Honda Insight Forum banner
881 - 900 of 945 Posts

·
Registered
Joined
·
101 Posts
I'm getting an upload error when I try to load software into LiBCM. It seems to be working okay, but I don't like this error message. I tried uploading the sw several times and with 2 different laptops. No joy. Here's the error window:

Reading | ################################################## | 100% 0.01s

avrdude: Device signature = 0x1e9801 (probably m2560)
avrdude: reading input file "/tmp/arduino_build_453929/MVP.ino.hex"
avrdude: writing flash (33568 bytes):

Writing | ################################################## | 100% 5.41s

avrdude: 33568 bytes of flash written
avrdude: verifying flash memory against /tmp/arduino_build_453929/MVP.ino.hex:
avrdude: load data flash data from input file /tmp/arduino_build_453929/MVP.ino.hex:
avrdude: input file /tmp/arduino_build_453929/MVP.ino.hex contains 33568 bytes
avrdude: reading on-chip flash data:

Reading | ################################################An error occurred while uploading the sketch
## | 100% 4.33s

avrdude: verifying ...
avrdude: verification error, first mismatch at byte 0x33d4
0xf8 != 0xfc
avrdude: verification error; content mismatch

====

Edit: I've got a spare AT Mega on the bench. It doesn't have any problems loading the LiBCM software. This makes me think the problem is either the USB cable or the AT Mega in the car. Groan. I do not want to take the battery out.

Another update: I've now got 0.7.9 loaded again. It took many tries, but it finally loaded without errors. I was unable to get 0.8.4 to load, even after dozens of attempts. So I'm at a loss. It still seems like I've got a hardware problem, because 0.7.9 doesn't always load either. But at least my car is operational with 0.7.9.
 

·
Engine-Off-Coast
Joined
·
2,459 Posts
Send me connection info. :) quick edit. The one I have is a more expensive 7" NX8048P070-011C-Y. I can get the smaller ones listed as supported.
I don't know if I can compile for that version or if I need to rebuild the display layout for it to work.

If you get the normal size the wire connection info is in this video @mudder made:

This video is from a month ago; the display looks a little different now and the grid charging page works and it has a warning page if you are grid charging and have the key on.
 

·
Premium Member
Joined
·
219 Posts
I don't know if I can compile for that version or if I need to rebuild the display layout for it to work.

If you get the normal size the wire connection info is in this video @mudder made

This video is from a month ago; the display looks a little different now and the grid charging page works and it has a warning page if you are grid charging and have the key on.
Awesome, thank you! I ordered the normal size, wire connections are same for 'bigger' one. I'll have the 3.5 in hand Wednesday, so last silly question, where is the code for it :). I looked around in the repo and it didn't jump out at me, but I'm not hitting on all cylinders.
 

·
Engine-Off-Coast
Joined
·
2,459 Posts
Awesome, thank you! I ordered the normal size, wire connections are same for 'bigger' one. I'll have the 3.5 in hand Wednesday, so last silly question, where is the code for it :). I looked around in the repo and it didn't jump out at me, but I'm not hitting on all cylinders.
It would have been hard to find:


####
Things you gotta do:
Double-check config.h before you compile for the Arduino

  • Check the current hack flag is the correct one for your car (0%, 40%, etc)
  • LCD_4X20_CONNECTED is commented out by default. So the LCD won't turn on. But if you keep your LCD as well you can uncomment this line. NOTE: I haven't tried running the LCD and LiDisplay at the same time so you'd be a guinea pig for that. You would need an extra USB cable.

