A Forum run by Enthusiasts of MidNite Solar

Inverters => Midnite DIY Series => Topic started by: ewilhelm on January 13, 2022, 08:26:19 PM

Title: MN3548DIY battery current
Post by: ewilhelm on January 13, 2022, 08:26:19 PM
Hi,

I have a MN3548DIY with 2 kW of solar and 5.1 kWh of LiFePO4 storage. The MN3548DIY reports higher battery current than the battery's BMS reports or I measure with a clamp-on ammeter. Here are some example values:

MN3548DIY batt current (+ input to battery), BMS and ammeter current
19, 12
7, 2
5, 0
-2, -3

For input, it looks like there's about a 5 A offset. Is it possible I'm misunderstanding something?

This is relevant for my setup because I want to set the max charge current at 20 A (as specified by the battery), and I am not sure the MN3548DIY is actually reaching that current.

Thanks,

Eric



Title: Re: MN3548DIY battery current
Post by: boB on January 13, 2022, 10:43:54 PM

We will take a look at this !

Thanks for the report.

boB
Title: Re: MN3548DIY battery current
Post by: ewilhelm on January 15, 2022, 02:08:02 PM
Thanks for your help.

On this topic, the PV current display doesn't match what I measure either. Here's an example:
PV input reported on MN3548DIY display
102 V, 9.5 A -- based on sun conditions and what another, nearby solar system is producing, this seems very high
measured with handheld meter
103 V, 2.3 A -- seems accurate based on conditions
PV power reported on MN3548DIY display
0.500 kW -- note sure how this matches either the voltage and current display nor the measured voltage and current

It probably doesn't help, but I've included a picture (I ran the PV wire to the left of the intended hole in the case so I could land some flexible conduit there to protect the PV wire).
Title: Re: MN3548DIY battery current
Post by: Wizbandit on January 18, 2022, 11:21:19 AM
I have a 3548 and 5048 here I will test.  One thing I know from the past experiences with Charge Controllers, our Classic for sure,  is the noise on the PV wires will mess with the Hall Sensor used in DC Clamp meters and the reading will not be correct.  I will setup a shunt & Wbjr and run the DIY Inverters through it and see what readings I get...

I have contacted the SRNE Engineers to see if we can calibrate the voltage and current readings.
Title: Re: MN3548DIY battery current
Post by: ewilhelm on January 19, 2022, 12:17:49 AM
I typically run my system on "only solar charging" and "inverter priority mode". Today I switched it to "hybrid charging" and "mains priority mode" to fully charge a new battery. I noticed the MN3548DIY showed 20 A of battery input current, which was confirmed by the BMS and my ammeter.

Perhaps the incorrect battery current and PV current are related and only happens in "only solar charging"?

Thanks for your help.
Title: Re: MN3548DIY battery current
Post by: boB on January 19, 2022, 02:14:37 PM

I thought of an idea for this wrong current reporting for both PV and battery current.   Maybe....

For the situations where the currents are measuring wrong, are you sharing a common negative wire for battery AND PV ?

If the inverter/charger measures current on the negative legs only, then it may be necessary to run separate PV and battery negative wires to the DIY3548DIY.

Does the documentation say anything about this ?

boB
Title: Re: MN3548DIY battery current
Post by: ewilhelm on January 19, 2022, 04:52:31 PM
The negative battery and negative PV are only connected to the inverter/charger's terminal blocks. I have not connected them to each other.
Title: Re: MN3548DIY battery current
Post by: boB on January 19, 2022, 08:07:08 PM
Quote from: ewilhelm on January 19, 2022, 04:52:31 PM
The negative battery and negative PV are only connected to the inverter/charger's terminal blocks. I have not connected them to each other.

OK, great !   Just thought I would check.

We are talking to SRNE to get this corrected.   Maybe it is just a software update or an adjustment  ?

boB
Title: Re: MN3548DIY battery current
Post by: ewilhelm on January 19, 2022, 09:00:08 PM
And, here's more, hopefully helpful:

Output source: UTI, Charge source: OSO
PV current 12.2 on MidNite, 8.8 on ammeter
Battery current 12 on MidNite, 9.4 on ammeter (different battery without a display on its BMS, so no third data point)

Output source: UTI, Charge source: SNU
PV same as above (consistent, sunny day)
Battery current 25 (hitting the combined max charge current) on MidNite, 22.2 on ammeter

Seems like the "lost" current is related to the PV as current from the Mains through the inverter didn't change the offset between the MidNite's current and the ammeter's current.
Title: Re: MN3548DIY battery current
Post by: Wizbandit on January 20, 2022, 06:38:46 AM
I wonder if it's doing what you are asking from your settings...
Look at this parameter readout from the DIY3548.  The BLUE are charging parmeters.

