Driving the Iwasaki H50 with resistive ballast

Today i got the question by the Chemistry department 'Do we have another mercury vapor lamp'. Well, kinda, i just ordered those from CP lighting... But i'm still waiting for the mercury vapor street lighting to be torn down from the building so i can take out the ballasts and use them for the new lamps. Anywhere between now and januari next year...

But at least that prompted me to try out alternatives for if we really need the Iwasakis at short notice. The Iwasaki h50 is specified at 95v and 0,62a. I first characterised some incandescent lamps to see what current they pass at which voltage. The incandescents would be running at about 135v with the H50 dropping 95v.

At first i tried an 100w + a 60w in parallel. This resulted in too little power - the arc voltage didn't go past 80v, and the current not past 0,45a. Then 2x 100w - taken from my secret hoard of incandescents at work. They must truly be the last ones surviving at work.
This resulted in an almost perfect voltage and current - 0,61a and 92v. The current on warmup is 0,82a, a bit on the high side, but if you want you can drop it to half during warmup by using only one of the two incandescents. Gotta say, i've never measured how much current a HPMV draws when warming up on a conventional ballast choke. Might be even more - i'll do it soon.

It goes without saying that this is crazy inefficient - but for this purpose, efficiency is of less importance than just having something that works.

Either way - if you have one of those 50w HPMV lamps and you live in a 220-240v country, using two 100w incandescents in parallel appears to be a very good option. If you want the 'Appears to be' to disappear, you'll have to wait until i repeat the measurement with a true RMS multimeter. Bloody Reichelt still didn't ship my order... Then again, i have no clue if the current specified in Iwasaki's datasheet is measured with a conventional meter or a TRMS meter.