Can a electronic ballast be controlled with a standard relay?

[electricus]

I would like to have separate control to each light fixture of a bank of fixtures. These are the basic standard multivolt ballasts from Sylvania. The relays I would like to use are rated at 5 amps, the one ballast draws about 0.5 amps for a 2 lamp T-8. Would this work, could the relay contacts handle it?

 

[Haji]

I would like to have separate control to each light fixture of a bank of fixtures. These are the basic standard multivolt ballasts from Sylvania. The relays I would like to use are rated at 5 amps, the one ballast draws about 0.5 amps for a 2 lamp T-8. Would this work, could the relay contacts handle it?

Look for UL rated relay for such application, for Electronic ballasts have a high ?inrush? (typically 50X to 100X normal operating current) that will damage any connected switching component and the UL Electronic Ballast Rating is given to products that outperform the rigorous testing to withstand the excess inrush current.

 

[kwired]

Look for UL rated relay for such application, for Electronic ballasts have a high ?inrush? (typically 50X to 100X normal operating current) that will damage any connected switching component and the UL Electronic Ballast Rating is given to products that outperform the rigorous testing to withstand the excess inrush current.

Wouldn't that high of an inrush give us a lot of instantaneous trip problems with circuit breakers, even with lightly loaded circuits? I have never had that problem with electronic ballasts, but have with magnetic ballasts before.

 

[Haji]

Wouldn't that high of an inrush give us a lot of instantaneous trip problems with circuit breakers, even with lightly loaded circuits?

The inrush may not trip the breaker, if its characteristic is properly chosen.

 

[iwire]

http://europe.lutron.com/product_technical/pdf/362-649.pdf

 

[kwired]

Look for UL rated relay for such application, for Electronic ballasts have a high ?inrush? (typically 50X to 100X normal operating current) that will damage any connected switching component and the UL Electronic Ballast Rating is given to products that outperform the rigorous testing to withstand the excess inrush current.

The inrush may not trip the breaker, if its characteristic is properly chosen.

From another recent thread I gather that the instantaneous trip of a typical 15 or 20amp inverse time breaker would be around 300 amps.

You are saying to expect 50 to 100 times normal current for inrush for an electronic ballast. That would mean a 20 amp circuit with 15 amps of connected ballasts would have an inrush of 750 to 1500 amps. If that were the case we would not be able to use typical inverse time breakers on these circuits yet that is almost exclusively what is used with little problems.

There has to be more to it. If this is potential inrush with no impedance in the supply circuit - I could understand that a little more, still seems a little high though. What is duration of this inrush? Does it only happen during part of a cycle, full cycle, or over multiple cycles?

 

[Haji]

kwired:
Please read the article at post#5 and then come back for further discussion.

 

[iwire]

There has to be more to it.

The quality of the ballast ....

Check out the link I posted but I am sure Lutron is not the only one making ballasts with low inrush.

 

[kwired]

Ok, so necessary core components may draw an inrush that high, but many (probably most) incorporate devices to limit the inrush.

I have never encountered any that have given me starting inrush issues. And I will frequently put 15 amps of lighting on a 20 amp circuit which should result in well over instantaneous trip of a 20 amp device if they didn't have a limiting device.

This problem may not be all that common, and chances are is not a problem unless you purchase the cheapest equipment you can find, Then it may cost you more to solve the problem than it would have to purchase good quality items. Live and learn I guess.

 

[Haji]

I have never encountered any that have given me starting inrush issues.

Perhaps even the instantaneous circuit breaker takes some few tens of milliseconds of unintentional time delay to trip whereas the duration of inrush of electronic ballast is around 500 micro seconds only. So before the breaker is able to trip, the inrush dies down.

 

[kwired]

Perhaps even the instantaneous circuit breaker takes some few tens of milliseconds of unintentional time delay to trip whereas the duration of inrush of electronic ballast is around 500 micro seconds only. So before the breaker is able to trip, the inrush dies down.

500 microseconds if I calculate correctly is a little under a third of a 60 Hz cycle isn't it?

I do agree this may be faster than the breaker can respond, but is likely also limited even further by circuit impedance.

 

[Haji]

So it appears that the current limit feature of electronic ballast brand of article in post#5 is redundant.

