Fun with European power

[porkchop]

Touring gear never ceases to provide fun situations. We travel 8 Lycian 1290 XLT's. They are rated for 105V to 125V AC @ 50/60Hz or 200V to 250V AC @ 50/60Hz. You adjust the voltage rating by moving a jumper, options are 105V-125V, 205V-220V, and 220V-250V. We're in the UK currently (240V @ 50Hz expected power) so they are all set for the 220V-250V range.

Ok that should be enough detail here's the situation. Of the 8 lights we are having unexplainable problems with 4 of them. If we plug them into house power (240V @ 50Hz) they will run for hours during load in (think 11 am till 2 or 3 pm). Once we get into the show (7pm) they last maybe twenty minutes then they destrike showing symptoms of over heating (indicator lights on the back of the ballast are off until the unit sits for a while with a fan on, then the lights turn back on, and the unit strikes back up). So bum deal our lights get hot here's where it gets odd.

If we run power (240V/415V 3Ø @ 50Hz from the house transformed down to 120V/208V 3Ø @ 50Hz) from our distro rack to the lights they work great set on the 205V-220V or 220V-250V jumper. So if I transform the power down and run 208V through and extra 300' of copper it works great event thought there is no power conditioner in the line to effect the power curve. For the life of me I can't figure out what is the difference in power (other than the obvious 32V) that is causing the lights to overheat.

Some additional info: I would buy that ambient temperature goes up when audience members are present, but we did a glorified dress rehearsal today and they still over heated all the time. The staff actually complained that it got colder in the bowl as the show went on. The operators at the lights that over heated said the lamp housing wasn't very hot to the touch. For the lights that stayed on lamp housing was pretty hot to the touch. We've replaced fans, ballasts, igniter cards, and lamps none of this seems to change the lights performance. This is the third venue this has happened at. Any ideas anyone?

1290 XLT manual for any interested:
http://www.lycian.com/Parts/1290/1290%20Parts%20Manual.pdf

 

[WooferHound]

What does the amp meter on the back of the spot read out ?
Do you make the adjustments to keep your Arc Amperage between 80 and 85 amps ?

 

[porkchop]

I will check that again tomorrow, but every time I've noticed the amperage it;s been pretty close to 80

 

[JD]

4 work, 4 don't. Such an even split! Any chance that they are grouped that way on dif phase legs, and one leg has some irregularity when the show lighting is on? You mentioned they work fine during load in, but I assume at that time the stage is not lit and the distro is not loaded.

Sounds like a ballast shut down.

EDIT: Just a note, the lamp amp meter would not show this irregularity as the E-ballast would compensate until it was outside of operating limits, at which point the lamp would drop arc and the meter would go to 0.

 

[David Ashton]

This is almost certainly a frequency related issue, check that they are set for 50Hz as the impedance is lower than at 60Hz, this may be the problem, just to check, measure the working volts on the lamp,[carefully]

 

[porkchop]

Frequency is what I would like to blame but there is no adjustment for frequency according to the manual, and the power through my transformer is still 50Hz and the lights work fine on that power. Both of those make it hard to believe that is the problem. As far as JD's ideas on load, I thought a bit about that too, but we did have the whole rig stuck and operating while the lights where running during load in, and when this first started happening we really quizzed the building about their power. All spots had house supplied power in the 230V -240V range and the buildings power is set up in sectors and none of the lights in any given sector were either all working or all having issues, there was a mix everywhere.

JD when you say ballast shut down, are you talking about tripping a safety switch (i.e. overheating like we think) or are you talking about something else?

This week we're playing the O2 in London, one would have faith that a building this size built right around the turn of the century would have pretty decent power. We are using 4 house spots (4k Lancelots that are amazing for what it's worth) and 4 of ours. 3 of them have issues 1 doesn't.

The only thing I can think of that is any different between the power supplies is the obvious 32V that we account for by setting the jumper down and that 208V comes from two hots that are out of phase and 240V comes from one hot and a neutral. But looking at the power curves and remembering what I can about how power works that doesn't matter because they both oscillate around 0V in the same manner.

