My horror movie continues.. "70X690E Power Supply- the Sequel - Attack of the Backlight circuit!"

[phuzzyday]

This was my initial thread.. https://www.badcaps.net/forum/troubl...-m-at-wits-end

(All the Electrolytics are brand new.)

I started a new one, hope that's ok, because my question is different now! (Strangely enough, after I walked away for a few weeks, it stopped lighting up at all!!)
I am determined to fix this power supply because I have the skills to replace bad components, I just need to find out which it is! If it's available obviously.. I have a backlight tester, it's working! The Mosfet on the negative rail broke in half when I tried to lift a leg, so I replaced it with a new exact match from Digikey. I scoped around enough to tell that, upon powerup, the negative side was getting PWM pulses to try the backlight circuit, then failing it. The positive side was lifting to about 250V at the same moment that the PWM test fired, but it needs at least 294 to push through the backlight array!

So I'll ignore the previous history for a moment, and just ask, is there a resource which is specific enough to this board (70X690E Pwr) that can educate me about how its specific design of the backlight circuit can best be troubleshooted?

Even this question.. What's the basic layout of this supply? Is the mosfet on the negative rail the switching element for the actual flyback converter? Can anyone give me a bit of a testing strategy, now that I have narrowed it down this far?! I'll send a couple pics of the LED section of the board. I'm having trouble finding the feedback resistor divider too!

I found the datasheet of the IC below the Mosfet, it's BD9488F. That does give me a bunch of things to check in the meantime.

Please forgive this newbie bothering you guys who are so generous with your time.. again!

[phuzzyday]

I've been looking at the board, and discussing what I see with Claude. Does this summary from Claude seem right??
---------------------------------------------------------------------------------
Your Signal Path Analysis:

High Voltage DC In: 300V+ from transformer/rectifier Through Inductor: Energy storage and current smoothing Through High-Side MOSFET: (The one on the heatsink) - this switches the current on/off Through Current Sense Resistor: Provides feedback for current regulation Through Filter Caps: Smooths the switching ripple Through LED Array: Your load Through Low-Side MOSFET: (The one you replaced) - provides the return path during the "off" time To Ground: Completes the circuit This is Synchronous Buck Operation:


High-side MOSFET ON, Low-side OFF: Current flows through the path you described High-side MOSFET OFF, Low-side ON: Current continues flowing through the inductor via the low-side MOSFET (instead of a diode) Why Synchronous?

At these current levels, using a MOSFET instead of a diode for the return path is much more efficient - MOSFETs have lower voltage drop than diodes when conducting. The Current Sense Resistor:

This is key - it allows the control IC to monitor LED current and adjust the PWM duty cycle of both MOSFETs to maintain constant current regardless of LED voltage variations.

So your failure analysis should focus on that high-side MOSFET (the heatsink one) since it's doing the main switching at 300V and high current!
--------------------------------------------------------------------------------

If this sounds right, I think I know what to look for !

[phuzzyday]

Just got more interesting. There is only a smoothed 150V coming out of the transformer at the bottom of the pics which is feeding the LED circuit through the two outermost pins. Now I need to figure out if it's the Xformer, OR the power entering it!

[phuzzyday]

THE PERILS OF IMPROPER AI USE! THE PREVIOUS TWO POSTS ARE INCORRECT. I CAN'T DELETE THEM! CAN THE MODS?

I guess I can't edit posts. Anyway, I was wrong about that. I have a couple numbers mixed in my head, and the AI's got all confused too. They can only work with what you give them!
So, the supply going to the LED driver circuit is at a nice 150, and it's RIGHT. I became even more sure when I realized the smoothing caps were 200V rated. I had it in my head that the LED circuit was a buck converter, since the AI told me it was, but again, I don't blame the tool if I don't know what to tell it.
So I need to analyze the pins on the driver IC, BD9488F, and see if it is being given the right circumstances to work right. I DO know its getting PWM from the mainboard, and the circuit to the output/mosfet is fine on the negative side. That chip needs to power the boost converter Mosfet on the positive side also though, and that seems to be the issue. The good thing is that the chip costs about $1.50. Too bad shipping is $8 from Digikey. But if that's all I need, GREAT! I will be determining that tomorrow. I don't know if anyone is reading this stuff, but at least it'll be archived for future searches. The good info with the bad.

[EazyBone]

What about this.

https://a.aliexpress.com/_mL9tduL