2

SOLVED

Problem: When my UPS does the switch-over from line power loss, the pellet stove runs but you can hear the stove motors slow down, and the wattage meter on the UPS drops from ~150w nominal to ~100w nominal. I say "nominal" because the auger is running every few seconds and adds ~25w to the load.

Setup: I have a pellet stove to heat my house, it draws at most 450w on 120vac. However once it's running good usually consumes ~175w.

I got a 1300w Uninterruptible Power Supply used on eBay, an APC BR1300G. This unit should be able to power over 1000w(rated at 1300w) easily and do it for days given you have enough battery backup. It uses 24vdc for backup power. It came without batteries-- I hooked up two group 31 marine deep cycle car batteries in series to provide this power, and the UPS is charging correctly. The DC wires are crimped tight, only draw about 15A 24vdc at this load, and do not get warm.

Edit to answer some questions: I'm not asking this UPS to turn on the motors from stopped, I'm only asking it to keep them running at a tiny wattage requirement. I get that start-up current is much more than running current.

Line voltage is between 118v and 120vac depending on the day. Output voltage on backup power is 118-119vac, according to the UPS screen. (This, however is incorrect as I just found out. 55v!) Hertz is 60 from line and 60 from backup power. I don't know what you mean by "measured the affects of the motors", nor do I know what high phase angle loads are.

Battery wires are 12ga, and the longest is about 2.5ft long. Which is just fine for less than 20A sustained load @24vdc.

"You are expecting a used device with external batteries meant to drive computers with a pseudo sine wave to drive motors perfectly. That seems unrealistic to me. – StainlessSteelRat" I'm expecting a 120vac appliance to run on 120vac provided by a device designed to power stuff with 120vac. If there's a bunch of flavors of AC and they are not all compatible with each other then this is where I get to learn about it. Believe me, if I could have found this answer by googling it I wouldn't be here. But I couldn't. Now that I know motors are somehow special maybe I could search using different criteria.

A big thank you to Tim Wescott for you answer! I probed the UPS output on line power: 120v. Probed it on backup power: 55v!!! Turns out this was wrong, I measured the voltage wrong and figured it out later when my sine wave UPS arrived. It was putting out around 120v. But the stepped sine wave was totally messing with the motors and were drawing 15A @24v for a 100w load...bad stuff. My sine wave UPS now takes over using only 6-7A @24v. The above reading was taken with a True RMS multimeter, which according to APS is the only way to get an accurate reading of the stepped-sine wave output when running on battery power. Link: https://www.apc.com/us/en/faqs/FA157483/

9
  • 1
    Is the line V 120 or slightly more? What is the UPS output? And have you measured the effect of the motors - some UPS don't like high phase angle loads.
    – Solar Mike
    Nov 21, 2021 at 16:51
  • 1
    Model number of UPS? Suspect the Sine Wave output is too distorted for motor to spin at full speed. Nov 21, 2021 at 16:54
  • 1
    What size battery cables are you using? Other than the issue of motors creating a phase shift because they are a heavy inductance load (power factor less than 1) verify the voltage under load is the same as when powered at the input to the board (true UPS usually run on DC all the time to prevent any bumps) my guess would be the inductive load is the main issue even true sign units have an problems with motor loads. Note I just looked up that discontinued model, it is a 780w unit 1300 va input as motors take ~3 times the starting current the unit may not be able to produce enough true power
    – Ed Beal
    Nov 21, 2021 at 18:10
  • 3
    You are expecting a used device with external batteries meant to drive computers with a pseudo sine wave to drive motors perfectly. That seems unrealistic to me. Nov 21, 2021 at 18:14
  • 2
    Probably because your voltmeter (or the UPS's or both) is inaccurate, due to the AC power not being a sine wave. Lesser voltmeters make assumptions about waveform when measuring voltage. Nov 21, 2021 at 22:18

1 Answer 1

7

I can't answer why your UPS won't do the job, but I can give some reasons why some random UPS may not do the job. Basically, mostly it's because a UPS is designed to run a computer, not a bunch of motors -- but here goes:

  • Motors "expect" a sine wave power input. That thing has a stepped sine wave (I looked at the specs). That's not going to keep the motors from trying to run (motors don't have "brains"), but they may draw excessive current while doing so.
  • As mentioned, motors tend to have low power factors. That's a complicated concept, but basically it means that (A) they don't appear to the UPS as a computer, and (B) they draw excess current for the amount of actual power going into the load. This could cause your UPS to dial down the voltage.
  • Motors usually draw a lot more current when they're starting than when they're running. This isn't a big deal when they're attached to the mains, because everything in that circuit is designed to handle an overload that lasts for half a second or so. A UPS might cut out (more likely, so probably not your problem), or it might go into some limping low-voltage mode (which would account for what you're seeing).
  • Used parts are used for a reason. That could have ended up on eBay because the previous owner had a policy of replacing old with new on a routine basis. It could have ended up on eBay because the previous owner had a policy of milking their UPS's for as long as humanly possible, and only selling them when they were obviously crap (as, for example, persistently putting out a low voltage).
    • Personally I'd measure the output voltage of the UPS with a voltmeter. But unless you're in the habit of sticking voltmeter leads into wall sockets and know how to avoid the dangers, or have a gizmo that lets you do so with 100% safety, I'd suggest you refrain.

You're only asking "why" -- but my recommendation for going forward would be to start by searching the web for UPS's that are known to successfully run motors.

Speaking as a circuit designer, I can tell you that there's a lot of different ways one could design the output stage of a UPS, and some of them would drive AC motors just fine; there may be some combination of the requirements of the market and the whim of the design team that made a UPS that can, indeed, run motors. But unless a UPS was specifically designed and marketed to run motors it won't be in the advertising literature.

So you either need something specifically designed to run motors, or something that the DIY community has found that runs motors, "off label" as it were. Now that you know it's an issue, you can search for it.

8
  • "True sine wave" is the term to look for in the UPS (or inverter) advertising. You also need adequate capacity for starting the motor(s) in question, but that's most of it.
    – Ecnerwal
    Nov 21, 2021 at 19:20
  • You seem to switch from the term UPS to UPC partway through this answer. Is that intentional? If so, you may want to explain it--and if not, you may want to correct it.
    – Hearth
    Nov 21, 2021 at 19:41
  • Gaaah! "UPC" keeps popping out. Fixed, and thank you.
    – TimWescott
    Nov 21, 2021 at 20:18
  • 1
    @ChrisH: "limping low-voltage" yep, you're right. I was thinking board-level power supply design, where the next safety mode after "just let it burn things up" is going into a perpetual start-up mode with a current limit (i.e., limping low voltage). Now that you mention it, I would expect that a decent UPS would monitor the output voltage and shut itself off if it was low, though -- which implies a fault in reading the output voltage back to the controller, rather than a bad driver. But -- who knows what design decisions were made?
    – TimWescott
    Nov 22, 2021 at 15:48
  • 1
    I've even looked at the manual, and it's not clear from that whether the output voltage reported is measured or intended. Even if measured it could be some trick scaling from the positive peak and ignoring the negative, or vice versa. I'd like to see the waveform and would rig something up if it was mine (involving big resistors and a battery oscilloscope) but won't suggest the OP risks that as it's not likely to lead to a repair anyway.
    – Chris H
    Nov 22, 2021 at 15:56

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.