HMI Circuit Tripping Question

[Tommy Kraft]

Howdy ya'll

So, basic question since I'm still fairly new to HMI. Does anyone here ever have difficulty with an Arrisun 12 and 575/1200 electronic ballast occasionally tripping a 15amp household circuit when striking a 1200 watt lamp (with nothing else plugged in)? Sometimes the house breaker for the room trips, other times it strikes fine.

In interest of full disclosure, I ask because I'm using a Chinese clone of the Arrisun 12 (so let's skip the "don't buy Chinese stuff," I know the risks). So, I'd like to know if this issue is also common in the Arri version. I've never had this issue with their "economic" (read: cheaper) 1200 watt fresnel, but am occasionally seeing it with the PAR.

Thanks in advance for any insights!

 

[Les Hillis]

Some info here might be relevant -

http://www.lightlineelectronics.com/faq_hmi.html

"...Power Factor and Harmonic Current on the Neutral Wire
Until relatively recently many electronic ballasts were made without power factor correction circuits. Because of its heavily capacitive front end components, an electronic ballast puts current and voltage out of phase with one another. As a result, it typically has a power factor of 0.6 or less, meaning the ballast has to draw 40% more power than it uses. Associated with poor power factor are harmonic currents which build up on the neutral wire. Sixty-five to 80% of the current does not cancel out between phases, even when all hot legs are evenly loaded. This means that when operating a large number of electronic ballasts, the neutral wire will need to be doubled or even tripled to carry the additive current.

Power Gems ballasts (used by Strand, Leonetti, some Cinemills and LTM) and new Arri ballasts have power factor correction circuits. Lightmaker and Desisti (except the 18k and older ballasts by the others) do not. Power factor correction reduces heat, there-by increasing reliability; it uses power more efficiently and minimizes return current. It also weighs and costs a little more..."

If your ballast isn't properly power-factor-correcting and if, upon striking the lamp, the ballast is pulling 40% (or more?) power than it uses to then run the lamp, that could be tripping your circuit breakers.

Which ballast in particular is it? Among the Chinese ones there are some that are dedicated 1.2K and some that are dual 575/1.2K, that might make a difference too.

If the genuine 1.2K HMI PAR units never have these problems, then it's probably safe to assume there's something wrong with or lacking in the cheaper copies, and you'd hope it's not something that will result in quick or catastrophic failure.

 

[Tommy Kraft]

Some info here might be relevant -

http://www.lightlineelectronics.com/faq_hmi.html

"...Power Factor and Harmonic Current on the Neutral Wire
Until relatively recently many electronic ballasts were made without power factor correction circuits. Because of its heavily capacitive front end components, an electronic ballast puts current and voltage out of phase with one another. As a result, it typically has a power factor of 0.6 or less, meaning the ballast has to draw 40% more power than it uses. Associated with poor power factor are harmonic currents which build up on the neutral wire. Sixty-five to 80% of the current does not cancel out between phases, even when all hot legs are evenly loaded. This means that when operating a large number of electronic ballasts, the neutral wire will need to be doubled or even tripled to carry the additive current.

Power Gems ballasts (used by Strand, Leonetti, some Cinemills and LTM) and new Arri ballasts have power factor correction circuits. Lightmaker and Desisti (except the 18k and older ballasts by the others) do not. Power factor correction reduces heat, there-by increasing reliability; it uses power more efficiently and minimizes return current. It also weighs and costs a little more..."

If your ballast isn't properly power-factor-correcting and if, upon striking the lamp, the ballast is pulling 40% (or more?) power than it uses to then run the lamp, that could be tripping your circuit breakers.

Which ballast in particular is it? Among the Chinese ones there are some that are dedicated 1.2K and some that are dual 575/1.2K, that might make a difference too.

If the genuine 1.2K HMI PAR units never have these problems, then it's probably safe to assume there's something wrong with or lacking in the cheaper copies, and you'd hope it's not something that will result in quick or catastrophic failure.

Thanks, Les! It's a clone of the 1200/575 Arri electronic ballast, however the ballast is labeled as universal 575/1200/1800, which would indicate (if true) that they've engineered some part of it themselves, since Arri (as far as I know) doesn't offer a ballast that could power 575/1200/1800.

