SLC 5/04 - Added 1746-BAS faults CPU

[stevendt]

Well, the poll has the PSU (7 votes) just edging it from Operator Error (5 votes) - so, on that basis alone, it could still be either

Although still not ruling out the Operator Error option, I am erring towards it being the PSU as the cause of the problem myself now though.

As much as I didn't like doing it, I powered the PLC off/on lots of times today and managed to get some fairly reproduceable results.

It would help if I knew what was supposed to happen, but anyway..... According to the manual the P2 PSU is supposed to be able to hold up power for between 20ms and 3000ms depending on the load. Given that I have just the CPU, BAS, 1 IB16 and 1 OW16 installed, I'd expect the hold up time for me to be at the higher end of the scale. (The CPU Backup battery is reported as being good).

Test Results

Powered off the PSU for 20 seconds on 10 consecutive occasions - the system failed to recover every time and "hard" faulted. During this time, I discovered an apparently reliable way of recovering the PLC after each failure. If I power down for just as long as it takes for the PSU power light to go out (~1sec), the system recovers every time. (This is what I did between tests).

Powered off the PSU for 60-61 seconds on 10 consecutive occasions - the system recovered every time !

Powered off the PSU for 3 seconds on 10 consecutive occasions - the system failed to recover every time.

So, either power it off/on very quicky (1s) or leave a longer time before powering back on (>60s).

One anomoly in the testing though, when powering off for ~5 seconds on 10 consecutive occasions, the system successfully recovered ~50% of the time !

Don't know if I'm missing something about how the system recovers from power failure, but these tests lead me to believe that it is a PSU fault,

Regards
Dave

 

[JesperMP]

I voted the CPU. And it could well be the CPU.
Why does the capacitor not keep the program ? It should be able to do so for 1 hour or so.

 

[stevendt]

Jesper,

the capacitor and/or CPU battery DOES keep the program. The problem is that the system does not always start correctly - as I described, recovery (or not) seems to depend on how long the power is interrupted for.

When I do eventually get it started, the program is still in there and able to run when commanded (I currently have it set to halt on power failure, but as soon as the reset key is used, the program starts provided that the hard fault is not present). Does that make sense ?

504bloke is sending me a CPU and PSU to try this week, so one way or another, we'll know by the weekend.

Regards
Dave

 

[JesperMP]

What "reset key" ?
Do you mean you turn the keyswitch from RUN to PROG and back to RUN ?
If that works, you should also be able to go online and check the Major Error Status word.

 

[stevendt]

Yes, using the PROG/REM/RUN key to clear the "User Error". At that point though, the hardware fault is clear and there is no major fault to see when online

 

[JesperMP]

Take a gander at S:54 "Last Major error". It may tell what the problem was.

 

[stevendt]

Thanks Jesper,

I'll check in the morning, have gone to bed now !

Goodnight

 

[504bloke]

A picture is worth a 1000 words

 

[stevendt]

Thanks Dave,

I have a slightly different display in my Version RS-Logix (7.1) - no S54 displayed, I guess there is probably another way to get to it though.

But now I DO need to go to bed - early flight in the morning,

Regards
Dave

 

[stevendt]

Via CDM, S:54 = 0

(But the software does no recognise this address - it leaves the Description Field empty, whereas for "known" addresses, it puts the Register Description in the Description field - if you see what I mean

Goodnight

 

[JesperMP]

I have an OS401 Ser C, you have an OS401 Ser B.
It seems S:54 came with OS401 Ser C. Too bad.
And you really cannot get online when it is still faulted ?

 

[stevendt]

Yes, really

 

[rta53]

Had an interesting experience today. Had to go to a customer's site and try and fix a problem with a SLC 5/01 system. They had actually removed the entire 7-slot rack and had it in the shop on a table. After reinstalling the unit and powering up, the fault light came on steady and I noticed that the PSU led was off. Even after Pulling the PSU out of the chassis the led would still not come on. When I checked the user 24vdc it was only 21. I could not go online with the processor and even tried a different processor and still no joy. I was able to find a psu and rack locally ( I had flown to the site) and after installing these I still had the fault but it was flashing now. Was able to go online and clear the fault and get it in run mode again. Fortunately program was still in memory and customer was back up and running. The old rack may not have been bad but I replaced it anyway just to be sure.

 

[tomalbright]

I had a 5/03 faulting. Two other guys had gone out, tried reloading the program, and decided the EEPROM was bad. I took out a new EEPROM, installed it, and the CPU was still in fault, and stayed in fault. While the backup on EEPROM is a great idea, they could do without it. I left both EEPROM modules on the shelf, and the system ran.

While the PSU seems to be an issue, it is possible there are other factors involved. PSU would be cheapest, but don't be surprised if you "blow" the new one.

See post above mine. I've run into problems with the power sorta being right, enough to power some things but not others. Good news, I KNOW the 5/05's will run to 17VDC with 3 modules installed!