Honda Insight Forum banner

LiBCM Open Beta Support Thread

69830 Views 1492 Replies 60 Participants Last post by  ghaze24
This thread is for any and all questions related to LiBCM during the Open Beta.
After browsing www.linsight.org, if you still have questions, please post them here.

I will answer questions posted in this thread by updating the FAQ/Instructions/Documenation/etc at www.linsight.org.
I will probably respond to questions posted in this thread by replying with a hyperlink to the answer (which I may have just written).
  • Like
Reactions: retepsnikrep
1301 - 1320 of 1493 Posts

· Registered
Joined
·
11 Posts
Automotive tire Motor vehicle Vehicle Tread Automotive exterior


Hi and I hope that this is the appropriate place to post this?

During the cold UK Xmas holidays, it seems that we left our LiBCM gen1 parked up for a little too long without turning the IMA off!

The firmware had expired so I have re-installed that and have attempted to grid charge the lithium battery pack but the IMA battery is not being recognised or recalibrated.

I am also getting the MCM error code, P1648 I think.

Is there anything that I can do to recharge / save the lithium battery modules because the grid charger doesn't seem to be doing anything at the moment?

Would it help to warm the car up whilst the grid charger is connected?
 

· Linsight Designer
Joined
·
4,812 Posts
Discussion Starter · #1,302 ·
Do you know which version of the firmware you were running previously?
There was a bug in older firmware versions that prevented LiBCM from turning off once the firmware expired.
However, that issue was fixed many months back (probably August).

Have you been downloading the latest released firmware each time you reinstall? Or are you just reuploading old firmware every 40 days? If so, that defeats the purpose of installing the latest firmware ;).

...

Which firmware version are you running now? The latest released version is 0.8.5... if you're running anything older than that, update to the latest and try again.

You can determine the version by opening the Arduino Serial Monitor Window (see the second video on this page).
LiBCM might also spit out some more useful debugging information. Please copy the output and paste it here so I can review it (again, see the video for details on how to do this).

...

Based on the picture you posted, while your battery SoC is certainly very low, it's not damaged (yet). If the battery SoC continues to decrease, it could certainly get damaged... but right now Vcell(min) is 3.276 volts, which is within the specified limits to prevent rapid cell degradation. To be sure, it's not ideal, but it's not dangerous either.

Given that the pack is 9 degC and the cell voltages are all good, I don't see any reason why LiBCM wouldn't want to charge the pack. Do note that once the SoC drops below 10%, LiBCM will turn itself off after a few minutes... so you'll need to plug the grid charger in before that happens... otherwise LiBCM will be off, hence it won't turn the grid charger on. FYI: If you have a laptop plugged into the USB port, LiBCM will NEVER turn off... so that's a good troubleshooting option.

I'm very interested in reading your log file to provide further guidance.

...

P1648 is LiBCM's way of telling the car it's unhappy. Eventually I'll add more IMA Pcodes (like OEM)... for now it's basically P1648 or nothing.

...

While it's fresh on your mind:
-Do you remember final SoC the last time you drove it prior to winter break?
-How long did the car sit over the break?
-How cold was the car the day you stopped driving it?
-What was the coldest the car got while parked?
 

· Linsight Designer
Joined
·
4,812 Posts
Discussion Starter · #1,303 ·
Just wanted to give some updates on the firmware changes I've made this week:
-Changed manual temperature query ($TESTT) to be more user friendly
-Lots of formatting changes for readability
-Complete fan controller rewrite
-Complete grid charger controller rewrite
-Added Heater controller code
-Fixed a bug where temperature sensor values might not be valid for the first several minutes when LiBCM is first powered up (e.g. every 40 days after updating firmware)... this was an annoying race condition between the temperature controller and the power-on initialization handler... now completely isolated.
-Complete temperature controller rewrite
-Added low level getPinMode() & getPinState() functions and related logic to read actual pin states when needed for decision making. Previously LiBCM used separate state variables in RAM, which was easier/faster, but theoretically prone to random bit flips. This is a safety improvement.

I still have some work to do, but firmware is my primary focus right now... spent another ten hours on it today. One of the recommendations from the low level safety critical video lectures I watched earlier this month recommends writing at most QTY3 lines of code per hour (including all the conceptual white boarding before actually writing code). At first I laughed, but now I'm starting to think that's what it takes to write safety critical code.