I guess if it *always* tripped the breaker it wouldn't concern me, it's the fact that sometimes it trips the breaker and sometimes it doesn't. And also fun fact, I did some measurements. After it had been running for about 20 minutes the average draw of the current at full power on a 1200 watt bulb stabilized to around 14 amps. On the cheaper Chinese ballast I have, current stabilizes to between 12.5-13.5. I've measured both the cheap ballast and this "clone" of the Arri as having a power factor of around .69-.72, which is definitely a far cry from Arri's .98.

I should also say that otherwise the light functions pretty great.

 

[Les Hillis]

There's a PowerGems 575/1.2/1.8K ballast they could have copied.

Even if the lamp starts most of the time, I'd be concerned about stresses being put on internal components of the ballast and lamp whenever it triggers your circuit-breaker.

All I can suggest, with my limited knowledge, is making sure you follow the same start up and shut-down procedure every time (ballast main switch on/off before wall power), and checking inside the lamp for burn marks or discolorations that could be signs of arcing/short-circuiting in the wiring (that might be happening when the voltage is boosted on start-up).

 

[Tommy Kraft]

Thanks much for the feedback, Les! Anyone else have any input?

 

[Josiah Duncan]

I use the off-brand Chinese version of the D12 pretty often. Replacing the bulb with a name brand such as Osram or Phillips is a must, but as far as cons you're only looking at the ballast not lasting as long as the Arri (expected) and it's not as efficient, pulling around 1500watts to provide 1200w worth of hmi. As Tommy said the Arri ballast efficiency is about .98 where I've measured mine to be about .80. I can live with this for now to save the extra $6000. Be careful not to hot re-strike though or you could damage the lamp or ballast.

Edit: To answer your original question, I've had no problems operating it on a regular 15amp household circuit without other appliances on the same circuit.

 

[Ulf Krentz]

Thanks much for the feedback, Les! Anyone else have any input?

Running mine on 240V so this is no real insight ;-) It is quite common that a circuit running at the edge of a limit like in your case sometimes triggers the fuse and sometimes not. Every starting process of an HMI bulb is a bit "random" until it reaches its regulated operating condition. Missing PFC of your unit is definitely in charge here. You might try a new (not higher rated!) fuse and see if it solves your problem. It is not advised to interrupt the starting process of a HMI device as it will shorten lamp life, always let it reach the full output for at least a few minutes.

Cheers, Ulf

 

[Tommy Kraft]

Running mine on 240V so this is no real insight ;-) It is quite common that a circuit running at the edge of a limit like in your case sometimes triggers the fuse and sometimes not. Every starting process of an HMI bulb is a bit "random" until it reaches its regulated operating condition. Missing PFC of your unit is definitely in charge here. You might try a new (not higher rated!) fuse and see if it solves your problem. It is not advised to interrupt the starting process of a HMI device as it will shorten lamp life, always let it reach the full output for at least a few minutes.

Cheers, Ulf

Thanks, Ulf! That makes me feel a little better. I will say that when the breaker trips it's always before the bulb actually ignites. I turn the ballast on, it runs for a second or two, and then the breaker trips. When the breaker doesn't trip, it still normally takes a couple of seconds between flipping the switch and the light actually igniting.

Regarding the fuse, I don't think that would apply here since we have a circuit breaker in our house rather than fused breaker.

 

[Tommy Kraft]

I use the off-brand Chinese version of the D12 pretty often. Replacing the bulb with a name brand such as Osram or Phillips is a must, but as far as cons you're only looking at the ballast not lasting as long as the Arri (expected) and it's not as efficient, pulling around 1500watts to provide 1200w worth of hmi. As Tommy said the Arri ballast efficiency is about .98 where I've measured mine to be about .80. I can live with this for now to save the extra $6000. Be careful not to hot re-strike though or you could damage the lamp or ballast.

Edit: To answer your original question, I've had no problems operating it on a regular 15amp household circuit without other appliances on the same circuit.

Thanks, Josiah! Mine actually came with an Osram UVS bulb so there's no issues there. The bulb they sent me seemed pretty new, as if they'd only fired it to do a test ignition.

Interestingly enough, their cheaper stamped-steel ballast actually draws less power than their supposed (and more expensive) Arri clone. And that one never trips my 15 amp breaker. They claim a 0.98 PFC on this "higher end" clone which is obviously not the case.

