Yesterday I successfully repaired a Middle Atlantic UPS2200R-IP that had inadvertently been used as a surge suppressor at our transmitter site. It has already been replaced with an updated one but it's coming home with me now instead of being e-cycled. To be specific, this is the MA model that has the blue backlit display and has been discontinued, although this design may have been used in many UPS units including the smaller Middle Atlantic ones in this series.
Here's what happened: We experienced a brownout/surge at the site, which damaged the UPS and took our broadcasting facility off the air. While it was not powering the transmitter, it was powering some signal delivery equipment and kept it on during the transfer to and from the generator. The engineer on duty had to hike in part of the way and found the UPS powered on but showing "E021" with no power to the rear outlets.
Unit was removed from service. Bench testing revealed the following symptoms:
When plugged and cold started the unit gave one relay click and then showed "E021" with no voltage at the rear. IIRC the battery charging circuit operated normally.
When unplugged and cold started the unit powered up and delivered voltage from the batteries to the rear outlets as normal, like if the shore power went out.
If plugged in while running the unit would constantly fire a relay as if to try to switch over to shore power. Power to rear outlets would be interrupted when this was happening.
Middle Atlantic/Legrand was of little to no help saying that they do not offer repair service on these units as they are over 3 years old. They did drop a nugget, suggesting that the input relay may have failed.
With that I pulled off the top cover to have a look at the unit. Removing the main circuit board looked to be a daunting affair and it was. There are a considerable number of connections to this main board, so take copious notes about what connections go where. All but one of the connector points on the board are labeled so use those markings when you are taking your notes.
It goes without saying, unplug the unit and remove the battery pack. Discharge the two main capacitors by shorting across the big heat sink, otherwise you will get more than a tingle when you brush across this heat sink.
Start by cutting any zip ties that look like they should be removed. Squeeze tabs on and push in circuit breaker at upper left rear of unit. Pull up black and white incoming power wires and undo ground screw near rear of unit. Uncoil wires from ferrite cores and pull power input cables over back of unit. This will give you more access to unit and the large strain relief does not need to be removed. Remove large wires going to transformer and pull back, then remove large wires going to UPS rear outlets and pull back. Remove housing for IP ethernet connection board and pull out. From there pull out board mounted plugs and large wires going to heat sink. Again, make notes and take pictures of where everything hooks up. Eventually you will get all the connections off and 5 screws hold board in place. Remove
those and pull the board out.
While I had the board out I did open/shorted testing of all power handling devices such as MOSFET's and diodes. All checked OK. Input MOV's and capacitors were OK. At that point I recalled the hint given by MA Tech Support- the input relay. Here you can see my Sharpie notes on what wires go to the heatsink. The relay has been removed at this point.
I started checking relays and found that one along the input line checked backwards- the normally open contacts were closed and the normally closed contacts were opened. The surge must have fused the contacts this way causing the unit to go into an error mode when plugged in- it could not go to normal mode when the power input was applied.
Discovered that these relays have been officially discontinued or are about to be but eBay was thick with them. 793-P-1C is the Song Chaun part number and Panasonic seems to have one that will work. I ordered two from eBay thinking i was getting two but it was two packs of two so now I have three extras. They are only a few dollars each. While the board is out I would check all of them. The failed relay was RL2 in this case.
I'm a desolder braid kind of guy. That usually works well but removing this relay was a bear. I did remove C22 to allow better access when pulling up on the relay as I was heating up the pins. This is at minimum a two sided board so perhaps a solder sucker would have given better results. Eventually I was able to work the relay free. The heat involved did cause the relay contacts to un-fuse but I did not, of course, re-use this relay.
Soldering in the new relay was easy along with re-installing C22. I slowly and carefully re-installed the board and all connections. I powered the unit up and success! Here is the new relay in place. You can see that the "12VDC" marking is slightly different. The part number I used has an "F" as a suffix; they are electrically identical. The "F" denotes how well the relay is sealed at the bottom. Ignore the "X" on the adjacent relay; it's OK.
