480 Volt 3 PH Soft Starter By Pass 5

[jburn]

Hi, All

I just read the 12 or so, different post about soft starters, and I have a different question reguarding the use of these devices.

We have installed several SMC soft starters over the past six months. They are on a 25 HP motor, 480 volts 3 PH. The 480 volt power system is an ungrounded system. IE the unit sub is a 2500 KVA delta - delta transformer. They had been running OK for several months. The motor FLA is 31, but the normal running load is only about 12 amps. Very lightly loaded.

Last week, we had several of the soft starters shut down, within seconds of each other. We noticed, that at about the same time we had a single phase ground, coming and going. Several hits every few seconds, lasting for up to 20 min.

When we looked at the phase to ground voltages, using a Dranetz analyzer, we see transients up to 900 volts, when these grounds occure. Apparently the soft starters don't like the "Noise" on the line, and shut down.

My plan is to install a 3 PH contactor in parallel with the soft starter, and have it by pass the soft starter aftr the motor is up to speed.

Question, do I need to size the by pass contactor, based on the motor HP, or can I use a smaller unit, sense it will only see runing current and not starting currents?
The soft starter unit itself, has 3 small relay contacts, used to by pass the SCR's, after the motor gets up to speed.

The bypass contactor will be used to disconnect the motor from the line when the motor is stopped, and / or when the motor overloads take the motor off line.

Thanks for your response.

 

[busbar]

…HRG units …not have as many interminit grounds…

jburn, that is a most likely a very precise assumption. Although anecdotal, I have found a fair number of cases in the last three decades that affirmed your statement—from unbelievably frequent stator-winding insulation punchthrough on ungrounded systems. With an incredible motor “burn rate” [one facility had fourteen 480V-wye unit substations,] all but one were operated ungrounded while serving a hundreds of motors over the range of 2-500hp. In thirteen of those particular 750-1500kVA unit subs, the XO bushings were all “flying in the breeze.”

OTOH, I have seen a number of other ungrounded plant systems with strikingly improved insulation-failure rates after high-resistance upgrades.

The grounded-wye/broken-delta transfomer arrangement is cited repeatedly in IEEE publications; applied at all voltages—although at higher voltage levels it is used for annunciation and/or protective-relaying functions, and not the sole means of directly or independently limiting transient overvoltages.

 

[RJF]

Jburn - IS the ground alarm system that you mentioned having on your subs a commercial system? If so, who made it? We have looked at designing our own as I have not found a commercial system. We just use the built in alarms for HRG equipped systems. I agree that an arcing fault can give very high and weird voltages on an ungrounded sub - an HRG is supposed to help limit voltage swings.

 

[busbar]

Made an error in my 30-Jan posting. I referred to SMC starters as “VFDs.” My apologies. SMCs do not have "load-side transistors."

 

[d23]

jburn

As a question or personal education I've always wondered about multiple failures on an ungrounded system in a large manufacturing facility. Are you having any catastrophic failures (SCR's Motors etc...) due to the phase to ground faults? If you are do they occur in multiples? Do you know the source of the grounds?

 

[jburn]

RJF, and d23

The ground alarm system is a home grown system but using all standard components. I do not have HRG on my 480 volt Delta unit subs, so I used a ground detector made by Erickson Electrical Equipment. The Erickson unit gives me a dry contact. I hooked up the Erickson to the closes, already existing, remote PLC I/O card. Because most of our PLC's are on a network, I then used a data historian system called PI, which already existed in the plant. The PI system scans the PLC remote racks, and records each time the Erickson sees a ground. The PI system time stamps each event. Using the PI system, I can look at all of the history of the grounds, and plot each event, on my desk computer. The PI system also sends me an E-Mail and a page, on each ground event. The system works quite well, as I can overlay any of the other 18,000 data points in the PI system, to look for any corralation. The system is fast enough that it picks up interminit arcing grounds.

d23,
I have worked in several different industrial plants for 32 years. All of them had unground delta 480 volt unit subs. I have seen catastrophic failures due to the phase to ground faults, but I have never seen multiple failures, as described in Beeman's book. The problem I have just had, shut down 5 machines at the same time. In this case there was no apperent damage. I do know the source of the grounds, and they vary. We are very agressive in tracking down each ground and fixing the problem.

The alarm system as I decscribed above, has shown me that we are having more interminit arcing grounds than we ever knew about. Before we installed the alarm system, we had no way of knowing when we had an interminit arcing (coming and going) grounds. We were only seeing grounds that were permenit. Another benifit of the alarm system is that because we can plot the exace time all of our ground come and go, we have been able to track down the ground, with the help of operations, just knowing what equipment is cycleing on and off at the same time the grounds do. The system has well paid for itself.

 

[GGOSS]

Hello jburn,

I can sympothise with you as I have experienced similar problems and have only just managed to resolve them. Please advise if the starters to which you refer to are SMC3 or SMC Delta. Your location (country & state) would also be beneficial as this will assist me to identify persons able to provide localised support if need be.

Once I have the above details we can move to solve your problems.

Regards,
GGOSS

 

[jburn]

GGoss

Thanks for the offer of help. I would perfer not to name the make and/or model of the soft start controler (SMC)that we are using. I am receiving help and support from two different manufactures of SMC's, including the manufacture of the one I am using, as well as from some power quality experts, looking at system grounding, arcing ground faults, issues etc, etc.

I would like to keep the discussion on this forum generic, dealing with power quality and solid state/ microprocesser controled equipment issues in general, on ungrounded power systems.

The real fix to the problem has more to do with power quality, system grounding, etc, and not any one make or model of device. All of the units on the market will work fine if your systems are properly grounded, no ground loops etc, and if you address all of your power quality issues, which is where I am focusing my time.

If you do not address grounding, and power quality issues on your power systems, from a reliability standpoint, all of the equipment on the system, like all motors, and the cables themselves are subject to failure, not just SMC's

 

[firozengr]

As I understand a soft starter left online without bypassing can also perform as an energy saving device by reducing the operating voltage until a point that the required running torque is maintained. Thus the current being the same, the product is the reduced power consumption.
Selecting a bypass contactor shall be of the motor name plate rating, since the bypass is not a closed transition.
As we are aware that any electronic power device introduces harmonics into a power system, a harmonic measure of the power network at the MDB shall be done.

 

[jbartos]

Suggestion: Visit

http://www.engineeringtalk.com/news/edl/edl100.html

for soft starter features

http://www.dialelectronics.com.au/articles/74/0c010f74.asp

for soft starter motor management system features
etc.