Alright, I have one more question regarding my 4005 that I’m hoping I can get answered. Hopefully this should be it for a while!
Ever since I received my 4005, it has had an earth/ground trouble on it. It’s always something I’ve wanted to get resolved, but I haven’t reached out for help, because I figured the reason I was getting the trouble is because I don’t have anything plugged into the Earth terminal on the power supply.
I’ve gotten a couple helpful comments on my YouTube videos from the same user, one of them I just got tonight, saying that having a connection to that terminal has nothing to do with the trouble, and that the trouble is either coming from a positive wire being shorted to the metal housing or a short somewhere else.
I have checked the whole panel and all the wiring, and there doesn’t seem to be anything wrong. If I remember correctly, I was still getting the trouble before I had any other wires (except the AC power wires) plugged in.
So all I can assume is that either I need to plug in a wire to the Earth terminal and run it wherever it’s supposed to go, or my power supply is bad, because all the other boards on the panel (except the DACT) have been replaced. Any ideas?
The ground trouble can be caused by either a positive wire or a negative wire touching ground. The system checks for both. The NAC/relay card has some MOVs near the terminal strip that are grounded by the foil on the bottom side of the board touching the chassis. If an MOV is leaking it could cause a ground trouble. Remove the NAC/Relay card to see if that clears the trouble. It is also possible there is a wire fragment or drilling chip caught under the power supply board.
To determine if there is a real ground or the system is lying make the check in the drawing below. The voltages shown are from a real panel in normal condition. They can be different from those shown by a volt or two and be OK. If they are off by many volts there is a real ground fault somewhere in the system.
Be careful when working around the power supply. It is a switching power supply so there can be as much as 350 volts in a few places.
Alright, so I tried everything you said.
I first removed the NAC/Relay card, and the trouble did not clear.
I got a voltmeter and measured the two terminals with the black test lead attached to the bare part of the chassis. On the top terminal, I was getting 22 volts, and on the third one down, I was getting -2 volts, so something is definitely not right.
What is the next step I should take?
That voltage difference definitely indicates a ground fault on the negative side of the system. I’ll check my resources to see what the next moves are.
Just curious, does your have the expansion power supply on the left side or just the standard power supply on the right side?
There are only a few other places where there could be damaged components causing a ground fault. I will have more details later.
Nope, I’m just running one power supply.
Not sure if I’m offering any helpful information here, but the panel was partially fried when I purchased it used off of eBay a couple years ago. I think you made the suggestion it might have been struck by lightning. Additionally, when I received the panel, the I/O cards that originally were with the panel were not packaged at all and were floating around inside the panel, not plugged in. They all had components broken off of them (I assume during shipping) that would have had to have been re-soldered.
As far as damage, I had a chip blown out on the CPU, a chip blown out on the Power Distribution Board, and no I/O cards that looked like they weren’t damaged.
So I went ahead and spent some money to replace everything on the panel, except for the power supply, and the DACT, which I’m not using. The ground fault was there both before and after I replaced everything.
I remember that now. That leaves the power supply as suspect. I will concentrate on the PS to find components to take a close look at. I bet we find one burned, cracked, broken, or with its insides spewed out.
I found the components on the power supply that are the most vulnerable given the previous damage we found in your system. The picture below is a little fuzzy but it will give the locations to look at. D2, D5, and D9 are 40 volt transorbs. Any one or several could be damaged because they connect between system power rails and ground. C8 (the orange cap) also has the same connection but is less suspect. Towards the top of the picture is U3. That chip is the ground fault sensor. It is possibly damaged. Look at all the components in the area because any of them could be damaged, but the ones I named are most likely.
A quick check is to remove the screw in the lower left corner of the board. This is a plated through hole so slide a little piece of paper under that corner. This will break the ground connection to these components. If one or more are damaged and leaking the ground should clear or the voltages measured in the previous troubleshooting should change more towards normal.
Thank you very much for taking the time to do that.
When I get home tonight, I will give those components of the power supply a good look over, and then I’ll do the check with the paper and report back.
I didn’t have a whole lot of time last night before I had to head to bed, but I checked the components and did the paper check really quick.
