General > General Technical Chat

Lightswitch trashed in only a few weeks...

(1/3) > >>

peteb2:
So the pic is a standard PDL fully taken to bits to reveal the arc damage that occurs now that the ceiling full of room flouro lights have been substituted for multiple standard batten LED. There seems to be a huge a back emf upon lights switchoff as the (i'll assume) buck converters in each light that produce the 400VDC plus from Mains 230VAC to feed the large cob strip of LEDS. Anyway, the poor lightswitch soon gives up the ghost & the dissection reveals the destruction. Back in the day flouro lights would put a toll of the switch too when their ballasts or starters went sick. Can anyone suggest a fix ie an X or Y class capacitor & maybe a resistor in one leg of the capacitor across the Line & Neutral to snub out the arc potential & let a switch last a bit longer? TIA!

Jeff eelcr:
Use a higher current switch. (Much higher)
Jeff

radar_macgyver:
You could use a quencharc device, these include an RC network with the appropriate mains ratings and are specifically for this purpose.

https://www.cde.com/resources/catalogs/Q-QRL.pdf

Whales:

--- Quote from: peteb2 on August 22, 2022, 03:38:46 am ---There seems to be a huge a back emf upon lights switchoff as the (i'll assume) buck converters in each light that produce the 400VDC plus from Mains 230VAC to feed the large cob strip of LEDS.

--- End quote ---

Huh?  All of the power supply schematics I've ever seen start with a fuse, perhaps some protection components, perhaps some filtering (X & Y caps, maybe a common mode choke) and then a diode bridge (or active bridge) before that 400V caps.  That shouldn't act very inductive overall or be able to backfeed the contents of the 400V cap, ie it shouldn't fry the switch on turn-off.  If it did then every switchmode power supply of similar construction would fry your switches, including computer power supplies.

EDIT: But perhaps if they have bad diodes (or equivalent active parts) the DC could backfeed?  That would wear an AC rated wallswitch out pretty fast if timed correctly.

Are you still running the odl ballasts in series with the new power supplies?  ie was it a replace-tube-and-remove-starter conversion to LED, or were the ceiling boxes taken down and the ballasts bypassed/removed too?   I wonder if that situation could cause your problem... perhaps the LED PSUs bypass these big ballast spikes via their protection components and then the switch contacts take the hit instead?  But why would these spikes be bigger than with the original tubes -- super low impedance when activated style protection?

wraper:
I rather suspect this switch was at deaths doors already and changing the lights was a coincidence. Back EMF from LED bulbs is basically impossible FWIW. But with inductive ballasts of fluorescent bulbs this should be possible.

Navigation

[0] Message Index

[#] Next page

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod