I recently purchased an Osram BSH400ZC-K 400W HPSV/MH dual-rated pulse start ballast as seen below (image not mine).


And I've heard you can run most MV lamps on probe-start metal halide ballasts as they have similar arc characteristics. A probe-start ballast is a simple inductor with only one winding and no ignitor. So it should be possible to run the pulse-start ballast as PBMH-equivalent by using only the main winding without the tap in series and removing the ignitor to avoid destroying the PBMH electrodes.

I haven't yet bought the 400W merc cause I don't want to have a lamp and no way of running it, and being forced to spend a ton to get a good ballast for it. As per my observation, the ballast seems to run quite a bit hotter to the touch with a sodium vapor lamp so I take it that it's better suited for the arc of a MH lamp, ie. a slightly modified MV lamp; which is to say the ballast seems better optimized for metal halides. Note: as per the ignitor (SN58/02)'s diagram I'm following, I don't get a say in what voltage goes to the ballast (220 or 240), since that part of the winding is used to generate a pulse.

So my question today is, how safe and/or realistic is it to run a 400W merc using a 400W HPS/MH ballast as a series inductor? Has anyone tried it before?

This ballast is for 400W HPS lamps and MH lamps that operates on HPS ballasts, like the Osram HQI-BT 400W/D. It would overdrive probe-start MH lamps and pulse-start MH lamps that working on MV gear like the Philips HPI-BUS 400W.

Well, it would work, but over ride the mercury arc, resulting in the lamp running notably more green, plus the arc voltages are all wrong, average running specs are below....
400 W mercury: 135 arc volts at 3.25A
400 W sodium:   105 arc volts at 4.4A 👍

Even though the visibly greener arc is tempting, I'll refrain from it this time. That seems to be a massive difference (30V) and would probably damage the arc tube.

MH lamps that operates on HPS ballasts, like the Osram HQI-BT 400W/D

Yep this is accurate, as per this lamp's datasheet it operates much closer to the 100V of an HPS lamp.

Thank you to @AngryHorse and @dor123 for the insight   .

I've run plenty of 250 mercury lamps on 250 watt HPS gear, even decent older makes such as a Trulite, it didn't do them any specific harm, but for short testing purposes I wouldn't worry too much 😎

I think if you want to run 400W mercury vapor lamps on a 400W high pressure sodium reactor ballast, I think the best way to do that is to use a variac to adjust the voltage such that the mercury vapor lamp runs at about 3.3A or so.

That seems to be a massive difference (30V) and would probably damage the arc tube.
Thank you to @AngryHorse and @dor123 for the insight   .

You seems to misunderstand this. All things the same, higher arctube voltage means lower circuit current. BUT HPS ballast also has circa 15% lower impedance vs HPM one (39 Ohms vs 45 Ohms for 220V line). How much lamp current will increase is *notoriously* hard to calculate numerically, but let's say about 10-15%. HPM lamps are quite tolerant, but it is no good for the long term. You can look up this handy table from Chinese manufacturer, or dig up proper IEC papers, IEC60188 and IEC60662

Note that increased arctube voltage is good for the ballast, meaning less voltage is dropping across the ballast, when things are in reverse you have to account for possible core saturation and even more current increase!