I am a bigtime electronics nerd, and I can shamelessly spend way too much time digging into things that other people would deem silly. I found myself having a very weird behaving 2x49W T5 HO fixture recently. It uses a Philips HF Performer, HF-P 249 TL5 220-240. It works absolutely flawlessly when it's warm, but as soon as the temperature drops, it refuses to start. Not even a blink, or tube glowage, nothing. It does not even draw any wattage.

My obvious suspect is failing capacitors. So I took it apart, but to my surprise, the majority of capacitors are actually polyfilm, which more or less never go bad. There is just two electrolytics in the whole thing, the main filter cap for the rectified mains, and then a small one presumebly for filtering the supply going to the PFC chip. I yanked the small 47uF 35V cap which acts as a filter for the PFC chip to measure it, but it appears absolutely fine, so I will likely put it back. Now my suspect leans more towards a transistor or FET or something like that on the thing which may not like the cold. I plan on troubleshooting it more tomorrow after work, I think I will put it back together, and try to start it in the cold again. Assuming that it still does it, which I think it most certainly will, I think i will take my heatgun and heat around a little and see if we can pinpoint exactly what is wrong!

I just wanted to share a little about this, I am curious if there is anyone else out there who has repaired ballasts before, and I am also curious if anyone else has had a similar failure mode! this is definetely a peculiar failure!

At first I would thought maybe the tube was having cold temperature striking issues, but even that would probably draw some measurable current from the mains. Maybe it has some sort of protection circuit that turns off the ballast if the tube doesn't strike? What kind of temperatures are we talking about here in terms of cold and warm?

Likely to be the main electrolytic capacitor that's gone, I had this with a Tridonic 18W TC-D ballast, it worked perfectly when warm but refused to do anything when cold. When I held the ballast PCB over a heater for a few moments and then tried it, it came back to life immediately.
The EL cap did run noticeably hot when in use, but replacing it for a suitable equivalent restored the ballast to normal again.

There is a lot about the ESR of the caps and this changing when they dry out, and of course you ned to get a replacement one of the correct values, though at the moment a lot of that is above my understanding of the subject.
I do also wonder if this may become an issue with old NOS HF ballasts that have been stored for a long time, if the caps eventually dry out.
Hopefully your ballast should be repairable though as it has failed like this rather than with a "bang" which usually means something more serious!

The film capacitors in these ballasts actually do wear out, mainly the resonant capacitor parallel to the lamp (few nF rated at some 600..2000V). They tend to loose capacitance over time. Dunno the exact lamp circuit connection in this ballast, but it either causes insufficient preheat (so then the cold electrode operation triggers the EOL protection) or even inability to reach the resonance, so generally insufficient voltage to strike the lamp.
Fortunately this can be usually measured without disassembling the ballast (you just have to subtract few 100pF; capacitance reduced by some 20..30% is no uissue, problems start when it droops below 50% of the rated value and that you will see in circuit).

Also look carefuly for small electrolytics - some ballasts use them for EOL protection debouncer/filter. If these dry out, the protection kicks in prematurely, e.g. when the lamp is harder to start due to temperature.


And problem could be with the fixture construction and wiring: These lamps rely on the grounded metalwork near the lamp to provide an auxiliary external electrode that helps ignition. Without it, again mainly at low temperatures, the ignition voltage may become too high or the ignition take too long for the protection to kick in.

But also truth is, warming electrolytic capacitors reduces their ESR, so if some is dying, it pops up as a problem at low temperatures first.

The temperature is about -5°C right now outside, so we're not talking about crazy low temps. They should work down to -15°C, probably even colder, without any issue whatsoever. So we're definitely not outside of the temperature range here. The 49W tubes are pretty easy to strike in the cold from experience with other fixtures, and they seem to warm up pretty quick in enclosed luminaries like mine at home which is the one behaving weird.

I think my approach will be checking the main electrolytes with the heat gun. If that doesn't give any positive result, I'll try the resonance capacitor. The resonance cap is actually somewhat suspect already, because the tubes have what I'd call severe end blackening after just about 1000-2000 hours. That's literally nothing. I'd expect this level of end blackening at maybe 40000-50000 hours. I'll dig into it more after work today, I really want to learn what's wrong with it! and I want to fix it!


