Chauvet LED 300 LED Driver Problems

[posty07]

Hi all,

I'm looking at a few 300 LED Barrels which have very low output on the LED.

The LED itself is fine as I've swapped the driver from a working one and it comes back to life. 

Whatever the problem is, it seems to be common to these as they're all acting the same.

I've done a fair bit of testing between the working and broken one and most voltages, resistances etc seem to be very similar if not exactly the same. These boards use 2 MOSFETs from the -ve pin of the led to ground. Each shares half the load. 

My question is.. is it likely that the MOSFETs are bad? Just normally I'd expect a MOSFET to fail short circuit.

 

Cheers

Adam

Edited October 31, 2022 by posty07
Grammar

[indyld]

Post decent pics of the board. Perhaps only half the frequency is resulting in half the on time. Either MOSFET not opening or gate driver not driving. 

 

Might be able to diagnose using a camera and a working unit to compare flicker. 

Edited October 31, 2022 by indyld

[posty07]

 

16 hours ago, indyld said:

Post decent pics of the board. Perhaps only half the frequency is resulting in half the on time. Either MOSFET not opening or gate driver not driving. 

 

Might be able to diagnose using a camera and a working unit to compare flicker. 

Ok I'll get some pics later today - I hadn't thought of that as a troubleshooting step, will give it a go!

I think they might be flicker free but it's worth a try 

[indyld]

Although the PWM speeds are probably pretty high (even with only one MOSFET firing) you may find an older camera phone can still show running lines particularly if the faulty unit has reduced on time. Max on/off (e.g 50%) may be the most effective but you can try at different dimmer levels.

You can also make a ghetto PWM frequency detector with an LED and an oscilloscope but if you already have the latter then you might as well just probe the device.

I used to demo PWM on stuff like Schnick Schnack by waving the LEDs strips back and forth which has reverse zoetrope effect and shows the LEDs as dashed lines in mid air. I have no idea how to make this work with a large fixture (or perhaps an actual zoetrope). However, all this might be just a bit of fun. Ultimately, if stuff is less bright than it should be and other symptoms as described then a prolonged off time would be suspected. 

Edited November 1, 2022 by indyld

[DrV]

35 minutes ago, indyld said:

I have no idea how to make this work with a large fixture (or perhaps an actual zoetrope)

You can actually get a degree of this effect by shaking your head while looking at a flickering light source. It certainly works with neon indicators (100 Hz) but I don't know that you'd see it at the frequency of an LED PWM drive.

[Brian]

If you are serious about wanting to fix things then it's time to buy a scope.

£100 on ebay will get you something worth owning.

[posty07]

30 minutes ago, Brian said:

If you are serious about wanting to fix things then it's time to buy a scope.

£100 on ebay will get you something worth owning.

I've been looking actually and really want one but I don't know what specs or models are going to be useful!

Any recommendations?

[themadhippy]

Quote

Any recommendations?

if you can increase the ole budget

https://www.rigol-uk.co.uk/product/rigol-ds1054z-50mhz-digital-oscilloscope/

been around for a few years,but still pretty good bang for buck.

few quid more,but newer and slightly better specs

https://siglent.co.uk/products/digital-oscilloscopes/siglent-sds1000x-u-series-digital- oscilloscopes/

Ive got my eye on one of theses to replace my old trusty analogue instek,but other expenses keep getting in the way

[posty07]

I mean they look great but unfortunately it's just a hobby so I can't justify it 😢

Would one of these little things be any good for basic troubleshooting?

https://www.ebay.co.uk/itm/385040468593?mkcid=16&mkevt=1&mkrid=711-127632-2357-0&ssspo=-4xkiftwsz6&sssrc=2349624&ssuid=5Ff1NvmRTVm&var=&widget_ver=artemis&media=COPY

[posty07]

A few photos of the board. I've swapped the MOSFETs on this one for a similar spec but I've since realised that their fully 'on' voltage is a little higher. 

To me this circuit seems to not use pwm to dim the LED and actually seems to use current limiting using the MOSFETs.. of course I could be very wrong.

I'm waiting on exact replacements so hopefully they will do the trick as the led is now brighter, but not completely full, with these replacements.

P.S. ignore the missing pad underneath.. too much heat applied - it doesn't connect to anything as it's the pad on the top that connects to its trace

[indyld]

Look up the differences between constant current and constant voltage drivers.

What are the numbers on the two SOIC8s?

PS. Use lots of added 60/40, flux (and in some cases low melt solder) and some decent wick to make a better removal job. Having said that, a low quality iron will also often mess pads up. Or a poorly sized tip. 

Edited November 2, 2022 by indyld

[Brian]

15 hours ago, posty07 said:

I mean they look great but unfortunately it's just a hobby so I can't justify it 😢

Would one of these little things be any good for basic troubleshooting?

https://www.ebay.co.uk/itm/385040468593?mkcid=16&mkevt=1&mkrid=711-127632-2357-0&ssspo=-4xkiftwsz6&sssrc=2349624&ssuid=5Ff1NvmRTVm&var=&widget_ver=artemis&media=COPY

Unlikely...

Maximum Bandwidth: 200kHz Bandwidth

... is far too low to be of any real use.

[posty07]

6 hours ago, indyld said:

Look up the differences between constant current and constant voltage drivers.

What are the numbers on the two SOIC8s?

PS. Use lots of added 60/40, flux (and in some cases low melt solder) and some decent wick to make a better removal job. Having said that, a low quality iron will also often mess pads up. Or a poorly sized tip. 

Thanks I will - to be fair most of that burn was there already as these appear to get very very hot..

 

The IC on the right is an LM393 and left is LM358. Both double op amps which appear to be working. Though I'm having trouble working out how the output is calculated using the 2 inputs.

[indyld]

Total guess but perhaps one is in a voltage comparator circuit that the trim pot is also part of.

Edited November 2, 2022 by indyld

[posty07]

2 hours ago, indyld said:

Total guess but perhaps one is in a voltage comparator circuit that the trim pot is also part of.

Yes I think it is, I've checked the pot and it's working well.

The 3 pin connector comes from a control board which gives it a 5v power, ground and a control PIN which seems to change voltage to control the brightness - which I'm guessing is what the comparator circuit is for