MARTIN MAC 101 Power Issues. Help me please??

[partyanimallighting]

Hi all! I'm having an issue with one of my MAC 101's. I have 4 units and despite little issues such as blown diodes, faulty sensors, glitchy displays (which I've been able to repair myself) they've been true little workhorses despite being discontinued by Martin. Parts are really scarce and really pricey for these units so I'm hoping the fault I've encountered can be resolved. The problematic unit does not power up at all, I get only a light spin on the fan and a flash on the display. I've pinpointed the problem to the mainboard and not the main power supply in the base (which outputs 24VDC and 48VDC to the mainboard) by simply swapping out the base from a working unit and the problem is constant with the mainboard. I'm assuming that the 48VDC leg powers the LED diode plate and the 24VDC drives the mainboard circuitry but I am not seeing any burnt or scorched components on the mainboard. I will be very grateful for any assistance to resolve this issue and get this unit up and running again. Help!!!!

[indyld]

Check for a short on the main dc input on the problem board first. If so, look for an open TVS diode near the DC in. It should be reverse biased to ground.

If no short on the DC in, find the next power supply on the the offending board and try again. A quick check is beeping any inductor you see to ground. If shorted, find the buck converter that creates the supply. 

Edited October 17, 2022 by indyld

[partyanimallighting]

OK indyld, I'm not that versed in the electronic jargon but based on my limited experience I'm interpreting this to mean to check the  four main spade terminals coming in at 24VDC and 48VDC for shorts at the solder points to the mainboard so I could just reheat and get a good solder flow to fix any possible shorts. As for the diodes near the DC in, off of the picture I'm seeing D21 and D22 in the general area and D20 a bit further along to the bottom left of the image. As for the reference to "next power supply", "inductor" and "buck converter"???? By power supply do you mean voltage regulator? Inductor ~ Choke (630 2350 in the image)? Would the buck converter be listed as an IC on the mainboard (such as IC27)? Yes....it's a slow learning process with me but you have helped me a lot in the past.

[indyld]

OK, so an overview of any board in this situation.

There is a two terminal DC input. (In this case, I would first test to see which of the two voltages inputs has the problem. It will be whichever one prevents the main PSU from outputting so try them one at a time and measure for voltage at their respective inputs when plugged in. If you haven't already, unplug every other connector from the board but the main DC in you are checking.)

With the board disconnected the two terminals of the input should not read shorted together. One is ground and it's helpful to find out which one, it should have continuity to one of the other DC input terminals and also appear all over the board. 

If the input reads as a dead short, the next place to look is for a TVS diode across these two terminals. This is often used for overvoltage protection and can fail short to ground if pushed hard enough.

The main DC input voltage will then be used to create lower voltages that are needed by the board. These are also referred to as power supplies and comprise of a way of making the lower voltage either an IC, MOSFETs (buck convertor),sometimes a linear regulator or a zener diode. In most cases, these supplies have their own inductor - on this board it looks like a bobbin coil style package. As a good starting point, neither terminal of such a coil should be reading short to ground. Also, beep across any linear regulators you find to look for a short - a 3 legged regulator should only have 1 leg that beeps to ground. If any found, then there is an issue either with the mini-supply itself or what it powers which is the next part.

These are all things that I start with when presented with a board that, as described, is shutting down the main PSU. Given the symptoms, I do not actually think the issue is with the main DC in but we do need to check that first. Further diagnostic methods are for the future, but the goal for now is to identify which 'section' of the circuit (e.g which mini-power supply) has the problem.

Start with which DC in causes the issue, 24v or 48v, then trace from there. There is a small chance that when two main voltages are used, one only switches on once the other is deemed OK but on a lot of the Martin stuff I've worked on the main power rails are just on and everywhere from switch on.

I would be surprised if this was fixed by touching up solder joints and you are correct there is a fair bit to learn when it comes to board diagnosis but this is fixable in all but the most extreme cases.

Edited October 19, 2022 by indyld

[partyanimallighting]

Hi indyld, thanks for this great response. It appeals to my basic learning skills when it comes to troubleshooting and repairs. My skillset so far consists of basic parts replacement (easily recognizable scorched or burnt out components makes this easy), continuity testing etc and I am slowly learning as I go along. The board receives two DC voltages, 24VDC and 48VDC and I might be correct in thinking that the 24VDC drives the mainboard circuitry and the 48VDC drives the diode plate. The voltages at these wires disconnected from the mainboard are correct so I will follow your instructions and report back to you as your student ready and willing to learn a little more day by day. I do know that the input voltage, be it 24VDC will be broken down by a voltage regulator, probably to 5VDC, to drive the circuitry as in other PCB's. I'm accustomed to them being called voltage regulators and when I'm troubleshooting voltage issues I usually start there but I did not know they were also referred to as buck converters. I'm assuming that the linear regulator or zener diode goes across the current flow to also reduce the voltage. You are also correct that the issue is not with the main DC from the power supply in the base of the unit as I swapped a known working base and the problem follows this mainboard and probably one of the "mini power supplies". I will report back to you soon and thanks again for the advice.

