Even Newer 4 to 24-cell Battery Management System (BMS)

It looks like I will just replace the IEC C-16 receptacle/plug currently used on the scooter with a 4-pin XLR connector with 2 pins jumped on the plug.

The C-16 computer/hot-pot connector for charging isn't a good idea anyway, as an uninformed person could plug 120VAC into the charger plug - imagine the exploding, flaming cells!

But, how much DC amperage can an XLR connector handle?
 
I think some are more capable than others but, from the web:

I use these XLR connectors every day in the Sound and Lighting business.
3 pin - for Audio inputs & outputs like Microphones and Mixers
4 pin - for Automated color changers that fit theatrical lights
5 pin - For the digital DMX control signal between console and Lighting

These connectors are very reliable, if there is a problem it is almost always something wrong with the wire or the solder connection, but not the connector. I never knew the amp capacity of the pin connections but the PDF I found on the Neutrik.com site that dinges posted has the following information...
3 pin: 16 amp
4 pin: 10 amp
5-6 pin: 7.5 amp

The PDF also says that they are rated to Connect/Disconnect 1000 times. I would say that it is 3 to 5 times better than that. They would not last very long if you plugged them while under load. Arcs will pit them pretty bad. Microphones output millivolts, Sound or Lighting consoles output up to 10 volts, Color Changers output up to 24 volts. The amps on all these outputs is quite low, less than .1 amp. If you tried to disconnect a hot load of 10 amps it would eventually become unusable.
 
Assuming the batteries can handle it, what is the maximum charging current this BMS can handle? What is the limiting factor?

Thanks!
 
I've used the 3 pin XLRs at 20 amps with no problem. You do have to use the high quality ones, since some of the really cheap ones use very thin metal in the contacts. A 4 pin XLR should be perfect for this application.
 
Gary,

The chargers are just 8 amps. but I'd like to go to a more powerful charger at some point. The Thunderskys can be charged at 0.5C or 20 amps.

Jeffkay,

I was assuming those ratings were for AC RMS amps since they are normally used for stage and audio stuff. How does this relate to DC amps? Then again, amperage ratings are usually very conservative.
 
As with any wire or connector, if it is getting hot. you will know so you can test it. Personally I think 20A into an XLR is call for a different style connector. I use Anderson's large forklift style on "Deep Cycle". They are good to 175A.

Here is a FUNNY for you guys. Due to funding parts for the new 3-wheel EV I am building, I am selling an extra ICE engine for my 1960 Porsche 356. Since it is semi-restored, there is no line from the fuel tank. So I rig a 1 gal. gas can near the engine. I went to the gas station to fill it--ON THE ELECTRIC MOTORCYCLE! Felt weird.

Jeff K.

RMS-- I don't think the pins are rated RMS because the original spec was for power connections. I could be wrong and most usually I am. See above.
 
I use alternator harness connectors, they are robust and can certainly handle high amperage. They are working well, but I keep buying the last of the stock. I have andersons also, but honestly I don't really like them very much.
 
Johnbear said:
Assuming the batteries can handle it, what is the maximum charging current this BMS can handle? What is the limiting factor?

Thanks!
Click to expand...

I was designing for 20 amps, but I don't have a test setup that can go that high yet.
The limiting factor will be heating on the FET. If the FET runs too hot, it would just be a matter of adding a heatsink. The FET can handle up to 70 amps with a big heatsink.
 
Instead of XLR's, I opted for 4-conductor Neutrik "Speakon" connectors which are designed for commercial amplifier/loudspeaker connections and are good for 30 amps. My Chinese scooter, which was deceptively sold as being "German" (they had a German WMI code in the VIN, but they got ordered to change it to a Chinese one a just before they went out of business) will now at least have some Lichtenstein content. Close to Germany anyway...

As far as the amperage capacity of conductors and charging a battery pack, I have found that before you reach the physical amperage limit (excessive heating) of the conductors and connectors, you reach a limit where the voltage drop between the charger and the pack is so large that the pack doesn't charge optimally - the voltage at the cells can still be a bit low when he voltage at the charger reaches the switching point. This results in the charger prematurely switching to CV mode, and longer charging time.

So, as far as the BMS - is the PC board trace between "pack negative", the FET and the charger adequate, or can it be beefed up, to carry larger charging current without much voltage drop - say 300 mv? At least it looks like the positive side of the charging circuit doesn't have to through the PC board.
 
The trace on the board is very short and goes right to the FET. The wire connection points are very beefy. Most of the resistance will be in the wire.
 
We're using a 4110 as the FET, and it has a very low "on" resistance. I've tested up to about 10A, so far, and the FET never got above 95F, and that was with an ambient temp of 89.5A.

-- Gary
 
PJD said:
So, when will the kits be available? And Price?
Click to expand...

