Hey everyone, long time no see!

I'm trying to figure out what correct readings should I expect to get from my multi-meter at the socket terminals of a 175W Mercury Vapor set-up.

I recently acquired a Thomas Industries "Mini-Merc" fixture, and another member here I believe got one too. It's rated at 120V operation, both on the fixture label, H39 GE ballast and metal tag on the line wire.
The fixture won't fire up and doesn't even produces a buzz, it's dead silent. There is no visible damage or rust anywhere inside the can, everything looks like new.

At first I thought it was the two caps, so I bypassed them to no difference. They're wired together in parallel, and the cap bank in series with ballast and lamp. See diagram below.

Caps or no caps, I'm reading 120-125 volts at the socket terminals.

My only guess is something in the ballast came loose and isn't going through a winding or producing a magnetic field? Again, no buzz, no kick. But the ballast wires are tight, flexible and high-temp rated.

Neutral connection to the ballast or the coil section towards the Neutral is disconnected

I do not know what is wrong with your ballast but I'd like to say nice drawing skills....

Neutral connection to the ballast or the coil section towards the Neutral is disconnected

The common wirenut that connects mains, socket and ballast neutral is ok, I'm assuming you're talking about the tap / solder on the ballast that may have snapped or failed. I'm going to have a closer look at that

EDIT: If I can't get a good view / access of the tap, is there any other test I can try with the multi-meter to make sure the ballast is or isn't the problem?

Check if the Neutral wire from the ballast sees any connection at all to the others. I'd expect no more than 10 Ohms for good or >few kOhms...infinity for bad

If the Neutral end of the coil is damaged but is in the upper winding layer, you could unwind one turn to get to the wire to receonnnect it - It won't be too big difference. Dont let it touch the core though

You were absolutely right man, thank you!

Solder at neutral points under the sheet of paper / mica was the issue. There were scorch marks as evidence. Solder got two solid copper wires from the winding and a stranded copper wire bridging from one winding to the next.

The connection wasn't loose or broken but it looked like it tried to make contact before the solder point, and since the solid copper is varnished at that point, it sparked and arced a bit. I removed the solder and crimped them together and it's running quiet.

Funny thing, if the mercury lamp is running and you separate those neutral leads, the arc thins out and start to swirl and snake inside the tube.

Think what you are doing when you separate the wires...

The ballast have 2 coils :

 - Primary between Phase(P) and Neutral(N)

 - Secondary between some point on the primary to the Lamp(L). Let's assume (it isn't allways the case, but accurate enough for getting the concept), that the seconrary starts at P

Disregard the capacitors at this stage of the explanation



The ballast serves as 2 devices :

1. A transformer. The 120V applied at the primary (between P and N) is transformed into some voltage in the secondary (between L and P). The (initially unlit) lamp is connected between L and N, so it gets the sum L-P voltage + P-N voltage. This is the ballast Voc, on the order of 300..500V

2. An inductor. The secondary (between L and P) is not strongly coupled to the primary, so it does have its own magnetic circuit with inductance in it, which is to some extent not strictly tied to the primary

You could, for the most part, imagine it as 2 devices : a step up transformer stepping 120V to something like 300V..500V, and then a plain choke (like low power HPS choke) between the higher V and the lamp

Without Neutral connected in the ballast, the 1st function is disabled, only the 2nd remains. So without Neutral, the ballast behaves as a simple choke on 120V



When the Mercury lamp is not started yet, and ballast is without Neutral, it puts out 120V through a choke. As long as no current flows in the circuit, the choke have no effect at all, so on the output are unchanged 120V. They are not sufficient to start a lamp that was designed to probe-start at >300V

When the lamp jsut started, the pressure in it is low, and low pressure arcs take the most of the arctube volume

As the lamp is warming up, the pressure rises and the arc becomes a thin line

The warmed up lamp takes about 120V. The transformer function of the ballast supplies >300V. There is plenty of voltage headroom - voltage to drop on the choke part, which results in high current (it is equal to the voltage dropped on the choke divided by the choke impedance), and in stable operation (small dips in the line voltage will become small dips in the votlage on the choke, which dont matter much)

When you disconnect the Neutral, you turn the ballast back into a plain choke on 120V. If the arc would need slightly more than 120V, The lamp would have extinguished completely. But the voltage across the arc is still on the order of 120V, so there is virtually no voltage left to drop on the choke. Accordingly, the current becomes small too, resulting in a thinner and less stable arc. The lack of voltage headroom means additional impact on stability

I think you make 3 of us on here who own these fixtures.

This is why I love Ash explanations. Thanks so much for that detailed post!

Hey Miles!