APC SMX1500RM2U - Clicks constantly, never actually turns on

[JOHN_FL]

I've read many of the "it clicks constantly" threads regarding this APC, as well as others.  I may have missed where any of them had an exact answer, other than "it's broken", but let me ask anyway:

1. Upon plugging my unit in (It was allegedly new in the box, but had been in storage for many years), it clicks constantly, a cadence of about once a second.
2. All LEDs flash in synch with the click
3. This happens with the battery plugged in.  (But in troubleshooting, I've done it without the battery also)
4. There's no response from pressing and holding the power button to attempt a brain-dead restart
5. I tried leaving it clicking for a while to see if eventually it changed.
6. The battery pack is definitely dead - voltage builds up to about 3 volts, and if I leave the multimeter for a few seconds, the voltage goes down constantly, 1/100th of a volt at a time.

Knowing the battery pack is dead creates the temptation, Hey, try with a new battery pack.  But at $100 plus (that's if I pull it apart and replace each of the 4 12 volt 9ah batteries), that's a gamble to take, knowing the device itself may be malfunctioning.

Key questions:

1. Is it normal to click?  I would assume there are hundreds and thousands of UPS devices the world over with dead batteries.  Given that there is a warning indicator for dead batteries, I feel confident in saying "Hey, it shouldn't click, even if the batteries are dead..
2. Are there any other troubleshooting ideas? 
3. I'm just an end user.  The device was given to me, and it'd be nice if I could get it to work. 

Thanks in advance!

[Teken]

The SMX Series UPS is hot swap and cold start capable. As such it can be turned on with battery only when AC Mains isn’t present.

It also allows the battery cartridge to be replaced (Hot Swap) while the unit is running on AC Mains.

The system in normal operation will (click) several times upon power up. The click you hear are the relays opening / closing for the input / output / inverter circuit.

Based on your description there are a couple of possibilities from most likely to possible.

1. Relay Weld (Failed internal relays)

2. Bricked system from firmware load

3. MCU is in a unknown state

4. Internal hardware failure of various components from relays, power transistors, capacitors, etc.

If the system is connected to a computer running APC Power Chute Software you might be able to see what the active fault / error codes are. If there is a NMC installed you can login and view the Event / Data logs to see the history.

If the system came with the EPO plug you may consider connecting the same as outlined in the user manual. Once connected you can short the cable to engage the EPO circuit.

The system should immediately shutdown and disconnect itself from AC Mains (internally).

Remove the short from the EPO circuit and disconnect the AC Mains from the wall outlet. Wait another 30 seconds and than press and hold the power button.

Typically you would hear a click / beep as this would complete the Brain Dead procedure. In this case you won’t but it doesn’t matter because this step will still discharge the capacitors inside the unit.

Let me know what you see and observe. Lastly, I’m going to assume you never had this unit working or saw it working since opening the box?!?

What is the serial number for this unit?!?

[JOHN_FL]

@Teken, thanks for the reply.  I'm not home to try the suggestions out yet (I admit I didn't have high hopes for diagnostics since it never fully turned on, but it is worth a try.)  I do have the Serial Number handy from a photo, it is AS1538324833

[JOHN_FL]

Also, no, never have seen the unit in operation.  There's still an inspection tag on it (from 2015!) saying Ok/Passed to everything. 

[Teken]

Let me know results when you get them.

As an aside if you’re truly located in Florida (User Name) hope very much you, family, and home were fine from the impact of the two hurricanes! 👍

[JOHN_FL]

In addition to last Monday, I will attempt a reply (again.) 

Thanks very much @Teken, for the help.

The powerchute software didn't get too far.  The device never actually reached a state of being turned on.  (I also didn't keep trying the install when the unit couldn't be detected.  Perhaps I'll try harder.)

Using the EPO plug to short it out gave a momentary glimmer of hope, since after the shutdown, pressing and holding the power button after unplugging/waiting 30 seconds, the unit did click just once, as if something different to before happened.  But once plugged back in, it was back to constant clicking.

[Teken]

Hi John,

I normally would not suggest the following but you can try this if you want to.

Meter: If you have a True RMS multimeter available please measure the output at the UPS. What voltage is seen at those UPS outlets?!?

Case: Ensure the power outlet is disconnected from the wall outlet. Press and hold the power button for about 30 seconds.

Remove the screws from the top cover. You’ll see a row of brown / black rectangular relays. Take a plastic tool and tap / strike the relays 4-5 times not too hard but enough at the top / sides.

Reapply AC Mains to the unit and let me know if the clicking frequency changes.

IE. It used to click on-off every second. Now it clicks less or in a different pattern.

Let me know what happens keeping in mind you don’t need to do this step if you’re not comfortable in doing so. 

