[mudder]

If it's not too much trouble, maybe add:

CHARGE_DISABLE_BELOW_DEGC //LiBCM disables regen & charging below this value

The concept behind this feature will exist in a future firmware update. However, I don't intend to offer it as a user-configurable option... basically when the pack is below freezing, the battery won't charge until it's warm enough (or cool enough in hot weather). So if you don't have the heater PCB installed and the pack is below freezing, then the grid charger won't do anything until the sun and/or cabin air warms it back up.

 

[joeaax1j]

I haven't talked about it much, but I actually developed and tested the FoMoCo heater solution in parallel to the 5AhG3... it's presently installed in my car and working well. Next time I take the pack out, I'm going to see about adding a thin insulation layer underneath the aluminum tray... but otherwise the concept works and just needs some fine tuning.

This is awesome! Questions/concerns about the 47Ah heater solution were the only "pre-sales" question/concern that I wrote down on the 47Ah LiBCM kit order form, as I knew I would also need a heater solution with that kit. Thanks!

 

[trickle]

well i don't feel guilty anymore, but i'm still not going to step on mudders toes, at least until he get's everything inside his scope dialed in. thank you for the positive feedback

 

[tervic]

More parts shortage headaches... based on my test results, the 18S heater PCBs need either 6k34 or 6k48 2010 resistors... and they just aren't in stock anywhere. The only thing in stock anywhere is a $0.30 precision part, which adds an unexpected additional $107 in BOM cost per kit. This is a non-starter, but even if it wasn't, there are only enough $$$ parts for sale worldwide to build QTY3 kits.

It blows my mind that in 2022 I still can't purchase discrete passive SMT resistors.
I haven't exhausted my search yet, but this is just ridiculous.

...

Update: I ended up modifying the other heater kit resistor values in order to use an in-stock 6k2 resistor. Purchased enough parts to build QTY5 5AhG3 Heater Kits... for the five people who have reserved this time-suck, money pit kit. If you haven't signed up yet, you won't be in the first batch. If there's no real demand, there might not be a second batch, which is fine because I'd much rather work on other LiBCM stuff.

Mudder,

As you know, I have two Insights with the 5ahG3 LiBCM kits now. I signed up for only one heater kit because I didn't want to "hog" all the good stuff... I do want two heater kits if a second becomes available or if you are able to source more parts.

I am still interested in one FOMOCO kit/heater as well when they are available.

 

[mudder]

I signed up for only one heater kit because I didn't want to "hog" all the good stuff... I do want two heater kits if a second becomes available or if you are able to source more parts.

I've put you down for QTY2 kits, but you'll only receive QTY1 in the first batch.

I am still interested in one FOMOCO kit/heater as well when they are available.

Both the 'DIY FoMoCo Kit' and 'FoMoCo Heater Kit' will release at the same time, so that customers don't end up having to take apart their FoMoCo installations after-the-fact to install the heaters. Installing the FoMoCo kit takes much more time than the 5AhG3 kit. It's not difficult, but there are several more steps. I'm guessing there will be QTY8 total pages of installation videos on Linsight.org (compared to QTY4 for 5AhG3).

The one silver lining with FoMoCo installs is the modules are installed/removed separately, so there's less back strain compared to hauling the complete 5AhG3 assembly in/out. During installation, the heaviest single subassembly is 60 pounds, whereas during removal the heaviest single subassembly is 26 pounds. The weight difference will make sense once you watch the FoMoCo installation/removal videos.

 

[mudder]

Just finished writing the low level pack heater functions. LiBCM automatically detects whether the pack heater is installed. If so, LiBCM will use the heater - instead of the fans - when the pack temperature is below the user-defined setpoints. This code is present in the latest prerelease firmware (0.8.4n).

Today I also manufactured QTY22 12S pack heater PCBs. I thought about just manufacturing QTY6, but honestly it doesn't take that much more time to manufacture QTY22 versus QTY6, so I just went ahead and made a bunch. These PCBs are MUCH faster to manufacture than LiBCM. Once I got going, it took 11 minutes to manufacture each 12S PCB.

