CHAUVET Arena Hazer 3 DMX issue

[indyld]

Plug the DMX board in and measure the Vcc at the chip relative to the GND pin. Also check your 0v ground connection. 4.2 seems pretty low for a 7805 even allowing for tolerances during measurement. 

 

Datasheet suggests a minimum 4.75v for the chip. 

Edited December 3, 2022 by indyld

[DrV]

If I've managed to turn the picture round correctly those two pins supply the 5volts to the MCU side of the opto i.e. the non isolated side.  The white wire is the DMX signal out of the opto going to the MCU. I can't see what the green and red wires do as they're obscured in the photo. The supply for the 75176 comes from elsewhere, and I think that's the 3 pin connector. As Indyld says, check the voltage directly across pins 5 and 8 of the 75176. And check the voltage on the 3pin connector.

[partyanimallighting]

So based on your advice, even though the voltage is a little low, 4+ volts compared to 5+ volts, I should concentrate on the DMX PCB and ignore the mainboard. I'll power up and check voltages on the optocoupler first. According to the datasheet, pins 5 and 8 are ground and voltage in so I should get 5v (or rather 4+ volts) at these pins. If I don't get voltage at pins 5 and 8 of the 75178 that will point to a faulty optocoupler. Hope this sounds correct...... Is checking the voltage on the 3 pin Molex really necessary? It seems to be only a harness to feed the DMX signal from the 3 pin XLR input/output to the 5 pin input/output below.

 

[DrV]

Nope. The voltage on the black and yellow pins is the supply voltage for the optocoupler. You can check it on the optocoupler pins if you want but you'll get the same voltage as you're seeing on the black and yellow. You can actually see the tracks connecting the black and yellow wires to the opto on the underside of the board (see C3).

The supply to the 75176 comes from a separate source, nothing to do with the optocoupler. As I said before, it appears to come from the 3 pin connector. I suggest you verify that by checking continuity between pin 8 on the 75176 and the + pin on the 3 pin connector, then check continuity between pin 5 on the 75176 and either the - pin or the gnd pin. If there is continuity then we've worked out the circuit correctly. That means that with the power on there should be a voltage on those pins. If there isn't then the main board (which should be supplying that voltage) is at fault or the 75176 has gone short circuit and is dragging the voltage down.

[partyanimallighting]

Hi DrV, I did as requested and the voltage to the optocoupler was the same as those on the yellow and black harness wires as expected. I know that the 3 pin Molex is just a signal feed from the 3 pin DMX PCB to the 5 Pin DMX PCB and there were no voltage readings on this harness and there are no components on the 5 pin DMX PCB but I've still attached pictures. I suggest you verify that by checking continuity between pin 8 on the 75176 and the + pin on the 3 pin connector, (PROBE - at RED terminal wire OL. I'm assuming this is because the solder trace passes through diode D3. Probe + at RED terminal 612 ohms and continuity is fine across all solder traces from the 3 wire harness back to DATA+ input and output) then check continuity between pin 5 on the 75176 and either the - pin or the gnd pin. (PROBE - at BLACK terminal 701 ohms. PROBE + at BLACK terminal OL). I also removed the 75176 and the optocoupler to take pictures of the solder traces if this will help you guys. So from what I can ascertain, the 3 pin harness has nothing to with powering the PCB and only carries the DMX signal to the 5 pin DMX INPUT/OUTPUT PCB but with my limited knowledge of components and how they work I could be wrong. If the 75176 feeds the DMX signal directly to the harness wires then I should a reading from pin 5 on the 75176 to GROUND and I'm getting 200 ohms as the solder trace for DMX PIN 1 passes through R1 (200). I also checked all the small components and I'm getting continuity in one direction only on R3 and R8 and no continuity in either direction on R6 (202) but I'm getting a reading on R6 of 1.98 Kohms and not 0.202 and this I find to be strange. Why would I get a resistance reading but no continuity or am I not taking a correct reading? Please go easy on me. Every time I post it's a learning experience for me.

 

 

Edited December 5, 2022 by partyanimallighting

[indyld]

Does the 75176 have 5v at it's Vcc pin relative to it's GND pin? 

Sorry if this has been confirmed already somewhere but there is a lot of information here that is of less importance than the IC having power. 

[partyanimallighting]

I just reconnected the PCB to the unit to retest and the readings with the 5 pin harness attached to the PCB are fluctuating around 1V  and at pin 1 0.9V and at pin 8 1.1V on the 75176. When I powered down the unit and removed the PCB completely and retested at the black and yellow harness wires, I got a stable 5.07V. Once the PCB was reconnected that voltage dropped back down to >1V so there's something sketchy on the PCB itself. I hope this helps.

[DrV]

You still didn't answer Indyld's question though. With the - probe on pin 5 and the + probe on pin 8 what voltage do you get?

[partyanimallighting]

Sorry about that. I wasn't clear above. I'm not getting anything near 5V from pin 5 (gnd) to pin 1 or pin 8. From pin 5 to pin 1, 0.9V, from pin 5 to pin 8, 1.1V with the -tive probe at pin 5 (gnd). This is a 75176 that was taken off a tested working PCB so it should be fine but.... If required, I could replace this 75176 with another tested working one but I'll still retest this original component when it's removed. Obviously, with the 75176 not outputting 5V, nothing will work and something on the PCB is causing the short as the terminal outputs a steady 5V when the PCB is not connected and all you guys suspect that 75176.

Edited December 5, 2022 by partyanimallighting

[indyld]

Take the IC off and re measure the same pads on the PCB. You are now  trying to find out if the problem is the end point (the DMX chip) or something else on the DMX board. The latter sounds more likely but needs verifying. 

