Leave charger connected to BMS? Diodes?

[NeilP]

Just wondering about fitting a friends little chinese bike with the charger permanently connected to the BMS. Then bike can just be charged where ever she is just by finding a kettle/PC power lead. They are everywhere generally

I just repacked the 3 x 10 Ah SLA box with 14 x Headway 10Ah cells and a BMS unit.

Charge connects via the BMS. usual setup, positive feed always to charger, negative goes through the BMS.

OK, so there may be issues with charger vibration damage..but she has been carrying the old SLA charger around in the front basket of the bike for the last year with no issues. I plan to mount charger securely, with maybe a bit of damping of some sort.

I guess diodes will need to be fitted. Or do you reckon there will be some discharge prevention built in to either charger or BMS any way?
If need diodes, just on negative ? or on positive too?

What sort of diode is best for this application? some sort of power diode with low voltage drop I guess...any suggestions.

I suppose final stage would be to turn up the charger voltage a bit to take the diode voltage drop in to account.

Have I missed anything ?

 

[dnmun]

why do you need diodes? who told you that?

if you leave the battery connected to the charger then the battery will leak down through the drain down resistor across the output caps on the charger.

you have to remove this drain down resistor if you leave it wired up permanently or you have to put a switch on the charger output.

 

[NeilP]

I could see that happening with charger direct to battety, but since this is via a BMS, I wondered if the BMS would prevent this. suppose this is dependant on BMS design.
I'll have to put an ammeter in series and see.

Guessing a diode or two would do the same job

 

[dnmun]

but why would anyone deliberately put a diode in the main current path?

the only time you do that is on the output of the charger transformer. once the current is in the back end you should not be using more diodes.

if you don't know how to remove the drain down resistor you should use a switch. or just leave it disconnected. the BMS has nothing to do with it.

 

[NeilP]

but why would anyone deliberately put a diode in the main current path?

To block discharge back through the charger. just to be clear I am referring to the charge current path, not the main battery to controller current path

if you don't know how to remove the drain down resistor

Could well do that, had not considered that option. I'll pop the charger open and find it.

you should use a switch. or just leave it disconnected

This is for a dippy woman that gets confused with things like switches and plugs, and seems to get confused even plugging the 'IEC 'kettle' plug in to the charger. So want to make it as simple as possible so that she does not even have to look at the charger. I want it so she can get off the bike, unwind a cable and connect it straight in to the wall socket, so she can't do any damage.

She has only had it a day, and I left her to unplug the pair of Anderson's, that connect the charger to the battery pack charge point. I had super glued the red and black connector together, filled the back end of the Andersons with hot melt glues, and heat shrunk the lot together. Did this at both ends (pack end and charger end) . Showed her how to plug and un plug them. I go there this morning, and she had managed to pull just the wire and pulled the terminals out of the Anderson shells. Lucky it was the charger end and she had powered it off first . if it had been the battery end, the two ends would have shorted..Bye bye BMS i am sure

the BMS has nothing to do with it.

Why not ?..charge current goes through the BMS in to the battery pack FROM the charger. so if the charger is going to drain down the pack, it has to do it via the BMS also.

 

[dnmun]

there is a 5k resistor across the output caps on the charger. you only have to cut one end of the resistor and not even remove it from the pcb. i remove them.

 

[NeilP]

I am back there tomorrow to change the Andersons to something more secure , so I'll do that. Want to change the OEM IEC plugs too. They have done the usual of using them for the battery power, too great a risk of the owner accidently shoving 240 volt straight into the battery

 

[DAND214]

This is not related to the charger setup.
But why 14s haadways not 16 or 12?
Just wondering.
Seems an odd numbr for LiFEPo.

Dan

 

[NeilP]

Because the OEM controller on the bike has a HVC above 50 volts.
As soon as voltage gets above 50.5 or so, 5v line goes to zero. It won't run above that voltage.

.
It was originally 36 v SLA bike.

I had a pile of 18 Headway cells free after battery fire on someone elses bike. I got the good leftovers in exchange for rebuilding his pack

14 cells would fit in the original SLA box.
14 gave a little performance and capacity boost over a lesser number.
Paul ( Cell man) had a 16s LiFePO4 BMS that would work with 14 cells
I could have made it 16 but that would have meant building a new battery box and buying and fitting new controller.



Thompson Duo, a chinese scooter lookalike import
It was bought new, unused for 4 or 5 years, sat in shed with many many others. Import deal that went bad. A bloke bought a couple at auction £70 for the pair, brought them to me to fix. I told him the price just for New SLA's not even my labour. He said that even SLA price was too much, he was not going to pay for Ping battery either that was for sure.
So I got one working, and chopped the othef cor the wheels for myself for BMX project..

While the good ( battery less ) one was sat in my shed, a work friend came by. She said she had similar one but wanted this one as well. As I knew I had these spare Headway cells, I said I'd build them into a pack for her to save her cost of Ping pack. If it works, great, If not nothing lost, apart from price of the BMS.

 

[NeilP]

Here they are

http://endless-sphere.com/forums/viewtopic.php?f=9&t=47031&hilit=+Thompson


First pic is one of the pair.
Second style is my work friends original


Then there were these

http://endless-sphere.com/forums/viewtopic.php?f=3&t=25399&hilit=+Is+this+bike


All ( plus other designs) got imported to the UK under the 'Thopmson 'brand

A local Jersey businessman then brought in a load of the from UK to our island. They never sold/ were not popular. So everynow and then I get a dead one to fix.

 

[dnmun]

i built a 14S pack out of what was left after it got too expensive to replace the pouches in a 10Ah pack.

i still had 14 with good tabs so i hacked the BMS to 14S and adjusted the ping 36V3A charger brick to put out 51.2V at 3.7A so i could charge the 14S.

it works good and i am thinking of making a hybrid 16S/14S/12S assembly of pouch packs connected in parallel through diodes so they will work in succession with the higher count shutting down for LVC as the voltage drops. with the 12S being the last to discharge. just an idea now.

 

[NeilP]

Cell man has some older BMS's that do not need hacking for lower cell counts. It is built to work with lower counts as it comes from him.