low pressure sodium lamps are really a constant wattage once run up, the combination of V and I equals the lamp power. a 36w on a 55w ballast, voltage is much lower as current is increased. same applies here. i did not measure the arc voltage as the current is very close

Might have to try this and see what happens.

james will know the properties of these lamps, the xenon must make a difference? to the running current of the discharge. what it says is the ballasts on their own do not provide the 145v required to keep the lamp going. always and usually half the supply voltage or less on sox gear - the higher wattages are different, but 90/35/55 can run on this

no ignitor here just a 155/800 as per flu tube setup

the ho lamp - the ignitor shares some current with the lamp and switches on and off at a higher frequency than the supply, it induces an induced higher voltage across the lamp. inductor - ac - fast switching equals induced voltage. btw this setup alone with thorn 90w sox ballast ignitor is the same, the lamp 90w sox extingushes it the ignitor is taken out of the circuit. only 35w can run without and ignitor, 55/90/36/66 need them

Is it running a la boost converter (multiple impulses throughout the half period, providing continuous high voltage) or just helping restrike the lamp with 1 pulse after each zero crossing ?

no just the two ballasts in parallel, i thought in theory 0.94A + 0.6A would be close, no ignitor used

I mean in the 55/90/36/66 SOX circuits you mentioned above

i am not sure i don't have a scope anymore, maybe fsb would know? i would guess it would depend on the ignitor, thorn ones use an scr? if i remember. philips triac / diac possibly. scr would be the positive half, triac would be both. correct me if i am wrong

i know for sure the old g53282/a & b use a c106m,

Half wave activation is sufficient as long as the ignitor is really only for starting the lamp. If the ignitor's assistance is required to keep the lamp running, then it must be working symmetrically in both half cycles, or else the lamp will rectify

The type of switch alone does not imply half wave operation, since the entire ignitor circuit could be placed after a rectifier bridge, so do both halfwaves with whatever type of switch

Triac would be suitable for HID ignitors, but it seems odd to me to have found one in a "2 wire" ignitor - The triac goes out only at the current zero crossing, so that would not be very useful to produce a back EMF from the ballast. In HID ignitors the pulse is ultimately ended by the emptying of a capacitor inside the ignitor, so it is not tied to the line zero crossing