An analogy may seal its fate: Lightning current does not operate fuse or circuit breaker despite its high magnitude, because of its microsecond duration of individual flashes.

 

[electricus]

Good article, I understand what is said there. I don't know if the ballasts I have or they relays I want to use are that specifically rated for the purpose. I'm going to make an attempt at getting more information from the manufactures. The other opption is to purchase one of each and do some testing.

 

[GoldDigger]

500 microseconds if I calculate correctly is a little under a third of a 60 Hz cycle isn't it?

It seems to me that 60 Hz would have a period of 16+ milliseconds. Since 500 microseconds is .5 milliseconds, I think that would be more like 1/32 of a cycle, not 1/3 of a cycle.

 

[copper chopper]

what type of fixture and relay are we talking about flourescent or led and as long as the voltage and amperage are within the specs of the ballast it will be fine. we hook those up to lighting control panels all the time and thats all they are is relays...

 

[kwired]

It seems to me that 60 Hz would have a period of 16+ milliseconds. Since 500 microseconds is .5 milliseconds, I think that would be more like 1/32 of a cycle, not 1/3 of a cycle.

even shorter time yet, One or even both of us probably has the decimal in the wrong place, too many zeros and I am more used to putting them on the other side of the decimal with most of my calculations.

 

[electricus]

what type of fixture and relay are we talking about flourescent or led and as long as the voltage and amperage are within the specs of the ballast it will be fine. we hook those up to lighting control panels all the time and thats all they are is relays...

Fixture/Ballast =

Lithonia 1 x 4 IC rated troffer, (2) T8 lamps, p/n GT8 232 A12 MVOLT GEB10IS (instant start multivolt electronic ballast). I'm not sure what brand ballast they use. From experience I figure 0.5 amps draw, inrush?
​

Relay =

Automation Direct RS4N-DE Slim Card Relay. Silver alloy gold plated contacts rated at 5A, Max. Make/Break Current (resistive load) 250VAC/5A. Spec. Link. One relay with be used per fixture/ballast.
​

 

[ELA]

Here is a pictorial that might help some envision what is meant when they talk about a large inrush for electronic ballasts. At the top is a single CFL that may only draw approx. 0.2A RMS. You can see that while the peak Inrush is very large compared to the RMS it is for a short duration. In the next shots you can see the effect of adding several devices together. The peak does not add or increase linearly.




We are taking about a capacitive inrush. The bigger the capacitor the larger the peak and/or time to charge it up (also dependent upon source impedance).
In most cases tripping the breaker (Instantaneous) would not be an issue with the relatively small capacitor values in CFLs or electronic ballasts.

This capacitive Inrush is present in many devices that incorporate an AC/DC rectification stage.

We had to add inrush limiters to some large Servo drives that had very large input capacitors ( much much larger than would be present in a common ballast). The Servo drive would randomly trip the breaker on power up due to the much longer charging time required for the much larger caps.

You do have to be careful about your selection of a relay. This large current spike will stress the contacts. The relay pointed to may not handle that current spike all that well since they are gold plated and intended for lower currents. No real way to predict what your inrush situation might be without measuring it ( unless you have a spec. sheet for the ballast that states inrush value, or states that it limits inrush.)

If you do indeed draw 0.5 amps RMS the inrush could be quite a bit above the 5 A rating. Note the "Resistive" qualifier. It is possible that it would appear to work just fine initially but end up failing prematurely with welded contacts.

 

[broadgage]

Here in the UK we use any standard, off the shelf industrial relay for switching ballasts.
In practice the inrush current does not seem to be a problem, after all in the absence of relay control, the lights would be controlled by a standard light switch without any concern re inrush current.

 

[hurk27]

It seems to me that 60 Hz would have a period of 16+ milliseconds. Since 500 microseconds is .5 milliseconds, I think that would be more like 1/32 of a cycle, not 1/3 of a cycle.

Or 16000 microseconds which you are correct 16000/500 is 32

even shorter time yet, One or even both of us probably has the decimal in the wrong place, too many zeros and I am more used to putting them on the other side of the decimal with most of my calculations.

Yep it took me alittle to get my head around that a second if 1,000 milliseconds, and a millisecond if 1,000 microseconds, don't use these small figures very often anymore.