Yes indeed film caps do fail, and to be honest, I think that's what has happened somewhere! The elevtrolytics both checked fine :O

You said -5degC, that is not a "normal" range for generic electronics.
Is the ballast rated for this temperature? If not, that could be the main problem.
Because standard electrolytics (those explicitely not rated for freezing temperatures) increase their ESR rapidly once the temperature crosses 0degC (the electrolyte inside freezes, so stop being conductive, so won't provide the connection to the functional dielectric surface). Even when they are perfectly good - it is just their functional property.
So in such case the 0degC is really a limit where the function tends to deteriorate very steeply, it is literally water freezing inside of the components...

There are electrolytic capacitors rated down to -40degC, but they are explicitly rated for that range.

And even if so, to test they are good, you need to cool down the electrolytics to that low temperature (at least -5degC, where the system is acting up), higher temperature may mask out the high ESR failure and the capacitor may appear good when tested, even when in reality it is completely bad.

I believe the HF performers are rated for operation down to -20°C.

Today I bought two new ones, I gave up troubleshooting the one that was acting weird heh, instead, I've desoldered and tested every suspect component. I can't find anything wrong with any of them. I even checked the resonance capacitor, the main switching transistors, etc, and they're all fine. I think there might be a micro fracture in one of the solder traces on the board, because in some spots it looks slightly delaminated. Weird, this ballast has only about 2000 hours on it! But it is one of the earliest HF performers they made. I know they had some problems with the early generation.

The ones I got today are generation 3, which supposedly doesn't have any issues like this

I'll mount one of them and get the fixture back up and running this weekend.

But, I still want so badly to know what happened to my original HF performer.

Do you have an oscilloscope?

What is the voltage measured over main DC bus capacitor?

What is the name of main control chip, some Philps UBAxxxx have open datasheets.

The capacitor and ballast may have specific temperature ratings, but most effects  of temperature (ie. short of the electrolyte freezing solid) are gradual, and the point of cutoff chosen for the rating is fairly arbitrary

Together with additional factors - age of components, resistive condensate films forming (including on the tube), and other abnormal conditions, the actual non working range may end up within the rated range



For example :

Consider a ciruict where the small chip Vcc electrolytic is charging from the rectified line voltage through some high (in the Megs range) resistor. The current supplied by the resistor is not sufficient for the continuous operation of the chip

The chip has undervoltage with hysteresis. Once the cap charged to the high level of this hysteresis window, the chip starts using it, and there is enough energy in the cap, that the chip can start running the circuit before the cap discharges to the low level

What if the cap discharges so quickly (e.e due to capacity loss) or has so high voltage drop (due to ESR), that the chip never gets its starting voltage for long enough time to even generate the 1st output pulse. The circuit will appear dead, the power draw of the circuit will remain basically VRectifiedLine / Resistor



Measure whats up with the voltage on this cap. Evan if you dont have a scope, just DMM can give a general indication

I think there might be a micro fracture in one of the solder traces on the board, because in some spots it looks slightly delaminated. Weird, this ballast has only about 2000 hours on it! But it is one of the earliest HF performers they made. I know they had some problems with the early generation.

Yep, I also think a solder crack (and consequently also trace crack) is indeed quite hot candidate for a fault. It checks a lot of boxes:

Is acting up only at some temperatures
It appeared after some time
Mainly when the thing was operated outdoors, so went through wide temperature swing cycles (cold winter nights vs hot summer days or evenings) or even worse mechanical vibration (wind vibrating a pole,...)
Sounds like part early production using lead free solder (it took about the two decades to iron the reliability problems out and settle to some usable solder composition, still still there is some engineering work to do)
And it is the most frequent case of things failing on their own with modern electronic. And ironically one of the last to think about when diagnosing (I'm also guilty of overlooking that :-) )...

I think here is described what to look for: faint line circles around leads in solder joints are fatigue crack forming. Most prone are heavy components, but small regular ones aren't immune either, mainly when they have short stiff leads not allowing any give for thermal expansion of the board.
Reflowing them using quality flux and solder tends to solve the problem.
Of course be on watch for broken traces. If nothing exploded yet and mainly when the ballast is working at least sometimes, there is high chance the components are still good so the fix could be successful.