Looking at MME LIN BUS - 12V BMS Intelligent Battery Sensor

[Eric_C_Boston]

If you have a program to read the extended PIDs, like Torque Pro, you can compare the LIN readings to the CAN readings. Toque Pro can also log PIDs.

LVB Age,LvbAge,0x224027,INT16(A:B),0,65535,Days,726,,,1
LVB Current,LvbA,0x22402b,A-127,-127,128,Amps,726,,,1
LVB DCDC Enable,DCEn,0x22483d,BIT(A:0),0,1,,746,,,1
LVB DCDC HV Current,DCHVA,0x22483a,A*0.1,0,25,Amps,746,,,1
LVB DCDC LV Current,DCLVA,0x224836,A,0,255,Amps,746,,,1
LVB State of Charge,LvbSoc,0x224028,A,0,100,%,726,,,1
LVB Voltage,LvbV,0x22402a,A*0.05+6,6,20,Volts,726,,,1

I noticed LVB Current can be zero with a non-zero LVB DCDC LV Current when the DCDC Converter is powering the 12 Volt Bus, but not charging the battery.

Sponsored

 

[Mach-Lee]

Well, I definitely turned off auto-off recently, but just now it was on again. Good idea for a several hour ON test.

It defaults back after every key cycle, you need to change it for every instance you want the car to stay on longer.

 

[louibluey]

Interesting, you should do the same in regular ON mode for several hours to see what's happening.

Wow, I lost track of time, 4 hours (3.992 hr), Louiebluey is definitely nice and dry now! Everything on, headlights, heat, radio. MME seems to run the LVB all the away up to 100% soc, just like an old ICE alternator car. Of course, this is just this one test so far, and only one car, my Dec 2020 FE with the latest module code according to FDRS. The DC/DC converter was on the entire time. 7.3 MB of just ascii data! (0x21 @ 25 Hz, 40 mS reporting). About 32F at start, 24F now, started this morning. Began at Sun Jan 02 11:14:53 AM 2022. MME was ON (motor "on", regular drive mode).

Unfortunately, I only recorded BMS 0x21 LIN bus #4 (I think I might have finally found the LVB current on BMS 0x2D, not confirmed, and no calibration constants yet).

Sample of raw BMS 0x21 data, left column time in seconds. Here, we are using Bytes 3,2 (voltage), and byte 8 (soc) here. (Count from byte 8, all the way in the right most column. Or, byte 1 is just to the right of "Rx d 8".) But, note at byte 5, how FF roles over into 00 corresponding to byte 6 rolling over from 01, to 02, so there is something at bytes 6,5 too where 6 is the MSB. Unclear if byte 5 is the LSB or the LSB is found in part or all of the 16 bits (two 8 bit nibbles) of byte 4. Feels like trying to solve a Rubik's cube.

 

[louibluey]

Two short trips today, first was 8 miles out, some secondary highway driving 65/55 mph, out and back, and then a quick round robin, drive through window, MME not off, 2.4 miles out, 4.7 miles back, so about 7 miles. Voltage is as now expected, DC/DC converter running while driving. soc ranged from about 70% to 87%, the variation seems consistent with the periods of driving and parking.

It was about 16F on departure, probably near the small city is warmer. I think (?) I found the BMS temperature in 0x21 byte 7. Time in decimal hours. Notice that LIN bus #4 went into that sleep schedule towards the end of that first stop in the grocery store parking lot.

I think this temperature note is incorrect, temps did not go lower on very cold days. Temp is somewhere in the data for sure, but I have not found it yet.

 

[louibluey]

I recorded another one hour drive (LVB, BMS, LIN bus, Dec. 2020 build FE, 80% soc hv battery), but what turned out to be more interesting was the 0x21 data from the plugged in (L2, 11 kW) idle time from before the drive, about 28 hours worth. May not matter, but this is with the AC Delco 50 Ah battery. Note how the DC/DC converter came on five times (possibly a couple more) to maintain the LVB (Update, actual LVB charge sessions are minutes to over an hour, most likely the shorter spikes are MME wake ups, without significant LVB charging, probably the kind of wake up where the lights come on momentarily). Top graph is LVB Voltage vs time (decimal hours), bottom, LVB soc vs. time. Note also, that LVB voltage is probably no load, or very light load where the LIN bus woke up to check LVB voltage. That means you would see a lower voltage on a plugged in monitor when MME is awake.

