Supply distortion by VFD

[Marke]

Hi
I am involved in an existing installation with a large pump that shares a supply transformer with two consumers houses. The inverter is a significant load and well exceeds the combined loads of the houses. The transformer is fed by a long overhead 11KV line.

When the pump is running, one house has problems with the harmonics generated by the VFD but the other does not. Both houses are fed directly from the transformer terminals with underground cables.
The inverter is fitted with DC Bus chokes and also with a coupled three phase inductor on the input.
The inverter is rated at about 70% of the transformer rating.
One solution to the problem is to install a second transformer to supply the two houses but that is an expensive option. Is there a cheaper and effective option that is guaranteed to work?? Mark Empson

http://www.lmphotonics.com

 

[gjones33]

Hi,

I suggest that since only one house is affected, the houses are fed from different phases and the solution might be to connect both houses to the same phase or the problem house to the third phase.

Regards,
gjones33

 

[cbarn24050]

Hi, are you sure that the problem is harmonic distortion and not RFI?. A filter in the house supply might cure the problem.

 

[kspiroff]

Marke

There are a number of FREE homonics analyzers available on the web. These programs will allow you to "dry lab" a number of scenarios. I have found these situations to be very individualized. You may have to "turn over a lot of rocks" to solve the problem most economically.

The basic solutions are:

Passive Filter - This could be a choke of the THD is not too large. It could also be a tuned filter.

Active Filter - These always work, but are very expensive.

Put in a phase shifting Xfmr - This works if you have an EVEN number of loads.

Change your control scheme. If you have more than one VFD you may be able to start with a single motor. sitch it across the line and then bring up the next motor, and so on.

Replace the VFD with an 18 or maybe a 12-pulse model.


Check to se if your local utility has adopted IEEE-519, and if it has, which release. that will provide more guidance.

I've just gone thru this w/ four 200 HP motors. Please feel free to contact me a kspiroff@stantec.com

Kerry

 

[TheDOG]

What about a Shunt passive filter (commonly know as a tuned harmonic filter)?
The Shunt passive filter is placed across the incoming line and is designed for electrical energy flow at fundamental frequency, but provides a low impedence path for harmonic frequencies. The fundamental frequency energy component flowing into the shunt filter also provides leading VARs which can be used for power factor correction, but may cause network resonances (and other problems).
For this reason, an in-depth power analysis study is required to determine if the SPF is suitable.
It is not as effective as an active harmonic filter, but is a relatively cheap option.

 

[Marke]

Hello TheDOG.
Long over head 11KV lines would lead to considerable supply resonance problems I expect. This is a potential problem with series filters also! Mark Empson

http://www.lmphotonics.com

 

[electricuwe]

I'm wondering how many experts can give their tips withoaut even knowing what type of drive it is.

The most important question in this case is if the drive has a DC-Voltage link or a DC-Current link.

 

[dpc]

My suggestion is to just go ahead and put the houses on a different transformer. It will probably be more cost-effective than any other filtering solution you may come up with.

This can't be guaranteed to solve the problem, but it is probably a good idea regardless.

 

[Marke]

electricuwe, standard DC voltage link drive. Problems experienced are partially RFI but largely harmonic.
Another transformer is an option but it is a very expensive option as it will be required to mount on a different pole, so there is additional cost for a long length of underground cable x 2. We are looking for a more economic solution if possible.
The distortion is equal across all three phases. Mark Empson

http://www.lmphotonics.com

 

[nm1970nm]

The information available to answer this question is not adequate.Please furnish the following data_
1. Length of 11kV line and capacitance to ground.
2. Type of drive , make , no of pulse , current source or voltage source.Forced commutated or line commutated?
3.Description of the system with SLD mentioning various fault level.
4.What is the exact problme ?Which type of other loads are there in two houses ? is the load in problmatic house is defferent than the normal? You have to make several observations in defferent permutation and combination of the three major load.
4.Have you tried to do some experiment in order to solve this problme untill now?If yes,then give the outcome .
5. Have you installed power factor improving capacitor at the common bus?
6.Is there any hormonic present in the supply line itself?
7.When you start the drive with motor do you experiance voltage dip in the system?
We can not reach on any perticular solution untill full analysis is made.You can contact me at guddu1969@hotmail.com

 

[thinker]

I think the most important question has not been asked yet.
What is the front end of your drive system? Is this
PWM or 6 pulse converter? The cure will strongly depend on
the type of front end. If you have an active (PWM) regenerative converter the harmonic spectrum will be multiples of switching frequency and their sidebands. This
frequency range is different from classic 6-pulse harmonics,
and you are dealing with high frequency voltage harmonics
rather than with current harmonics.

 

[Marke]

The inverter is a standard DC voltage link inverter with an uncontrolled rectifier front end. so it is a 6 pulse input.
The consumer is having problems, particularly with his computer, but also with other appliances due to the distortion of the supply voltage. The measured THD on the supply before we added the three phase reactor was 6%. I am waiting for the printout showing the changes since we added the three phase input choke but these are tied up in bureaucracy and will appear eventually. The harmonics without the inverter running are in the order of 0.5 - 1% so the inverter is the major source of problems. When the inverter is not running, all is well.

Because this is at the end of a long overhead line, I am reluctanct to add any capacitance as this will almost surely create line resonance problems (very common in this type of environment.)

I am not yet sure where the major source of the impedance is, either in the transformer or the lines. The power supply authority are not particularly cooperative in this regard. Mark Empson

http://www.lmphotonics.com

 

[nm1970nm]

Normally for correct analysis of the complex situation like this one should clearly present all the facts and figure at one time .I understand that there are other issues than technical for this problme .
In my first reply I mentioned about line capacitance, thanks GOD there is no resonance , as it appears from the symptoms described . However the same can occure if the load on the VFD is changed or there is change in the supply side impedance.
I am sure you must be in touch with VFD manufacturer in order to solve this problme and most likely they might not be co-operating with you.
Despite the indefferent approach by the power supply authority you should find out the supply side short circuit level at the incoming voltage level . This will be required in order to evaluate the filter requirement at the VFD input.The problme you are facing may mostly because of poor supply system .Please ask the VFD manufacture about the supply side requirement in respect of the short circuit level.
Have you got power factor correction capacitor in the affected system ? if yes is it tuned to some frequency ?
Untill now the SYSTEM ONE LINE DIGRAM is not very clear . Do u connect 11kV directly to VFD via series inductor, or it is connected via stepdown transformer . Please let me know the various order of hormonic in the supply line when VFD is running.
I think ( just imiginate because hormonic data are not available ) your problme will be solved by installing shunt compensating capacitor tuned to 5th or 7th hormonic.
thanks

 

[electricuwe]

Judging from Marke's last post there is some doubt to me if really the drive's harmonics are the origin of the interference problem ! A THD of 6% is not extraordinary high and the line chokes will have led to a further reduction. Maybe the other equipment is to sensitive regarding the harmoncis or the problem is not related to harmonics but to RFI.

But, keep in mind that THD is only an integral value, that isn't very helpful for a detailed analysis. Marke, can you post the values of all harmoncis, if possible.

 

[jbartos]

Suggestions:
1. Hybrid harmonic filtering (combination of active and passive) is recommended.
2. Active Front End at your VFD could also be considered.
3. A power line conditioner (AC-DC-AC) for the affected house could suffice if there are no other problems with the high harmonic content caused by the 6-pulse VFD.
4. Get all prices for the above options first.