Here are a couple videos that cover it pretty thoroughly.
@b264 is pretty familiar with FOC as well.
2 Likes
But that’s fine since he’s adding a pack so one of the existing packs can already carry the entire load.
I did this for years paralleling a boosted V1 12S lifepo4 pack with some 18650 10s packs. They have similar voltages and discharge curves so it worked fine. Just had to check that the voltages were exactly the same after charging them separately before paralleling them again which was a bit annoying.
2 Likes
this is the first I’ve heard of 12S LiFePO4 in parallel with 10S li-ion
I’d imagine the li-ion will carry most of the load for the first quarter-ish of discharge, then the LiFePO4 would carry most of the load for half the discharge, then the li-ion again for the last quarter-ish of discharge, assuming similar capacity packs. Interesting.
@Battery_Mooch what do you think about that?
1 Like
anyone tried rivet nuts instead of inserts?
interested to try it with some m4 blind rivet nuts and to buy the appropriate tool
Hm that is a weird situation for sure. Obviously it worked out okay but I wonder what the actual discharge looked like from each battery, especially on acceleration
2 Likes
The issue with rivnuts is that they have a very limited panel thickness that they can grip, and they stick out the back. Inserts are better for thicker things like skateboard decks, while rivnuts are better for thin sheetmetal.
2 Likes
ah ok
sadly the build im doing is with carbon fibre and though i could install inserts the whole way the through… the deck is only 5mm thick and the inserts i have are way larger
decided to go with connecting bolts and have countersunk ones at the top side of the deck
gonna use loctite to stop them coming loose
Ooohhh…tough one.
During the discharge each will supply a certain share of the current based on their internal resistance. Either pack could supply more or less current depending on the starting internal resistance (this includes the spot-weld, nickel strip, solder joint, and wiring resistances) and how the cells change their internal resistance as they warm up.
But as @b264 alluded to, each pack will try to drop their voltages at different rates due to the discharge characteristics for the different cell chemistries. This will cause inter-pack current flow that will change the overall current from each pack.
If run to a low enough voltage then whichever has the greater capacity will eventually start supplying more current to make up for what the almost-empty lower capacity pack can no longer supply.
My head hurts trying to think this one through. I was thinking about running a simulation to see what happened for different setups but my battery models aren’t detailed enough for this.
Bottom line…I’m not sure what will happen as each pack discharges. The current from each will shift but I don’t know by how much and what particular cells are used could have a bigger effect than the cell chemistry (and their respective discharge characteristics).
Interesting topic though!
I don’t see any big issues as long as neither pack is being run near its current rating (since we don’t know how well they will share the current) and as long as the packs are being charged to 42.0V.
That will charge the “standard” Li-ion pack at 4.2V/cell and the LFP pack at 3.5V/cell.
3 Likes
My theory was that the packs won’t charge each other as long as you never exceed the max draw of the totality of the packs. It worked fine although the battery guage on the remote didn’t give you an accurate reading anymore. Not sure if it’s because the columb-counting was off or because the voltage-vs-remaining capacity no longer perfectly matched the ramp for LiFE batteries.
Only did it because Justin from ebikes.ca (who also founded endless-sphere) said it should be fine - he’s a very bright dude.
My thread about my hack:
https://endless-sphere.com/forums/viewtopic.php?f=35&t=70234
Packs don’t have a set “proper” current rating though. You can rate a pack at 5A or 500A depending on the cycle life, performance, and safety thresholds you want.
There will be inter-pack current if you discharge them at any level. They will still share current unevenly and differently as the discharge proceeds. The inter-pack current will be lower for lower overall discharge current levels, of course, but it will still be there.
I made some assumptions about your setup that maybe I should not have. Each pack must have its own BMS or no BMS’ at all (which I do not recommend).
Both packs should be charged together to the same voltage, 42V. Otherwise the 43.2V LFP pack will force current into the “standard” 42.0V Li-Ion pack when you connect them. This will force the standard Li-Ion pack to overcharge since it was already fully charged.
Make sure you low voltage cutoff is high enough for both chemistries.
This sounds very likely. Unless it does the coulomb counting you mentioned and the true effective capacity of the combined packs wasn’t set right.
You can parallel different chemistry packs but, as mentioned above, there are things that need to be taken into account.
Yup I was using a few of these:
https://ebikes.ca/product-info/grin-products/ligo-batteries.html
Which each have their own BMS (plus the boosted built in batteries which of course have their own BMS).
Yup - that’s what did for the most part.
There were times when I hadn’t been using the extra packs and I’d charged the board to 43V and I’d need to run down the board to 42V before paralleling in the packs to avoid the disaster scenario.
should i still use the antispark switch with trampa vesc 6? do i need to buy a new ppm cable? also how do you convert the vesc 4 remote to vesc 6
anyone have experience with these type of right angle bullet connectors?
im thinking of using some Turnigy hardcase lipo pack, and it seems these connectors are the way to go to keep it thin, i won’t be pushing mega amps, but want to know if anyone have use them or not
1 Like
Aren’t these just regular bullets but not soldered straight?
4 Likes
actually i don’t know, i’ve been seeing plug like this
but they stick out quite a bit, then i found something like these and wondering if that’s what the Turnigy connectors are
they seems to have a low enough profile to make a thin pack
5 Likes
Uuh interesting, haven’t seen those before!
2 Likes
If I use the maximum duty cycle to set speed limit, can I use the real time data in the VESC tool to see the speed i would get? Or do I have to acutally drive the board and see for myself?
1 Like
the switch cables go in here right?
use the forum search function, none of these are hidden. its also publicly available on vendor’s website
4 Likes
i dont get that pinout, theres no switch port