Still waiting on parts to arrive for the 18S pack heater PCBs.
Also waiting on a couple other parts for the 5AhG3 Heater Kit, one of which won't be here until Friday... so kits won't go out this week... but soon.

 

[joeaax1j]

Nice, thanks!

 

[mudder]

Today I:
-manufactured QTY22 18S pack heater PCBs.
-assembled QTY5 complete kits, minus the one remaining part I'm missing (ETA Friday).
-verified everything fits together correctly (cables, connectors, etc).
-finalized the BOM. This little side project ended up having QTY27 unique parts.
-storyboarded the assembly videos.

Tomorrow I'm going to film and edit the installation videos.

 

[joeaax1j]

Getting excited!

 

[Tynow]

While I don't have much to add to the conversation at this point, I just wanted to chime in and say that I'm still quite excited for the progress being made on this project.

After getting multiple days of sub-zero (F) temperature a couple weeks ago, I am fully on board and interested in a heater to keep the packs warm and safe. Even if it means that my FoMoCo batteries have to wait longer before being used, I know they'll live a longer happier life as a result.

 

[mudder]

Today I:
-Realized that after logging all that temperature data last week, I forgot to actually update the file on the pick and place machine with my results... which means I had to replace QTY5 parts on each of QTY22 18S PCBs (QTY110 parts total). That took about three hours, which is about 90 seconds per part. For each replaced part, I had to remove the incorrect part, remove the solder from the pads, add solder back to one pad, place and solder one pad on each correct part, solder the other pad, and then clean off all the flux. Before doing this, I made sure to update the file on the pick and place machine.
-Cut out the foam pieces that go on either side of the battery... forgot to do this last night.
-Added an outer jacket to each wire leaving each PCB.
-Started filming the installation videos.

Tomorrow I'll finish filming the installation videos, and might also begin (and finish) editing the videos together. I forgot how much work it is to make these videos.

 

[joeaax1j]

Thanks, John!

 

[tervic]

Sorry that the "time suck" project sucked up even more! Thank you from me too!

 

[davismltc]

Same for me! Sorry for the trouble, but a major Thank You for making this available to the community!

 

[mudder]

The final part I'm waiting on is now expected to arrive tomorrow by 8PM, which means these kits should ship out Saturday.

I've sent paypal payment requests to the six people that pre-ordered them. Right now I only have enough cables to ship QTY5 kits, so they will ship as follows:
-if all six people pay by Saturday, then I'll ship in the order they were reserved. Sixth in line won't get their kit for another week (when more cables arrive).
-if at least one person fails to pay by Saturday, then I'll ship kits to whomever has paid.

Notes:
-The kit costs $225 for those who reserved a kit prior to this post. This price includes shipping.

-In addition to the above price, there is a $50 core charge, which is fully refundable after I receive the old temperature sensors and aluminum end plates. I hate to have to charge this fee up front, but several LiBCM customers have failed to return old parts (after modifying their orders)... and honestly I'm just tired of dealing with it. There is a return shipping label inside the box, so all you have to do is drop the package off at any USPS location.

-Making each kit is more time consuming than I had anticipated, so the price will increase for future orders. I do not know by how much at this point, but it will only increase enough to cover my labor to physically manufacture each kit... and specifically won't include the 100+ hours I spent actually designing this product.

-I don't anticipate making a profit selling these heater kits.

...

I just finished recording the raw videos... now it's off to cobble them all into an intelligible installation video.

 

[mudder]

Final part arrived this afternoon.
QTY5 kits are shipping out tomorrow morning.

Things I need to do before the packages are delivered next week:
-firmware changes so the heater will actually work in the car.
-finish editing the installation videos.
-build heater install webpage on linsight.org

 

[IamIan]

Next time I take the pack out, I'm going to see about adding a thin insulation layer underneath the aluminum tray.

Bumped into today .. passing along FYI

Although pretty expensive per sqft.
McMaster sells 3/16" thick R1.9 flexible Aerogel sheets.