Overall I've modified QTY17 files and created QTY2 new files.
 

· Premium Member
Joined
·
3,255 Posts
So just something to note for everyone. I'm helping out @ghaze24 by passing on one of my spare batteries to him, and I just pulled it out of storage. This battery has been sitting for about a year (or so, probably more like 15mo) in an non-climate controlled shed which has seen temperatures as low as 20F and as high as 140F. I'm pleased to report it's at exactly the same charge level as it was when I originally received them. It's at 69.5V, which is about 3.86v per cell. I vaguely recall when I built the original pack for my CVT car, that's about what they were at.

Seems to me like long term storage doesn't cause these cells to lose any SoC and won't be in danger of 'going bad'.
 

· Registered
Joined
·
11 Posts
Which firmware version are you running now? The latest released version is 0.8.5... if you're running anything older than that, update to the latest and try again.
I confess that I have been re-loading the version of the firmware that I downloaded on 24th June 2022. This is because I was really happy with the performance of the assist and re-gen that that version of the firmware gave me. Anyway, lesson learned and from now on, I will always update to the most recent version when the 40 days is coming up.

I am very relieved to report that after uploading the latest firmware this morning, the grid charger kicked in again straight away and has taken the battery pack from 5% to 76% during the day. I have just switched the charger off and will see what the display says in the morning before clearing the P1648 and going for a test drive.

Based on the picture you posted, while your battery SoC is certainly very low, it's not damaged (yet).
I am extremely relieved and grateful that we seem to have caught it just in time with sincere thanks and respect to Mudder. ;-)

P1648 is LiBCM's way of telling the car it's unhappy...
Understood and I guess that I should be pleased that is the only code that I have right now.

-Do you remember final SoC the last time you drove it prior to winter break?
Soc 60 I think.

-How long did the car sit over the break?
Due to illness and Xmas commitments, the car was sat for a couple of weeks and was then driven a few times with the expired firmware so in total, the firmware was probably expired for almost a month. I am not the main driver of the gen1 during the winter so I had not had sight of the display for about a month which was somewhat foolish of me!

-How cold was the car the day you stopped driving it?
Here in the UK we had a cold snap just before Xmas when temperatures were below freezing and down to -5degrees C for about a week.

-What was the coldest the car got while parked?

-5c and possibly -7c during one night.

Mudder - Thanks again for your excellent prompt advice and support - I was really thinking that I had done a Peter Perkins airport Job on my lovely new lithium battery pack!

Cheers,
Steve V in the UK
 

· Linsight Designer
Joined
·
4,812 Posts
Discussion Starter · #1,308 ·
Do you have a link for those videos? Would like to watch and learn.
It's basically every single video Professor Phil Koopman has on youtube.

Here are my screens from today and everything seems to be working again. ;o)
Thanks again to Mudder for the excellent technical and morale support.
Glad to see we found the root cause and you're back up and running!
Do you have any feedback on the new versus old assist/regen levels?
Also, is your 4x20 display actually broken, or is it just something your camera is doing?
 

· Registered
Joined
·
11 Posts
Do you have any feedback on the new versus old assist/regen levels?
I've not had a chance to drive it on the open road yet with the latest firmware but will report back when I have.

Also, is your 4x20 display actually broken, or is it just something your camera is doing?
Yeah, the screen comes and goes and I was thinking that it might be a temperature thing?
Maybe I can get a spare one from Peter P?
 

· Linsight Designer
Joined
·
4,812 Posts
Discussion Starter · #1,310 ·
It's an off-the-shelf part, so you can buy a spare one on Amazon.

Given that you're in Europe, you might need to search for the equivalent version that's available in the UK... it's possible Amazon will automatically do that for you. When I load it here in the states it's $9.99 delivered. Hard to beat that price (unless you're me and buying QTY50 at a time in bulk).

Sorry the screen crapped out. They're made without love... out of each QTY50 units I buy, I only end up shipping maybe QTY45 of them. But they're so cheap I don't mind... unless they break out in the field. I never intended for people to use these displays long term; my goal is that once we exit Beta, those people that still want the display will switch over to the Nextion LCD replacement that @Natalya is working on right now. Of course, my original plan was to exit Open Beta maybe six months after I started selling these things... a deadline I've long since blown past.
 

