Remote ballast problem

[mengelman]

I was called to a site today with what I thought would be a simple problem. I am the second electrician to work on this. The owner said the other guys put in 5 hours and gave up.
It is a two sided sign with one 8' HO lamp on each side. The ballast is mounted about 40 to 50 feet away with #12 wires from the ballast to the sockets. The ballast is an electronic unit for 2 8' T12 HO. I have checked all the wiring I have checked continuity from one socket lead through the lamp filiments and back to the other lead. I have replaced both lamps and ballast. The thing will not light up. Any ideas?
Mike

 

[kwired]

What happens if you move the ballast (a lot) closer to the lamps?

I don't think you are supposed to have them that far apart, but don't really know the reason why.

 

[Earl Grey]

Ballast case grounded?

 

[mengelman]

What happens if you move the ballast (a lot) closer to the lamps?

I don't think you are supposed to have them that far apart, but don't really know the reason why.

There is nowhere to mount the ballast close to the lamps. I have seen remote ballast in HID applications but I cant see why this would not work. It worked at one time with an old magnetic ballast.

 

[TNBaer]

Have you replaced the magnetic ballast with an electronic ballast?

 

[GoldDigger]

I was called to a site today with what I thought would be a simple problem. I am the second electrician to work on this. The owner said the other guys put in 5 hours and gave up.
It is a two sided sign with one 8' HO lamp on each side. The ballast is mounted about 40 to 50 feet away with #12 wires from the ballast to the sockets. The ballast is an electronic unit for 2 8' T12 HO. I have checked all the wiring I have checked continuity from one socket lead through the lamp filiments and back to the other lead. I have replaced both lamps and ballast. The thing will not light up. Any ideas?
Mike

Some lighting fixture/ballast combinations rely on the presence of grounded metal near the full length of the tube in order to get it to strike initially. If the T12 tubes are mounted in empty space or there is no effective ground connection to metal parts close to the tube this could cause a problem.
The only other question I would ask is whether the ballast is one which powers the filaments or is a rapid-start type that starts the tube with high voltage without first heating the filaments. There may be a mismatch among the type of tube selected, the type of ballast selected, and the way the tubes are physically mounted.
That is, IMHO, more likely than a problem simply from the remote mounting. However, if the ballast is trying to generate a high voltage pulse to the tube or is using high frequency drive to the tube, the extra capacitance and inductance of the long run of #12 wire may be filtering out that pulse.

Replacing the electronic HO ballast with a magnetic HO ballast if you can find one would be a good thing to try if the existing electronic ballast is high frequency.

Can you try temporarily moving the ballast close to the tubes using external temporary wiring to test whether the distance really is part of the problem?

 

[kwired]

Read page 1-12 here.

As Golddigger mentioned high frequency and long lead length will have increased capacitance in the conductors causing operation issues with electronic ballasts.

 

[GoldDigger]

Read page 1-12 here[/URL

For the most forgiving of the Phillips electronic ballasts, the longest lead length (page 1-15 ff) is 20' and for most ballasts some of the wires cannot be longer than 4'. The OP's 50' separation is just too far.
A magnetic ballast will not have this limitation, but may have more visible 120Hz flicker in the sign when a single tube is used, will have more ballast noise at the point where it is located, and may have lower efficiency than an electronic ballast mounted within its distance specs. (Clearly the efficiency of the one he has now is questionable, since no light is produced.)

 

[broadgage]

Agree, the distance is much too great for an electronic ballast to work correctly, or even at all.
The output of an electronic ballast is at a very high frequency and therefore very liable to capacitive losses between conductors and inductive losses along the conductor.
It may have worked for a bit until the extra losses killed the ballast or it might never have worked.

Options include going back to a magnetic ballast which should be fine, or putting a new electronic ballast in the sign, or in a nearby weatherproof enclosure if there is not room in the sign.