Voltage Increase, failed component?

[ajp8868]

Hi All,

I'm currently repairing a DJ light which uses two voltages, a 12v and 28v supply.

Currently, a couple of the steppers on the 28v side aren't working at all and I've been swapping chips for a few days to no avail.

I've just noticed that the 28v side is being pulled up to ~31v. I've checked another unit and it doesn't do this, I've also checked with a known good supply to make sure it's not that either.

So I'm a bit puzzled, could this be a shorted component? Any suggestions as to what would cause that?

Picture attached of the board. The 28v is the brown wire and supplies the LED circuit, and two of the steppers controllers.

Thanks!

[fzabkar]

Where is the 28V supply being generated? Is it a raw supply, or is it regulated?

[ajp8868]

Direct from a dual smps, I believe it's regulated...

Just for info, I tested the same supply with a working board and it doesn't do it. I also tested another supply with this faulty board and it does the same.

[fzabkar]

What is the component on the vertical heatsink?

FWIW, I notice that one of the 4-pin connectors in the following photos is labelled 24V, Gnd, Gnd, 12V.


SHOWTEC SPCI126 PHANTOM 50 SPOT MAIN PCB:
https://www.indiamart.com/proddetail/showtec-spci126-phantom-50-spot-main-pcb-2852762037848.html

https://5.imimg.com/data5/SELLER/Default/2023/10/356541585/SZ/DP/QW/10222995/showtec-spci126-phantom-50-spot-main-pcb.jpg

https://5.imimg.com/data5/SELLER/Default/2023/10/356541509/RP/GS/XX/10222995/showtec-spci126-phantom-50-spot-main-pcb.jpg

https://5.imimg.com/data5/SELLER/Default/2023/10/356541492/IZ/PO/NW/10222995/showtec-spci126-phantom-50-spot-main-pcb.jpg

https://5.imimg.com/data5/SELLER/Default/2023/10/356541474/RN/TE/TB/10222995/showtec-spci126-phantom-50-spot-main-pcb.jpg

https://5.imimg.com/data5/SELLER/Default/2023/10/356541848/AW/RQ/SG/10222995/showtec-spci126-phantom-50-spot-main-pcb.jpg

https://5.imimg.com/data5/SELLER/Default/2023/10/356541781/HN/SY/UP/10222995/showtec-spci126-phantom-50-spot-main-pcb.jpg

https://5.imimg.com/data5/SELLER/Default/2023/10/356541666/QJ/QQ/AS/10222995/showtec-spci126-phantom-50-spot-main-pcb.jpg


User manual:

https://www.farnell.com/datasheets/1518634.pdf

[ajp8868]

Yes the boards labelled 24 but the PSU supplied in these runs 28 and 12v. It might be that these same boards are used for different lights.

That's a 5v regulator and is connected to the 12v rail

[fzabkar]

I'm trying to understand how these lights work. IIUC there are two steppers for moving the light in the horizontal and vertical planes. There are two additional steppers for rotating the Colorwheel and Gobo-wheel discs. Is that about right?

I'm also wondering why this simple(?) device would need two Megawin 82FG216 micros. :-?

You say that you have been swapping chips. Which chips have you swapped?

[ajp8868]

Ah ok, yep there's 4 steppers for colour, gobo, gobo rotate and prism, these have drivers connected to the 12v supply and 5v for the controllers.

Then there are 2 steppers for pan/tilt which are connected to the 28v line.

The LED is also connected to the 28v

[fzabkar]

I would start by drawing up a block diagram which tells me which micro is responsible for a particular function. Does one control the colour/gobo wheels while the other controls the pan/tilt? What does the DAC do?

AFAICT, the two 24-pin ICs under the long thin heatsinks are the pan and tilt stepper motor controllers/drivers. They appear to be driven via a 14-pin IC, and this IC would presumably be controlled by one of the micros. I would measure the inputs and outputs of this IC (what is it?).

Datasheet for MG82FG216:

https://semiconductors.es/semiconductors.php?id=829616

Edit:

ISTM that the micro nearest the crystal is a slave and controls all the steppers. The other micro appears to be the master and is responsible for controlling the display and processing the user inputs. It also drives the quad TLC7226 DAC (whose function I don't understand).

https://www.ti.com/lit/gpn/TLC7226


Just for reference ...

HV9910B - PWM high-efficiency LED driver control IC:
https://ww1.microchip.com/downloads/en/DeviceDoc/20005344A.pdf

[ajp8868]

I believe that's correct that the micro close to the stepper drivers is responsible for all stepper control.

It must be partially working because the colour, gobo and gobo rotate steppers are working fine.

Do you think it's likely that the micro could be partially faulty? I've checked to make sure all pins are connected.