I'm pretty sure by the end of this coming week. I sent Richard a new board yesterday, and he's going to test a fully builtup board with his new 16s5p a123 pack. He also has changed the value of one of the resistors that improves the current regulation.

-- Gary
 
Gary or Richard or anyone,

This is actually something to add to "EV Basics" but I've never found a clear definition of the pack nomenclature you use here. By "16s5p" do you mean five strings of 16 cells connected in parallel, or parallel-connected groups of five cells connected in a series of 16? In this case, I presume the latter, but there a definite ambiguity in a lot of cases.

Paul D.
 
The nomenclature, which is commonly used in the RC world, can really apply to either configuration, but I always put the cells in parallel blocks first, and then the blocks in series. I have some a123-based packs that have some "stressed" cells that aren't quite as strong as the healthy ones, so I like to include a weak one with multiple strong ones, so that the stronger ones can "help" the weak one. I've found that under load, there's less voltage sag this way, than running multiple series strings in parallel.

-- Gary
 
The nomenclature, which is commonly used in the RC world, can really apply to either configuration...
Click to expand...

So, I don't want to belabor the point, but If that is the case, there is a fairly significant ambiguity unless there is some context. I propose calling Richard's pack "5p16s". While "16s5p" would be five parallel connected groups of 16 in series - admittedly not a practical setup in this case...
 
I'm not sure if there is a standard for that nomenclature, but the actual wiring configuration is usually determined by the battery chemistry. For lithium and lead-acid, you generally want to parallel the cell groups, then put the groups in series. For Nimh or NiCd, you have to make series strings, then put the strings in parallel in such a way that you can isolate them for charging.
 
Looks like this gentleman could use a few of your most excellent BMS:
8)
 
PJD said:
So, when will the kits be available? And Price?
Click to expand...

Question: What is the price estimate to build a 48 volt 9-10ah pack w/charging & LVC protection kit?

1) How many DeWalt 36v packs will be needed to tear down? (Any extra cells left over for replacement-repair?)

2) Cost *estimate range* (if final price not known yet) of TPpacks/Gary complete kit(s) to finish final assembly?

3) What other items are needed to purchase?
 
I have just become the excited owner of 40 x a123 cells for use on my Cyclone-powered e-bike and came across this thread while searching for a suitable charging system for my proposed 8s5p pack.
Bike-helmets off to Fechter and GGoodrum for this design :D

It is probably too late, but maybe for the next pcb layout, you may want to consider changing the board layout slightly to make it easy for people to physically cut down the board to make a circuit for smaller systems. Maybe layout the board to allow cutting down to 4, 8, 12 and then the full board for 16 series packs. This would probably be cheaper than getting different board sizes made, but would allow reducing the size of the PCB for smaller packs. You could then just sell kits with the appropriate number of components for the size ordered.

Do you have a ballpark estimate of the cost of this kit yet?

Keep up the excellent work.
JB_EB
 
I just finished putting together one of the new boards, with a full 16-channels. Here's what it looks like:

16-Cell%20LiFePO4%20BMS-v1.4c-02.jpg


It took me about 3 hours, but I was taking pictures as I went, for the instructions, which I will finsh up tomorrow. Next, I will run some "final" tests, to verify that everything works right, and to get a baseline on the heat. This version has the big 6.8 ohm/5W shunt resistors.

I haven't added up al the costs yet, but I have a spreadsheet with all the part costs that I need to update. If I had to guess right now, I'd say it will be between $80-$90, including a matching connection plug/harness.

-- Gary
 
Yes, that looks very professional. I like it!
(Starts looking for a suitable box with a clear lid so all the LEDs can be seen.)

So just to confirm, this board takes care of:
- Charging and equalising of the individual cells in the pack.
- And provides a Low Voltage Cutoff signal that can be connected to the e-brake switch on the controller to prevent over discharge?

In addition, all that is needed is a power supply that is a few volts greater than 3.65v x (Number of cells in series) and with sufficient Current to charge the pack in your required time, plus some high-current connectors. Is that correct?

JohnB
 
You must log in or register to reply here.

Similar threads

amberwolf
The HI-Lebowski: a Lebowski SMD brain running a zombified Honda IMA Inverter: *a HOW-TO guide*
2 3
Replies
64
Views
15,358
amberwolf
amberwolf
D
Open Source 18650 Lithium Battery Characterization System
Replies
6
Views
2,773
justin_le
justin_le
S
DeWalt 20V Max 3.0Ah Battery Pack Teardown & Analysis
Replies
10
Views
24,371
qwertyES
Q
Teslafly
[DEAD PROJECT] OSBMS - A scalable, low cost, and full featured BMS
2 3
Replies
58
Views
29,246
Vano
V
battman
Barron Greig's MAX14920 Based Perfboard BMS
Replies
12
Views
4,485
Madhi2023
M
Back
Top