 

 

[indyld]

For terminology reference, a linear regulator and a DC to DC convertor (buck / boost / whatever) can appear to do similar things but aren't the same. 

When I say I think it's unlikely to be the main rail, I'm not referencing the base PSU. If the fan flicks on etc. it is less likely to be a dead short on the main DC in end of the problem board. Something is trying to start and can't. A short on the input will normally shut down the PSU before anything happens at all. 

Edited October 19, 2022 by indyld

[partyanimallighting]

Hi indyld, So far I did the following as per your instructions. I disconnected all the terminals (pan and tilt motors, sensors etc.) with the exception of the display and the fan and connected both 24VDC and 48VDC legs from the PSU. No startup at all, as expected. I then disconnected the 48VDC leg and everything immediately started up fine, fan, display with reset process and so on. Reconnected the 48VDC leg again, no function. So now I'm figuring that the short is on the 48VDC leg as I suspected and that's the leg that drives the LED diode plate. I then checked for continuity from the RED, GREEN and BLUE LED outputs to ground and I got continuity on all three output grounds and a reading on the + side on the RED and BLUE outputs. However, I got continuity on both + and - on the GREEN leg to ground. Am I to assume that there's a short on the GREEN output leg of the mainboard? I need some sleep but I hope there's a response and further guidance when I try some more troubleshooting later today. Thanks for all the help so far.

[indyld]

Given that you see a difference between the 'normal' of R and B and a short both sides on G, it's a fair assumption that something is shorting to ground on that leg. I'm struggling to visualise the topology of the LED circuits in this case as I don't think I've ever looked inside a 101, but the next question to wonder about is where the problem is in the overall stream. 

In general, I would investigate if there was somewhere that I could 'divide and decide' on this leg of the circuit. Or even just remove something that would isolate the green leg and verify that R and B legs are actually fine and that 48V will work without the G connected.

In practice, the things that short at the LED end of things are usually whatever is driving them - usually either a dedicated driver IC per LED or series of LEDs, or a MOSFET. If a MOSFET, workout which legs are which and check for a short Gate to Source, Gate to Drain, Drain to Source. A driver IC is basically a MOSFET in a package with some other bits and bobs, but can show similar faults if you can work out which pin is which. Generally, only one pin is GND although some may have another pin pulled low too.

As you no doubt are aware, just because you measure a short somewhere doesn't mean it's that component. But it's a start. After that, it's a case of isolating different parts of the circuit to narrow it down. You can also use a DC power supply to identify the source of the short, either with the 'old skool' method of soaking suspected components with IPA and seeing which one heats up first or using a thermal camera which I'm increasing relying on these days.

Edited October 19, 2022 by indyld

[partyanimallighting]

Hi indyld, Some good news. I took some readings again and everything pointed to there being an issue on the GREEN leg so I took your advice and I decided to 'divide and decide'. Each leg has it's individual components so I removed C25 I3310 listed as D33 (diode right?) and ZXMP 10A16 which is a P Channel Enhancement Mode MOSFET (not sure what that means) from the GREEN leg, connected both PSU inputs and the unit started up fine, without any terminals connected. I then connected all the terminals and again, the unit started up fine of course with ERROR messages because the PAN and TILT motor leads are disconnected. Sooooo, I took the one old mainboard that I have and I removed the two components and installed them in place. The unit started up fine once again and I could initiate TEST MODE without an issue but there is now no GREEN output (RED and BLUE are fine). So I'm assuming that either C25 I3310 and ZXMP 10A16 (or both) were causing the original short. So now it looks like all I have to do is figure out which component on the GREEN leg is causing the output failure but I think I've made some really good progress under your guidance.

"In general, I would investigate if there was somewhere that I could 'divide and decide' on this leg of the circuit. Or even just remove something that would isolate the green leg and verify that R and B legs are actually fine and that 48V will work without the G connected."

[indyld]

So, there is a possibility that the MOSFET shorted but you still have an LED out. Check the old MOSFET if you haven't already (Google for how to test a MOSFET with a multimeter) and, while the MOSFET was removed you should have no longer had a short to ground at the coil for Green.

