Hi,

I have an RCT Axpert 5kVA inverter connected to a Pylontech US3000 battery. The inverter has a tamper sticker on the side marked May 2015, the battery is 12-18 months old.

The problem I'm having is that the inverter turns off when the battery is at 49.8V which is long before the low battery cutoff voltage in program 29 of 47.5V?

Its happened a couple of times in the last few days, but specifically, this morning when I woke up (about an hour into load-shedding) the inverter was off, with the Pylontech showing 5 out of 6 green lights (ie: mostly full?). Turning the inverter off then on again with its little rocker switch got it to start up and run for a few seconds, during which time I paged through the status screens and saw battery voltage as 49.8V and the little battery icon on the inverter showing 2 bars (out of what looks like 4 total). I fiddled with Program 29 a little to reduce the cutoff voltage progressively down to 46.9V, but each time I turned the inverter on, it turned itself off after a few seconds - without apparently even attempting to energize the AC load (I see 0V AC out on the inverter display panel). I don't understand this behaviour?

As further context, this is a recent problem. A few days back I changed the cutoff voltage to the recommended 47.5V after having inadvertently been running it at 45V since installation. What made me change it is that the Pylontech battery was (I believe) shutting off the inverter due to low SOC. When loadshedding ended I always had to pad out to the garage to push the button on the Pylontech to turn the house back on, which was getting old with the extended Stage 6 schedules. I was surprised I had 45V configured on the Inverter, so changed it to 47.5V so that it would be the inverter deciding when to turn off, but hopefully still be able to turn itself back on when mains came back. Now however, the inverter usually turns itself off too soon. But last night we had an extended power outage during which the backup power ran for nearly 3 hours (which is what I expect for my average load and the battery capacity). So it doesn't *always* turn itself off.

The inverter doesn't drive the fridge or the heatpump or the pool pump or washing machines - those are all mains-connected and go off during load-shedding. So I can't imagine its some inrush current/voltage droop issue from big inductive loads. The slightly random nature does maybe hint there's something like a fridge turning on and off at different times that triggers an overload and prevents recovery, but I can't think of anything like that - its pretty much PCs, TVs and other light electronics plugged in. What I'm gonna try when this happens again is to shut off the AC load output breaker on the inverter to see if it turns on with no load whatsoever and therefore if its something about the load that it gets unhappy about?

The other settings on the inverter are per the recommendation in the Downloads section of this forum:

Program 2 - 30A (although recommended is 25)

Program 4 - SdS

Program 5 - USE

Program 12 - 48V

Program 13 - 51V

Program 26 - 53.2V

Program 27 - 50.5V

Program 29 - 46.9V (current value. I'd like to change this back to 47.5V)

 

Thanks in advance for any assistance!

Edited July 4, 20224 yr by jholler
Typo

Oops. I wrote RCS when I meant RCT! Its the RCT Axpert 5kVA. Model number RCT-AXPERT 5K. Jammer.

Edited July 4, 20224 yr by jholler

I can recall this was a software issue(might be mistaken) I also get the same and needed to lower the setting for the inverter to operate normally.

Quote

The problem I'm having is that the inverter turns off when the battery is at 49.8V which is long before the low battery cutoff voltage in program 29 of 47.5V?

Was the inverter / battery under load?
The load will push the voltage down.
When the load is removed (cut off), the voltage will climb and settle.

Hi,

Thx for the replies.

Yes, the inverter was under load when it shut off. Having thought about it some more a bit, I could also imagine that some of the randomness could be when a family member turns on a heater without realizing we're on battery backup, the Axpert tries to bypass to mains, and turns off.

I think the title for the topic was poorly framed by myself. Its less the inverter turning off that bothers me ('cause it might just be some transient overload) - what I really don't understand is that it won't turn on again when the battery looks decently charged and there is no hectic load as far as I can make tell (this morning there was nothing much happening in the house with everyone still asleep). If I power-cycle the rocker switch it will turn on for about 30 seconds or so - make no attempt to energize the AC load (that I can see - maybe this happens too fast for me to see with the eye) and turns off again.

When this problem happens again I'll try lower the cutoff voltage a bit more per Namreh's suggestion, and I'll also try disconnect all AC load and see if it wants to turn on then (and if so, I guess I'll then enable breaker-by-breaker and see if I can track down some long-forgotten industrial-grade load 😄).