As we can skip to another possible route assuming it’s not the relays. 👍

[JOHN_FL]

Thanks, yet again.  

voltage fluctuates from 0, 0.8, and 1.4 Volts at more than one outlet.  And I was impressed that my $15 multimeter from that place is True RMS!

Pictures attached.  

Upon tapping, all the relays felt the same and seemed firmly seated.  

no obvious capacitor swelling or leakage 

the connector at J4-10 (dead center left to right and about 60 percent toward the top) next to the vertical/edge toward the camera board had a small amount of wiggle but was probably seated ok. 
but.  

And I don’t know what I did wrong in my past life to deserve loose parts in APC UPSs, in the two wooden-handle pics, note the loose magnet.  (I plopped it in place for the full board picture). There is dried glue on the inside of the loose magnet, clearly it was on that transformer once. (But I haven’t messed with trying to plug in / turn on again yet. )

 

 

 

 

[JOHN_FL]

I should clarify:  the last picture is where the magnet was, holding itself against the frame of the ups. 

[Teken · Accepted solution]

Hi John,

Put the magnet back where it’s supposed to be in the photo. Fire up the system and listen carefully as to if the relays click differently.

If they make the same timed clicking noise. Take that Woden handle and strike the rows of relays.

If any of them are stuck while the system is powered on they will react if struck.

DO NOT ATTEMPT to do this if you’re not sure if this can be done safely! As there are very high voltage present when the system is running and energized. ☝️

[JOHN_FL]

Thanks again.   I’m searching to see if the various glues I have at hand are high temp enough.  

also unsure so far if the magnet is polarized or not.  It drops into the slots whether right side up or down. 

knowing it’ll be another 24 hours for my reply to post (if I’m lucky) I may have more news to report before tomorrow.  

[JOHN_FL]

There was a momentary turn-on moment, but all too brief.  The fan even spun up for a second, then it went back to clicking.

Tapping all the relays didn’t yield any striking new info.  

However, some new discoveries (new to me, anyway.)

Trying it very safely and carefully, I did notice the fan does actually spin quietly/slowly while connected to AC power.  (Not as fast as the brief moment where it seemed to be turning on though.)

In the close-up pictures, it seems clear to me that for the assembly in question (what’s it called, anyway?) and what I’ve been wrongly calling the magnet (it’s not a magnet, but it is magnetic aka magnets stick to it), the bottom part (under the assembly) should reach up and touch the “magnet”.  (You can see in the closeups where the glue lines indicate they were once connected.)

They do not reach one another unless the glue/etc keeps them together and thereby keeps the lower part raised,

 

 

 

 

As you can see with the half-dinglehopper photo (some call it a fork, but they’re wrong), the assembly to the bottom of the transformer can rise (and I’m guessing should rise) to touch the top.

I haven’t yet tried re-gluing, etc.  It seems logical, but at the same time, I can and will wait to hear more of what you think. Thanks yet again

[JOHN_FL]

To add a little context.   The core is in two pieces?  

 if I end up trying to glue them back together I’m not sure what type of glue I would feel like I could count on.  

[JOHN_FL]

This took forever to find... here's a photo of what a normal one looks like, found in a post on the internet's most famous auction site:

I'm still wondering what to glue or tape that with in a reliable way.  A sneeze seems like it would shake this loose.

[Teken · Accepted solution]

Hi John,

I would use any Epoxy glue you have on hand to secure the two pieces together. This piece is a ferrite magnet which is intended to reduce RFI / EMI while focusing the EMF back into the coil. Once you have that part repaired there is really only three options.

Relay: If we assume the relays are the main culprit than replacing them is the solution.

Firmware: The behavior you see of the system clicking its relays is also very similar to a unit that was flashed with firmware and it went south and bricked the entire system. If you had a healthy battery cartridge and the supplied OEM serial cable.

You could try to see if flashing the firmware restores the system.

Given, you don't have any new batteries on hand that option is off the table for now - thoughts? 

[JOHN_FL]

SUCCESS!!! (plus one final discovery): There's a safety circuit built in, so after the epoxy fix holding the top and bottom parts of that transformer together, AND (key point, "and") actually putting the cover back in place, it fired up, turned on, and proceeded to tell me the batteries are dead.  (Before putting the cover back on, it was doing a couple things differently than before, but ultimately not working: It was going about four seconds between clicks instead of every second, but still clicking.  With the cover on, it's up and running!   

As a computer hobbyist, where no self respecting hobbyist actually has the cover (desktops I mean) in place (or if they do, it's not screwed on), I had to laugh. 

But it is on and operational!  Now, I'll go ahead and buy new batteries for it.  Thank you @Teken for your expertise, patience, courtesy, integrity, perseverance, self-control and indomitable spirit!  FYI, I can't send private messages as a non-corporate member, but if you can send me one, please do.  