Would love to know if this helps or if you have any questions.
Val
Here's what happened: We experienced a brownout/surge at the site, which damaged the UPS and took our broadcasting facility off the air. While it was not powering the transmitter, it was powering some signal delivery equipment and kept it on during the transfer to and from the generator. The engineer on duty had to hike in part of the way and found the UPS powered on but showing "E021" with no power to the rear outlets.
Unit was removed from service. Bench testing revealed the following symptoms:
When plugged and cold started the unit gave one relay click and then showed "E021" with no voltage at the rear. IIRC the battery charging circuit operated normally.
When unplugged and cold started the unit powered up and delivered voltage from the batteries to the rear outlets as normal, like if the shore power went out.
If plugged in while running the unit would constantly fire a relay as if to try to switch over to shore power. Power to rear outlets would be interrupted when this was happening.
Middle Atlantic/Legrand was of little to no help saying that they do not offer repair service on these units as they are over 3 years old. They did drop a nugget, suggesting that the input relay may have failed.
With that I pulled off the top cover to have a look at the unit. Removing the main circuit board looked to be a daunting affair and it was. There are a considerable number of connections to this main board, so take copious notes about what connections go where. All but one of the connector points on the board are labeled so use those markings when you are taking your notes.
It goes without saying, unplug the unit and remove the battery pack. Discharge the two main capacitors by shorting across the big heat sink, otherwise you will get more than a tingle when you brush across this heat sink.
Start by cutting any zip ties that look like they should be removed. Squeeze tabs on and push in circuit breaker at upper left rear of unit. Pull up black and white incoming power wires and undo ground screw near rear of unit. Uncoil wires from ferrite cores and pull power input cables over back of unit. This will give you more access to unit and the large strain relief does not need to be removed. Remove large wires going to transformer and pull back, then remove large wires going to UPS rear outlets and pull back. Remove housing for IP ethernet connection board and pull out. From there pull out board mounted plugs and large wires going to heat sink. Again, make notes and take pictures of where everything hooks up. Eventually you will get all the connections off and 5 screws hold board in place. Remove
those and pull the board out.
While I had the board out I did open/shorted testing of all power handling devices such as MOSFET's and diodes. All checked OK. Input MOV's and capacitors were OK. At that point I recalled the hint given by MA Tech Support- the input relay. Here you can see my Sharpie notes on what wires go to the heatsink. The relay has been removed at this point.
I started checking relays and found that one along the input line checked backwards- the normally open contacts were closed and the normally closed contacts were opened. The surge must have fused the contacts this way causing the unit to go into an error mode when plugged in- it could not go to normal mode when the power input was applied.
Discovered that these relays have been officially discontinued or are about to be but eBay was thick with them. 793-P-1C is the Song Chaun part number and Panasonic seems to have one that will work. I ordered two from eBay thinking i was getting two but it was two packs of two so now I have three extras. They are only a few dollars each. While the board is out I would check all of them. The failed relay was RL2 in this case.
I'm a desolder braid kind of guy. That usually works well but removing this relay was a bear. I did remove C22 to allow better access when pulling up on the relay as I was heating up the pins. This is at minimum a two sided board so perhaps a solder sucker would have given better results. Eventually I was able to work the relay free. The heat involved did cause the relay contacts to un-fuse but I did not, of course, re-use this relay.
Soldering in the new relay was easy along with re-installing C22. I slowly and carefully re-installed the board and all connections. I powered the unit up and success! Here is the new relay in place. You can see that the "12VDC" marking is slightly different. The part number I used has an "F" as a suffix; they are electrically identical. The "F" denotes how well the relay is sealed at the bottom. Ignore the "X" on the adjacent relay; it's OK.
Would love to know if this helps or if you have any questions.
Val
Attached Files
Comment