Firing up 5k and 10k tungstens without a dimmer

[belford]

Hey guys

Is there any way to start 5k and 10k tungsten units (fresnels) without a dimmer?

Usually these will run over a dimmer board. But you don't have that on location. And carrying a small (haha) 10k dimmer with you also somewhat negates the benetits of not using HMI (tungsten: cheaper, no bulky ballasts required).

Starting the 5k is possible by flipping the breaker a few times in rapid succession until it's not tripping any more, but I consider this an absolute emergeny procedure. It's not good for the lamp and not good for the breakers. And it's just not possible with a 10k (this is pulling more than 43 Amps at 230 Volts when it's running! Not even to speak of the inrush current when starting it cold....)

So, how do you fire those up the easiest, cheapest and most portable way?

belford.

 

[JD]

Put it on a long cable. (An extra 100 foot #6 cable) The concept is simple, inrush is ten times operating current. A long cable will serve as a resistor and will drop the voltage during the start-up making it easier on the filament and delaying the breaker trip so that you get through the inrush.

 

[FMEng]

There are circuit breakers and fuses with slower overload response that should easily hold through the inrush of a cold tungsten filament. It's just a matter of using the correct fuse or breaker.

I would probably use a fused disconnect because they will tend to be more affordable. High current time delay fuses are pretty common.

 

[ariggs]

I know that Mole Richardson makes a 120V 100Amp switch. It's Type 5001762.

It's a lot cheaper than a dedicated 6k or 12k dimmer pack

Picture on the left side
-Riggs

 

[ship]

Factory stuff is often more economical - can contact other suppliers for distro if too expensive and they I'm sure can pre-figure the inrush currents to the breaker type used also. Mole often not cheap.

High inrush breakers on the other hand a good solution. Didn't do so on a 5K single fixture power supply I recently made and have not had problems with it so far but very likely a delay or high inrush breaker will have been a better choice in breaker.

One concept on the long run however, a switch unless designed specifically as it ain't a circuit breaker. Use a circiut breaker as a switch too much and you tend to burn it up. Better to have both as per design of each.

 

[belford]

Hi guys

Using long cables is not an option: those need to be lugged around as well and are heavy, just like dimmers. Besides, this will mean that the lamp won't run at 100% output - bad when you need constant color temperature.

I'm not so much concerned about the tripping breaker per se, more about the stress it puts on A) the power grid (the voltage dip from the inrush current can extiguish HMI lights running on the same phase, for example) and B) the lamp itself. AFAIK, especially 10 kWs can fail early and in a spectacular way due to the (mechanical) forces generated by the inrush current. Also this puts a lot of (electrical) stress on all contacts and connectors. I've always wondered whether the switches on the 5k and 10k fresnels are actually just for switching them off or taking them off the dimmer grid for lamp changes? Anyone knows anything definitive here?

Is there something like a dedicated soft-start circuit available that is not a full fledged dimmer?

belford.

 

[fredthe]

Is there something like a dedicated soft-start circuit available that is not a full fledged dimmer?

Perhaps just a simple diode, with a manual bypass switch? That would at least cut the power in half.

-Fred

 

[Chris15]

Inductance.
Lots of it.

Since the voltage a cross an inductor is equal to L di/dt, the current through an inductor must remain continuous. It can't jump suddenly. This is why they are used on the outputs of phase chopping dimmers, to smoothe the curve...

Inductor goes in series with the load. You may need to add a parallel or series capacitance to return the whole thing to having a reasonable power factor.. The value and configuration will require some analysis but it's easily enough achieved...

 

[mrb]

i think youre looking into it too much. Movie guys run these lights on straight power more often than not. The thing is making sure your supply is sufficient. You dont want to be firing up a 10k light off something tapped into a 100amp panel. And forget about supplying them from any residential power.... get a generator if you need it.

