Hi @bluetti_jebrroprotonmailcom, When the device is turned on, even in standby mode, it will still consume a certain amount of self-consumption power. Both the AC300 and B300 require power to stay operational.
When the AC output of the AC300 is turned on, the self-consumption is 43.5W per hour. If not in use, we recommend turning off the device. If it’s inconvenient to power it off, turning off the AC output is also a good option.
A loss of 1500Wh within 12 hours is unusual. Could you please provide the SN number and BMS version of your AC300? We want to confirm if you have upgraded to the latest firmware. How many B300 batteries are connected to your AC300, and does recalibration help resolve the issue?
I have attached two images that hopefully show the information you are requesting. There are two (2) B300 batteries connected to my AC300. I will also perform the recalibration very soon and let you know if the excessive amount of power being lost decreases.
Hi @bluetti_jebrroprotonmailcom, We found that there is a new BMS v1021.13 for your PACK 3.
I have now applied to upgrade your firmware.
Our IT department will push the new BMS to your unit in 1 workday.
Please upgrade the firmware tomorrow. Please do not load any device when upgrading it, meanwhile, please only connect PACK 3 with the AC300 when upgrading.
We suggest you recalibrate it after that.
Please let us know if it does help.
Pack2 shows a current version of v1014.16 and pack3 shows a current version of v1021.14. I upgraded this morning (8am EST).
Why do the batteries have different firmware versions?
Why can’t I upgrade pack2? It says “no new firmware version” when I check for update. It appears to be an older version and yet I can’t update??
I have not yet recalibrated but will do so later today or tomorrow.
Hi @bluetti_jebrroprotonmailcom, Yes, this new firmware is specifically designed for Pack3. The Pack2 version is different and cannot be upgraded with this firmware.
The BMS version start with 1014 and 1021 involve different hardware batches, so the latest firmware versions are also different, but there is no functional difference between them.
The same applies to ARM or other firmware. These are separate firmware versions renamed based on hardware and batch differences, but each version has the latest firmware.
Please test the new firmware’s performance after completing the calibration.
This is the data charted in Excel. Nothing was being powered other than AC is on.
I did reconnect the batteries during this period because the display dropped down to 50% from near 100%. When the connections were moved to different ports the reported value changed on the main screen to a higher value. There seems to still be excessive power lost when nothing is being powered and AC is on. Also there seems to be some problem with the ports as changing where things are plugged in should not changed the home screen SOC value. This is not a good condition. Below are the pictures of the screens.
Could you please see my late reply to this thread. I have an update on the status of my issue with SOC.
Hi @bluetti_jebrroprotonmailcom, Apologies for the delay in responding, as I wasn’t tagged in the earlier comments and missed them.
We’ve checked your firmware, and it is all up-to-date, so further upgrades are not possible.
We’ve reviewed your data table, but the timing isn’t entirely clear. For example, in the first row, “22hr 14min,” is that the recorded time or the runtime with AC on?
If you test the capacity of the AC300 + B300 by turning on the AC output without connecting any load, the results will not be accurate.
Additionally, please note that ambient temperature can also affect the test results.
AC300+2 B300 combination is 6144Wh. To protect the battery and avoid over-discharge, the battery is not drained 100%. The discharge formula is: capacity * DOD * n ÷ (load + self-consumption). DOD is calculated at 90%, and n is 90%.
The self-consumption when the AC is on is 43.5W. Your load power is too low, so the effect of self-consumption is quite noticeable. We recommend using a load of 1000W-2000W to test.
What is the ambient temperature where the machine is placed? Extremely high or low temperatures can have an impact.
We also observed that the SOC of the two B300 units is becoming more close, indicating that the BMS system is functioning as expected.
Thank you for your reply.
- I recorded the date and time of the values display. That is the 22hr 14min time you show. You can see this from the pictures.
- I don’t understand why the results would not be accurate if I do not load the AC circuit? (BTW I have a garage door opener plugged in 24/7 so it is technically loaded to a small degree.) Why would the result only be accurate if I load the AC circuit? How much is it supposed to be “loaded” to display a correct result? This is very confusing. It is like saying I have to find things to power all the time otherwise my equipment won’t report the correct values? Please clarify what you mean here?
- Ambient temperature in the electrical closet has been ranging from mid 30’s to high 40’s.
- Again very confused by your comment that my AC load power is too low to get valid results. I consider it a valid test if I want to power a small workshop and keep my garage door opener plugged in on a continous basis. This is real-world usage. BTW I have PV input during all this time. It is rather low due to the time of year and the overcast skies. So if anything the power lost is showing less than actual because I turned the PV input on and it still loses a considerable amount of power when AC input is ON and there is little to no load on the AC circuit. Hope that made sense.
thank you
Also want to point out again that the SOC percent on main screen changed when I connected the power cords into different ports. This seems like a problem. Is this normal? Pretty sure the values in GREEN were after I connected differently.
Another reply posted below.
Thank you for your answer
Hi @bluetti_jebrroprotonmailcom, Yes, you will need to load some devices, and it’s best if the total power is above 1000W, as this will make it easier to test the battery capacity.
Since the AC300+B300 is a large-scale energy station, its own power consumption is quite high. Without a load, the self-consumption will significantly impact the test results.
With a high-power load, the influence of self-consumption becomes very small and can be ignored. Additionally, this will greatly reduce your testing time.
By the way, please note that if you need my assistance, please @BLUETTI_CARE, otherwise, I might not see your message and be able to help you promptly.
That is the reason why we can’t reply to you in time.
