Softstarter, VFD and Capacitor banks

[SilverArc]

Hi,
We are selecting a softstarter for a mine ventilation fan application, I guess the decision was based on the manf. sales guy recomendations.

He also mentioned that while using softstarter and VFD, it is not advisable to use capacitor banks.

Can you kindly put a word to this.

Regards,

 

[jraef]

#1 - There is no need for power factor correction caps on a VFD application, the displacement power factor presented to the line by the VFD is alrerady as corrected as you need it to be. The distortion power factor is worse, but that is a different issue.

#2 - You can use capacitors on a soft starter motor application, but they can NEVER be connected to the load side (downstream) of the soft starter. The charging current of the capacitors looks like a short circuit to the SCRs, so it can cause them to self-commutate and misfire, causing damage to themselves and other equipment. In addition, the harmonics created by the phase angle firing of the SCRs can cause the capacitors to overheat, swell and sometimes even explode. it is usually a race to see which component, the SCRs or the caps< fail first. If you want to use PFC caps however, just wire them on the line side of the soft starter and have a separate capacitor contactor keep them off-line until after the soft starter has brought the motor to full speed and preferably bypassed. If you have (unfortunately) selected one of the phony baloney "energy saver" soft starters because someone fooled you into believing they will work, you MUST turn the "energy saver" feature off if you want the PFC caps. When that feature is in use, the SCRs are firing continuously and the SCR/capacitor damage will occur.

 

[LionelHutz]

Well, it's not avisable to use a cap bank on any power system without first evaluating the harmonics and the possiblity they will cause bad interactions between the power system and the cap bank.

Soft-starters produce (6 x n +/- 1, where n = 1,2,3..) harmonics during starting.

In the end, it depends on how much money you want to spend vs how much risk you want to take.

We supplied 2 soft-starters with 1800kVAR power factor correction banks to a mine with cycloconverters on the hoists at the surface (another big harmonic producer). I don't remember the motor size but these banks were oversized for the motor.

We used 5610V rated caps for a 4160V system. The bank was 250 amps @ 4160V or 454A rated @ 5610V rated.

We supplied fuses and overload relays on the caps. We also did a double Y bank with a current relay on the Y connection to detect a shorted "roll" in a cap. The overload relays were set to operate on a current of 400A. We basically provided enough protection that the caps should not do anything really bad no matter what was done to them.

The starters were around 5000' down the mine and each starter was about 1500' from the other one. During certain points in the operation of the cycloconverters on the surface, the overload relays would begin to pick-up and indicate an overload. If I remember correctly, the cycloconverter harmonics were also passed through 2 transformers to get to the cap bank.

The good part was that earlier in the project I got in a somewhat heated discussion with the mine engineer/electrician who felt that the caps would never interact with the cycloconverter 5000' away and that we could just provide any old bank and it would work. So, when the relays began showing an overload they never contacted us about it but rather hired in another company to measure the cap current during the overloads to see if any long term damage would result to the caps. It was determined they were safe.

Overall, they were saved by the use of 5610V rated caps because they'd likely be replacing 4160V rated caps regularily. I was saved from being dragged into a post intallation discussion about the harmonics because I'd expressed concern beforehand.

As an interesting side note, the mine personel felt that a lot of the other equipment in the mine worked better when those cap banks were on-line.

 

[jraef]

But Lionel, you didn't have those caps on-line while the soft starter was ramping did you? Or are you just mentioning this because of the unforeseen consequences of having the caps in the system with the cycloconverters on the same grid? If so I totally sympathize.

I had a meat plant in Colorado start going through SCRs on soft starters and DC drives like they were McDonald's Hamburgers one year, along with repeated failures on the bulk PFC control system caps. Everyone was blaming the SCRs, I simply asked them if they added any machinery at about the same time as when the failures began. Turns out some different section of the plant had added 2 giant older Westphalia separators that had 4 speed motors. Not only were the motors starting Y-Delta (in some complicated scheme I didn't bother to investigate), but they had their own PFC caps on each speed setup and the operators were switching speeds constantly. Every time they did, the caps would interact with the bulk PF correction. You could see a massive voltage spike on the scope every time they switched, usually in excess of 2kV (the rage of my scope at the time) and this was apparently taking out the SCRs and caps elsewhere in the plant. They removed the separator PFC caps and all the failures ceased.

 

[LionelHutz]

No, in this case the caps were energized once the motor was at full-speed. We have done installations having capacitors on-line while the starter was ramping though.

So yes, it's just an example of how caps can cause problems on the power system. Transformers don't necessarily make it safe. Distance doesn't necessarily make it safe. Energizing the power factor capacitors after a soft-starter has reached full-speed doesn't necessarily make it safe.

Any harmonic generating piece of equipment installed in combination with capacitors can potentially be a problem.

 

[SilverArc]

Hi ,

Kinda old post, but this issue has surfaced again in one of my projects.

LionelHutz,
You mentioned using a 5160 V cap. bank on a 4160 system.
What advantage do I have ?

Jraef,

You explained this meat plant situation and what I infered that a dedicated cap. bank for those new four speed motors in a different part of the plant when switched were interfering with the main bulk PFC system(I beleive installed at the main switchgear or service entrance of the premises).

In my application, I have a 1000 KVAR bank on surface in the mine, a fixed one and we are planning to install a 9000 KVAR deep down in the mine, may be 2500 ft below the surface.

How can I analytically prove or otherwise, anticipate installations of cap, banks near pumps or main buses in the mine deep down inside, will not interfere with the PFC systems at the mine main switchgear on surface or other harmonics creating devices like fan DC drives located on surface.

A long question, I would appreciate a word.

Regards,

 

[LionelHutz]

I'm talking 3-phase capacitor cans here. I used those 5610VAC rated capacitors because they give a margin of safety when used on a system with harmonics. If you take a look at line voltage rated caps and then calculate the effect harmonics can have on the caps you will often find that it doesn't take a lot of harmonics on the system to go over the capacitor ratings. Besides, we have a good source so those caps are readily available.

Anyways, there are really ony 2 solutions if you want to know what will happen.

The first is to do a system study. You need fairly detailed knowledge of the power system and the harmonic generating devices on the system. This can take months and many tens of thousands of dollars. A consultant that has enough knowledge to do this typically cost about $150-$200/hr. Then, any time you add a large piece of equipment to the system you should really go through the study again.

The second way is to just install the bank. Do enough engineering on the bank that the protection systems you install will indicate there is a problem and trip before a catastropic capacitor failure occurs. Once installed, you can also monitor the bank (or banks) during operations to see what occurs.

A 9000kVAR bank is big. I would expect as a minimum to use phase overcurrent relays, a double Y connection and 3 fuses for each capacitor can. Actually knowing the