The dips in voltage at each DC/DC converter turn on are likely more modules coming on (and/or lights) which causes a 0.3 to 0.4 V drop. Some or all of the wake ups may be my key fob or iPhone nearby.

In this 28 hour example, MME (plugged in) kept LVB soc between about 70% and 80% soc.

~24 hour temps in unheated garage (blue)

 

[BillPitman]

louibluey,
I'm starting to get a headache.....lol Bill

 

[louibluey]

Per @Mach-Lee 's request, I did a verification run to compare the LIN decoded LVB voltage and soc to what Forscan reports (from the CAN bus). This is a long report, so I'll post here so as not to go too far off thread in the other discussion. I will link back to here.

At about 9:05 pm began an new LIN log
started MME plugged in at about 9:06 pm

at 9:30 (or 9:35 notes unclear) measured 13.5V at battery jump terminals, 13.4 indicated by Forscan (Fluke 376, unverified at 28F garage, 4F outside).

9:38 turned off MME, and back on in Aux mode, 13.4V Forscan

unplugged MME from L2 at 9:57 (trying to get into a situation that would drain LVB)

by 10:32 pm Fluke measured 12.1V at the jump terminals, 11.9V by 10:47.

Headlights were on the entire time (26A), ran max defrost for a bit (40A). MME would not give me any heat most of the time in Aux mode, unplugged. Load shed included seat heat, wheel heat, and cabin heat.

I probably have enough forscan screen prints to graph voltage vs. soc, the soc fell off with little relation to the AGM graph that we usually refer too.

Here are the LIN plots and a couple of samples of Forscan tables. Values compared well, I am confident that my LIN plots are accurately reporting BMS LVB voltage and soc.

One aside, as MME LVB voltage dropped off, the DC/DC converter setpoint increased (a parameter, DC/DC converter was OFF).

The DC/DC converter was on until that last ramp, after that now well known period of about 24 minutes (0.4 hours) in Aux mode. I finally went to AUX mode unplugged as a known way to run the battery down to get some comparison of soc well below 90%.

By the end MME was really upset with me. After I plugged it in, it took a minute before it would come back on.

Note that MME BMS reports LVB soc directly, it is just a straight conversion from one byte hex to decimal.

This time I did remember to turn auto-off OFF, MME still made no attempt to charge LVB a second time in AUX, unplugged (79% soc on HVB), it just gave up and turned off.

I expect Fordpass to be upset by tomorrow morning.

10:23 is the top of that last discharge ramp unplugged in AUX mode

10:47 is near the bottom of that last discharge ramp unplugged in AUX mode

Probably too many Forscan screenshots, for a post, happy send them if anyone wants to see them (.png files, about 250k each). The correlation was good. The time is Laptop time lower right, the Forscan voltage lower right hand corner is not updated and usually wrong in this series.

 

[louibluey]

louibluey,
I'm starting to get a headache.....lol Bill

Hey, all I have been able to find in the LIN data so far, is voltage and soc (there is a lot more information hidden in there), plus I can't get the damn thing to send its files over WIFI like it is supposed to be able to do (hours wasted on that one). Talk about a headache! couldn't agree more!

I'd be pissed that I paid more for WIFI, except that even though I can't figure out file transfer yet, it does seem to check the time the over WIFI, so that alone is a nice feature.

 

[Shayne]

How did you come up with your correction coefficients again? Voltage vs SOC. I would like to see your graph V vs SOC for the one on your bench (the original one).

40% is 12.1, 12.4 and 12.5V depending where you look for an agm? Such a small voltage zone and small shifts have a large impact 12.5v is 60% for the one that is 12.4 @ 40%.

Your graph at Preparing for Drive (charging) looks about right. Think that part is when something gets stuck in the middle of the night. Too much resistance when cold? -30 C this morning and it came on at 2 - 3 am Preparing for Drive. Made sure the battery was up yesterday. Your graph above charges for about an hour.

Have you seen what happens with the 12V when it is running a departure schedule when cold and set to warm? So far waking it up with departure seems to be keeping the 12V healthier with the least interaction? Monitoring but not because I love 12V's . I know if I let it sit for a couple of days in the cold it will error out on me with a possible 12V drain outcome. Don't think the 12V is the cause. It can do it appears it is just bugging out once and a while.