 

[Edrick]

I was going to say we constantly run 10k lights on set and off location. As far as connecting it to a longer cable (100 AMP Whip) and it effecting color temperature I don't believe I've ever heard of it being an issue and in film we're more concerned about lighting temp but we regularly plug it into 25 / 50 foot whips. The issue of voltage drop is an issue over distance with a small conductor but if it's sized right which it should be anyways it wont be an issue.

 

[belford]

ricky, sure... I don't doubt that.... but when you use a cable actually long (and thin) enough (i.e. relatively high resistance) to "soft start" a 10k (as suggested by JD), I guess you _will_ have quite some voltage drop...?

So you're starting the 10Ks directly off the power grid, no problems (except breakers)?

belford

 

[mrb]

another thing worth mentioning, is proper movie lighting power distros have high inrush hydraulic magnetic breakers which are able to start a cold 10k lamp without tripping. If you are using a home made power distro with standard (thermal magnetic) breakers, or a distro from one of the many places making power distros who are also using standard circuit breakers you will have problems. If you are using the proper gear, it should work fine.

 

[JD]

The voltage drop would be very minimal as you would use cable of the proper gauge. The drop would occur at inrush as compared to operation. Remember, a lamp drawing 43 amps run current will draw in excess of 430 amps at turn on. Under no circumstances am I recommending that you use anything thinner than the required gauge.

The biggest danger in turning the units directly on is to your budget! Lamps in the 5k to 10k range are not cheap! An extra 50 to 100 feet of cable will soften the blow on the lamp as well as help you thread the needle on circuit breaker response times.

I used to run into this problem with ACLs (4559) in rock shows which put a full 2400 watt load on the dimmer of the same capacity. In most cases, we ran the chains off of larger dimmers, but on occasion had to run them off a 20 amp dimmer. Many EDI packs of the day (1980s) used a very responsive breaker that would trip. Throwing an extra 50 foot loop of 12/3 on the circuit was a pretty reliable fix. No apparent color change.

 

[belford]

JD, yes they are expensive, especially the 10ks...
Ok so I had misunderstood your original suggestion... makes sense then. Just wasn't sure whether a cable of appropriate gauge would suffice to soften the blow.

mrb, of course you're right... with proper distro boxes this is not a problem. I was more thinking of cases where you run the lights directly off the grid; not as uncommon as one might think, especially on smaller sets. I've seen 18Ks (HMI) being run directly from a nearby 3-phase 64 Amp CEE socket, with no distro box in between... That's a scenario where you probably would have some problems with a 10k tungsten if you weren't really careful, for example...

Thanks for the suggestions so far!

belford.

 

[FMEng]

I was just thinking about this again. The gentlest way to switch these big lamps on is to do it at the zero crossing of the AC cycle. Doing it that way would reduce the inrush as the filament would heat on the low part of the voltage waveform and get up to its hot resistance before the voltage nears the peak. I suspect that filaments are less likely to blow at turn on that way, too.

There is a relatively easy way to do this. There are a number of industrial, solid state relays on the market that are designed to switch on at the zero crossing. They could be built into a distro panel or other enclosure, and controlled by low voltage. The SSR will cost around $125 to $300. The Newark Electronics catalog has a pile of them.

 

[mrb]

I was just thinking about this again. The gentlest way to switch these big lamps on is to do it at the zero crossing of the AC cycle. Doing it that way would reduce the inrush as the filament would heat on the low part of the voltage waveform and get up to its hot resistance before the voltage nears the peak. I suspect that filaments are less likely to blow at turn on that way, too.

There is a relatively easy way to do this. There are a number of industrial, solid state relays on the market that are designed to switch on at the zero crossing. They could be built into a distro panel or other enclosure, and controlled by low voltage. The SSR will cost around $125 to $300. The Newark Electronics catalog has a pile of them.

I dont think that would do too much. The warmup (and time to full resistance) time of the filament is many, many cycles. I guess there is a benefit of switching it at the zero crossing vs closing a breaker into the load -but the proper breakers are made for it. Come to think of it, last time I used a fixture this large it had a breaker built into it as a switch -it was a mole i think.