| Assume all components are actually good by themself (lamps not EOL, connections good, ballast not fried, ignitor not degraded and within its specs,...) and assume the problem is just the chosen combination of them.
What ignitor type and ballast are you using? I mean the configuration: Supetrimposed (3-wire box connected between the ballast output and lamp and 3'rd wire to N) or semiparallel (3-wire thing connected to Neutral, lamp-ballast connection and tap within the ballast)? My guess is the second (you in indicate the US in your profile). With that you can not simply take "a higher wattage ignitor" and expect it to work woth lower wattage ballast or lamps. With these the ignitor box is actually just part of the ignitor circuit (the part detecting lamp has ignited and generating low voltage pulses if not), the second part is the ballast winding (transforming the low voltage pulses to the high voltage needed for lamp ignition), so the "ignitor" needs to be matched with the "ballast".
I assume you are trying to use S56 ballast (that would be the closest I can think of to the electrical characteristics of two 70W 55V HPS in series). In your case there could actually be 3 possible problems:
- the wire is too long so has too high capacitance so the ignitor is not able to generate high enough voltage on it to actually strike the lamps. That would demonstrate as the ignitor firing, so the ballast will generate the characteristic ignitor buzz/clicking, but there is no traces of even faint discharge mainly at the lamp on the Neutral side. For this a long range ignitor should help, if available for the exact type of ballast you are using.
- The ballast OCV is too low to support two pair of electrodes holding cold cathode discharge. After ignition, a high current discharge needs to form with electrodes still cold (so exhibiting high cathode fall), only this discharge warms them up and transition the lamp into low voltage hot electrode operation (normally this takes about half second after ignition). The 70W HPS are designed to work at 120V OCV (because of the series choke ballast these are normally using), so the series combination would need 240V OCV. You can check that with a V-meter, but with the ignitor deactivated (its Neutral wire disconnected; so it won't flashover the multimeter). If the OCV is 230V and above, I would think this should not be the problem, but lower voltage could be. This would also correspond to the fact it worked. And if this is the case, at least some thin sparks from the ignitor should be present in both lamps.
- Ignitors normally sense whether the lamp has ignited via the voltage they see across the lamp. Lamp not ignited means the voltage is the ballast OCV, the ignitor sees it and fires. If the lamp ignites, it clamps the voltage low, below the ignitor threshold, so the ignitor stops firing and let the lamp burn on its own. The 1000V ignitor is designed for higher arc voltage, so its triggering voltage is set higher. It is possible your ballast OCV is below the threshold your ignitor is set for, so the ignitor "thinks" the lamp has already ignited and does not fire. This is another reason why you should always use ignitors designed to operate with the particular ballast type (not just the wattage, really the type; normally one ignitor is rated for range of ballasts, but that does not mean all other ballast, yours must be within that range). If this is the case, you won't see any ignitor activity at all (no characteristic sharp buzz, no traces of any sparks in any lamp,...).
|