Thx again.

2 hours ago, jholler said:

Yes, the inverter was under load when it shut off.

So the battery voltage when it shut off would have been lower than what you saw when turning it back on.

With this setup, setting 29 should be much lower, around 46.0 V. I know this is too low to save the battery, but the BMS will do that. The high cutoff setting causes problems.

2 hours ago, jholler said:

the Axpert tries to bypass to mains, and turns off.

No, it knows if there is AC-in present or not. So it won't try to bypass to mains if there isn't any.

2 hours ago, jholler said:

what I really don't understand is that it won't turn on again when the battery looks decently charged and there is no hectic load

These are old models, though I suspect that the logic may not be much better when battery type us USE in later models. So the firmware uses "lead-acid thinking", and the voltage spreads are huge. In particular for your situation, the battery is considered "low" if it's not over 2.0 V higher than the low battery cutoff setting (setting 29). So if that's 46.9 V, then the battery voltage has to be 48.9 V for the battery to no longer be considered "low". That could be a considerably decently charged state of charge.

Patched firmware uses a more reasonable (for LFP chemistry) 0.5 V instead of 2.0 V here. I suspect that this is the cause of your surprise.

Consider using the LFP flavour of the patched firmware for your model. If it's 64 V max (check the upper range of setting 26), then use 72.20e; if it's 58.4 V, use 74.40e. These fully patched firmwares will also implement KettleKomp™ and other features that are handy. Update instructions are in the same huge topic; see the index in the first page.

Thanks for the detailed info Coulomb. That makes sense that its using 'lead-acid thinking' 'cause lithium-based batteries were pretty new on the scene when I bought this guy.

My inverter goes up to 58.4V for Prog 26, so I'll go checkout the recommended firmware with interest. (And brace myself for robust conversations with the wife if I turn what's an annoying situation into a permanent outage 😅)

 

If you have the insight/time/inclination to explain, what I still don't understand is this:

* I hear what you're saying about the 2.0V spread, and how my setup might droop below the cutoff level with this spread under load.

* On startup, however: The no-load voltage of my battery this morning was 49.8V, which is 2.9V above the cutoff voltage of 46.9V. I would still expect the inverter to 'try' energize the AC load (and lights to at least momentarily flicker right? maybe even a relay inside the inverter to click or something), before the voltage drooped again under this load and with the 2.0V spread to then dip below my cut-off point and turn off again? Maybe this does happen but in milliseconds and so I don't see it, or maybe deep in the firmware the code works on a 2.0V allowance under running conditions and, like, 3.0V during cold-start to handle inrush current? I find it surprising (and interesting) that the inverter doesn't seem to do anything but turn off and sulk.

I think I will risk the family's patience and fiddle with Prog 29 again (set it back to the 45V cutoff point I had before) and see at what point it turns on again, for science; then do the firmware updates, for quality of life.

Thx again!

 

For anyone else in the future reading this thread, what is now working for me is a cut-off voltage in Program 29 of 46.5V.

In last night's extended load-shedding the inverter ran the house for 3 hours while the under-load voltage gradually dropped. The inverter shut itself off at around 46.5V as expected (I was manually watching it and last saw it at 46.9V before it dies a few minutes later). Power-cycling the inverter with its rocker switch had it turn on again and show the no-load battery voltage as 47.7V, before turning off again. (The inverter not turning on again is now what I expect, since the battery is flat at this point - unlike when I posted this topic wondering why the inverter wouldn't stay on with a reasonable charge level in the battery.)

Most importantly for me, when mains power came back, the inverter was able to turn itself on again. This is the behaviour I was missing before when I had the cut-off voltage at 45V and the batteries themselves cut power to the inverter - forcing me to go manually start things up again at the end of each extended load-shedding event.

So there might be a better magic number than 46.5V, but its working for me, and hopefully will:

* prevent premature shutdown of the inverter during load-shedding

* allow auto-recovery of the inverter after load-shedding

* and go some way (but not all the way) towards not fully depleting the battery each time and shortening its life.

 

Thanks to all for the replies and help!

