Soft Starters & ELCB's or RCD's

[GGOSS]

Hello all,

 

I would greatly appreciate advice as to what causes Earth Leakage Circuit Breakers (ELCB's) and Risidual Current Devices (RCD's) to 'trip' when a motor is started via a soft starter.

 

Note: The above does not occur in all applications and does not appear to be limited to ELCB's and RCD's without harmonic compensation. Also, our company distributes some 3000 soft starters per annum, but we only experience two or three problem installations each year.

 

Any advice would be appreciated.

 

Regards,

GGOSS

[marke]

Hello GGOSS

 

This is usually, from my experience, related to the capacitive currents that flow to ground every time an SCR switchs ON. The is quite a high capacitance between the stator windings and the frame of the motor. The switching of an SCR during start or during stop will result in a moderately fast switching edge. This will be coupled through the internal capacitance of the motor to the frame and cause a current pulse. Similarly, the capacitance of the cable to earth will be charged/discharged when the SCRs switch ON. The result is a series of short pulses at a 300Hz rate (360 for 60 Hz). These current pulses are very short, typically less than 10 uS but can be stretched by circuit resonances etc. If the ELCB is very fast operating, and/or the capacitance is very high, then this can occur. You will probably find that this occurs on submersible pumps (high capacitance to earth) or installations with a long neutral screened (or armoured) cable between the starter and the motor.

Best regards,

[GGOSS]

Many thanks Marke,

 

As the most recent case involves 2 x 15kW submesible pumps at a public swimming pool, you may have hit the nail squarely on its head. The devices in question are presently set for 300mA and are still tripping. The end client would like to set these for 30mA sensitivity without nuisance tripping!

 

In another recent application (300kW mine shaft vent fan), the device is set at 2A and trips on every 4th or 5th start. The end clients preferred sensitivity here is 100mA. He has another identical starter panel in which the device is set at 100mA and is not experiencing nuisance tripping. From your previous post this would imply differences in motor characteristics and/or cabling are the root cause of his problems.

 

That leaves me with 2 questions:

 

1. Are you able to provide detail as to how the level/magnitude of leakage current can be calculated?

 

2. As the device cannot be inhibited during start for safety reasons and it is unlikely that the clients will pursue the option of replacing motor and/or motor cables, are there any possible solutions?

 

Regards,

GGOSS

[marke]

The actual magnitude of the current is a function of the capacitance and inductance of the system and the rate of rise of voltage as the SCRs switch. This becomes very difficult to quantify and is probably not particularly easy to relate to the characteristics of the RCD.

The RCD has a trip point and also a response time. Because the leakage current is essentially a series of impulses, the magnitude will be reduced by the input filtering of the RCD. The results will be dependant on both the magnitude and the duration of the current impulse in addition to the time constant of the input circuitry of the RCD. A slower operating device will ignore these impulses unless they are very severe. A faster input filter in the RCD will allow it to see the impulses and could let enough through to trip. I expect that you will find that the results are influenced by the type of RCD and possibly in some situations, you may need to use a particular type (with a slower input filter) to overcome the problem without setting the trip current too high.

At this point, I am unable to make recomendations on best uints to use.

 

Have fun!!

Best regards,

[GGOSS]

Many thanks Marke,

 

I think there's enough information there to assist in solving any future problems.

 

Regards,

GGOSS

[Guest]

I have an earth problem.

 

I have done earthing solution using the best metrials for my ICT hardware, yet I am getting a voltage of 67V between the nueutral and earth terminals. This voltage can destroy low voltage components in devices like switches and routers.

 

If I also bridge the nuetral and earth cables at the ground is there a risk?

Please reply soonest.

 

Thanks guys

[marke]

It is difficult to help with out very good information.

One of the important isues is what constitutes a good earthing solution.

It is not satisfactory to have a good DC path, you must have a good HF path and this means that the length, path and cross sectional area are all very important.

For a start, what is your "ICT hardware"?

 

Best regards

[niallnz]

GGOSS

 

Do your starters have 2nd Environment EMC filters, theses are of course capacitors to earth and a good source of leakage current.

 

Cheers

Niallnz

[GGOSS]

Thanks for your interest niallnz,

 

I suspect however that you might be confusing soft starters and variable speed drives. The issues with both products are quite different.

 

Thanks again & regards,

GGOSS