florescent starter circuit collapsing power supply?

[jack4192]

Hi guys I'm after some help. The device is in an ophthalmic piece of equipment which has 4 15 watt small fluorescent tubes powered from a board mounted electronic ballast. It has a 12vdc class 2 power supply. Problem is, even with new tubes it will fail to ignite the tubes if it's recently been turned off and on, leading to us having to leave it on all day. A few different power supplies have been tried, and even tried a ferrite core around the lead from the supply to the board as an oscilloscope shows lots of noise on the supply lead when the lamps are on. Any ideas? Thanks in advance.

 

[Gryd3]

Hi guys I'm after some help. The device is in an ophthalmic piece of equipment which has 4 15 watt small fluorescent tubes powered from a board mounted electronic ballast. It has a 12vdc class 2 power supply. Problem is, even with new tubes it will fail to ignite the tubes if it's recently been turned off and on, leading to us having to leave it on all day. A few different power supplies have been tried, and even tried a ferrite core around the lead from the supply to the board as an oscilloscope shows lots of noise on the supply lead when the lamps are on. Any ideas? Thanks in advance.

Sounds temperature related to me...
How long must it be left off to reignite?

 

[jack4192]

Sounds temperature related to me...
How long must it be left off to reignite?

The longer the more reliably it will strike at the flick of the switch. At least 3-5 minutes else it just flickers and never ignites the tubes.

 

[Arouse1973]

Does the tube have a starter?
Adam

 

[jack4192]

Does the tube have a starter?
Adam

No, there are 8 wires coming from the starter circuit, 2 each straight to the tubes. More scope work.today...

 

[Colin Mitchell]

The output from the flyback needs to be separated into 4 . . . so each tube starts separately.
You may be able to do with with 4 x 10n caps
One of the tubes is zenering the supply.

 

[KrisBlueNZ]

The output from the flyback needs to be separated into 4 . . . so each tube starts separately.

I guess there will be four independent oscillators on the ballast board, one for each tube. Are there four identical circuits, each with its own inductor, jack4192?

 

[jack4192]

This is the two starter boards, each one powering two tubes. Lots of emi going back down the power line to the dc power supply. Peak currently draw on startup measured at 14amps, the supply is 4amp max... Cheers.

 

[KrisBlueNZ]

OK, so the problem seems to occur if something is still warm when you try to start it up. Would it be useful to know whether it's the tubes or the ballasts that are causing problems when warm? You could try this: Disconnect the power to one ballast. Run the other ballast and its two lamps for a while, then quickly unplug the unit, reconnect the power connection to the disconnected ballast, swap the lamp plugs over, and power back on. So now the warm ballast will be trying to start the cold lamps, and the cold ballast will be trying to start the warm lamps. This might tell you whether it's a warm ballast, or warm lamps, that is causing the hard starting.

That won't be conclusive, and in any case you've said that it happens with brand new tubes, but it could be worth trying just in case it reveals something unexpected.

You said you've tried different power supplies. Do you mean different electronic ballasts? Or did you replace the 12V power supply? Are the ballasts powered from this 12V power supply?

Have you checked the DC output voltage of the 12V power supply?