Here's an illustration. I don't know if this is usual/acceptable or not. But I believe that numbers don't lie, so this is what it is.

With a return to the office and thus an early morning commute imminent, I want to run my heat pump earlier in the morning.

I used to do this. It was the situation when my system was installed in mid 2019. Here's what happened 23rd July 2019

SOC at midnight was 82%. Pump turns on at 4:30 with SOC 75%. The thermostat turns the pump off at 6:10 with SOC now 41%.

 

Here's what the portal shows me for 25th Jan 2020

70% SOC at midnight. Heat pump kicks in at 4:30. SOC when the heat pump turns on is 58%. At 5:15 the thermostat kicks in, heatpump turns off and SOC is 46% at that point.

Whilst I've been working from home I have set the heat pump to turn on later as I take exercise in the morning and don't have to leave early to get the bus. Now with a return to the office looming I want to start running it early again. So this is what happened this morning.

SOC at midnight was 76%. Heat pump turns on 5:30 with SOC at 60%. 25 minutes later SOC has hit 40% (my maximum on-grid DOD is 60%) and the system starts drawing from grid.

Load in 2020 and 2022 varies around the 1.8 kw mark.

So these batteries seem to have lost quite a bit of woema in 2 years.
 

Edited January 25, 20223 yr by Bobster
Correct dates and add in loading

Your last discharge curve looks strange - the slope changes dramatically. I think you have one or more dud cells, or your cells aren't balancing.

Edited January 25, 20223 yr by P1000

10 minutes ago, P1000 said:

Your last discharge curve looks strange - the slope changes dramatically. I think you have one or more dud cells, or your cells aren't balancing.

A recurring event in my solar life. Remembering that the green trace is SOC, look what happened later today (this is not uncommon)

 

Looks like the same issue that the Hubbles are having...

I called Revov. They told me how to check if the batteries are 100% and if there is cell balancing going on. However this involves watching LEDs on the the BMS - who has the time? But I am grateful for the information. 

4 minutes ago, Bobster said:

I called Revov. They told me how to check if the batteries are 100% and if there is cell balancing going on. However this involves watching LEDs on the the BMS - who has the time? But I am grateful for the information. 

Use a cell phone video/camera app to time-lapse the BMS-LED.

3 hours ago, Bobster said:

I called Revov. They told me how to check if the batteries are 100% and if there is cell balancing going on. However this involves watching LEDs on the the BMS - who has the time? But I am grateful for the information. 

get the software the techs used, and log it all for a week or so (even so don't turn it off, keep at it), then send that to them. leds aren't going to help, they're an indicator at best.

if really needed allow them to remote log in to install said software.

also greetings  

17 hours ago, system32 said:

Use a cell phone video/camera app to time-lapse the BMS-LED.

It's a good idea, thanks, but I don't think it  will do much good. There are 4 LEDs, so 25%, 50% and so on. When it's DISCHARGING and you see 3 LEDs you know it's somewhere north of 50%. When it's charging, you will get one flashing light, that's the "segment" that's charging now. So two LEDs on and a third flashing tells you that it's past 50%, not yet at 75%. So the chances of catching it at exactly the right moment would be slim.

But back to my original question - 2 years ago my batteries could power that heat pump in the morning. Now the batteries last about 20 minutes doing the same job. I know that the batteries will degrade over time, but is this exceptional degradation or not? If not then I just have to learn to live with this increasing decline in performance.

BTW, all the firmware involved is not much help here. Ever since I've been paying attention (at least a year), the "state of health" for the batteries has been reported as 103%. Huh?
 

Edited January 26, 20223 yr by Bobster
grammar

That degradation is not acceptable, and unless the problem is fixed, it will go down hill very sharply from here (assuming my previous assumptions are correct).

Update: Revov had paid two house visits now. The first was just perfunctory: Plug a PC into the BMS and query it's internal logs. The guy who did this said the batteries were fine. 