PARM[07] is total charge current into the battery, PARM[28] is how much allowed from the GRID (AC IN) and PARM[36] is how much to allow from the PV IN.

The difference you are seeing may be because your battery charge current from PV might be minus what is being used to power the inverter and/or loads.  I'm reorganizing my "test wall" so I don't have a DIY Inverter powered up working just yet but I will test as soon as I get everything setup.  We are used to seeing separate readings from the charge controllers and inverters but remember this unit is integrated and shares the same control computer.
Title: Re: MN3548DIY battery current
Post by: ewilhelm on January 20, 2022, 01:11:05 PM
Thanks for your continued help. Here are my settings:

Parameter: setting
7: 25 A
28: 20 A
36: 25 A

Under Output source: UTI, Charge source: OSO, the mains are powering everything on the AC output, and only solar is charging the battery. In this configuration, I still see different battery and PV currents between the MN3548DIY's display on what I measure with an ammeter or the battery's BMS reports. In UTI and OSO, parameter 7 and 36 are effectively the same because no charging is permitted from the grid, and the inverter AC output is off since the loads are bypassed.

As noted in an earlier note (perhaps not clearly), when the battery is charging only from the mains without solar charging, the display, BMS, and ammeter current all match. There's something wrong with the PV input and/or battery from PV current.
Title: Re: MN3548DIY battery current
Post by: ewilhelm on January 26, 2022, 07:30:28 PM
Right now I see:

MidNite
PV 500 W
Battery current 5 A
Battery input 250 W
Inverter output 250 W - confirmed by measuring power on AC line

Battery BMS 0 A
Ammeter on battery 0 A (battery is at 100% SOC)

This leads to the curious situation where the MN3548DIY thinks it is charging the battery, so is holding voltage at 56 V, the charge voltage. As soon as a cloud goes in front of the solar array or the AC load changes, the MN3548DIY's battery current will fall to zero and it will turn off the solar until the battery reaches the fully charged recovery point as described in my recent post:
http://midniteftp.com/forum/index.php?topic=5769.0

Title: Re: MN3548DIY battery current
Post by: boB on January 27, 2022, 01:10:57 AM
Quote from: ewilhelm on January 26, 2022, 07:30:28 PM
Right now I see:

MidNite
PV 500 W
Battery current 5 A
Battery input 250 W
Inverter output 250 W - confirmed by measuring power on AC line

Battery BMS 0 A
Ammeter on battery 0 A (battery is at 100% SOC)

This leads to the curious situation where the MN3548DIY thinks it is charging the battery, so is holding voltage at 56 V, the charge voltage. As soon as a cloud goes in front of the solar array or the AC load changes, the MN3548DIY's battery current will fall to zero and it will turn off the solar until the battery reaches the fully charged recovery point as described in my recent post:
http://midniteftp.com/forum/index.php?topic=5769.0

Battery current will change with battery voltage some for a particular output wattage...  In this case, 250 watts.

What is the battery voltage in this case ?   Where were you seeing 5 amps ?  Your meter of the DIY's ?

boB

Title: Re: MN3548DIY battery current
Post by: ewilhelm on January 27, 2022, 10:01:54 AM
Battery voltage reported by the MN3548DIY: 56 V
Voltage at terminals of battery: 56 V
Voltage reported by the battery's BMS: 55 V and slowing dropping as the pack relaxed after being fully charged.

Battery input amperage reported by the MN3548DIY: 5 A
Amperage at battery terminals (ammeter): 0 A
Amperage reported by the battery's BMS: 0 A

I assume the BMS is reporting the voltage behind a switch or FET. The difference in voltage doesn't bother me; the difference in amperage does (or at least I don't understand).

