THORN OM CSI 1000W metal halide lamp

I believe you may want to determine the cause of the "noise". Could it be that the control voltage is too low for the troublesome relay? And while relays are inexpensive, you may have better luck with a solid state (electronic) relay which can be more forgiving about variations in control voltage.

If on the other hand, you mean to say ballast instead of relay then we are talking about two distinctly different items. And the ballast and starter (igniter) need to be matched up to the lamp type.

I'd be happy to advise but I need to be sure we are talking about the same thing. Here is a site about how metal halide lamps work. They are part of a family of High Intensity Discharge (HID) lamps.

thanks very much Joe.

the problem is relay ( which i think is acting as ignitor here) I guess because if I change it back to old relay, the whole things works and we dont have any information on old relay as the lamp is very old and most of old relay have started failing.

the components of the holder are:

1.High Voltage Pulse Transformer

2.2X capacitors (230 nF, 1000V)

3. 1X capacitor (4700PF, 3KV) , these I put new one, earlier 5000PF, 800V)

4. 1XResistor (4700ohm, 2W)

5. 1XResistor (5500000ohm, 23 W)

I have drawn a circuit diagram as well. The lamp holder type is G38 and globe used is SPL 1000, CSI 1000 PAR 64, Mogul Bi Post type.

I will really appreciate any help.

regards.

I want to know what are the alternatives part I can use in place of old relay which can work because th elamp is not working with new relay.

Regards.

Hello zoombash,

Sorry for the delay. I was off to a seminar. I was looking at your schematic and even switch to my bifocal glasses but I am still having a little trouble making out some of the details. But for now, let me address a couple of general issues that may accelerate your progress.

You mentioned replacing a 5000 PF capacitor rated at 3000 volt with one rated at 800 volts. That component may have already failed due to its lower rating. The dielectric (insulating layer) will frequently develop short circuits internally when the dielectric strength (rating) is exceeded. So you may need to replace it again. You can always go higher than the original equipment rating, but going lower is likely to backfire since the original manufacturer probably was using the cheapest components he could find.

It looked to me like the relay in your drawing was the small square device in the upper left corner. If so, I am puzzled as to the connections. Generally a smaller voltage is used to power the electromagnetic coil in order to control a larger (or isolated) voltage of another circuit. In that case, a wire would be missing. It could, however, share a common connection which would make a three wire device as drawn.

The general description of the whole circuit would be as follows. There are two circuits that are passively selected by the voltage applied to the lamp. That is to say that before the lamp starts, the impedance is so high that it can be excluded momentarily from consideration. That leaves only the starter circuit (using the relay). But at the same time the open circuit voltage at the lamp terminals would be the same as the supply line voltage (unless the ballast is wired to step it up to match the lamp characteristics during normal operation)

It appears that this relay may be powered by an RC (resistor-capacitor) timing circuit that will open and close the circuit that sends power through the pulse transformer. This transformer is also the ballast which must limit the current that the lamp can draw after the arc has been established. But before the lamp starts, the opening and closing of the relay contacts would send power through a section of the ballast that would step up the voltage (due to the sudden interruption of current) and hopefully initiate the arc. The spike in voltage during starting would easily exceed 1000 volts.

Referring to your first description, the noise you heard may have been the relay buzzing which would make it difficult to start the lamp because each time the contacts close it steals power from the "run" section of the circuit in order to set up a pulse spike necessary to ignite the arc. I would expect the relay to open and close no more than a few times per second.

Other types of starters uses a small bi-metallic strip of metal enclosed in a small glass envelop. The bi-metalic strip is heated when the gas around it ionizes. As the bi-metalic strip heats up, it bends and contacts another electrode inside the glass envelop, causing a short circuit inside the envelop and extinguishing the glow or discharge. This short circuit also is also across the lamp (which would estinguish it, had it been running at the time) so that the ballast is now in maximum current mode. When the bi-metalic contact cools and opens the circuit, the interruption of current causes the inductor (i.e. ballast) to raise the voltage (trying to maintain the current flow) and a very high open circuit voltage is applied to both the lamp and the starter circuit. If the lamp does start, the normal lamp voltage is lower than the starter voltage so it won't ionize and heat up the bi-metal strip.

I think the relay in your circuit works in a similar fashion. When the lamp starts, there will be insufficient voltage to the relay coil to cause it to operate. But if the capacitors have aged then the normal firing of that relay may be at a faster rate (i.e. the buzz instead of a click - click -click) as it attempts to initiate the arc.

You may have to set up some off line tests to see if the new relay has the same coil characteristics that you need for this circuit. Otherwise, you may need to change the value of the capacitors to make it "click" rapidly, not buzz. I hope this helps.

Thanks Joe, Ealrier there was 800V capapcitor,I replaced it with 3KV one.

I putting a new and old relay photograph. The thing is if I replace old relay with new it does not work but agin if I put old one it works.The new one is rated as 240V, 10 Amps...Can I use pulse ignitor(tridonic atco 1000W) in my scheme of things. I am mechanical engineer not electrical.

Thanks again for your help.

Hello again zoombash,

Ah, good! I feel better about the capacitor. I don't know anything about that particular ignitor. I would say that as long as it is a two wire device, you can try it. The worst thing that could happen is that it would start the lamp and then attempt to keep running, causing it to heat up and maybe begin the start cycle again. Or, it just might not be designed to operate at your supply voltage.

Speaking of which, please examine the writing on the coil of both of your relays and confirm they are the same operating voltage. The operating voltage is not the same as the rated voltage and current. Ratings are about what the contacts can handle. Note that there is a functional difference in AC and DC coil designs in relays. While they may be the same voltage, there are sometimes choices in the coil design that use lighter wire (higher impedance) or heaver wire (lower impedance).

If you can determine the impedance (inductance) of the old relay coil, then you should be able to match it. It looked to me that the inductive reactance of the relay coil was part of the timing circuit used to drive the ignition circuit.

Also, you might consider visiting a commercial supplier of lighting supplies for some more immediate assistance. They can tell you if your tridonic atco 1000W ignitor will work. I hope this will help you make some progress!

Thanks alot Joe..will get back to u..

Im a Engineer at a International Sports stadium in Sri lanka. I want to purchase a quantity of 100 Ignitors (AME 47077) for 1000w Thron metal halide lamp type No. KW CSI -99-1222. In Sri lanka, I couldn't find a supplier. It's a great help if somebody can mention a supplier or a method I can find these ignitors.