One of our fire engines has a FRC LED floodlight (240V, 20K Lumens) that is no longer working. It flashes on/off about 1/second. Power is confirmed good. Replacement cost is $2.5K+, so I'm looking to see if there's any common problem that might cause this (I'm an EE, but with no schematic and no background in LED power control, I'm flying blind). Is there any common thing that fails with these kind of lights to check? Board and components don't appear to have any visible damage.

However, in the lower left corner of the reverse side of the board (bottom 2 pictures) is a MF72 10D11. The part number comes back as a power NTC thermister. Apparently used for initial inrush current limiting. It's hard to tell, but the shell appears to have a small crack in it. Or it could just be the normal way the device's case looks.

Any help would be appreciated. Thanks.

 

It uses a LM3464 led driver, check the supplies on the pcb especially the chip.

https://www.ti.com/product/LM3464

 

Based on the three electrolytic capacitors rated at 120µF 350V it looks like the unit contains a bespoke switch mode power supply to provide the dc voltage to the control/LEDs. Electrolytic capacitors are well known for their short life expectancy, especially when operated close to their voltage/current/temperature limits.

The right-hand of the three electrolytics looks a bit suspect with a dark area/slight deformation at the lower right.

If you have access to a capacitance meter, it will normally give an indication of the correct value even with the components in circuit, or better still cut one of the capacitor leads and measure.

The data sheets provided by Texas Instruments are pretty useless since they don’t show the IC pin out numbers.

 

Pinout of the chip... It will be a switch mode psu or module of some sort, voltage is 20 to 80V,

 

With a multimeter measure the voltage between Vin (pin 15) of the IC and PGND – take care when doing this as there are hazardous voltages accessible with the unit open and powered.

 

Fortunately pin 15 is bottom left on IC U6 in the 3rd photo (so easy to probe without the danger of shorting adjacent pins) – PGND (on pin 22) is probably common with the unit’s chassis.

 

Have you asked whether FRC can repair or refurbish it? A reputable manufacturer might want to find out what went wrong so they can improve their product.

 

Based on the three electrolytic capacitors rated at 120µF 350V it looks like the unit contains a bespoke switch mode power supply to provide the dc voltage to the control/LEDs. Electrolytic capacitors are well known for their short life expectancy, especially when operated close to their voltage/current/temperature limits.

The right-hand of the three electrolytics looks a bit suspect with a dark area/slight deformation at the lower right.

If you have access to a capacitance meter, it will normally give an indication of the correct value even with the components in circuit, or better still cut one of the capacitor leads and measure.

The data sheets provided by Texas Instruments are pretty useless since they don’t show the IC pin out numbers.

The 3 large caps are in parallel. I took a good look at all 3, and what you're seeing appears to be an artifact of the photo, not an issue with the cap. Also, after I initially ran it and then unplugged it, I waited a minute or so and then shorted across the cap (didn't want to get zapped while disassembling things) and got a decent spark, so the caps are definitely holding a charge.

I also took a closer look at the MF72 and there sure does appear to be a crack. Based on the description it appears provides initial inrush surge protection. Initially 10 ohms, then resistance drops:

" The MF72 series Power NTC Thermistors provide inrush current suppression for sensitive electronics. Connecting a MF72 in series with the power source will limit the current surges typically created at turn on. Once the circuit is energized the resistance of the MF72 will decrease rapidly to a very low value, power consumption can be ignored and there will be no effect on normal operating current. "

I'm wondering if I would cause any problems just shorting across it to see if the LED Driver starts.

 

Replace NTC MF72 to a wirewound resistor of 5 Ohms or so and test. This will limit current surge

 

Have you asked whether FRC can repair or refurbish it? A reputable manufacturer might want to find out what went wrong so they can improve their product.

They will repair or replace, cost is up to half of a new unit (could be less depending on what they find). The issue is, they require the whole unit. The whole unit includes a screw-on backplate with an integrated pigtail which then goes through the roof. Removing the pigtail from the truck will require dropping the headliner and possibly lifting the AC heat exchanger (above the roof) that kind of covers the roof entrance to get the pigtail out. FRC nicely designs the unit so you can remove the backplate, pull 3 quick connects, and be on your way. But then requires everything be returned before they'll look at your light.

 

Do as Pyrex suggests, which will eliminate the NTC as the faulty component (or not).
A 5Ω resistor passing 0.4A will dissipate less than 1 Watt.

 

Do as Pyrex suggests, which will eliminate the NTC as the faulty component (or not).
A 5Ω resistor passing 0.4A will dissipate less than 1 Watt.

Did so, still no joy. NTC component is not the issue. But it was a good try. Thanks.

 

Have you measured the voltage on pin 15 of IC U6?

If it is good, you can eliminate the PSU as the problem.