Autosensing charger?

Soon In my “fleet” I will have a 10s,12s and 16s board. I really would like to avoid multiple chargers for each one of them.

Is there any possibility to make an autosensing charger they will readout the voltage of the battery and accordingly adjust the output voltage so we can have one charger to rule them all?

Is it even possible to make such a thing? If so I would really like to fund such a project. Or even better if there is already such a thing, where to find one?

3 Likes

i think @Battery_Mooch was working on something? or was bms instead of charger? i don’t remember now, fuzzy memory :smiling_face_with_tear:

5 Likes

Possible and not very difficult to do. :slightly_smiling_face:
The problem is that different packs can have overlapping voltage ranges, for example (assuming 3V/cell minimum)…
10S = 30V - 42V
12S = 36V - 50.4V

If the pack is between 36V - 42V the charger won’t know if you have a 10S or 12S pack. The charger could use the safer setting, 10S (42V), but if the user doesn’t notice that they’ll probably be pissed when they find out later that the pack is only partially charged.

You can have voltage settings on the charger (switches, a display and buttons, etc.) but the user would have to remember to change the settings when necessary. The charger won’t be able to auto-detect all of the possible bad switch settings (for example, using a 12S setting on a 10S pack).

8 Likes

So the only solution then is to somehow connect the charger to communicate to the BMS, like metr is doing, fetch the BMS config and then adjust the output.

That could be done by connecting to CAN bus I suppose?

3 Likes

Making the whole thing another link in a buggy, unreliable, all-kinds-of-incompatibility-issues chain that everyone already goes crazy trying to deal with. :slightly_smiling_face: Flipping one of 3-5 switches on the charger begins to have real appeal. Especially considering the additional development time and cost to implement a “smart” charger like that.

The FreeSK8 system could probably do something like that successfully though.

6 Likes

@WavRX implementing communication between the ilogger, the LTT smart BMS and a hypothetical wavrx smart charger would make the complete Esk8 echo system :grin:

You could even implement a notification in the app or by email (if the ilogger it’s connected to a wifi network) that tells you when your board it’s finished charging.

5 Likes

IMO the charger should have switches (or some other method to select voltage) anyway since there will be a very limited number of people with the particular “smart” setup it would be using.

The amortized development costs would be huge unless there is a good sized market for the charger.

5 Likes

What happens if I turn the 16s switch and plug it in 10s BMS?

Hopefully the BMS turns off the charge at 42V. If not the pack could catch fire. It would not be a complete charge since the pack wasn’t allowed to sit at 42V to “top off” but it would be a pretty decent charge.

A possible problem…if the BMS was turned off when the charger is connected at 16S it could burn out a BMS that was not rated for 67V

3 Likes

That might be the next thing I should get going, who knows :grinning:

If I was to tackle this, I think there should be some sort of a universal handshake that happens as soon as you plug a smart charger in using some sort of combined CAN/POWER plug.

We already have a great open source smart BMS (DieBieMS and derivatives) capable to communicating through CAN, which iLogger can detect and interact with.

Handshake would give the smart charger all the information it needs to adjust itself and start charging at the correct voltage level, while periodically receiving status and safety updates.

5 Likes

It’s not auto sensing but the Cycle Satiator can save profiles etc so it’s just clicking a button for diff voltages

2 Likes

I think the simplest way to go about the detection of the cell count is to use a 3 pin connector and use the 3rd pin to provide and “ID”. Kinda like with USB. You could just have a particular resistance for each different cell count, then the charger would measure that resistance and adjust its output setting accordingly.

I think this could all be done with an Arduino and a 16s charger. You would just have the arduino monitor the voltage and switch the output on/off based on the cell count and if it has reached its max charge voltage yet.

3 Likes
Wouldn't the BMS possibly pop if you supply it 16s instead of 10s/12s?

Nevermind. Since it’s sensing first, it wouldn’t send the full voltage at all times.

After reading all this I feel a whole lot better about just buying a new charger for ever voltage I got. Which so far is 12s,16s, and 20s. If each charger is around 120 then that’s $360. I’m imagining this system would (at the end) cost more than $360?

What’s the size of the potential market? How many people have this really cool problem lol

7 Likes

If you implemented the arduino based solution I described, it should only cost you one 20S charger and less than $50 in other components.

3 Likes

Eh, if you were DIYing it all you’d probably need is a 120V adjustable DC power supply and an arduino which would add up to about $120 if you bought a cheap power supply. Messing around with 120V power supplies is not something I really like though as they’re actually dangerous. On both ends and just about everywhere inside assuming you get one capable of any reasonable current.

What about big squishy buttons with S counts on them. And we can just go BLOOP BLOOP BLOOP…BLOOOOOOOP

and then it does magic science stuff and we can ride moar sket.

:sunglasses::test_tube::dna::alembic::magnet::magnet::goggles:

9 Likes

The cycle satiator is $330, so savings! I currently run 12, 13, 16, 18, and 20s depending on the EV, so an adjustable was the best play financially

4 Likes

This reminds me of the charger to end it all thread, still haven’t gotten around to making one though

FYI

4 Likes

I wanted to post that it would be awesome if the satiator could be modded to have a DC input as well as AC.

2 Likes