[Teken]

Hi John,

That is absolutely fantastic to hear it’s back up and running. Can you confirm the system allows you to turn the output on / off?

If so please confirm with your True RMS Meter the output voltage. Confirm the voltage drops off when you command it to do so via the LCD.

LCD: Please confirm the LCD indicates the input / output voltage is what you see on the multimeter (approximately).

Power: Remove AC Mains from the unit the system should power down immediately. Apply power to the unit what happens? Does the system turn on or does it require you to press the power button?

EPO: When the system is running engage the EPO circuit by shorting the two leads. Confirm the system immediately powers down. Your meter should indicate there is no voltage on the outputs.

Remove AC Mains from the wall. Apply AC Mains to the unit. The LCD should indicate EPO something and to press and button - do so.

What happens?

A few gentle reminders when the new battery cartridge arrives.

Battery: Insert the cable into the system. About 10-15 seconds the system will prompt you to confirm a new battery cartridge is in place. Toggle from No to yes and enter the month / year and hit OK.

Scroll through the LCD to confirm the battery charge and voltage and report back what you see.

Once the battery is 100% charged manually initiate a Self Test. Once it passes wait for the battery to show 100% charge state.

Follow up by initiating a battery calibration with at least 30% attached load. Let this process complete and let the system recharge the battery cartridge to 100%.

Confirm the operational runtime is close to the APC Runtime Calculator based on a defined load say 50~100% load.

Ill send you a PM now - Congrats! 👍🔋🍺🍺

[Teken]

Hi John,

The system indicates you don’t have PM enabled please check. 

[JOHN_FL]

So far everything looks good.  I can't mess with it too thoroughly right now, but it turned on again today (a good sign!) and reads 120.5 at the outlets.  On unplugging and replugging, it returns to being on.  After triggering the EPO (which I've done a lot of during the course of this,) I have to turn it on manually. 

Regarding the other stuff, I remain what Schneider must believe to be a "not worthy" non corporate account/home/retail customer, whose posts all wait until overnight to post, and who doesn't get to even see the Private Message Icon.  

I did have a private message inbox visible briefly yesterday, after a moderator messaged me about "sorry we marked your post for deletion," and I even attempted sending you a message while I had the ability.  Not sure if it reached though.  But I don't have the inbox icon at all (this, copied from the "how to" video)

Thank you again, @Teken!  I'll have batteries some time next week, I bet.  My only issue is if I'll be suffering from a serious case of overkill having way more of a battery backup than I realistically need.   I'll check again tomorrow morning, when this post is approved.

[JOHN_FL]

I’m not sure whether to be proud or ashamed of this.   But I will say one thing, this epoxy bond isn’t going anywhere!

 

 

[Teken]

Epoxy -> 🤣 🤦‍♂️ 👍

I tried to send you a PM but the system indicates it’s not turned on?!?

I also did receive your PM but could not reply back given the above. 👎

[JOHN_FL]

I briefly had my envelope icon/PM available for a bit this morning again, but gone now.  

As long as you saw the thank-you picture I made for you of the UPS in it’s up and running glory!  That was a battle!

[Teken]

Hi John,

Unfortunately, the system filtered that out from your message to me. ☹️

[JOHN_FL]

I'm trying!  😉   As it stands, my replies take a day to be posted, three days if it's a Friday, and my Private Messages have briefly worked only twice in the time I've been here, seems to only be days with a "T" in the name between 6 and 7 am.  😉 

[JOHN_FL]

Perhaps someday.   Thank you again!

[IVI]

Guys,

great job resolving John's problem.

Now, when you guys have experience with this model, I need your help as I have the exact same problem.  🙂

I also have SMX1500RM2U that won't turn on. It worked for an year or so, and then it stopped. 

When I try to plug it in:

- I hear a clicking sound from those 3 relays

- The fan starts turning for about a second

- the display flashes for few microseconds, and then goes out again.

- The output voltage jumps to about 118 volts for few microseconds

I have opened the device and:

- Did not see any missing, broken, loose, or burned parts.

- I tired hitting the relays a couple of times to cause them to release

Is there anything else you guys could recommend me to check?

I don't believe that it is a firmware/software problem as nobody have messed with that, and it was working without issues before it suddenly stopped.

Do you guys think that there is anything I can check? I don't really want to give up on it.

Thanks i advance!

I.V.I.

[Teken]

Start a new thread with the following details:

UPS serial number and battery production date. Firmware installed on the UPS. Provide the data / event logs from the UPS, NMC, APC PCBE / PCSS Software.

Lastly, call out any active faults, errors, LCD LED colours.

[IVI]

Will do. Thank you Teken!