repair desktop fluorescent lamp

[sender jones]

I have a National desktop fluorescent lamp that failed to work. Every time I press the power button, it blink once and doesn't light up. When I then press the button again (intended to put it to off state), it blink once again after 8 seconds. I trace the PCB and found that it is almost the same as this one.

except that the values of the components are different. The C3 is 1.25kV 5.1nF while the C6 is 400V 470nF. I suspect that it is the C3 that failed. Photo here

I don't have exact replacement, but I have two 630V 10nF

and thinking of connecting them in series to replace C3. I think the calculation is correct, but I don't know if the type of capacitor and the current loading match as replacement or not. The 10nF is much smaller in size. Can somebody please help me on this.

 

[Nigel Goodwin]

Why do you suspect C3? - far more likely would be a duff tube, have you tried replacing the tube?.

 

[sender jones]

Thanks for the reply. The old tube was broken (one filament always open) and I replaced it with a new one but encoutered the problem in my post. Searched the web and found this http://www.pavouk.org/hw/lamp/en_index.html . It mentioned a common cause of failure is C3. Besides, it blink but didn't light up, so I guess it's the resonant circuit that stopped to oscilate. Plese let me know if I should look for something else.

 

[Nigel Goodwin]

It would seem unlikely that two faults occurred simultaneously, perhaps the new tube is faulty? (quality on modern tubes is pretty poor).

Another likely suspect would be C4, electrolytic's dry out and die frequently, so are always something to check.

 

[sender jones]

Nigel, not simultaneously, the first tube failed, I bought a new one, tested it at the shop before buying. Then put it in the lamp and found that with the new tube the lamp started a blink but didn't sustain. I'll check C4 tonight when I return home.

 

[Nigel Goodwin]

As you've got the two capacitors you could put in series, give them a try - easy to do, and nothing to lose.

 

[sender jones]

Nigel, you are right. I measured C4 (22uF 400V) output 293V which I don't know if it is normal or not. But I got a 100uF 400V on hand anyway. So I just replace it to give it a try (I believe is much safer than my original idea). It gives 303V and the lamp works as new. I guess the 22uF is of low quality and old and dried, charges leak through its internal resistance causing output to drop lower than required. Is this the correct explanation?

Thank you for your help.

 

[Nigel Goodwin]

Nigel, you are right. I measured C4 (22uF 400V) output 293V which I don't know if it is normal or not. But I got a 100uF 400V on hand anyway. So I just replace it to give it a try (I believe is much safer than my original idea). It gives 303V and the lamp works as new. I guess the 22uF is of low quality and old and dried, charges leak through its internal resistance causing output to drop lower than required. Is this the correct explanation?

Not really

They just dry up, go high ESR and eventually completely O/C - essentially it's like you don't have a capacitor there at all.