MN3548DIY battery charge settings

Started by ewilhelm, January 26, 2022, 06:49:13 PM

Previous topic - Next topic

ewilhelm

Hi,

I have a my MN3548DIY running with LiFePO4 batteries. My settings are:
01 SBU
06 OSO
08 L16
09 56 V
10 120 min
11 54 V
37 52.8 V

Once the battery reaches 56 V and 100% SOC, and its BMS reduces the input current to zero, the MN3548DIY turns off the solar (PV icon stays on, dashed line from the PV icon turns off), and doesn't turn it back on until the battery depletes to 52.8 V. This doesn't match my expectations in two ways:

1) It ignores the float voltage setting and lets the battery discharge right through this parameter. There appears to be some debate over float charge of LiFePO4 batteries. Maybe the MN3548DIY has chosen not to float charge this type of battery? If so, I'm confused why this parameter is still available. For example, the equalization voltage parameter is not accessible with battery type set to L16.

2) It turns off the solar even if there is an output load on the inverter. I would prefer the solar to remain active and powering this load while putting no current into the battery rather than discharging the battery to the fully charged recovery point before turning back on.

Are there different settings of the parameters to achieve what I want in 2)? Or is there another way I should be thinking about the batteries -- perhaps setting an even lower charge voltage so they don't fully reach 100% SOC?

Thanks,

Eric

boB


Kind of strange...

There IS a setting I hear of about charging priority in their menu.

Have you seen that one ?   I think It can work with either or both.  Maybe the priority is grid ?

I'm not exactly sure.   If the float voltage is set to lower than the battery is sitting at, then I would expect nothing to charge until the battery voltage goes below that float set point...   But that is not where things are set in your case, right ?

No EQ for an L16 flooded lead acid battery ?  That is kind of weird.
K7IQ 🌛  He/She/Me

FNG

You say the BMS backs down the charge, Usually the BMS is a switch that opens if the battery is being over charged. If the BMS is opening the DIY wouldnt have anything to feed if that makes sense. Can you tell us the make and model of the battery you have?

ewilhelm

I have a 48 V 100 AH LiFePower4 Battery by EG4 and a JK48V100 100 AH LiFePO4 Jakiper Battery in parallel.

FNG

Ok so from what I am reading the BMS is just a safety and the DIY inverter should be regulating. Are you saying the battery actually disconnects or is it actually the DIY backing down?

FNG

I think Parameter 37 may be getting you, Try turning it up to 54v

ewilhelm

52.8 V is the maximum I can set parameter 37. (I should have put that in my original message.) If 37 interacts with other variables that might allow it to go higher, I haven't found them yet.


boB


There have been some issues with these batteries not powering up inverters   evidently because of their BMS disconnecting from over-current surge from either charging the inverter's capacitors and/or the inverter charging up.

You probably don't have this issue because it does charge in Bulk and Absorb cycles.

Just in case, here is a thread on DIY Forums that you may already know about...

https://diysolarforum.com/threads/schneider-xw-pro-6848-not-powering-up-with-eg4-lithium-battery.33250/page-20#post-429224

There are many pages to this thread but it doesn't necessarily have anything to do with this issue.

K7IQ 🌛  He/She/Me

ewilhelm

Thanks, I've been following that thread, too!

The LiFePower4 seems to do a cold start with the MN3548DIY without problems. I was getting an alarm condition on that battery (and not on the Jakiper) when I was charging to 58 V (as specified in the spec sheets; manual suggests a different voltage). Once the battery reached 100% SOC and the battery current went to zero, but before the MN3548DIY thought the battery was fully charged and/or thought it was still sending current into the battery, I noticed the battery voltage on the MN3548DIY and on the battery terminals would wildly fluctuate, +/- 2 V. In this configuration, I assume the MN3548DIY is trying to keep the voltage constant on its input capacitors, but this is difficult because they are small compared to the battery, which is rejecting current. I think this voltage was capable of tripping the over voltage alarm on the LiFePower4. I reduced the charge voltage to 56 V, and haven't seen this condition repeat.

FNG

I would assume that you are seeing the BMS open and close due to high dc voltage.

ewilhelm

When I found the LiFePower4 in an alarm condition, I always fussed with it: press reset button, switch off breaker, etc... So, I never saw the alarm condition resolve on its own, as the battery's manual indicates it will. In any case, I'm optimistic this won't repeat.

Let me know if my reports need more details, videos, or something else. I'd like you guys to be able to duplicate the problems and suggest solutions.

boB


And let us know if you find more weirdness !

:D
K7IQ 🌛  He/She/Me

ewilhelm

The maximum of 52.8 V of parameter 37, battery fully charged recovery point, is now causing me problems.

The voltage drop of my 10 kWh of LiFePO4 batteries is shallow enough that 52.8 V represents a significant amount of energy. Over the past three days it has been sunny, and my system reached 100% SOC near the end of day 1 (56 V absorption). Over night and during the next full day, the batteries didn't discharge to 52.8 V until the end of day 2, and the MN3548DIY rejected all the solar and didn't charge the batteries or operate the inverter from solar (only from batteries). Once the batteries discharged to the recovery point and started charging again, the sun was going down. On day 3, it charged as normal, but didn't reach full SOC. You can see the battery voltage in the image.

I've sized my system to have about three days of autonomy (and I have grid backup). I'm disappointed that the parameter limitations create a situation where I burn through one day of autonomy even while sunny. I'd like to see the battery fully charged recovery point go at least to 53.2 V.

Thanks,

Eric

Wizbandit

PARM#37 is tied to PARM#11 (Float)
Move Float to 55.2 and you can take PARM#37 up to 54

Don't ask me WHY, I don't speak Chinese...

ewilhelm

It worked! Recovery voltage is not set to 53.2 V.