Connect Nextion to LiBCM
  • mudder's stock cable for the LCD works but the coloured wires change position so just make sure you get the pinouts correct
  • Once you got the pinout correct turn the key to ON
  • If display lights up white you've probably done it correctly

Install the software onto your Nextion:
  • Download the .tft file from here: GitHub - Lady-Natalya/LiDisplay: LiDisplay for LiBCM (choose the folder for your version of the display -- you don't need to download the repository just the .tft file)
  • Using Windows computer format a MicroSD card with FAT32
  • Paste the .tft file into the flash drive (not needed to use a subfolder or anything)
  • Key Off
  • When display is off put the MicroSD inside the Nextion
  • Key On
  • It should have like a % uploading
  • Update Successded (idk its spelled funny)
  • Key Off
  • Remove SD when screen is off

Test LiDisplay Works
  • Key On
  • Verify LiDisplay running, and shows SoC
  • Try grid charging screen too
  • Verify when you drive that the CHRG and ASST gauges work
  • If the values don't update properly there is a big chance one of your wires is attached to the wrong pin, try changing them until it works
  • Neither the Fan or Screen buttons work right now, so don't be alarmed if they do nothing.
  • Grid Charge page doesn't show any Amps, but it does say CHARGING or NOT CHARGING appropriately

Disclaimer:
Keep your eyes on the road. This is under development beta software, neither I nor mudder can guarantee//warranty that it works. It hasn't been evaluated by the government. You are using LiDisplay at your own risk. Drive responsibly.
 

·
Registered
Joined
·
310 Posts
I saw this in the jazz hybrid manual and thought it was relevant for LIBCM.

If a crash occurs
● Be careful of electric shock hazard.
If a severe crash damages your vehicle’s High Voltage system, there is a possibility of electrical
shock due to exposed High Voltage components or wires. If this happens, do not touch any of
the High Voltage system components or any of its orange wires.
● Avoid contact with High Voltage battery fluid.
The High Voltage battery contains a flammable electrolyte that could leak as a result of a
severe crash. Avoid skin or eye contact with the electrolyte as it is corrosive. If you accidentally
come into contact with the electrolyte, rinse the exposed skin or flush your eyes with copious
amounts of water for at least five minutes, and seek medical attention immediately.
● Use a fire extinguisher for an electrical fire.
Attempting to extinguish an electrical fire with even a small quantity of water, from a garden
hose for instance, can be dangerous.
● Anytime the vehicle is damaged in a crash, have it repaired by a dealer.
 

·
Premium Member
Joined
·
219 Posts

·
Linsight Designer
Joined
·
4,143 Posts
Discussion Starter · #889 · (Edited)
I'm getting an upload error when I try to load software into LiBCM. It seems to be working okay, but I don't like this error message. I tried uploading the sw several times and with 2 different laptops. No joy. Here's the error window:

avrdude: verification error, first mismatch at byte 0x33d4
0xf8 != 0xfc
avrdude: verification error; content mismatch
Does the error always occur at address 0x33d4? If so, then it's probably the Atmega2560. If the address changes, then it's a signal integrity issue. This is the first I've heard of this issue.

Edit: I've got a spare AT Mega on the bench. It doesn't have any problems loading the LiBCM software. This makes me think the problem is either the USB cable or the AT Mega in the car. Groan. I do not want to take the battery out.
FYI: The Atmega2560 that ships with LiBCM has a custom bootloader. If you replace with an off-the-shelf Atmega2560 then LiBCM will eventually stop working (specifically whenever I get around to enabling the power saving features).

Obviously I can send you another Atmega2560... or, if you know how to flash bootloaders, then you can just load the bootloader source code onto any Atmega2560 using your favorite debugger.

Don't have a debugger? Here's a standalone Arduino Uno flasher... Here's the readme.