· Registered
Joined
·
11 Posts
It's an off-the-shelf part, so you can buy a spare one on Amazon.
Is this one compatible?

Sorry the screen crapped out.
No worries and entirely not your fault!
 

· Premium Member
Joined
·
3,255 Posts
It's an off-the-shelf part, so you can buy a spare one on Amazon.

Given that you're in Europe, you might need to search for the equivalent version that's available in the UK... it's possible Amazon will automatically do that for you. When I load it here in the states it's $9.99 delivered. Hard to beat that price (unless you're me and buying QTY50 at a time in bulk).

Sorry the screen crapped out. They're made without love... out of each QTY50 units I buy, I only end up shipping maybe QTY45 of them. But they're so cheap I don't mind... unless they break out in the field. I never intended for people to use these displays long term; my goal is that once we exit Beta, those people that still want the display will switch over to the Nextion LCD replacement that @Natalya is working on right now. Of course, my original plan was to exit Open Beta maybe six months after I started selling these things... a deadline I've long since blown past.
I really need to make the move over to the nexon display. I love my 4x20, but Natalya has done such amazing work it looks extremely compelling.
 
  • Like
Reactions: 100percentjake

· Registered
Joined
·
16 Posts
I uploaded the latest software yesterday and had grid charged last week. This morning it was cold and frosty as I set off for work. I noticed soc was low around 24%. As I drove the display started to flash and I noticed state of charge was 0%. It was. Not charging down hills either, I pulled over and switch off to allow reset. Still the same I notice cell voltage was 1.8v and temp of cell was climbing. I switch the pack off by the isolator in the boot and then proceed on my journey. I plan to put on grid charge tomorrow during the day when it is a bit warmer.
any advice most welcome.
 

· Linsight Designer
Joined
·
4,812 Posts
Discussion Starter · #1,314 ·
1.8 volts is not good news.
LiBCM isn't going to allow grid charging with a cell (or cells) that low.

The very first thing we need to do is figure out all cell voltages. To do that:
0: Turn the key off and the IMA switch on. LiBCM might continuously beep because a cell voltage is low. Get earplugs if that bothers you.

1: Plug a laptop into the USB cable, then open an Arduino Serial Monitor Window. If you haven't done that before, follow the steps in the second video on this page.

2: Plug in the grid charger. It won't actually charge if any cell is below 3 volts, but it will start spitting out all 48 cell voltages. Copy and paste that output text into this thread so I can see what's going on.

...

Next, please answer the following questions to the best of your abilities:
-What firmware were you running previously?
-What firmware does the 4x20 display show right now when you turn the key off?
-When you grid charged last week, do you remember the final delta voltage (between highest and lowest cells, top-right most item displayed on 4x20 display)?
-How "cold and frosty" was it this morning?
-Do you remember the SoC when you parked the car? How long ago was that? Was it when you grid charged last week?
-When you say "the temp of the cell was climbing", what was the cell temp 'climbing' to? 50 degC? 10 degC?
-Anything else you can think of out of the ordinary in the past week?

...

Please reply ASAP, so that I can offer further guidance if I think your cells aren't at risk of permanent damage. The longer cell(s) sit at 1.8 volts, the more dangerous it is to recharge them. I'll need to review some notes before I can comment further about whether there's any chance these cells can be safely recharged. This might already be a 'recovery mission', but maybe we can save the pack. If that's the case, then you'll need to change a couple constants in the firmware to allow grid charging.
 

· Registered
Joined
·
16 Posts
Update.
I returned to the car tonight after work and switch the IMA back on and started the car. The regen began. This morning it didnt work at all. I drove a couple of miles and it was charging ok and SOC went put to 23%. I was worried that T was rising and as I didn’t know what a high value was I pulled over and switch IMA of again. When home I switch IMA back on.
I have taken two photos, 1st one is after a couple of miles before I switched off 2nd one when I got home and switched it back on.