Edited December 5, 2022 by indyld

[indyld]

Reading back through the mass of information in the thread, I *think* you have verified that the fault is probably not on the main board as everything works fine with a known good DMX board swapped in and that the 5V gets to the DMX board -  but is pulled low when the suspect board is plugged in.

It seems like you currently have a known good DMX chip on the board, yet the power rail is still being pulled low by something. Once you have removed the chip and verified that is not the reason that the Vcc is being pulled down by something, beep around the rest of the board and mark every other component that has a connection to this rail (pin 8 on the IC?).

It's only a small board, and you can check much of them in circuit but I'd start by looking at the main culprits for 'things going wrong' in this manner which would be anything with silicon junctions in it, diodes, opto, zener diodes or similar. You can also check that a resistor isn't way out of spec. Unless you measure a dead short across the rail (perhaps less likely given a 1v reading) then you might be hard pressed to easily identify a dodgy capacitor without injecting voltage. In general, when trying to find faulty capacitors I would start with electrolytics and those under most pressure - particularly the larger MLCC ones - in any circuit. There doesn't seem to be much that fits that criteria here, so I'd look at silicon first.

Having said that, there are so few caps on the board that you could just measure the rail resistance and temporarily remove one cap at a time checking to see if the reading changes.

Edited December 5, 2022 by indyld

[partyanimallighting]

Hi indyld, just a note, I did not test with a known working DMX PCB but everything basically is correct otherwise. This seems to be dragging on forever and consuming a lot of my time when available and more unreasonably, your time. I did pull the 75176 from the DMX PCB and voltages were 0.9V at pin 1 and 1.1V at pin 8. This was before I discovered what may be the reason for the low and fluctuating voltage coming into the DMX PCB. I got the problem without the DMX PCB plugged in with voltages ranging from 1V to 4.5V so I decided to change the LM7805. I then tested the voltages at the harness and got a steady 5V so I figured, problem solved. So I plugged in the DMX PCB and as I touched the harness the voltage dropped and started to fluctuate. I jiggled the harness wire and I got it into a position where the voltage remained at a steady 5V so the harness appeared to have a short of some kind and these types of shorts I think are the worst type to find. One minute you're fine. Next minute, nothing. So I tested the harness for continuity and flexed it every way but no shorts. I then pulled the mainboard and resoldered the solder points at the Molex terminal and all other components close by (3 capacitors and a transistor below the LM7805) and just to be sure, I replaced the yellow and black wires in the harness. I then reassembled and tested and got a steady 5V........until I touched the harness and the fluctuation started again. I finally got the harness into a position where the short disappeared and clamped it securely in place. I then put another meter into the harness to confirm a constant 5V into the DMX PCB and then connected the hazer via DMX to a console. With the second meter reading a steady 5V with the harness clamped, I tried control via DMX and nothing, no output. I then tested voltages at the 75176 and got 5V at pin 8 and 3V at pin 1 (pin 5 test probe -tive). Should I now think maybe it's the mainboard with that frustrating short? 5V in at pin 8 should be the correct reading but maybe 3V at pin 1 is a little low? Also, could a faulty component on the mainboard cause the fluctuating voltage at the harness? I can't see how but I'm really guessing here now. The 5 pins were resoldered and wires replaced and the short is still there but I could always remove the Molex connector completely and solder the harness directly to the mainboard. I did change only 2 of the 5 wires. Could a short in the other wires carrying DMX signal cause a faulty component to short and cause this? I'm really guessing here now and it seems that all I'm doing is passing on my frustration onto you guys with these way too lengthy posts.

[DrV]

I think we are missing a bit of vital info which is; where does the supply for the 75176 come from? It is unlikely to be the black and yellow wires as these supply the MCU side of the optoisolator. Can you check where the supply pins (5 and 8) to the 75176 come from as it's not obvious from the pictures. My favoured suspects are the green and red wires, but use your meter to verify this. Where they shouldn't come from is the same place as the supply pins to the opto (also pins 5 and 8) because this defeats the object of having an opto there in the first place. As you've removed the two chips this is a good time to buzz things out.

[indyld]

I'm happy to help. Doesn't take much of my time thinking and commenting.

I think here is an example of missing some important verification steps before making assumptions and altering the device under test too fast and in too many ways.

One thing to remember at the moment is that just because you can cobble together a steady 5v at the DMX chip, doesn't mean that your final test method, running the device via DMX, is testing that alone. Keep all potential possible faults (including multi-element ones) and only rule out exactly those that can be verified properly. 

Edited December 6, 2022 by indyld

[partyanimallighting]

Quick question. Shouldn't I first of all be sorting out the short in the harness that causes the fluctuating voltage at the yellow and black wires? I'll replace the existing Molex connector but I don't see how this will help and the original one seems to be fine and I'm figuring that the short when the harness is jiggled is caused by a bad solder point on the mainboard terminal but I touched these up already. Or maybe components feeding the RED and GREEN wires? I also just checked the voltages at the other 3 wires (with the second meter in the yellow/black on the shorting harness) and I'm getting 4.98V from both the RED and the GREEN to chassis ground on the mainboard and I'm also getting 4.6V from the RED and GREEN to pin 5 (-tive GND) on the optocoupler so that's the voltage that feeds the 75176, correct? . . . . . . . . . . I'm lost. I'm just not sure how to proceed. Try to find the cause of the short? Replace the Molex? What if it's a component feeding those 5 wires that's causing that fluctuating voltage?