Very nice.!

Good move, HID lamps are getting more and more expensive... supply is quickly waning while demand has not yet been fully crushed by the big LED changeover. This year marks the end of the production of standard (HPI) and sports (MHD/MHN/HPI) quartz metal halide lamps at Philips Turnout. All that will remain there next year is the UHP manufacturing base as LDs/LEDs have not replaced those lamps in larger beamers yet (and the production line is really high tech, involving clean rooms, lasers and ultra-pure materials and processing, so not easily transferrable to China).

I wouldn't call the Philips CDO the pinnacle of HID lamp technology, but its certainly a very good one. It's peculiarity and difference compared to other ceramic metal halide lamps lies in its full compatibility with HPS systems. This means that it has to start and operate reliably at relatively low OCV and ignitor voltages (220 V and 2.5 kV, respectively) and can be dimmed down to 50 % of nominal wattage without extinguishing (function certified on electronic control gears only). To that end Philips reduced the argon fill pressure, shortened the arc gap, raised the mercury pressure, and used a sodium-rich salt mix (initially). The latter explains the lamp's pleasant light color. The model you have here is a relatively late one, optimized for performance and reliability, using a shaped arc tube. The first CDOs released in 2003 had the classic 5-piece barrel burner and were characterized by a lower light color temperature of 2800 K (more sodium and less rare earths) that made them easily identifiable in the wild (the red output from Na-Hg quasi-molecules in the plasma was particularly strong). Philips changed the CDO burner and the fill chemistry in late 2010, resulting in your lamp type and ending that characteristic extra-warm white light color (neutral-white CDO variants were even introduced to replace the original CDM-TTs which predate CDOs).

Maybe just pinnacle in my lighting collection then . Certainly one of my newer lamps. That's interesting you mention the compromise for being suitable for HPS replacement. I have started both of the lamps I have on old SON magnetic gear and they both started on the second attempt. I did find this surprising with a brand new lamp as starting only becomes more difficult with ageing. My guess is that these lamps may fail to start before eol where used as retrofit.

No, it's normal for some new lamps to be hard starters (even HPS ones). In your case that's the result of impurity leftovers from a rushed production process, a consequence of cost cutting measures introduced in the 2010s. Those lamps may be difficult to start initially (it's even mentioned in some Philips notices), but they'll ignite and run perfectly fine after the impurities have cleared up after some minutes/hours of operation.

Wow, in a strange way that's a pleasing bit of information. I had visions of replacing the ignitor with a new type but was resisting it as the gear is all original in my Thorn industrial fitting. There's no difficulty in ignition with an electronic ballast I have but it does have a more aggressive start up as I discovered when it destroyed my oldest Philips SON within a minute of starting.

Ooh err vicar, that's a bright one

Tuopeek - You destroyed an old HPS lamp with an electronic ballast? What happened?

About your Thorn system, is this an old/used one? If so, then I also wouldn't rule out the possibility of weak HV pulses due to degraded insulation or components.

Ah, a sad story indeed. I had a 250W Philips SON manufactured 1973 in Holland. It had many hours on it and was reaching eol. It was a great demonstrator of rectification at start up for the first few seconds where it would flicker at 50Hz and the ballast would hum and rattle angrily. I decided to see how it would behave on an HF electronic ballast. It appeared to run up perfectly and smoothly then at about 80% run up the light output suddenly dropped and the outer bulb darkened. On inspection the arc tube had broken into two pieces. It may be partly my fault as the ballast I was using was actually switchable and 250W was the lowest setting. I think the regime may have been to run up all lamps the same and limit the load and this may have been too much for the lamp. However, I have tried it on less valuable lamps and this has never occurred again.

I did wonder if the Thorn ignitor may have been out of spec as it is in a well-used unit. However, it seems to start SON lamps normally.

Personally, I wouldn't use those cheap no-name multi-watt hydroponic lamp ballasts with any of my lamps. By your description, it's likely that your SON-T 250W suffered from a fatal resonance at the driver's operating frequency (which may depend on lamp voltage, etc).

Very nice lamp. We still have a pile of them in our depot stores for use, there's only one 3-way column and one floodlight that uses them now (to my knowledge) on the inventory. I've found that they don't like some electronic SON ballasts that don't state CDO / MH on them, and just don't strike. All EM gear I've tried works fine on them, MH rated or not.
There's still plenty of assets about that run 70W CityWhite, which may include a few SON lanterns that I stuck CDO in for the fun of it
It's quite a mesmerising lamp to watch warm up too, with the shift of white to cyan and back to white, especially when the pinkish flares sometimes happen.

Wonder why the don't start on some electronic SON ballast. I guess it the pulse maybe just a bit too low for a new lamp as they appear difficult for the first few starts. Some electronic ballasts will only attempt a few starts per cycle to prevent continual firing of a dead lamp.

Maybe it's due to a combination of a too weak ignition pulse and a too low OCV (never ran into that issue so far). It must really depend on the driver, they all differ, even within a type produced by a manufacturer. I have a stock of Philips's excellent HID-DV CDO 70W and the older Dutch-made ones have different specs than the later Polish ones...