Hi,

could someone please tell me if a standard Thorn 35 W SOX can be run on a Philips EXC 036 S-50, which is intended for 36 W SOX-E? If it can, are there any risks involved?

Many thanks!

As a general rule of thumb, no. SOX-E lamps can be run on standard SOX ballasts, for example a 26W SOX-E lamp could be run on a standard 35W SOX ballast (although at somewhat reduced efficiency), but not the other way around.

There may be exceptions, but I am not aware of any; hopefully someone more knowledgeable will be able to help.

The E36 is actually the 55s replacement, so yes it will run and should be ok, as the 35/55 were compatible on the same gear 

Some gear, but there were 35W and 55W ballasts that weren't dual rated, the one in our Gamma 6, for example.

Many thanks guys!

In the datasheet, it doesn't say anything about 55 W, so I'm not sure if this is really the same as a driver for both 35 W and 55 W SOX?

Philips EXC 036 S-50 Datasheet

Also I'm confused, since in the manual it says

Electronic drivers for SOX lamps; designed to save energy and reduce maintenance
costs

but a bit futher down,

Lamp Type SOX-E

and it just keeps on mentioning SOX-E, but never just SOX. If I give it a try, what could happen? Could the driver and/or the lamp get destroyed or lose some of its expected life time?

Yes all SOX-E lamps can be run on standard SOX ballasts, but their efficacy is reduced.  The thermal insulation of SOX-E lamps is so much better that they run slightly too hot on regular SOX ballasts.  However, a SOX-E lamp on SOX gear is still more efficient than SOX on SOX gear.

When the SOX-E was introduced by Philips in 1983, it was together with the dedicated new SOX-E ballasts.  However the new ballasts were not widely used in all countries.  For instance when the British Thorn and GEC introduced SOX-E, for the first few years they were promoted exclusively for use on standard SOX ballasts.  Then Philips also began listing both options in its catalogues and often gave as many as five different sets of data for its SOX-E lamps - relating to the quite different performance on both SOX and SOX-E gear of the leak transformer, inductive choke, optimal hybrid and squarewave electronic ballasts.

Yes all SOX-E lamps can be run on standard SOX ballasts

Thank you James, but my question is: Can I run a standard SOX lamp on a SOX-E ballast? To be more specific, can I run a 35 W SOX lamp on a Philips EXC 036 S-50 SOX-E ballast?

Ok, back home now with some electrical specs!
35 watt SOX run at: 0.63A (lamp current), 66V (lamp volts)
36 watt SOX-E at:    0.36A (lamp current), 114V (lamp volts) ‘Philips lighting catalogue 1994’
So not a perfect match, but should run the 35 fairly well.

35 watt SOX run at: 0.63A (lamp current), 66V (lamp volts)
36 watt SOX-E at:    0.36A (lamp current), 114V (lamp volts) ‘Philips lighting catalogue 1994’
So not a perfect match, but should run the 35 fairly well.

Thank you AngryHorse!

Now I'm even more confused, though. 

SOX: 0.63 A * 66 V = 41,58 W
SOX-E: 0.36 A * 114 V = 41.04 W

Neither of these is 35 or 36 W. Why not?

Also I wonder: When I run the SOX lamp with the SOX-E driver, then the lamp gets only about half the expected current, but about twice the expected voltage, right? Or do these values change when the driver sees an unexpected load? What could happen to lamp and driver?

The 41s are your gear losses, the catalogue gives figures for wire wound magnetic hybrid and optimal chokes.
A 35 watt SOX isn’t 35 watt!, they, (along with their running gear) actually draw 47 watts from your supply, (these are your ‘gear losses’, again this information comes from Philips’s own control gear schedules in the catalogue).
A lot of people don’t fully understand that discharge lamps are not as efficient as they make out to be!
For instance, a 35 watt SOX is something like 4600 lumens, 4600%35= 131 lm/W, but their not, including the gear losses it’s then 4600%47= only 97lm/W. So SOX lamps are not as efficient as they ‘look’ on paper!

When you say ‘driver’ do you mean HF ballast?, that changes the running figures again, but I don’t have specs for HF ballasts?

You can not calculate discharge lamp power just by multiplying lamp RMS voltage to lamp RMS current. Both waveforms are distorted, with voltage being specifically non-sinusoidal, so the result will be only a rough estimate. This have nothing to do with ballast losses.

For example, usual 40W T12 tube is often specified at 103V 0.43A, which will not give you 40W, either.

The only proper way to measure lamp power is to hook an expensive  high quality Watt-metter which integrates momentary Vlamp * Ilamp over a line period (or more) to get power value. May be a challenge for rapid start circuits, in the presence of HID ignitor, etc.

Sorry for misunderstanding the question!  Yes it’s also perfectly fine to run a SOX lamp on a SOX-E ballast but it will not run up fully.  The power, light output and efficacy will all be much lower, but it is a safe way to run SOX lamps in dimmed state.  Note also that the life will be somewhat reduced, because the discharge will take a lot longer to run up, which increases the stressful period for the lamp while the neon-argon gasfilling is being ionised.

Also although the current delivered by the SOX-E ballast is only about 60% of the full SOX current, this does not mean that the SOX power on SOX-E ballast will also drop to as low as 60%.  This is due to the fact that voltage of LPS lamps decreases during the later phases of run-up (whereas for allnother HID lamps the voltage increases).  Therefore, due to the proportionally higher voltage of a partially run-up SOX lamp the power saving of running a SOX lamp on a SOX-E ballast is not so great as using a SOX-E lamp.  And finally, as RRK said this cannot be calculated from a simple volts * amps calculation.  However, if you apply a correction factor of about 0.85 you will get closer to the actual lamp power - this correction factor does vary though as lamps are dimmed.

Thanks guys for your helpful answers!

including the gear losses it’s then 4600%47= only 97lm/W. So SOX lamps are not as efficient as they ‘look’ on paper!

I see, but doesn't the same apply to other things like LEDs, which also have lossy gear (i.e. constant current supplies)?

When you say ‘driver’ do you mean HF ballast?, that changes the running figures again, but I don’t have specs for HF ballasts?

Sorry, I was speaking of the Philips EXC 036 S-50. In the datasheet, Philips calls that an "electronic driver", I guess because it combines an electronic ballast and the starter electronics, is that right?

You can not calculate discharge lamp power just by multiplying lamp RMS voltage to lamp RMS current. Both waveforms are distorted, with voltage being specifically non-sinusoidal, so the result will be only a rough estimate. This have nothing to do with ballast losses.

Thank you RRK, now it makes sense to me!

Sorry for misunderstanding the question!  Yes it’s also perfectly fine to run a SOX lamp on a SOX-E ballast but it will not run up fully.  The power, light output and efficacy will all be much lower, but it is a safe way to run SOX lamps in dimmed state.  Note also that the life will be somewhat reduced, because the discharge will take a lot longer to run up, which increases the stressful period for the lamp while the neon-argon gasfilling is being ionised.

That is great, thank you James! I'm ok with the lamps lasting a little bit shorter. What would you say about the Philips EXC 036 S-50 when running a standard SOX? Could it get damaged or live shorter as well?

Yeah, when people say driver, I automatically think of either HF discharge gear or an LED driver, an LED driver is a slightly different ball game than a discharge driver so I think LED gear ‘losses’ are minimal to nothing as far as I understand?

I knew that James would come up with a detailed answer.