[davismltc]

@mudder
I've installed the heater boards and trying to test for the LED blink, but I'm not sure I'm doing it correctly. The installed firmware is 0.8.5
I have the main breaker "off" that splits the pack in half
Everything is installed per the videos and I've opened up the fan shroud to view the LED
I plugged in my laptop, opened arduino and selected serial monitor.
(Libcm display is lit and is warning about the plexi cover not being installed.)
I enter $testh and hit enter...also tried control+enter...also tried $TESTH
I don't see the green LED flash
You mention to reverse a connector....is it the two pin heater connector connected directly to the Mega board?

I haven't done much with serial monitor, so I'm not sure I'm doing this correctly...here's a pic where I would enter "$testh"
Should the main breaker be "ON" for this test?

 

[mudder]

Thanks for the feedback... looks like I need to make some corrections to the heater installation instructions. For now, here's some clarification:
-By default, LiBCM won't do anything if the cover is off when it first boots. You can either hold the keyboard button down with your finger while you open the serial monitor window, or you can comment out the line PREVENT_BOOT_WITHOUT_SAFETY_COVER in config.h. FYI: This is a safety feature to prevent people from installing LiBCM without the safety cover.

-The main breaker should be on. If the breaker is off, technically the test you're performing will work, but LiBCM might draw too much power from your laptop if the fans decide to turn on.

-You also need install version 0.8.5a or later, which as of this writing means you need to install the prerelease firmware branch (which is identical to installing the released (main) branch).

-Also, as of this writing I haven't finished writing the code to actually use the heater (only to make the LED blink). I'm doing that right now, and am behind schedule (as always). It is absolutely critical that the heater doesn't turn on when it should be off... which means I've had to rewrite a bunch of code, including some pretty low level architecture changes... for example, instead of using state variables, I'm actually querying the heater state from the memory mapped pin.

I certainly wish it was as simple as if(packTooCold) { heatPack(); }; so far I've created two new files, touched 12 other files, and changed quite a bit of code. The things we do for safety. The embedded safety lectures I audited over the past couple weeks were super informative.

-My goal is to prerelease heater firmware support tomorrow.

Thanks for your patience.

 

[davismltc]

Thanks for the speedy reply!
I'll play with it tomorrow morning and see if I can get the LED to blink.
No urgent rush for me on the firmware. I just had some time to mess with all the mechanicals and thought I'd get them out of the way while the weather was still warm.

 

[mudder]

Thanks for the speedy reply!
I'll play with it tomorrow morning and see if I can get the LED to blink.
No urgent rush for me on the firmware. I just had some time to mess with all the mechanicals and thought I'd get them out of the way while the weather was still warm.

Sweet, please let me know if you have any feedback regarding the heater install addenda videos. Did I get any steps out of sequence?

 

[davismltc]

I ended up commenting out the safety cover line in config.h and reloaded the firmware
The firmware uploaded is 0.8.5 ... 2023JAN02 build date (not sure about 0.8.5a?)
Breaker switch is "ON"
When I try the serial monitor, LiBCM seems to recognize the $testh command
(FYI.. In your video, you seemed to hit enter several times in a row to get the LED to flash. It's a one shot deal in windows and I have to use control+enter and then re-type $testh each time.)
When I enter $testh in serial monitor, LiBCM replies with "echo $TESTH", with the next line "temp:18", but I don't see any LED flash.
I tried reversing the 2 pin heater connector that's connected to the Atmega, with the same results (no LED blink)

Just to make sure I was visualizing the LED properly, I've since removed the aluminum plate of the 12 cell module with wires still connected. Still no LED blink either way the 2 pin connector is connected. I'm double checking all my wiring and haven't found any errors...any ideas?

As far as I can tell, the heater video is sequenced properly. I did have to remove the fan shroud to test for the LED blink, but that's just me not remembering to leave it unattached...no big deal.

Here's a pic..

 

[davismltc]

@mudder
Heater 12cell board wiring/connections:

 

[Balto]

I certainly wish it was as simple as if(packTooCold) { heatPack(); }; so far I've created two new files, touched 12 other files, and changed quite a bit of code. The things we do for safety. The embedded safety lectures I audited over the past couple weeks were super informative.

-My goal is to prerelease heater firmware support tomorrow.

Thanks for your patience.

🤣 🤣 🤣 🤣 🤣 🤣 🤣 🤣 🤣

Comments like this kill me, I have a friend that codes and I make jokes like this to him all the time.

 

[davismltc]

@mudder
I believe I've found the problem. When I enter $testh in the serial monitor, I now get an LED blink. HOORAY!!! (and it's with the two pin heater connector oriented as per your video)

It turns out that the 0.8.5 in "Main" (build date jan 02) is not the same as prerelease 0.8.5a (build date jan 05)

When I saw the two different dates, I thought it was worth a try to upload the prerelease and that solved the issue.

 

[joeaax1j]

I'll be tackling my heater install this weekend. This is good to know, thanks @davismltc

 

[davismltc]

When I re-install LiBCM in my car and perform data logging...is the temp displayed from the data log coming from the battery temp sensors?

I think you mentioned before that the temp on the LiBCM LCD display comes from the three OEM (air) temp sensors

Thanks!!