You see the issue as the battery capacity may be reduced and NOT as the AC300 system is drawing to much power when AC is ON and no load is present, correct? How would you rule out the latter case?
I will attempt to do what you ask here to help resolve this but I would point out that this requires a good deal of effort to troubleshoot the battery capacity. Is there not a simple way for customers to determine how much charge the batteries will hold?
For example, my original problem was simply that with no PV input, no grid power, AC on and zero loading I would see a big drop in the SOC reported on main screen in just a 12 hour period roughly. If you assume this is all related to battery capacity it seems possible to calculate that capacity from this test. It would seem to be possible to report the battery capacity without the customer having to test their system in the way you suggest (which I would appreciate and link to a clear set of instructions on this method to avoid mistakes).
Do you know if all AC300 units have the same of amount of idle state power consumption when AC is on and no load is present?
It may take me until the weekend to run a battery capacity test. I would like Bluetti to keep in mind that this cost customers $$$ to deal with these issues.
Is the power monitoring feature able to give us these numbers? It looks like you have to connect to the internet in order to use this feature which is unfortunate.
thanks for your help on this.
Also could you comment on the table I presented and the fact that the main panel SOC reading didn’t show the correct value until I changed the power cable connections. I.e. I switched them around to see if it would make a difference and it did.
Hi @bluetti_jebrroprotonmailcom, In our previous response, we provided the most suitable method for testing capacity. It would also be very helpful if the customer uses additional plugs or power testing accessories.
When the AC300 is without load and the AC output is turned on(DC output turned off), the self-consumption is 43.5W per hour, and the self-consumption of each B300 battery is 2W per hour.
We reviewed the screenshot of the table you provided and noticed that the total energy displayed in the yellow section is identical to that of Pack 3. We suspect that Pack 2 is not correctly connected or locked, which is why it has no impact on the total energy. Please try reconnecting and locking it again.
Additionally, please make sure to use the left-side method to connect the two B300 batteries in a symmetrical manner.
If the result remains the same after correctly connecting the batteries symmetrically, please swap Pack 2 and Pack 3. If the result changes, it indicates a fault with one of the AC300’s parallel ports, and repair will be necessary.
Update
-AC and DC output has been off and remains off.
-ECO mode is OFF and has always been OFF.
-Connections to batteries has not been touched. Recall previously that I was able to get home screen SOC to report correctly by switching the power cables. I am 100% sure nothing has been disturbed.
-System SOC was reporting 17% at 8:17am on 1/9/2025.
-Low SOC was result of numerous overcast days which greatly reduce PV charging.
-Sunny day all day 1/9/2025.
-SOC reported 49% at 6:15pm on 1/9/2025.
-1/10/2025 AC300 is unexpectedly found to be completely off. ECO mode is OFF and was never turned on.
-System was powered on.
-SOC reports as 47% at 6:50am on 1/10/2025 on home screen.
-BP2 is 95% charged. BP3 is 100% charged.
I would like to know your thoughts on this situation?
I have a second AC300 as I have been preparing to setup a split phase system (I already purchase the components) for my shop power and add two more batteries. Due to the various issues I am having with the equipment I am in doubt as to whether to continue with Bluetti products to do this. If it turns out that my battery capacity is low (which should not be the case because the system isn’t that old and it has very light usage) and the system turns off unexpectedly or reports SOC correctly and then out of the blue reverts back to an incorrect value then this would pose a lot of problems to deal with.
I will inspect the connections very carefully including looking at the cables to see if there is any indication that a loose connection exists. I don’t expect to find this as I was careful to fully install and lock the cables. Again the system was completely untouched. I will then swap out the existing AC300 with my slightly newer, unused AC300 to see how it affects these issues. I am not going to update the firmware at this time unless you know of a critical reason I should do so. I still need to check the cabling arrangement for the symmetrical configuration you mentioned. thank you.
Related question. If the cable isn’t locked then the BP SOC for that battery would show as unconnected correct? You indicated earlier you thought I hadn’t locked one of the battery cables.
Hi @bluetti_jebrroprotonmailcom, We believe the issue might be caused by the cable not being securely locked. You can try using a different cable to test, which will help rule out any faults with the cable itself. If it’s confirmed to be a cable issue, we can arrange warranty repair services for you.
As for the AC300 shutting down automatically while still powered, we guess this could be related to the ambient temperature. Can you confirm there were no loads connected? If there were no loads, this could also lead to an automatic shutdown. Please check the Alarm History for any past error codes.
Yes it is most likely the power cable that is responsible for the SOC misreporting on main screen. I will swap the two cables I have and see if it follows the cable. This morning, 1/13/2025 around 8:30am EST I observed the main screen SOC showing 49%. Thinking this is the suspected cable issue I carefully unlocked and relocked the switch on the cable several times without moving the cable itself. This resulted in the SOC on main screen again reading basically 99-100%. So I can confirm this is responsible for the SOC reporting issue on my system.
It seems odd that switch can lock and the firmware recognize it and then if the switch somehow “unlocks” electrically overnight, but does not move mechanically, the firmware doesn’t catch the changed state of the switch. Should it not still read 100% because it only recognized one (1) battery now being connected instead of two? I think you have a firmware issue here to go along with the cable problem. I don’t think it is reacting properly to the switch being mechanically in the lock position but electrically in the unlocked position.
Is there a chance this problem could be with the battery receptacle where the cable plugs into?
I will know more when I swap the cables. After that I will try and look at the battery capacity. Is there not an easier way to report battery capacity then for me to discharge my batteries completely under high load (>1000W)?
thank you