22 hours ago, jholler said:

On startup, however: The no-load voltage of my battery this morning was 49.8V, which is 2.9V above the cutoff voltage of 46.9V. I would still expect the inverter to 'try' energize the AC load (and lights to at least momentarily flicker right?

Yes, agreed. Sorry, I didn't read my notes thoroughly. The 2.0 V above cut-off is to trigger going into "low battery" state; to exit low battery state, you need a battery voltage of a whopping 4.0 V above the cut-off voltage. But I suspect that "lowness" isn't the criterion used for refusing to start altogether. I should go read the firmware again, but I've run out of time again tonight.

I'm glad you've found a compromise cut-off voltage for now at least.

Edit: What might be happening is that the inverter reads the battery voltage and says "yep, that passes", and turns on the inverter (DC-AC converter) to power whatever loads you have. Starting the inverter requires a significant surge of power. This surge of power causes the battery voltage to fall, possibly low enough that the BMS steps in and disconnects the battery, causing the complete shutoff. But it might also just be the battery voltage falling too low too fast for the internal logic power supply to cope. Perhaps this is exacerbated by some of the relevant capacitors having dried out due to a combination of heat and age.

A battery at low state of charge has a much higher internal resistance, resulting in a much higher voltage drop for a given power surge. By increasing your cut-off voltage setting, you are stopping the battery's discharge at a higher SoC, resulting in lower internal resistance, preventing this shut off. That's the theory at least.

Edited July 5, 20224 yr by Coulomb

5 hours ago, jholler said:

show the no-load battery voltage as 47.7V

So that's 47.7/15 = 3.18 VPC average. That's about as low as you'd want a rested LFP cell to get anyway. So I think you've found good settings, at least for your situation.

Hi, I have acquired a used RCT 5kv (Green cover not Axpert?) Inverter maybe 18months old, connected to 4 x 100AH GEL BATTERIES. Inverter output hardly peaks over 50%,and lasted up to 4 hours. It worked well until I tried to put entire household load through inverter. Assistant may have connected AC in and AC out to same supply, and Inverter shut down.  I reverted to essential supply liad of 25%. Since then, batteries drain out at 30 minutes with this load as when was initially configured? Is it possible that there could be damage to Inverter? Or would it have gone into a default mode?

Regards 

MANOJ

On 2022/07/05 at 9:57 AM, jholler said:

For anyone else in the future reading this thread, what is now working for me is a cut-off voltage in Program 29 of 46.5V.

In last night's extended load-shedding the inverter ran the house for 3 hours while the under-load voltage gradually dropped. The inverter shut itself off at around 46.5V as expected (I was manually watching it and last saw it at 46.9V before it dies a few minutes later). Power-cycling the inverter with its rocker switch had it turn on again and show the no-load battery voltage as 47.7V, before turning off again. (The inverter not turning on again is now what I expect, since the battery is flat at this point - unlike when I posted this topic wondering why the inverter wouldn't stay on with a reasonable charge level in the battery.)

Most importantly for me, when mains power came back, the inverter was able to turn itself on again. This is the behaviour I was missing before when I had the cut-off voltage at 45V and the batteries themselves cut power to the inverter - forcing me to go manually start things up again at the end of each extended load-shedding event.

So there might be a better magic number than 46.5V, but its working for me, and hopefully will:

* prevent premature shutdown of the inverter during load-shedding

* allow auto-recovery of the inverter after load-shedding

* and go some way (but not all the way) towards not fully depleting the battery each time and shortening its life.

 

Thanks to all for the replies and help!

Wish I'd followed this forum a bit more actively! I always assumed that the fact that my Axpert not turning back on by itself when AC returned (i.e. loadshedding ended) after it had shut off because of low battery was just how the unit worked. It absolutely sucked, because if you're on holiday and this happens, power stays off forever and you return to a house with spoiled food in the freezer and the alarm system has been off the whole time as well.

I think you're saying your setting means it's the inverter deciding to shut down instead of the battery's BMS shutting down power to the inverter, so the battery is still "on" in that case. So when it's the inverter's decision to shut down, it is able to switch back on when AC returns. But when the battery has shut down power to the inverter, there's no recovering from that because now the battery itself is off and the inverter needs at least *some* battery to turn itself on again. Still odd that the inverter can't just run on 

OK, I'm going to play around a bit with this. Could make a huge difference.