 

[mudder]

I ended up commenting out the safety cover line in config.h and reloaded the firmware
The firmware uploaded is 0.8.5 ... 2023JAN02 build date (not sure about 0.8.5a?)

You need to install the prerelease firmware branch, which is always located at this specific location.
If LiBCM doesn't report 0.8.5a or later, then this test will not work (because I didn't write the code to blink the LED until 0.8.5a).

(FYI.. In your video, you seemed to hit enter several times in a row to get the LED to flash. It's a one shot deal in windows and I have to use control+enter and then re-type $testh each time.)

You can resend the previous command by hitting the up arrow, then enter.

When I enter $testh in serial monitor, LiBCM replies with "echo $TESTH", with the next line "temp:18", but I don't see any LED flash.
I tried reversing the 2 pin heater connector that's connected to the Atmega, with the same results (no LED blink)

Until you're running 0.8.5a or later, the LED isn't going to blink.

Just to make sure I was visualizing the LED properly, I've since removed the aluminum plate of the 12 cell module with wires still connected. Still no LED blink either way the 2 pin connector is connected. I'm double checking all my wiring and haven't found any errors...any ideas?

Install 0.8.5a or later.

As far as I can tell, the heater video is sequenced properly. I did have to remove the fan shroud to test for the LED blink, but that's just me not remembering to leave it unattached...no big deal.

Whoops... I forgot that the shroud is already installed by the time I tell you to look for the LED blink. Maybe you can still spot the LED blink with the shroud installed (through the fan hole)? I'll reshoot that part of the video in the correct order. Thanks for the feedback

 

[mudder]

I believe I've found the problem. When I enter $testh in the serial monitor, I now get an LED blink. HOORAY!!! (and it's with the two pin heater connector oriented as per your video)

It turns out that the 0.8.5 in "Main" (build date jan 02) is not the same as prerelease 0.8.5a (build date jan 05)

Glad to see you were able to figure this out.

When I re-install LiBCM in my car and perform data logging...is the temp displayed from the data log coming from the battery temp sensors?

I think you mentioned before that the temp on the LiBCM LCD display comes from the three OEM (air) temp sensors

In both cases the displayed temperature is from the three battery temperature sensors.

FYI: The only time LiBCM displays the other temperature sensor values is if you type $TESTT, which outputs all temperature sensor values. Note: this is a debug tool I wrote for my own personal use. Note: When the key is off, you must send the command twice within one second; the first command turns the sensors on (but reports inaccurate values)... the second command reports accurate temperature values. I might improve this test code later, but it's not a priority.

 

[FlyingHood395]

Thank you for the fast and thorough reply as usual, we're so lucky to have you in this community. Really too bad for me that its the modules, suspected it but glad its not a weird Libcm issue that would cause you extra headache!

A. I got them from a user on the forum, I don't know too much about them other than they were claimed to be in good shape and stored at a reasonable temperature. I stored them in a dry 60-70ish degree place for a couple months. They were packaged well and didn't appear to have any damage. Once I got your good 12 module Libcm I was able to grid charge with no issues, initially my 12 module was so bad I couldn't grid charge. These were welded 18s modules if that potentially matters.

C. Right now this is the only picture I can find. I can take more once I find replacements and or install the heater kit.
View attachment 100918

D. Happy to send you one or more once they're out.

F. Thank you, I'll change that setting. I do plan to drive the car until I can find new modules.

G. I totally understand that, I wouldn't have placed blame on LIBCM in this scenario.

H/J. You're welcome and definitely unfortunate I've had such bad luck. Loving LIBCM anyway!


and on that note, I'm in the market for an 18s+ and or 18s- if anyone got any leads!

Another Frankenpack! You're not alone, I had a blue module for awhile as well. Bummer you have to go through this, but for me swapping out with the regular module fixed all the issues for me. So you should be fine when all is said and done.

 

[100percentjake]

I certainly wish it was as simple as if(packTooCold) { heatPack(); }; so far I've created two new files, touched 12 other files, and changed quite a bit of code.

 

[Balto]

View attachment 100969

Hahahaha

 

[mudder]

if( goingToCrashIntoEachOther ) { dont(); }

I would love to see the resulting chaotic flight patterns if this was guaranteed to always work.

 

[100percentjake]

I would love to see the resulting chaotic flight patterns if this was guaranteed to always work.

They would most likely equalize into a stable roughly homogenous density of drones within the given flight envelope. I'd assume each drone in a swarm is constantly reporting its absolute position in free space (relative to what I'm not sure) and an "air traffic controller" is maintaining a bubble of safety around each member of the swarm.

It is fascinating seeing the engineering and decisions that go into something as seemingly simple as "add heaters to a battery pack"

 

[Balto]

A space heater placed inside the vehicle would have been effective.

 

[mudder]

A space heater placed inside the vehicle would have been effective.

A couple people have shown off their kludged solutions. The version I just spent numerous weeks developing is safer, more efficient, and automatic. With a wide enough risk tolerance, either solution will work.

 

[Balto]

A couple people have shown off their kludged solutions. The version I just spent numerous weeks developing is safer, more efficient, and automatic. With a wide enough risk tolerance, either solution will work.

What you have done is nothing short of amazing, not to mention the time frame you did it in.