 

[Paul Dean]

Some info here might be relevant -
If your ballast isn't properly power-factor-correcting and if, upon striking the lamp, the ballast is pulling 40% (or more?) power than it uses to then run the lamp, that could be tripping your circuit breakers.

There seems to be a bit of confusion here. Electronic ballasts don’t strike 40% higher than their steady state load – magnetic ballasts do. Electronic ballasts ramp up to their steady state load. The load measurements don’t seem quite right either. Are you measuring with a true RMS meter? Are those measurements when the ballast is operating on the wall or on a generator? Non-pfc ballasts will draw more current on grid power than they will on a generator. For the best explanation for why that is, or for detailed information about HMIs, I would suggest you search the web for a white paper titled “Portable Generators in Motion Picture Production”. It has some of the best information on HMIs I have seen anywhere.

According to the nameplate on my Arri non-pfc 575/1200 ballast it draws at most 2290VA to generate 1200 Watts of light, which works out to be 19 Amps at 120 Volts (2290VA/120V = 19A) or a power factor of .52. I would be surprised if the Chinese knock-offs draw less. Check to see if your meter is measuring RMS current and let us know if it is.

Paul Dean

 

[Tommy Kraft]

There seems to be a bit of confusion here. Electronic ballasts don’t strike 40% higher than their steady state load – magnetic ballasts do. Electronic ballasts ramp up to their steady state load. The load measurements don’t seem quite right either. Are you measuring with a true RMS meter? Are those measurements when the ballast is operating on the wall or on a generator? Non-pfc ballasts will draw more current on grid power than they will on a generator. For the best explanation for why that is, or for detailed information about HMIs, I would suggest you search the web for a white paper titled “Portable Generators in Motion Picture Production”. It has some of the best information on HMIs I have seen anywhere.

According to the nameplate on my Arri non-pfc 575/1200 ballast it draws at most 2290VA to generate 1200 Watts of light, which works out to be 19 Amps at 120 Volts (2290VA/120V = 19A) or a power factor of .52. I would be surprised if the Chinese knock-offs draw less. Check to see if your meter is measuring RMS current and let us know if it is.

Paul Dean

Wow, that Arri ballast you named, is it a magnetic ballast? My Chinese one definitely doesn't draw that much. I'm measuring with a Kill A Watt, it seems to be fairly accurate. Granted, I haven't checked it against a high end ammeter, watt meter, or multimeter, but the numbers are believable, for whatever that's worth It claims to measure true RMS current, but I don't really know how to fact check that. When I plugged in my Chinese 1200 at full power these are the numbers it reported after running for about 20 minutes (this is from a couple days ago, so it's a more-or-less report ):

volts: 118
watts: 1150 (or real close to 1200)
amps: 14.06
voltamps: 1660
power factor: 0.72

As I said these are the rough numbers, I don't remember now exactly what they were, but it's pretty close. I've read that even electronic ballasts have a very short huge current inrush when you first turn them on, we're talking like less than a second of huge current intake. If that's the case, then perhaps sometimes when I'm turning it on using a 15 amp circuit, it's just luck of the draw for where I catch the AC cycle during that short and huge inrush that causes the breaker to trip.

I will also say it always trips before the light actually strikes, like it's charging up the capacitors to strike the light and it trips before firing.

 

[Josiah Duncan]

I use the off-brand Chinese version of the D12 pretty often. Replacing the bulb with a name brand such as Osram or Phillips is a must, but as far as cons you're only looking at the ballast not lasting as long as the Arri (expected) and it's not as efficient, pulling around 1500watts to provide 1200w worth of hmi. As Tommy said the Arri ballast efficiency is about .98 where I've measured mine to be about .80. I can live with this for now to save the extra $6000. Be careful not to hot re-strike though or you could damage the lamp or ballast.

Edit: To answer your original question, I've had no problems operating it on a regular 15amp household circuit without other appliances on the same circuit.

I should probably mention that this is an electronic ballast for clarity.

 

[Paul Dean]

Wow, that Arri ballast you named, is it a magnetic ballast? My Chinese one definitely doesn't draw that much. I'm measuring with a Kill A Watt, it seems to be fairly accurate. Granted, I haven't checked it against a high end ammeter, watt meter, or multimeter, but the numbers are believable, for whatever that's worth :

The specs I gave are for my Arri non-pfc electronic ballast. A magnetic ballast actually draws less after it has settled down from its’ high striking current. I wouldn’t put the greatest confidence in a $40 meter that claims it reads true RMS. You have to spend several hundred dollars to get a digital multi-meter that will read true RMS current. That white paper I mentioned has a very good section on why non-RMS reading meters will give you false readings.