Next, please answer the following questions to the best of your abilities:
-What firmware were you running previously? Uploaded 19/10/22 0.84
-What firmware does the 4x20 display show right now when you turn the key off? Uploaded 0.85 14/01/23
-When you grid charged last week, do you remember the final delta voltage (between highest and lowest cells, top-right most item displayed on 4x20 display)? No sorry SOC 62%
-How "cold and frosty" was it this morning? -6C 21F last night about -2C 28F when I set off for work
-Do you remember the SoC when you parked the car? How long ago was that? Was it when you grid charged last week? Grid charged 11/01/23 to 62% used a few short journeys after that.
-When you say "the temp of the cell was climbing", what was the cell temp 'climbing' to? 50 degC? 10 degC? Is that T on bottom left or T on right. See my photos
-Anything else you can think of out of the ordinary in the past week? No
 

Attachments

· Linsight Designer
Joined
·
4,812 Posts
Discussion Starter · #1,316 ·
Based on the information you've provided, I suspect one of the following issues:
A) at least one cell has substantially less capacity than the others, or;
B) one of the BMS leads is intermittently disconnected.

My recommended test procedure is the same for either condition:
1: Turn key off and IMA switch on
2: Connect a laptop to LiBCM's USB cable. You'll need to leave the laptop plugged in because this test will take hours.
3: Open the Arduino Serial Monitor Window.
4: In the bottom-left corner of the Serial Monitor Window, verify the button "Show timestamp" is checked. If not, click the button until the check mark appears.
5: Plug in the grid charger. LiBCM should start grid charging.
6: Let the test run for a few minutes, then copy and paste the entire contents of the Arduino Serial Monitor Window to a test file. Save this file as "00 Data.txt".
7: Allow LiBCM to grid charge the pack. This will take hours. You should stay nearby enough that you can respond to any beeping, if it occurs.
8: After every hour of charging, copy the entire contents of the Serial Monitor Window into a new text file. Save this new file as "01 Data.txt", "02 Data.txt", "03 Data.txt", etc.
9: Continue logging data until LiBCM is done charging ** and balancing *** the cells.

**LiBCM is done charging when high cell voltage parameter reads "H3.900" (the top left value on the 4x20 display).
***For the purposes of this test, LiBCM is done balancing when the cell voltage delta parameter reads "d0.005" or less (the top right value on 4x20 display).

If at any point during the test LiBCM starts beeping, immediately unplug the grid charger, then copy and paste all data from the Serial Monitor Window into a text file. Save this file as "Error-beeping.txt".

...

When this test is finished, place all the files into a .zip and upload here for analysis.
 

· Premium Member
Joined
·
466 Posts
@Campervanman - you asked about temp readings and which one... The lower left "T" on the display is the (T)ime in seconds that LiBCM has been running since last key on..
If you were looking at that and worrying about a climbing temp, that's not it!
 

· Registered
Joined
·
16 Posts
I have followed your instructions but the data in the serial monitor window, is not like it shows in the video. my data is not readable. Not sure what I am doing wrong.

Based on the information you've provided, I suspect one of the following issues:
A) at least one cell has substantially less capacity than the others, or;
B) one of the BMS leads is intermittently disconnected.

My recommended test procedure is the same for either condition:
1: Turn key off and IMA switch on
2: Connect a laptop to LiBCM's USB cable. You'll need to leave the laptop plugged in because this test will take hours.
3: Open the Arduino Serial Monitor Window.
4: In the bottom-left corner of the Serial Monitor Window, verify the button "Show timestamp" is checked. If not, click the button until the check mark appears.
5: Plug in the grid charger. LiBCM should start grid charging.
6: Let the test run for a few minutes, then copy and paste the entire contents of the Arduino Serial Monitor Window to a test file. Save this file as "00 Data.txt".
7: Allow LiBCM to grid charge the pack. This will take hours. You should stay nearby enough that you can respond to any beeping, if it occurs.
8: After every hour of charging, copy the entire contents of the Serial Monitor Window into a new text file. Save this new file as "01 Data.txt", "02 Data.txt", "03 Data.txt", etc.
9: Continue logging data until LiBCM is done charging ** and balancing *** the cells.

**LiBCM is done charging when high cell voltage parameter reads "H3.900" (the top left value on the 4x20 display).
***For the purposes of this test, LiBCM is done balancing when the cell voltage delta parameter reads "d0.005" or less (the top right value on 4x20 display).

If at any point during the test LiBCM starts beeping, immediately unplug the grid charger, then copy and paste all data from the Serial Monitor Window into a text file. Save this file as "Error-beeping.txt".

...

When this test is finished, place all the files into a .zip and upload here for analysis.
 

· Premium Member
Joined
·
1,550 Posts
  • Like
Reactions: mudder
1301 - 1320 of 1493 Posts
Top