The second visit was preceded by a phone conversation (see the comments above about them telling me what the LEDs on the BMS indicated) and a request for inverter settings. Then a visit. This time they said batteries are fine, settings are not and eventually there will be damage done to the batteries. The technician also suggested that the significantly higher charge rate and an incorrect float voltage were causing the BMS some confusion (I think he mentioned something about a balancing phase), hence the sharp drops and climbs in SOC that the portal shows. I said, well, it's getting worse, he said that the error is cumulative and getting worse all the time, but the batteries are not damaged YET.

OK... so we need to adjust the inverter settings. On the Goodwe this means using the "self define" battery settings which are tweakable. The inverter was set to use BYD B-BOX and you can see the settings but not change them (I presume they preset it according to data the battery manufacturers provide). 

So, with guidance from the installer, I put it into self-define mode, set everything as Revov would like, and promptly the inverter gives a warning that comms to the BMS are lost and starts reporting an SOC of zero. I switched back to the BYD settings.

I have no problem with Revov's response. They have been concerned and did check the batteries out. On the second visit they showed to me that the cells were all nicely balanced and talked me through some other data. But they cautioned me that the settings are incorrect, and the inverter is not theirs, and they can't be familiar with ever inverter on the market and how to configure it. So all they can do is provide settings and say "please set your system like this".

Trouble is that those settings are not easily implemented using the Goodwe.

I think that this underscores the importance of making sure, before the purchase of either, that there is good lithium battery and inverter integration, and comms. Also ensuring that there is an ongoing commitment from both battery and inverter manufacturers to support the duo with future firmware releases, because an update to either may affect something. This is particularly important when deciding on a battery or inverter that is a little less popular.

 

20 hours ago, Bobster said:

Update: Revov had paid two house visits now. The first was just perfunctory: Plug a PC into the BMS and query it's internal logs. The guy who did this said the batteries were fine. 

The second visit was preceded by a phone conversation (see the comments above about them telling me what the LEDs on the BMS indicated) and a request for inverter settings. Then a visit. This time they said batteries are fine, settings are not and eventually there will be damage done to the batteries. The technician also suggested that the significantly higher charge rate and an incorrect float voltage were causing the BMS some confusion (I think he mentioned something about a balancing phase), hence the sharp drops and climbs in SOC that the portal shows. I said, well, it's getting worse, he said that the error is cumulative and getting worse all the time, but the batteries are not damaged YET.

OK... so we need to adjust the inverter settings. On the Goodwe this means using the "self define" battery settings which are tweakable. The inverter was set to use BYD B-BOX and you can see the settings but not change them (I presume they preset it according to data the battery manufacturers provide). 

So, with guidance from the installer, I put it into self-define mode, set everything as Revov would like, and promptly the inverter gives a warning that comms to the BMS are lost and starts reporting an SOC of zero. I switched back to the BYD settings.

I have no problem with Revov's response. They have been concerned and did check the batteries out. On the second visit they showed to me that the cells were all nicely balanced and talked me through some other data. But they cautioned me that the settings are incorrect, and the inverter is not theirs, and they can't be familiar with ever inverter on the market and how to configure it. So all they can do is provide settings and say "please set your system like this".

Trouble is that those settings are not easily implemented using the Goodwe.

Yeah, so if you selected BYD and Goodwe reported comms with BMS were good, those settings were what the BMS requested. It sounds like Revov support has some room for improvement...

What is disturbing is the change in slope in SOC, could you not try drawing a known load say 1kwh and monitor if the batteries discharge by a similar amount?  The Goodwe software developed by CES records a reading every minute, in the data output there is a SOC, Vbat(V), IBat(A) field, (works for pylontech batteries) it will assist with quantifying power use, it may also show if the Goodwe actually communicates with you batteries. However if the power does match I dont know what the alternatives are. Also suggest you speak to the Goodwe agent he's quite helpful. 

What is strange is that it worked well in 2020 and is now playing up. Good luck with solving the problem