If you are confident that your readings are correct, one explanation for why they are so much lower than my Arri ballast is that the Chinese ballast is purposely under powering the lamp. I have been told by a very reputable LA service shop that the Chinese 575/1200/1800 knock-off ballasts are not power factor corrected. Since they will trip a 20A circuit with an 1800W globe without power factor correction, the ballast purposely under powers the lamp. So your 1800W lamp does not put out 1800W and likewise for the 1200W lamps, which would account for why it is drawing less than expected.

Paul Dean

 

[Tommy Kraft]

The specs I gave are for my Arri non-pfc electronic ballast. A magnetic ballast actually draws less after it has settled down from its’ high striking current. I wouldn’t put the greatest confidence in a $40 meter that claims it reads true RMS. You have to spend several hundred dollars to get a digital multi-meter that will read true RMS current. That white paper I mentioned has a very good section on why non-RMS reading meters will give you false readings.

If you are confident that your readings are correct, one explanation for why they are so much lower than my Arri ballast is that the Chinese ballast is purposely under powering the lamp. I have been told by a very reputable LA service shop that the Chinese 575/1200/1800 knock-off ballasts are not power factor corrected. Since they will trip a 20A circuit with an 1800W globe without power factor correction, the ballast purposely under powers the lamp. So your 1800W lamp does not put out 1800W and likewise for the 1200W lamps, which would account for why it is drawing less than expected.

Paul Dean

I wouldn't be surprised if that were the case! And it's great to have feedback from qualified people. I'm fairly confident the Kill A Watt is more-or-less correct. Every time I fire the light (with 1200W bulb) on a 20A circuit everything works fine and dandy, but often times will trip during ignition (but before the bulb actually strikes) on the 15A circuit. The conclusion I've come to is that the ballast draws a huge initial inrush current to charge it's capacitors, and depending upon how full they already are (ie mostly dead) it will trip the 15A circuit whilst charging the capacitors.

Given the heat I've measured on the cable and wall outlet (about 10-15f higher than ambient), I'm pretty sure the meter is accurate when it's telling me that continuous running amperage is 14amps with a 1200 watt bulb at full brightness (the meter reads about 1100-1150W draw). According to the meter I'm also running on a power factor of .72, which is not at all surprising though the Chinese do claim PFC for this specific ballast (surprise surprise).

All that said, the ballast and head do function great otherwise. It would just be nice to have a solid PFC ballast that draws around the rated 11 amps. To be fair to the Chinese, though, their ballast is labeled on the back at a 13.8A draw. You get what you pay for and I suppose if it saves me a few thousand dollars I can live with a couple extra amps as long as I'm careful about what I plug it into.

 

[Dan Kanes]

You could try a Power Gems ballast.

 

[Tommy Kraft]

You could try a Power Gems ballast.

They do look like nice units but aren't they basically around the same price as the Arri 575/1200 ballasts?

 

[Dan Kanes]

I've seen them for a lot less, just gotta shop around, but it seems like you're satisfied with the current ballast, so I think it will all work out.

 

[Tommy Kraft]

I've seen them for a lot less, just gotta shop around, but it seems like you're satisfied with the current ballast, so I think it will all work out.

It's too bad they're not a little more common, they look like really great units! But yeah I'm pretty happy with the Chinese ballast so far. I have two of varying designs. They're not perfect by any means but they're decent and get the job done. They haven't (yet) exploded on me.

 

[Josiah Duncan]

Thought I should chime back in here again. After a few months of light (excuse the pun) use my Chinese Arri clone Electronic ballast (1200w), it burnt out randomly this past Sunday in the middle of a shoot. Luckily I have some extra Lowell Tota lights I had brought with me to match the power output and save the day, but my opinion of the system has changed completely. It wasn't under stress and it was a cool breezy 65 degrees out. Just tore it apart to see if theres a chance of repairing it but everything is a burnt mess (ceramic capacitors crumbled to the touch, not sure what connects where to repair it since that whole section of the board is fried).
So really if you plan on using these for a shoot have a backup plan, time to save up for an Arri.

 

[Nick Gardner]

Yeah, that's what the chinese ones do.

Nick