Another update: I've now got 0.7.9 loaded again. It took many tries, but it finally loaded without errors. I was unable to get 0.8.4 to load, even after dozens of attempts. So I'm at a loss. It still seems like I've got a hardware problem, because 0.7.9 doesn't always load either. But at least my car is operational with 0.7.9.
I wonder if the USB cable/port got damaged when installing the top IMA cover?
We can indirectly test for USB issues by verifying the keyON datastream:
1: Turn key off.
2: Plug in laptop to LiBCM USB.
3: Open Arduino Serial Monitor window.
4: Turn key on.
5: Verify no data corruption. By default, you should see the following:
Code:
12:42:59.589 -> //Note: Gibberish before "METSCI BEGIN" is expected
12:42:59.589 -> METSCI BEGIN
12:42:59.589 -> BATTSCI BEGIN
12:42:59.589 -> LiDisplay BEGIN
12:42:59.589 -> LTC6804 BEGIN
12:42:59.589 -> HW Rev: D
12:42:59.589 ->
12:42:59.589 -> Welcome to LiBCM v0.8.5, 2022SEP06
12:42:59.589 -> Type '$HELP' for more info
12:42:59.589 ->
12:42:59.589 -> Key:OFF
12:42:59.627 -> Old SoC: 60%, New SoC:69%
12:42:59.627 -> Current Sensor 0A set to (counts): 332
12:42:59.627 -> Total hours since firmware last uploaded: 5 (960 Hours MAX)
12:43:05.645 -> Key:ON
12:43:05.645 -> Charger: OFF
12:43:05.682 -> 0,0,A, 183,169,V, 3.830,3.827,V, 3450,mAh, 0.0,kW, 23,C *
12:43:05.902 -> -1,0,A, 183,169,V, 3.830,3.827,V, 3450,mAh, -0.2,kW, 20,C
...
Code:
12:43:09.918 -> -1,0,A, 183,169,V, 3.830,3.827,V, 3450,mAh, -0.2,kW, 20,C
12:43:10.170 -> -1,0,A, 183,169,V, 3.830,3.827,V, 3450,mAh, -0.2,kW, 20,C
12:43:10.237 -> Key:OFF
12:43:10.237 -> Old SoC: 69%, New SoC:69%
12:43:10.237 -> Current Sensor 0A set to (counts): 332
12:43:10.237 -> Total hours since firmware last uploaded: 5 (960 Hours MAX)
If you get random extra/corrupt characters, then there's certainly a hardware issue.
 

·
Linsight Designer
Joined
·
4,143 Posts
Discussion Starter · #891 ·
With that level of reproducibility, this isn't a USB cable issue. So in both cases the errant byte has a bit error at bit(2):
At address 0x1ADA, 0x29 != 0x2D:
0x29 = 0b00101001
0x2D = 0b00101101

At address 0x3AD4, 0xB3 != 0xB7:
0xB3 = 0b10110011
0xB7 = 0b10110111

The error you're seeing is due to the bit(2) mismatch. On the Atmega2560, this memory is rated for at least QTY10000 write/erase cycles... but typically they last much longer than that. Your Atmega2560 had just QTY2 write/erase cycles when I sent it to you, so there might now be a couple dozen write/erase cycles now.

Regardless, looks like you need to replace the Atmega2560 PCB... I just shipped one to you (should arrive Saturday). Once it arrives, there's a return label in the box, so you can send me the defective one so I can verify the behavior. It's probably just defective, but I want to make sure.
 

·
Registered
Joined
·
9 Posts
Maybe my experience will help... I reprogrammed mine to v0.8.4 yesterday, and did have one oddity that may be different from a Mac: I had to actually "unzip" the file--it was NOT automatic! I use a Dell PC/Windows laptop, and thought things were going well: IMA battery switch ON, grid charger not plugged in, and Ign. switch OFF. Then, tap on Firmware, tap on MVP, open the Arduino installer, finally click on right-arrow "Upload" button--but then I got an error code, "libcm.h: no such file..." or some such. I tried again, no change. And again. Definition of insanity... Then I noticed a 'zipper' folder icon at the upper left and tapped that, and it actually unzipped the file I had downloaded. After 30 or 40 seconds, I got a message that I had extracted more than 900 files and--it worked! I successfully reprogrammed my Linsight BCM!! Whereas I thought the file had been automatically unzipped--after all, before I hit the 'unzip' icon, I already had multiple files and directories including the config.h file, the MVP file, the Arduino screen and the upload button, and everything seemed to be in order--but it would not work until after I hit the unzip icon. And now it's back to its sweet hot-rod self. HUGE difference from when the IMA quit working after I let it go more than 960 hours...
 

·
Registered
Joined
·
101 Posts
@mudder I got the replacement ATmega. Now I'm having new problems. Let me explain what I've done:

1) While I was waiting for the new board to arrive, I tried to burn the LiBCM bootloader from 0.8.4 onto my spare ATmega by using an Arduino Uno as the ISP. After much futzing around, it seemed to work. Unfortunately, no. Nothing would download to the board with this new bootloader. It always timed out. So I figured I messed something up, and waited for the new, preloaded ATmega.