McMaster-Carr

McMaster-Carr is the complete source for your plant with over 595,000 products. 98% of products ordered ship from stock and deliver same or next day.

www.mcmaster.com

These are about 2x the R value per inch of conventional foam board style insulation.

 

[mudder]

At that price it's cheap enough to play around with... but I'm not certain it's necessary, or worth the effort to have to deal with PPE. Right now the aluminum FoMoCo tray is basically in direct contact with the aluminum, so the R value is essentially zero. The stock anti-vibration adhesive might give us up to 0.05 R at best, but it's probably not that high.

Therefore, if we can add even 1.0 R between the two aluminum surfaces, that will give us at least twenty times more insulation. That's probably good enough. Unfortunately there are very few materials that meet the following:
-At least 1.0 R with just 1/4" thickness, and;
-achieve this value when compressed, and;
-do not emit toxic fumes when burned.

Silica aerosols (e.g. the aerogel you linked) do meet these requirements, but they are quite expensive and I have zero experience working with them. For example, I don't know how well they hold up to mechanical compression, or even if they retain any insulation value when compressed. Certainly more research is warranted.

Even a simple 1/4" wood panel is 0.6 R, which is still (probably) at least twelve times more insulation versus nothing at all. However, since wood is so rigid it would pose problems given how irregular the OEM aluminum surface is. So then we might consider recycled clothing insulation, which is compressible and would have around 0.5 R.

Another option that I'm leaning towards is EPDM... which meets NFPA 90A flame spread and smoke development requirements (per ASTM E84 25/50). However, 1/4" thick EPDM's is only 0.2 R... but that's still at least four times more insulation compared to nothing.

Overall I don't yet know how much thermal insulation I'll actually need to sandwich between the tray and the chassis. I did some bench testing last month, but until I get it all together in the car out in the cold it's unclear how much money we need to throw at this problem. I suppose I'l add one of these panels to my next McMaster order, but it's too early to tell whether it will end up in the shipping kit.

Thanks for bringing this product to my attention. It could be just the thermal barrier we need to insulate the pack from the chassis.

 

[trickle]

i would avoid product that can trap moisture since it will never get enough air to dry out once set up. anything to break the thermal bridge will help.. .. its not much of a bridge anyways with how uneven the floor is and its not like there is very much metal on metal surface area. i think the air gap would be better than a wet piece of wood or denim. as far as smoke.. what kind of smoke does the foam in the seats give off, i wouldn't lose sleep over it because flame needs o2. im certain whatever is between the two aluminum surfaces would be the last to go.. hypothetically kinda like tryin to burn a book without opening the covers. i'd still be willing to undercoat/insulate the underside of floorpan if installation of product inbetween the tray isnt effective or worth testing, and possibly save the space between the tray and floor for a 'ptc heat pad or similar' that would only be needed for prolonged storage.
maybe the tray has other cooling benefits in summer im unaware of, by having the air gap?
would be kind of neat if we could substitute traditional insulation for some type of tubing that was only 1/4" thick.. it would help decouple tray from chassis by ruducing surface area and could be made out of nonconductive material and have an option to pipe hot antifreeze through it.. if crushing was an issue could glue a couple shims to the floor.
alternativly something like a synthetic rope layed in an s-pattern might be enough to stay dry down there..
just thoughts, hope theres a simple solution in sight

 

[mudder]

Good point on moisture retention. That moves EPDM even higher on my list of potential solutions.

Based on my testing so far, there is actually quite a bit of heat transfer between the chassis and tray. For example, I typically park my car in a heated garage. However, when I do park my car outside, I find that the lower two modules cool down much faster than the upper three.

You're probably correct that insulation wouldn't burn much when sandwiched between two aluminum layers.

My proposed solution is to bolt several chassis resistors to the FoMoCo tray. Based on my testing last month, adding even just a little insulation between the tray and chassis allows the heat to spread pretty uniformly throughout the tray surface. Therefore, I don't think we'll need to place a heat pad between the tray and chassis. I also don't recommend it, due to how much compression it might see in use.