Or with the slight voltage increase, could it be a capacitor or resistor issue on that rail?

[ajp8868]

SHOWTEC SPCI126 PHANTOM 50 SPOT MAIN PCB:
https://www.indiamart.com/proddetail/showtec-spci126-phantom-50-spot-main-pcb-2852762037848.html

Btw the unit is a chauvet intimidator 255 irc, not the showtec phantom version which has the same board, different firmware. The boards are available for replacement for £165 but I'd love to be able to repair the board if it's not the micro controller at fault

[fzabkar]

There appears to be a 14-pin IC between the micro and the two stepper controllers. What is that IC?

[ajp8868]

It's an op-amp but can't remember the exact chip type/markings. I've swapped this chip from a working unit but it didn't change anything.

[ajp8868]

You say that you have been swapping chips. Which chips have you swapped?

Sorry I missed this question, so far I've swapped the TL7226 DAC and the Op-amp

I've also swapped the LED driver, but actually I've removed it from this board since I shorted a pin and blew it up..

[fzabkar]

If you select PAN or TILT functions from the menu, does the light behave as if it is responding to the command even though it isn't moving? In other words, does the MCU think that the light is moving even though it isn't? If you remove one of the stepper controller ICs, does the MCU complain?

How does the unit know when a stepper motor has reached end-of-travel? Does it simply count the steps, or is there some positional feedback? Limit switches? How many wires are there for each motor?

[ajp8868]

There are no errors shown and it behaves as normal with the controllers or without.

There is a limit switch on the pan and a limiting reed switch on the tilt. No positional feedback on these units

[fzabkar]

I think I would be breaking out the scope and looking for step pulses. Start with a working unit and then compare with yours.

I'm wondering how your high supply voltage observation relates to this fault. Can it really be significant, or is it a red herring? The only explanation I can think of is that the supply is unregulated, and the fault is unloading it in some way. I suspect that if you examine the power supply circuit, you will find that the 12V rail is regulated while the 28V rail just finds its level via the turns ratio in the transformer.

Is it possible that the light is stuck in a program where pan and tilt are disabled???

[ajp8868]

Yeah I'm going to try and get hold of a scope. I've never used one for repairs before but I think it's about time! Where should I be probing? Direct from the micro controller?

I tried probing the oscillator on a working unit with my DMM on frequency mode, but it made the light act crazy. Is there a way to do that properly?

It could be a red herring, but that supply doesn't do it on a working board..

It's a good thought but it doesn't have any functions where the pan/tilt could be disabled and even then it would at least perform it's startup routine.

[shakalnokturn]

I think I would be breaking out the scope and looking for step pulses. Start with a working unit and then compare with yours.

If you have access to a thermal camera that would also be of great help for comparison.

The only explanation I can think of is that the supply is unregulated, and the fault is unloading it in some way. I suspect that if you examine the power supply circuit, you will find that the 12V rail is regulated while the 28V rail just finds its level via the turns ratio in the transformer.

I was thinking of this too. Excessive loading on the regulated rail would bring the unregulated one higher than intended, so there again look for anything running hot.

[ajp8868]

I've tried with electronics freeze spray which doesn't seem to show anything obvious sadly.

I don't have access to a thermal camera unless I can rent one but even then it's quite expensive.

I'll keep looking to see if there's anything I can feel getting hot. Do you think possibly a short on the 12v side? Everything seems to be working fine on that side.

[themadhippy]

The most common fault on nodding buckets is broken wires in the arms,presume you've tested continuity whilst wiggling the wires

[ajp8868]

Swapped this board to a known working unit and it does the same, so it's 100% a board fault

[ajp8868]

Had another look and the lm7805 regulator seems to be getting very hot.. could this be a symptom of it failing and drawing too much current? Or is it more likely to be a failed component on the 5v rail?

Bearing in mind the 5v side is working perfectly

[fzabkar]

Had another look and the lm7805 regulator seems to be getting very hot.. could this be a symptom of it failing and drawing too much current? Or is it more likely to be a failed component on the 5v rail?

Bearing in mind the 5v side is working perfectly

Being a linear regulator, there would be a 7V differential between input and output. A load current of 140mA would dissipate 1W in the regulator.

[ajp8868]

Had another look and the lm7805 regulator seems to be getting very hot.. could this be a symptom of it failing and drawing too much current? Or is it more likely to be a failed component on the 5v rail?

Bearing in mind the 5v side is working perfectly

Being a linear regulator, there would be a 7V differential between input and output. A load current of 140mA would dissipate 1W in the regulator.

Meaning..?

[ajp8868]

I've tested the 28v side without the 12v side connected and it doesn't increase in voltage - so I'm assuming the 12v is possibly pulling more current than it should - I'll check with the DMM