2) When your replacement arrived, I tried to download the 0.8.4 LiBCM sketch. It timed out exactly like my spare board. Here's what I get:

Using Port : /dev/ttyACM0
Using Programmer : wiring
Overriding Baud Rate : 115200
avrdude: stk500v2_ReceiveMessage(): timeout
avrdude: stk500v2_ReceiveMessage(): timeout
avrdude: stk500v2_ReceiveMessage(): timeout
avrdude: stk500v2_ReceiveMessage(): timeout
avrdude: stk500v2_ReceiveMessage(): timeout
avrdude: stk500v2_ReceiveMessage(): timeout
avrdude: stk500v2_getsync(): timeout communicating with programmer

avrdude done. Thank you.


My first thought was that I messed up something in the IDE when I was burning the bootloader. But I've checked everything I can think of to check. The port and board type are correct. And I can still download stuff to my Arudino Uno.

So I'm stuck again.
 

·
Linsight Designer
Joined
·
4,143 Posts
Discussion Starter · #895 ·
Sounds like you successfully loaded the custom bootloader. The key you're missing is that with the custom bootloader installed, the mega2560 must be plugged into the LiBCM PCB.

The reason this is required is that during each power-up, the custom firmware looks to see if the ignition is 'on'. If so, the bootloader immediately exits. However, if the ignition is 'off', then the bootloader looks to see if the host is trying to upload firmware.

The reason this is required is that LiBCM must turn on immediately when the key turns on, or else you'll get a CEL.

So basically you need to plug the mega2560 into LiBCM and then it will work. Please keep track of which one initially shipped with LiBCM... so you can send it back to me (prepaid) so I can see what's wrong with it.
 

·
Premium Member
Joined
·
219 Posts
Story of the film so far: 1. really bad reaction to something used in surgery so going into 3rd week of meds with 'do not operate space craft while using'. I did manage to flash Lady Natalya's fork with Nextion display. I grabbed it all and compiled a version for my 7" as well, but the 7 is really too big.

Problem: (remember the meds, I'm truly sharp as a bowling ball right now). With LibCM turned on, I'm getting no start, as in no engine spin at all. No errors, same as if clutch wasn't pushed in or clutch switch bad. With LIBCM turned off, I get engine spin, but no start. Cranks but acts like no spark or no fuel.

The meds are great, and Elvis has stopped by to talk while I've been doing this, but he's not giving suggestions. Any advice on what I'm missing? Likely very basic...



Motor vehicle Communication Device Gadget Audio equipment Measuring instrument
 

·
Linsight Designer
Joined
·
4,143 Posts
Discussion Starter · #898 ·
Problem: (remember the meds, I'm truly sharp as a bowling ball right now). With LibCM turned on, I'm getting no start, as in no engine spin at all. No errors, same as if clutch wasn't pushed in or clutch switch bad.
Does issue persist with latest official LiBCM firmware?

With LIBCM turned off, I get engine spin, but no start. Cranks but acts like no spark or no fuel.
If the engine won't start on the backup start, that has nothing to do with the IMA system... something is wrong in the engine bay. Check for fuel/air/spark.
 

·
Premium Member
Joined
·
219 Posts
Does issue persist with latest official LiBCM firmware?


If the engine won't start on the backup start, that has nothing to do with the IMA system... something is wrong in the engine bay. Check for fuel/air/spark.
I'm going to let it sit a bit until I have more focus. The issue with backup no-start has come up before, and always in conjunction with LIBCM work (pack pull, and or/main cables disconnect/reconnect). Since it isn't directly an IMA relationship, it may have bearing on my power drain that killed the previous pack. To the question on LIBCM software, yeh haven't done my usual diff testing. Pulled the rpi out to put it into an AIO touchscreen, it was separate rpi/touchscreen with annoying cable sprawl. My endgame is to be able to push changes to LIBCM remotely through the rpi wifi, remote monitoring.
 
881 - 900 of 945 Posts
Top