Program rapid start fluorescent lights. Power disturbance tolerance between ballasts

[Electric-Light]

Each ballast is different. The difference in its tolerance to line dip can mean very angry customer vs not even getting noticed.

The best do not flicker at all or much with power disturbance.
Better ones just flicker
Worst ones reboot.

Rebooting is a problem unique to programmed start, because the cycle takes about a whole second and if it happened every time th ere is a disturbance, it is very annoying to the occupants.

I've found that some programmed rapid start ballasts are very sensitive to sudden line dip even if it doesn't dip below the rated range while others are very tolerant even those from the same brand.

Incandescent and fluorescent will flicker in response to sudden change in power (the more technical phrasing is high -dv/dt). Some programmed start ballast will reboot due to high -dv/dt if the magnitude of change is within tolerance and this means that it will go out for about a second.

If the permissible voltage range is +/- 12v and you have 120vrms nominal and a sudden loading causes the voltage to drop to 114vrms in one cycle and lasts a few cycles, you have a dv/dt of -360vrms/s (assuming 60Hz). This can happen when thyristor controlled loads are operated.

Instant start ballasts and incandescent lights do not suffer beyond visual nuisance.

Sometimes the spec sheets do not tell you anything about serious vulnerability.

Products tested are:
Incandescent and instant start system:
visible flickering

Sylvania QTP2x32T8/UNV PSN
Did not produce any noticeable response caused by laser printer fuser cycling on the same circuit.

http://assets.sylvania.com/assets/D...221).d9619c4c-3ef6-4cf5-ab15-6c9cf4fa3145.pdf

The newer QHE2x32T8/UNV PSN, premium efficiency model. Despite the fact it offers UL CC rated arc limitation, striation reduction circuitry and parallel operation, subjected to the same line dip, this thing will reboot.

It offers a saving of 3.4%(120v)/1.8%(277v) for 2 lamp model and 6%(120v)/4.5%(277v) on 4 lamp model.

Even though parallel operation allow working lamps to remain lit in the event of a failed lamp within a fixture and significant power saving for the 4 lamp model are lucrative, the reboot vulnerability far outweighs the benefits unless they won't experience power sags.

http://assets.sylvania.com/assets/D...218).2d59f1d4-f342-4187-868f-a209d1cb82a7.pdf

The difference between the old model and the new high efficiency model is that unnoticeable on old model vs one second rebooting whenever there is a disturbance on lighting bus caused by across the line or VSD elevators, compressor, high inertia HVAC fan, or large photocopy machines.

 

[cadpoint]

...
The difference between the old model and the new high efficiency model is that unnoticeable on old model vs one second rebooting whenever there is a disturbance on lighting bus caused by across the line or VSD elevators, compressor, high inertia HVAC fan, or large photocopy machines.

I don't beleive one can ups a power lighting control panel, nice thought but seems like a bad app...

I'll assume that its the "disturbance" over-all that is the problem.

The only real solve could be various at best. Just my opinion but you have a lot going on here! I didn't look at the links and the following don't address them either.(well maybe I did * )

First it seems that one needs to understand why there is a VD and where exactly it exists in your system or maybe down to the exact circuits and their assoication in a panel. Since everythings sees the drop then a total calculation might need to be redone to understand this drop at the panel.

The second cure is maybe just lift all light circuits and put them in a individual panel and get back to the main and not create a subpanel off of where the first problem is? If the whole system experiences VD, than this is not a cure.

* My third thought is your mixing apples and organges and the various flavors that are available! You've setting yourself up with this problem if your changing types of ballasts through either addition of refit's and of course not now matching what exists or for as to the lumanire was designed to do, if there now in the same area, of course they will be noticable.

Lastly; If it is just a line side issue or poor power supplied to your service point, then the POCO needs to be involved because they are responsible as well.
VD is only one possible problem, where it is exactly is could well be another.

Just my .02, hope that helps.

 

[Electric-Light]

Someone complained about this exact same thing in a different thread, so I performed a bench test. He was experiencing newly installed ballasts "rebooting".

Ideally notching shouldn't be there, but it happens. You've seen lights flicker in the storm in any type of building. Ballasts should be smart enough to not go through a reboot. That Sylvania ballast is dumb enough to actually shut off and reboot causing a loss of light for a full second.

Rebooting is a problem unique to programmed rapid start type in which a split second disturbance causes the ballast to go through the full start-up sequence and turn off the light for a full second.

Suppose you start a table saw on the same circuit as your lighting in a home shop. The expected action is dimming or flicker. Same deal when there's a quick dip in a machine shop from a compressor starting up.

I believe you should have this line in specification criteria when you're spec'ing programmed rapid start

"program rapid start ballast shall not power cycle or "reboot" in line voltage disturbance lasting less than 30 cycles" -addresses the problem I discussed above. It's not a huge deal if the lights flicker a bit in the elevator cab when the motor is started across the line. But if the ballast responds to it by rebooting, then I think its a problem.

"PRS ballast shall restore light instantly after a loss of power lasting less than 12 cycles" -prevents the ballast from having a noticeable one second reboot after a weather induced brownouts.