 

[louibluey]

How did you come up with your correction coefficients again? Voltage vs SOC. I would like to see your graph V vs SOC for the one on your bench (the original one).

40% is 12.1, 12.4 and 12.5V depending where you look for an agm? Such a small voltage zone and small shifts have a large impact 12.5v is 60% for the one that is 12.4 @ 40%.

Your graph at Preparing for Drive (charging) looks about right. Think that part is when something gets stuck in the middle of the night. Too much resistance when cold? -30 C this morning and it came on at 2 - 3 am Preparing for Drive. Made sure the battery was up yesterday. Your graph above charges for about an hour.

Have you seen what happens with the 12V when it is running a departure schedule when cold and set to warm? So far waking it up with departure seems to be keeping the 12V healthier with the least interaction? Monitoring but not because I love 12V's . I know if I let it sit for a couple of days in the cold it will error out on me with a possible 12V drain outcome. Don't think the 12V is the cause. It can do it appears it is just bugging out once and a while.

>>How did you come up with your correction coefficients again? The voltage is just the two bytes combined (concatenated) times .001 to place the decimal point. BMS reports with 1 mV resolution. It might not be "accurate" to 1 mV, but they probably wanted to see small short term changes for the software feedback and control loops. The soc is direct, just converting one byte to decimal, no scaling or offset. At times, as several have pointed out, it does not easily map to the soc graph. However, soc does seem to make sense, such as during high current loads over an hour or so, and also when charging back to 90% to 100%.

I am seriously tempted to just buy another BMS negative cable to see if my data changes, but they cost around $100, and my expenditure so far for this MME hobby testing is getting a bit high.

>>I would like to see your graph V vs SOC for the one on your bench (the original one). Not sure which one. I have used @Mach-Lee 's soc graph. We have been struggling with that one a bit, because of the issue of measuring voltage loaded or unloaded, which at least at the low end is typically .2 .3V higher unloaded (e.g. at rest after the frunk light goes out).

ohh, do you mean run that active load test again, but now with the new LIN monitored data? (It is cold in the garage, maybe when we get a warmer day, If I understand correctly, that is a good idea.)

There are times when BMS reports relatively low soc (e.g. 40%) when the voltage appears higher. It could be something else like low temperature being factored in.

I think my BMS is okay because it does seem to report full charging back to 90%, or even 100% during some drives. But, all I know is what it reports, so even that LIN verification run was somewhat dependent on my particular BMS.

I haven't completely understood the other comments, but I do think that the 12V LVB is okay in most of these cases, and as you say, something goes wrong with the monitoring and charing (care and feeding problems).

 

[Shayne]

There are times when BMS reports relatively low soc (e.g. 40%) when the voltage appears higher. It could be something else like low temperature being factored in.

Your SOC just looked a bit low 10%ish.

Hard pressed getting mine from 40% to 90% in an hour this season.

 

[louibluey]

Your SOC just looked a bit low 10%ish.

Hard pressed getting mine from 40% to 90% in an hour this season.

The Ford LVB BMS soc numbers seem to include other factors beyond voltage.

I don't calculate, or convert any of those, they are just literally what the Ford BMS reports. I suspect they are using temperature and possibly some accounting for current.

By contrast, both of my 12V chargers which also display soc, seem to literally have a look up table using a chart similar to what has been posted here.

 

[SnBGC]

Per @Mach-Lee 's request, I did a verification run to compare the LIN decoded LVB voltage and soc to what Forscan reports (from the CAN bus). This is a long report, so I'll post here so as not to go too far off thread in the other discussion. I will link back to here.

At about 9:05 pm began an new LIN log
started MME plugged in at about 9:06 pm

at 9:30 (or 9:35 notes unclear) measured 13.5V at battery jump terminals, 13.4 indicated by Forscan (Fluke 376, unverified at 28F garage, 4F outside).

9:38 turned off MME, and back on in Aux mode, 13.4V Forscan

unplugged MME from L2 at 9:57 (trying to get into a situation that would drain LVB)

by 10:32 pm Fluke measured 12.1V at the jump terminals, 11.9V by 10:47.

