Switch Failure/flourescent Light In Rush Current

Status
Not open for further replies.
P

Phil Devine

Member
We have a school that has been recently constructed and have experienced failures of the switches on the 277 volt fluorescent lighting. Some of the switches have failed and arced to the metal switch plates. There is an alleged instance of possible shock to an employee. In this particular school, we experience hearing a popping or snapping when the switches are operated. The original switches were rated for 15 amperes. We changed them all to 20 ampere rated switches of the same manufacturer. Both switches have experienced failures. When we dissected the switches, we found pitting on the switch contacts. The ones that failed seemed to be overheated and melted the insulation of the switch body. There are only 6 fixtures connected to each switch. We found the in rush current of each lighting circuit to be about 150 amperes.

We replaced some of the ballasts with some from a different manufacturer and the in rush current reduced to about 50 - 75% of what it was originally.

In other schools with the same type of ballast as the original ones we have in this school, we found the same levels of in rush current. We have not had any failures of the switches. The switches are of the same rating but made from a different manufacturer. We also do not hear any snapping or popping in these switches.

Has anyone ever experienced multiple switch failures in the same circumstances and if so, what were the causes(s)? We are stumped so far and are wondering if there are any other tests we should be performing. A lighting supplier recommended we install thermistors to limit the initial in rush current. We tried this on a few circuits and it seems to reduce the in rush but are concerned we are masking the problem.
 
This sounds pretty weird to me. I have been in the trade since 1980 and have never seen anything this severe with switches and florescent luminaires.

150 amps inrush seems way high. I would like to get more info on this. And how was the inrush measured?
 
What does the technical customer service department for the ballast manufacturer say about that particular model number?
 
We measured the in rush with a peak hold amp clamp and a testing company did it with a "wave sniffer." We found the same measurement at another school with the same ballast type installed. The duration was about .6 seconds before the system normalized to about 8 amperes. We too are baffled including the design engineers. I never ever measured ballast in rush current until now. The other school has the same readings but no issues with the switches. We would assume a bad batch of switches except we changed them all out with new 20 ampere switches with some of the same results. The school went on line this year so there is minimal use of the switches.
 
Phil Devine said:
We are told by the engineering firm that none of the ballast manufaturers we have will give an expected in rush current.
Click to expand...
The engineering firm is one thing.

I'd have to hear that said from the manufacturer's techs before I'd believe it that the inrush is "unknowable".
 
A transformer primary (ballast) can be very low impedance when first switched on. I have measured large inrush currents using an O'scope on many capacitive input devices (and have measured over 200 amps depending upon the source impedance).

So inrush is common on a lot of electrical devices.
Is your poping and snaping noise worse on turn on or turn off?
Could you be having an issue with transient voltages on turn off due to the inductive kick? If you could measure both voltage and current on a scope that would be revieling.
 
The snapping and popping is both switching on and off the circuit. We have not measured the inductive kick when the switch breaks the circuit, Since we have the same ballasts in other schools with similiar in rush currents, we don't suspect anything different here other than maybe the brand of switch. It seems odd that something as stright forward as a light switch and flourescent ballast could be incompatible. I would have thought there were industry tests and standards to prevent that. We are open to any and all suggestions to help us out.
 
Fluorescent Switching Problem

Fluorescent Switching Problem

This may be a bit simple but take one of the switches from the other school where you have the same inrush but no problem and install it on this site and see if the popping goes away. This may be a switch manufacture problem.

Just a thought..
 
Phil Devine said:
We have a school that has been recently constructed and have experienced failures of the switches on the 277 volt fluorescent lighting. Some of the switches have failed and arced to the metal switch plates. There is an alleged instance of possible shock to an employee.
Click to expand...
Sounds like a problem with a particular manufacturer of switch. One thing you should do right away is make sure that the metal switch plates are grounded or else replace them with plastic face plates. If the switches are in properly grounded metal boxes, you should be OK, but if the boxes are plastic, there may be a problem.
 
My first thought was poor grounding of the switch or light fixture. What is being used as the equipment grounding conductor? Was the same contractor used in this building as in the other? Has wiring been inspected in the panels and at the devices. What's the voltage between neutral and ground when the lights are on and off. Between hot and neutral?

My second thought is cheap switch manufacturer. Does it have a UL label? Is it name brand or generic? Are the switches rated for the right voltage? I just want to make sure they're not 24V DC switches. I've also heard about imitation products from China. It looks like the real thing but it sure doesn't function the same.

The arcing and pitting leads me to believe you're having high voltage problems. Every time a switch is opened or closed you'll create an arc. Because the arc makes an audible noise this makes me think the voltage is higher than normal. This would also leave larger pits on the contact surface. But a poor quality switch would create pitting also. And once it starts it will keep getting worse until the switch fails.

Having the switch physically melt might mean there is resistance in the switch where there shouldn't be any. Most likely from the pitted contact surface but it might have another source. Maybe a loose wire or a copper-aluminium connection somewhere.

Are any ballasts or lamps failing early? Any other things happening in the building?
 
I agree with Mr. Bill about the switch-body melting. Contact arcing alone won't do that unless it's continuous. Melting requires heat and time.

If you've found that switching (pardon the pun) brands helps, do so. I'd also try letting the poor-performing switchn amnufacturer know.
 
We experienced another failure of the 20 ampere switches with a different ballast manufacturer. It arced and threw sparks when opened.

We suspect the switch manufacturer has faulty switches because the 20 ampere switches we replaced the 15 ampere switches were from the same manufaturer. However, we use this brand of switch (it is common here at our suppliers) with other schools and have no issues that we know of.

The switch is UL rated and rated for the proper voltage. We are still waiting for some kind of recommendation from the design engineer.
 
Status
Not open for further replies.
Top