Judging by just 3 circuits running the RGB (and the MOSFET and large coil for each one) I'd assume that all the green LEDs are wired in series. In most LED fixtures, there isn't much after the coil so any faults are either a bad joint, 000 resistor blown, or in this case quite likely a LED where the Green part has blown. A LED fails, the momentary failure condition shorts the MOSFET to ground and you have the symptoms you currently have.

I've had success checking LEDs with the multimeter set to Diode mode and then using the voltage coming off the meter across the pins of the LED itself (with the correct polarity) to get a tiny bit of light from the chip for each colour in turn. 

If you do have a LED out, the next question is what to do about it. But see what you find first.

Edited October 20, 2022 by indyld

[partyanimallighting]

Hi indyld, that MOSFET is somewhere on the bench so I'll search for it and test it. Should the fact that the unit started up fine with the replaced parts convince me that these old parts are fine? As you said, in these units, all the LED's (each color) are wired in series and if one diode blows, all are out in that particular color. I also checked the mainboard with a known working diode plate and still no GREEN so it's still something on that leg of the mainboard so I'll do some more checking tomorrow (tonight). It was a lot easier to swap out the diode plate than to check each individual GREEN diode with the meter and these are those tiny 3535 CREE chips which are a pain in the @r$3 to change out but I've done so successfully in the past on these units. My suspicions are pointing me towards the 680 2350 induction coil. Other suspects would be the R22 resistor and another tiny "coil" labelled L31. Am I looking in the right direction?

 

 

 

[DrV]

Been following along with Indyld's excellent advice. Regarding failed inductors, this almost never happens so you can pretty much discount that. A possible scenario is that the MOSFET Q6 failed short circuit which would drive the LEDs too hard until one of them failed open circuit. If you can find it on your bench you may well find it is short between the drain and source (that's the middle pin and the one at the bottom on your pic of the board).

[indyld]

You can easily check the inductors and resistor in circuit but I'd be surprised if they weren't fine. If you run the unit up, you probably have full voltage at the start of the LED series on all colours and then it gets lost at a dead LED. 

I believe that the old MOSFET probably has shorted but it's worth testing it if you still have it around. It's not crucial but nice to know.

If you are interested different topologies of MOSFET switching and PWM dimming, you can probably find some general schematics online to give you an idea of how all the parts are laid out: MOSFETs, diodes, coils, LEDs etc. You may then be able to reverse engineer your own circuit to help understand how it worked and how it may have failed.

Edited October 20, 2022 by indyld

[indyld]

Just noticed you said you had already swapped the LED board out, not that you were going to as I'd originally read. In that case, there probably isn't much downstream of the FET and so once the coils and current sense resistor etc are checked, you might wish to trace back from the gate of the FET in case whatever is driving it has given up during all the unpleasantness. IC25 looks like it might be of interest purely due to its position on the layout. 

There are cases where inductors or resistors suddenly turn into fuses, and checking them in circuit is easy enough. Obviously, if any of those things are closed or if the current sense resistor has gone way out of spec, there will be no output. 

Edited October 20, 2022 by indyld

[partyanimallighting]

Hi indyld, Hi DrV, I took another stab at the mainboard a while ago. I swapped out the diode plate again with a known working one (just to be sure) and there was no GREEN output so that pointed back to an issue on the GREEN output leg. I then went to ground with a meter at the LED outputs and I got continuity on one leg only on both RED and BLUE but got on both of the GREEN legs so there was a short somewhere on that leg. So, I started to check individual components, comparing between the GREEN and the working RED or BLUE legs but I could find nothing (inductors, L31, resistors R22 those tiny 8 pin IC's [IC25 and IC26, not sure what these are]. I noticed a component numbered 918010 [Q28] but there was only one of these on the board and I was going to bypass it and then I turned the board over and DUH!!!!, there were two more 918010's on the other side so this was a component used on each color leg. Sooooooo, I tested the one on the GREEN leg and it was shorted to ground. I replaced it with one from the old mainboard that I got the other old parts from, started up and BOOM! GREEN output! The 918010 seems to be a MOSFET of some sort but I haven't found a datasheet for it with a cursory search. So, I have managed to repair this mainboard thanks mostly to all your assistance, knowledge and feedback. What I'll do now is order a few of these MOSFETs once I can find them and keep them in stock for future failures (hopefully this was the last one). I've attached images of the three components that were changed ZXMP10A16 (I did find the old one and tested it and it was shorted), C25I3310 (diode I believe?? This tested fine with a meter) and 918010 which is listed as a MOSFET but I haven't found a datasheet for this as yet. THANKS AGAIN!!!! 👍