I inspected the components in the mentioned area on the power supply very thoroughly (as good as I was able to without removing the while thing), and to my surprise, I did not find anything out of the ordinary. No damage I could see.
So I removed the screw in the lower left hand corner and slipped a piece of paper under the board. I’m honestly not sure if I did it correctly, because the paper wasn’t going in very far at all. I’m not sure if it was enough to break the ground. When I powered the panel up, I still had the ground fault, and I was still getting the irregular voltage readings on those two terminals.
You mentioned a possible culprit could be a tiny piece of metal stuck in the board. Should I remove the power supply and maybe use compressed air to blow off the front and back of the board? I suppose it would also be beneficial to remove the board just to get a better look at all the components.
Well darn, I thought we were on to something. If you got the paper under the board in the correct position you can see it through the screw hole. The idea is to separate the underside of the board from the mounting post.
I would pull the board for a closer look and maybe some ohmmeter checks on those components. And just to see if something falls out while you are doing it. It is 3 screws and that 13 pin connector which is a bear to take off. But you already know about that from replacing the power distribution board.
The only other components I could find that go to ground are 2 MOVs on the top of the CPU board. On the drawing I have they are located below the relay to the left of the EXT RUI/DET PWR terminal strip. They are labeled RV1 and RV3. They connect to the External RUI terminals. I guess they need a look at too.
Oh, okay. That’s probably what I was doing wrong. The paper didn’t make it up far enough to where I could see it through the hole. So should I actually pull the board forward and then slip it under? Because just slipping it under only allowed me to get it so far. It looked like it was hitting a tiny metal post. I’m obviously being overly careful, haha. Just really don’t want to damage anything more than it might already be damaged.
I should hope the problem doesn’t lie in the CPU; I just replaced that son of a gun, but I’ll check those as well.
Okay, so I got more time with the panel tonight. Here is my current information.
I pulled the board and made sure to slip the piece of paper under all the way, to where it was visible in the screw hole. I powered up the panel, and still received the ground trouble. However, when I tested the two terminals again with the volt meter, I got something different. I was getting 0 volts on both of them. I repositioned the test probes several times to make sure I had a good connection to each terminal and the bare metal on chassis. Still, I was getting 0 volts.
I powered the system down, removed the paper, put the screw back in, powered back up, and re-tested. This time, I was getting the same readings as before, 22 volts on the top and -2 volts on the third down.
Next, I removed the board just to do a visual inspection, since I do not have an ohmmeter, and I do not know how to test components with one and what I would be looking for as far as readings. Nothing fell out when I removed it that I could see; I still just used my breath and blew very hard on the front and back of the board to clear anything that might have been on there.
I took a flashlight and did a thorough and detailed inspection of every single component on the front of the board. Everything looked normal. No burns, cracks, or anything out of the ordinary.
However, when I turned the board over, I noticed a lot of white, powder-like stuff scattered throughout the whole board. On the top, the board almost looks like it is corroded. I doubt it’s anything, but I snapped a picture of it before I put the board back on. Once I can remember the website I was using to keep images on after I closed my Photobucket, I’ll add it.
Been studying that picture for a while. All I can do is come up with some “could be” things. In the past there have been boards that look like they had water spots on them. However, the spots were left by the cleaning process before going through the wave soldering machine. Those spots were not like the white looking residue on this board. It kind of looks like residue left if a board gets wet by water that leaks through masonry. That water picks up lime and other chemicals and can make white crusty deposits. All I can do is guess at the moment.
Does the insulator under the board or the sheet metal chassis show any signs of being wet?
Let me get this straight, you have a voltage only meter that does not have resistance or diode check capability. That is a handicap because I can think of some other ways to figure out where this ground is coming from. Let me know.
No signs that anything was wet. I found some dust on the insulator, but that’s about it.
As far as the volt meter, I have to check it again when I get home. It’s my dad’s. I know it only has three settings, one of them being off, one of them being AC/DC voltage, and the other one just says “AC” or something like that.
I can always pick up one with more capabilities at Harbor Freight Tools or something like that.
We’ve got one, so that’s covered.
Which components should I check?
OK. Is the panel back together or is the PS still out? Don’t do anything yet, just answer.
Is the meter analog or digital?