Repairing a Neewer 660 Studio light - How Hard Can It Be? -- Episode 594

  hifromkatie You are building a bigger box, for no reason. As an Integration & Sustaining Engineer, I used to work on all sorts of stuff,   And I had a problem, which you seem to have. STOP BUILDING BIGGER BOXES! It's a disease. please refer to Occam's razor. Where the simplest solution is most likely the best.  You started to build the bigger box:
     "I'm tempted to challenge myself and replace it with a 555-based circuit though, to output the PWM signal based on the pots to the LED drivers,              potentially still using the existing drivers, but if I redo them, I can get a single drop-in replacement PCB made with the 555s(/556), LED drivers and        power produced, which I think would be a pretty neat solution, just a little more time-consuming." 
I hope you don't. do easier things like wire a new CPU to the board, maybe something different if you have the toolset. You just need to get the I/O of the new CPU to the old CPU and start to program it. AND get rid of the LDO in favor of a better device like an LM780x but with one more component a Zener Diode across the input side of the regulator and you want it about .5vdc above the input voltage. This way a transient from the wall wart, will not do any damage, as when the voltage rises it will be directed to the ground. 

Thanks, I think a rebuild will be the most fun option. Currently looking at a 555 based control to replace the micro, as think that'd be a different way to solve it!

Smell tests, yes thats usually an obvious one, along with heat. Years ago I was fault finding regularly for work (thankfully with circuit diagrams and parts lists), and a very infrequently we'd get really stuck, and the last resort was to just power it up without a current limit and see if anywhere went pop, got hot, started to smoke etc

Yes, something to consider in a future redesign. I have many different barrel plugs lying around my workshop, with various voltages, they are all labelled up with what they should be connected to, but I can't for certain say that something didn't get plugged in incorrectly, or just that there wasn't some sort of power fluctuation at some point.

I immediately thought I'd just replace the microcontroller with a replacement, and write my own code, especially as the programming header is on the PCB and easy to connect in to.
I'm tempted to challenge myself and replace it with a 555 based circuit though, to output the PWM signal based on the pots to the LED drivers, potentially still using the existing drivers, but if I re do them, I can get a single drop in replacement PCB made with the 555s(/556), LED drivers and power produced, which I think would be a pretty neat solution, just a little more time consuming.

It seems to MCU was providing a PWM to the LED drivers circuit, so think it'd be fairly easy to use a 555 to replace this PWM control. A quick power up directly on the LED panel and it seems that is still working correctly, so I think it's just a case of build a new control circuit and new LED constant currant driver on a PCB that can be dropped into the case.

As beacon_dave says below there are 2 separate controls, which the micro converts into 2 seperate PWM outputs to LED drivers, 1 for yellow(ish) LEDs and the other for bright white LEDs, so you can adjust the brightness of each and therefore the colour temperature of the light output.

In hindsight removing the caps was unlikely to help, but I have had an occurrence where a ceramic cap had hairline cracked through it due to an impact, and I wondered if that had happened again.

My current idea is to use a 555 for control and avoid using a micro for a replacement control board, not really any reason other than to challenge myself to not use one in it's repair.

I sort of found the regulator and jumped straight in to that being the problem, I should have checked further along before assuming that was the problem. There was quite a bit of troubleshooting that got cut, as we almost had a 1 hour video of the highlights, unfortunately because I spent a while poking at it, there was also times I just wasn't recording.
My plan is to build a replacement controller, but will probably skip the micro and use a 555 to produce the PWM control for the LED drivers.

Oh also no battery charging, just the ability to power from them, I don't have any of the compatible Sony battery packs though so have never used those connectors

Lots of interesting ideas, in fact the replacement I got for this one is a new "upgraded" version which has a small LCD screen for displaying the brightness and colour temp, and is wirelessly controlled by a remote either individually or you can change all of them in a room at the same time!

The LED side works perfectly, I checked that, but I don't think it made the cut for the video, we almost ended up with a 1hour long video of the highlights!
Definitely will be reusing the external casing, barn doors, LED panel (It's a single PCB).
The pots are analogue inputs to the micro which is then PWM control to the LED drivers, I'm tempted to simplify the circuit and make a drop in replacement PCB which will line up with the holes for the pots, but these will be fed into a 555  (or 556, as theres 2 to control) and output a PWM to the LED drivers.