Headlights were on the entire time (26A), ran max defrost for a bit (40A). MME would not give me any heat most of the time in Aux mode, unplugged. Load shed included seat heat, wheel heat, and cabin heat.

I probably have enough forscan screen prints to graph voltage vs. soc, the soc fell off with little relation to the AGM graph that we usually refer too.

Here are the LIN plots and a couple of samples of Forscan tables. Values compared well, I am confident that my LIN plots are accurately reporting BMS LVB voltage and soc.

One aside, as MME LVB voltage dropped off, the DC/DC converter setpoint increased (a parameter, DC/DC converter was OFF).

The DC/DC converter was on until that last ramp, after that now well known period of about 24 minutes (0.4 hours) in Aux mode. I finally went to AUX mode unplugged as a known way to run the battery down to get some comparison of soc well below 90%.

By the end MME was really upset with me. After I plugged it in, it took a minute before it would come back on.

Note that MME BMS reports LVB soc directly, it is just a straight conversion from one byte hex to decimal.

This time I did remember to turn auto-off OFF, MME still made no attempt to charge LVB a second time in AUX, unplugged (79% soc on HVB), it just gave up and turned off.

I expect Fordpass to be upset by tomorrow morning.

10:23 is the top of that last discharge ramp unplugged in AUX mode

10:47 is near the bottom of that last discharge ramp unplugged in AUX mode

Probably too many Forscan screenshots, for a post, happy send them if anyone wants to see them (.png files, about 250k each). The correlation was good. The time is Laptop time lower right, the Forscan voltage lower right hand corner is not updated and usually wrong in this series.

I bet load shed also included battery heat. Are you able to see which modules load shed and in what order?

 

[Shayne]

The Ford LVB BMS soc numbers seem to include other factors beyond voltage.

I don't calculate, or convert any of those, they are just literally what the Ford BMS reports. I suspect they are using temperature and possibly some accounting for current.

By contrast, both of my 12V chargers which also display soc, seem to literally have a look up table using a chart similar to what has been posted here.

Can you do me a solid and tell me if it flips up to 15V when a departure schedule kicks in after its deep sleep?

Staying cold here and my tests are over as something inside is wrong here. Maybe I could leave a OBD2 plugged in but I have a warranty issue right now and need no finger pointing. Nothing left plugged in here no additional load. The 12V can get destroyed at any time. Can't have a warm fuzzy about that.

Seen departure kick on at varying intervals of time due to temp now. The graph above which got the 12V up 4% of SOC in 40 minutes was after it carried out a departure (precondition) for over an hour. Got in and started it for an additional 40 minutes (you can see where I drove it). Seen this thing drawing its own power for over 10 hours not charging. Then the next day same -28C it does nothing ?‍.

Would like to know what it is doing to maintain the 12V during a departure as just sitting is not going to work.

 

[louibluey]

Can you do me a solid and tell me if it flips up to 15V when a departure schedule kicks in after its deep sleep?

Staying cold here and my tests are over as something inside is wrong here. Maybe I could leave a OBD2 plugged in but I have a warranty issue right now and need no finger pointing. Nothing left plugged in here no additional load. The 12V can get destroyed at any time. Can't have a warm fuzzy about that.

Seen departure kick on at varying intervals of time due to temp now. The graph above which got the 12V up 4% of SOC in 40 minutes was after it carried out a departure (precondition) for over an hour. Got in and started it for an additional 40 minutes (you can see where I drove it). Seen this thing drawing its own power for over 10 hours not charging. Then the next day same -28C it does nothing ?‍.

Would like to know what it is doing to maintain the 12V during a departure as just sitting is not going to work.

I will see if I can set a short term departure for later this afternoon, otherwise will try for tomorrow morning.

okay, 12:33 pm, I set MME to departure (warm) at 3 pm and the Sync4 display then said that is the next departure. It will be in rest or sleep though, probably not deep sleep.

Both of the LIN monitor and dedicated Sense energy monitors are running. Put the fob in the small steel pot, and turned off BlueTooth on iPhone, hopefully MME stays at rest before the prestart. I should see it on the energy monitor right away, if it works at 3 pm.

okay, here you go, I hope I understood correctly what you asked for:
High Resolution look at a departure time time Precondition Power and LVB

Sponsored