Overall, I feel like the PV amperage is incorrect, which leads to incorrect PV power reports and battery current reports on the MN3548DIY. At the time of this measurement, 500 W from the solar array seems high by 50-100% as estimated by the power output of another neighboring solar array.
Title: Re: MN3548DIY battery current
Post by: ewilhelm on January 28, 2022, 06:15:22 PM
More, this time something working as expected, hopefully elucidating the issue:

PV: 0 W (sun is down)
Battery on MidNite: 4 A from battery
Battery BMS: 4 A
Ammeter: 4 A

The erroneous current measurements appear to be related to the PV current. From my observations, it seems like the first 200 - 250 W of PV reported by the MN3548DIY is "phantom". At 200 W of PV and 200 W of inverter output as reported by the MN3548DIY (and 200 W of output confirmed by by a clamp-on meter on the AC output), I see 200 W of power drawn from the batteries. At 400 W of PV and 200 W of inverter output, reported by the MN3548DIY, I see 0 W in or out of the batteries.

I understand the MN3548DIY consumes about 40 W. This matches with my measurements of AC output energy and battery consumption. I think the remaining 160 - 210 W of phantom power isn't real lost power but a measurement error of some sort since I don't think my solar array is producing this lost power. 
Title: Re: MN3548DIY battery current
Post by: boB on January 29, 2022, 12:44:16 AM

Well, it is good that there might be at least a little bit of sanity in these measurements !   :)

You are absolutely correct in that understanding what is going on is important.  Maybe I don't understand what I think I understand about how the PV and battery side are connected together ?

Maybe the 5A that is showing is actually coming from the PV and into the inverter input to power the inverter ?

40 watts of idle would only be around 1.5 amps or so though.

Must ponder some more.  I'm not up there right now so I cannot measure. 

Someone needs to verify these numbers etc.


Title: Re: MN3548DIY battery current
Post by: ewilhelm on February 24, 2022, 03:43:31 PM
Here's more: 100 - 200 W flowing out of the MN3548DIY as measured by an AC power meter. Same power flowing out of the batteries as reported by their BMS. Sun is just coming up, so solar is active but weak. MN3548DIY's LEDs are flashing reporting both that the inverter is operating (it is) and that the battery is charging (definitely not charging).



Title: Re: MN3548DIY battery current
Post by: boB on February 24, 2022, 10:51:18 PM

What was the battery voltage at that time ?

Title: Re: MN3548DIY battery current
Post by: ewilhelm on February 24, 2022, 11:20:07 PM
Around 52 V.
Title: Re: MN3548DIY battery current
Post by: boB on February 24, 2022, 11:50:27 PM

I wonder if the DIY had finished a charge and their message is just wrong ?

Title: Re: MN3548DIY battery current
Post by: ewilhelm on February 25, 2022, 10:05:19 AM
I think it's related to the overall issue of incorrect measurement of current on the PV, as described from different angles in this thread. When the sun is down, the numbers make sense: AC charger, battery in/out, and inverter. When the PV is doing anything, there appears to be a phantom 200 W that typically shows up as higher battery input current than seen on the BMS or an ammeter. When the real PV power is below 200 W, this results in weird scenarios where the MN3548DIY reports battery input current while the batteries are actually outputting current. When the real PV power is above 200 W, it looks like a measurement error or offset, but now the signs of current flow are at least correct.

Regardless of this problem, my overall system does appear to be operating within its design range. PV, batteries, and loads are all about matched given the time of year.

If you're not getting similar reports from others, perhaps I simply have a bad unit?
Title: Re: MN3548DIY battery current
Post by: boB on February 25, 2022, 04:52:54 PM

I think that I have heard of another instance of incorrect current measurement reporting.
Unless maybe you have reported other than just here ?

We are talking with SRNE about this.  I need to try and  round up all other reports of this.

Thank you !

boB
Title: Re: MN3548DIY battery current
Post by: ewilhelm on February 25, 2022, 08:51:35 PM
I've only reported here. Thank you for your help!
Title: Re: MN3548DIY battery current
Post by: boB on February 25, 2022, 11:06:49 PM
Quote from: ewilhelm on February 25, 2022, 08:51:35 PM
I've only reported here. Thank you for your help!

You bet !

Title: Re: MN3548DIY battery current
Post by: FNG on March 16, 2022, 09:07:53 AM
Quote from: ewilhelm on February 25, 2022, 08:51:35 PM
I've only reported here. Thank you for your help!

Can you email me ryan@midnitesolar.com I would like to replace this unit and see if the issue stays or goes