Smart UPS 600i

[Anonymous user]

Hi,

My trusty old UPS has got it's brains scrambled and I was wondering if anyone here has much knowledge of these vintage units?

The problems started when I changed the batteries and like an idiot I forgot it was still switched on, this caused a few sparks!
Now the UPS fires up but will only run on batteries after a few seconds of direct mains when switched on. It will quite happily run till the batteries discharge so I suspect all the hardware is OK.

On investigation it would appear that the EEPROM is possibly corrupted. Using APC's Powerchute software or apcupsd showed that it thinks it's an M model and not an I. This means the output voltage on battery is only 208v instead of 230v and consequently thinks the incoming mains is far too high hence switching immediately to battery. (at least that's my theory)

When I interrogated the EEPROM, apctest reports that it's not programmable although I have managed to write a name and battery date into it?
The other strange thing is it thinks it's a ups 250 and not a 600??

Trying to change the 'on battery' voltage in either software fails. In Powerchute it only shows 208v in the drop down box. In apctest if you try to write the new value the software says the reply received is 'NO'.

Any ideas?

Thanks in advance
Midibob

[Anonymous user]

After a bit more head scratching and Googling it would seem that these units have more than one copy of firmware written in to them and therefore it should be possible just to select the correct version.
With later units this is done in 'PROG' mode which is selected by 1 (delay) and 1, trouble is this unit just ignores this command?
I must have tried just about every other combination of characters with at least 3 second delays but no 'PROG' mode.

I'm sure someone knows what the magical command is to revert this unit back.

Happy Xmas to all.
Midibob

[Anonymous user]

It's amazing what you can find on the web with a bit of Russian translation!!

Apparently these early units (1st or 2nd Gen) need to have the temperature sensing pin on the ADC shorted to ground to enable the hidden modes.
The command structure is still a bit baffling and although I can now change various register settings I still can't change the model type.

I'm sure I will get there in the end but the odd pointer would be nice from someone in the know.

Midibob

[Anonymous user]

Things are looking up.......

There appears to be 7 registers that are adjustable or at least that's how many I've found.

Numeric keys 5, 6 and 0 are adjusted by keys 2, 3 and 4 respectively and they only shift in a round robin fashion so if you miss a setting you have to go all the way round again (00-FF)

'b' I was led to believe selected the firmware but I don't think it does as that appears to be done by 'H'

'a' and 'd' also do something but I don't know what?

Anyway for now the values of 'a', 'b' and 'd' have been left at 00. They may have something to do with voltage calibration but I can come back to these later.

The other values are as follows :-
5 - EA
6 - F4
0 - 9F
H - E5

'H' cycles through all the different model types 250/600/700/800/900 and various output voltages.

Anyway, I now have the UPS running at Firmware 6TI which means it now delivers 240v. Phew!!

Still love to know what the other settings do.
This must be documented somewhere..........anyone?

Midibob

[Anonymous user · Accepted solution]

Just a quick follow up to this problem.

After a bit more research this UPS doesn't have a separate EEPROM, from what I understand the 87C52 holds the data?
The firmware revision reports it's 2TM which kind of makes sense as that would make it a UPS 250 (2) and North American 208v server (M).
What I don't understand is even if the data was corrupted how come it thinks it's got the wrong firmware for the device.

The question still remains is it possible to recover from this?
I've tried clearing out the data with apcupsd and it replies with 'CLEAR' but doesn't appear to do anything, possibly as it hasn't got a separate EEPROM?