flourescent ballast resistance

[stj]

what should a 13w magnetic ballast look like?
i have a setup playing dead here - ballast is reading 148ohms
that seems high but i have no other ballasts to compare against anymore!

[redwire]

I would measure its inductance. F14T5 is 14W 0.265A lamp current, need around 900ohms impedance, so 148R from the coil and maybe ~2.5H is needed. Using 240VAC and 40V tube drop for rough math.
Does the ballast have a winding for the filaments, I would check that as well. It might have gone open in the tube. Philips tubes give some tube specs but magnetic ballasts I could not find info on.

[stj]

no, it uses a glow-starter, just an oldschool 2 wire ballast

[stj]

component tester says: 141.6 ohms 3.35 henrys

hmm - doesnt sound like shorted windings.

[CapLeaker]

I just switched out a whole bunch of these old magnetic style ballast lighting system and converted it over to LED. From what I recall, those old ballasts were like drenched in a tar like substance inside making a repair very difficult. If they break, it’s due to shorted winding inside mostly. At least mine where always shorted when I had to deal with it. What is it T8 or even a T12? Got an insulation tester?

[redwire]

That seems reasonable. You can check if there is a ground fault, winding shorted to case.
If the overall startup current is too low I think the starter will just sit there trying to get the filaments to warm up. You can see if they are glowing at the tube end.

I would suspect the tube. Some you can light up on your bench, just power the filaments with a few volts and put at least 30V current-limited end-end on the tube. Newer tubes run out of mercury and go open-circuit (end-end), they need too much high voltage to start and then don't stay ionized (on) and flicker.

[CapLeaker]

The one I dealt with had always turn on / flickering problems. The magnetic ballast would be getting so hot, I had to cut a piece of the AC cable off as the insulation would fall off and black stuff would or has ooze(d) out of the ballast. The winding had always continuity... but the resistance was lower than that. It was rare to find an open magnetic ballast, at least for me.

[stj]

it's a philips PL tube,
totally dead - maybe it's a bad lampholder or broken wire.
the problem is it fits into a tight housing - and the light is used in all directions.
so no room to use an electronic ballast and the tube only has 2 exposed pins

i thought about winding led strip onto some copper water pipe to give 360' illumination, but it's a last resort.

[CapLeaker]

I am sure you know that already. Lol

[eccerr0r]

I have an extra magnetic ballast for F15T8 that I swapped out, DCR=26Ω

[stj]

that sounds more reasonable for a coil - thats why i started the thread

[stj]

Originally Posted by CapLeaker

I am sure you know that already. Lol

you asked for it.
this is the lid of a countertop vending machine
notice how the lamp is between the top housing and a colour filter to illuminate the goods.
ignore the tape and led lights - thats the owner trying to bodge some illumination into it - failed

[CapLeaker]

For what it’s worth, GE still makes magnetic ballasts for the NA market. I thought nobody made those magnetic ballasts anymore?

Interesting light fixture. There is not much room inside. I’d be tempted doing a led conversion.

Just googled Osram D8/827… I can see D10’s and D13 also on eBay. I guess that lamp is a D8?

[stj]

i think it's 9w
but i have lots of spare PL in 7w - 11w, they used to be popular in retrofitted stuff.

i had an idea,
i will short the 2 pins on the ballast,
then connect the mains cable to a pcb from a CFL.
if the lamp lights - the ballast is fucked.

[Per Hansson]

In one of DiodeGoneWild latest videos he tests a 36w CFL ballast:

https://youtu.be/B4YM_ljELl4?t=723

[redwire]

Oh it's cold cathode? Then it's just a matter of the starter/ballast making a kV pulse. Sometimes holding the tube or having it next to grounded sheet metal helps it ionize at lower voltages.
There must be something breaking down preventing the HV spike from happening, either the ballast winding's insulation or the starter, CFL socket etc.
What a hassle.

[stj]

it's not cold cathode, the tube has a glowplug built into the base

[redwire]

Pic shows two wires to the tube and the Osram datasheet showed two and said DULUX S is CFL. Is that the problem, wrong tube? I'm konfuzed

[stj]

CFL = compact flourescent lamp
it has the starter built into the base.

[CapLeaker]

well stj... you started it! Today at work I done nothing else besides fixing ceiling lights. Changing ballasts and t12 F40 U shape tubes.
These are old magnetic style rapid start ballasts supporting two T12 40W u shaped tubes. Each bad ballast had a winding shorted to a different one and the black tar like substance running out of it.
Good thing is, I ran out of florescent tubes and ballasts.