LED TV repair, 250V but no lights ?

[MathWizard]

I have found an LED LCD TV and the backlght wasn't working. I took out the PSU board, since I never worked on these before really. And mapped out most of the PSU and the STBY, 12V , and Vaux, are the only main LV rails, and they seem fine, normal idle voltages.

So then I hotwired the LED driver chip on, I set BackLight Enable to 5V, and the idle voltage for the LEDout (from inductor not the chip) is sitting at 102V
MAP3301C
https://datasheetspdf.com/pdf-file/1108098/MagnaChip/MAP3301C/1
 
So I connected the LED connector, to the PSU PCB, then I applied 0V-5V PWM, @300Hz to the diode+resistor leading to pin15, for the PWMI / internal dimming mosfet.

And the output voltage for the LED's jumps to 250V, but there's still no light I can see, even at 75% duty cycle. I'm thinking also that the output is only that high, because it's a current source trying it's hardest to drive an open circuit if the LEDs are burnt out, and then hitting some peak voltage protection limit ?

For the output sense resistors of the PWMI / internal dimming mosfet, there's 3* 5R6, and I only read like 55mV there on pin 8. Too late to measure the pin 5 voltage, but it seems like no current.

The PowerGood pin7 is low, which is the fault signal, and I never saw it go high. So I'll have to check the other pin voltages, and hook up the Oscope.

And the only logic from into the PS PCB, at all, is PWR ON, BkLt-ON, BkLt-PWM, and it's a nicely labeled PCB.

What's this sound like ? I'll map the LED inductor circuit next, 250V and no light.

[james_s]

I fixed a similar fault that was due to an open LED. It became intermittent first and the spikes zapped a few other LEDs causing them to short and one of them eventually went open.

[PKTKS]

Like all LED driver ICs  they have at least 2 protections by default:

= OVP  which monitors the output network
= OCP which uses a typical current sense resistor.

In case any led opens typical thing is OVP tripped

In case any led dead shorts typical thing is OCP tripped

We hardly escape the tedious test of each strip..
and in many cases each led as well..

Paul

[MathWizard]

There's a voltage divider reading the 102V at idle, giving 1.7V to the comparator, with a Vref of 2.7V. Still the PGood is low.

I turned off the BK_ENable and still the PGood is low although it twitched to ~1V for a moment on the scope.

But that's without the LEDs hooked up, so I'll have to solder some wires around the chip and see what's happening w/ the load, but so far I've not seen the PGood go high ever.

This is a +40" LED TV, what sort of current might the lights draw? and what sort of voltage drop across the "bulbs", I've seen them on utube, so it's like 4 parralell rows of 10 or whatever for this size TV

[fzabkar]

I can't understand how these designs can balance the currents in each string. It only makes sense to me when each string has its own driver. :-?

[wraper]

I can't understand how these designs can balance the currents in each string. It only makes sense to me when each string has its own driver. :-?

Connect them in series.

[james_s]

LED backlit TVs that I've worked on have either 4-6 separate strings of LEDs each with their own drive channel, or in the case of my 65" TV there is one long series string that runs close to 300V at around 250mA.