 

[porkchop]

Checked lamp amperage, two are in the 80 - 85 range that they should be in. We turned two of them down to 70 and they don't over heat anymore. It's a solution I don't really like, but it is working

 

[JD]

I have not had the pleasure of taking one of them apart, but as they are a 2kw ballast that is mounted on the light, I assume they are electronic as compared to magnetic! Electronic ballasts generally have several safety circuits that will shut them down if something is out of range. (input current, output current, output voltage, input voltage, component temperature, etc.) Any of these shutdowns would produce the same symptom. (Time-out, or lock-down until the power is cycled off) Frequency is not as big a factor as the input of the unit is basically a rectifier feeding a big capacitor(s), as compared to a transformer. (Transformers are more finicky about frequency.) Electronic ballasts are more likely to see a sudden jump or sag of input voltage as a "problem." Since 50% of the units were doing this, I suspect that there was something about the power that was making them mad, as compared to a heat issue. They certainly were operating within their specifications, at least in theory. A brief "out-of-range" voltage change may have tricked them into believing they were being supplied with a different power then they were set for. Again, the 50% thing either suggests that things were very borderline, or that something was different about the power half of them were running off of.

The real proof will come when they move to the new venue. If there are no issues, then that's that. Of course, you will get the stock answer- "Nobody else has had any problems!"

 

[porkchop]

I have not had the pleasure of taking one of them apart, but as they are a 2kw ballast that is mounted on the light, I assume they are electronic as compared to magnetic! Electronic ballasts generally have several safety circuits that will shut them down if something is out of range. (input current, output current, output voltage, input voltage, component temperature, etc.) Any of these shutdowns would produce the same symptom. (Time-out, or lock-down until the power is cycled off) Frequency is not as big a factor as the input of the unit is basically a rectifier feeding a big capacitor(s), as compared to a transformer. (Transformers are more finicky about frequency.) Electronic ballasts are more likely to see a sudden jump or sag of input voltage as a "problem." Since 50% of the units were doing this, I suspect that there was something about the power that was making them mad, as compared to a heat issue. They certainly were operating within their specifications, at least in theory. A brief "out-of-range" voltage change may have tricked them into believing they were being supplied with a different power then they were set for. Again, the 50% thing either suggests that things were very borderline, or that something was different about the power half of them were running off of.

The real proof will come when they move to the new venue. If there are no issues, then that's that. Of course, you will get the stock answer- "Nobody else has had any problems!"

You're totally right about the ballast, but this is the third venue that we've set these up at and it's the third time it's happened, so if it is a power problem then this is going to be a very long tour. Surprisingly we're not completely getting the "nobody else has problems" we're just getting something more like "I've been told this happens, but I have no idea why"

 

[avkid]

It sounds like you need to rent a logging analyzer.

 

[JD]

Three venues, errrrr.....! Same spots each time? If so, and you feel comfortable doing it, try swapping the ballast tray from one of the known good ones with one of the known troublesome ones. (They are not hardwired and have two connectors that disconnect them.) If the problem follows the ballast, then there is a problem with some of them. If you have not been keeping track, simply add a piece of tape to one of the handles each time there is a problem. Reminds me of a problem I had on the old Ultra-arcs (or as derek calls them, Ultra-darks.) Three of the ballasts were just quarrelsome, and we ended up swapping them for rebuilds from Phoebus.

Since some of them work fine, I would look for what they have in common (age of lamps, grouping of serial numbers.) I have seen cases (not with this light) where ballasts from different manufacturing batches may be of different design and more / less tolerant of line voltage problems. Wish I could add more, but I have to stay generic on this one as I don't have any.

 

[SteveB]

And as follow on, I'd be on the horn to Lycian on Monday. They know stuff about their PS's that are alway's "educational" is my experience.
Steve B.

 

[mstaylor]

My first thought was they were all pulling from the same leg of power but since it was in three different venues then that's obviously not it. It will be interesting to find out the answer. Let us know after you talk to Lycian.

 

[porkchop]

We called Lycian a few weeks ago, they asked us your basic troubleshooting questions and they were really confused. Given the amount of spot problems we've had (this and others ones too) an oscilloscope and a bench DC power supply may be in our near future.