of the recent changes to the product, and therefore suggest you may need to 'encourage them' to consult the original manufacturer.

 

Regards,

GGOSS

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...