regen into a v28 battery pack?

[tno]

I hooked up a milwaukee v28 battery pack to a xootr ex3 and was wondering if it's ok to use regen braking with the v28 battery. I'm assuming the bms would limit the volts to 4.2v per cell.

the single v28 pack gets warm with the ex3 at full throttle so a another v28 in parallel would work better.

 

[Ypedal]

I would avoid it at the start of a ride....

But if you have a DVM hooked up to monitor peak volts allowed *( the scooter may be set for SLA or NIMH ??? )

 

[tno]

the ex3 is set up with the stock nimh battery on 1 side and a cable coming out of the other side that plugs into the v28 battery pack.

I've been using regen braking with the v28 connected and haven't seen any problems. But the thought occurred to me that the motor might supply too much volts to the battery. So I'm hoping the milwaukee bms would take care of that.

I'll try to hook up an astro whatt meter and see what the regen volts are. hope the whatt meter will show energy going in reverse.

 

[fechter]

Cool, Not too many EX3's still alive.

I'm not exactly sure how the charging BMS works, but I suspect if the voltage goes too high, you will suddenly lose regen as the BMS opens the circuit. With no batteries in the circuit, the regen voltage might go sky-high and fry something. As long as the batteries have been discharged a bit before using regen, it shouldn't be a problem.

Like Ypedal says, avoid regen at the beginning of a ride. My Zappy had the same problem, sometimes I parked at the top of a hill and started out with a full charge. Going downhill, it would over volt pretty fast. If I went uphill first, even just 1/2 a block, the voltage would never get too high.

Ideally, there would be some kind of dump load to handle regen current when the batteries couldn't take any more. There's not much room on a EX3 for additional circuitry.

 

[tno]

i'll try to monitor the voltage as it regens. by "dump load" do you mean connecting another battery in parallel to the v28

here's a picture of the setup. I have it mounted that way temporarily to test out the v28. the v28 is held in place by velco straps.
the ex3 can fit 13 emoli cells on the left battery compartment, 12 on the right, and 3 just below the left battery compartment. and there's still a little space left for a bms. so that's 4 v28 packs in parallel. I'll wait until emoli's are a bit cheaper before I try this setup. emoli's are 2.5mm taller than sub c cells so a 2.5mm spacer needs to go between the bottom pan and the frame (a rubber gasket will probably work).

 

[tno]

would a capacitor act as a dump load if the bms were to go open during regen?

if so, how many farads should the capacitor be?
the battery is 28v 3ah. the controller is up to 30v 35amp.

 

[jondoh]

Hi Tno,

I was wondering if you can take a close-up picture of the connections to the battery. Are you using a connector block? If not, how are you connecting wires to the battery?

 

[tno]

I use the standard blade connectors to connect to the v28 battery. you can see the female end on the charger below the battery.

it would be better to get the connection block and mount that on the ex3. then just snap the battery pack in and out quickly.

 

[fechter]

would a capacitor act as a dump load if the bms were to go open during regen?

if so, how many farads should the capacitor be?
the battery is 28v 3ah. the controller is up to 30v 35amp.

A capacitor would hold about 3 seconds worth of regen.
A bigger battery would be able to absorb much more energy before getting fully charged, but if you started out with a full charge and went down a hill, there would be nowhere for it to go.

I was thinking more like a giant resistor. I'm not sure how much regen current one of those generates, but I'd guess at least 250 watts worth.
Ever seen a 250W resistor? Anyway, a standard resistor would be impractical, but something like a bunch of wire wrapped around the edge of the deck or maybe some ni-chrome wire from a toaster or hair dryer mounted carefully might work.

That was a weakness in the orginal design. You can avoid the need for this if you don't have any hills around or are otherwise able to avoid an overcharge condition. If you start at the bottom of a hill and go up, then back down, you will never get enough regen to overcharge. Typical scooter regen systems might give you back 25 - 30% of the energy.

In order for the resistor to work, you'd need some kind of voltage detector and a big FET to switch in the resistor. If the voltage detector is set to trip just at upper voltage limit for the batteries, the BMS shouldn't have to do anything.

 

[tno]

that sounds too complicated. I think I'll just keep an eye on the volts and avoid using regen when the battery is nearly full.

 

[fechter]

Yes, that's probably why they didn't do it.

A somewhat simpler approach would be to have a voltage sensor that disables the regen when the voltage gets too high. You might suddenly lose regen on a long downhill, but that would be better than suddenly blowing up the controller.

 

[tno]

hooked up the astro whatt meter today. it regens at battery voltage and the arrow in the center of the meter switches direction to indicate a reversal. so it looks like it works fine as long as the battery has room. I hope the emoli's work better with regen than the saft nimh.

 

[xyster]

I hope the emoli's work better with regen than the saft nimh.

I haven't seen any data regarding lithium-manganese overcharge tolerance. I understand the a123 lithium iron phosphate cells are very tolerant. What happened with your saft's?

 

[tno]

The saft sub c's got really hot when running so I was hesitant to use regen with them. The emoli's are suppose to be able to recharge in 15min so I figured they would be more tolerant to a long downhill charge.
I chose the emoli's over the a123's because the v28's max volts is 29.4v which is about the same as the 21nimh cells in the ex3. that way I could use the v28 bms and charger.

 

[xyster]

I chose the emoli's over the a123's because the v28's max volts is 29.4v which is about the same as the 21nimh cells in the ex3. that way I could use the v28 bms and charger.

I haven't tried them yet, but for general small ev power, based on the specs the emoli's look better than the a123s. Compared to the a123's, the emoli's trade C rating (15C v. 30C) for added capacity (2.9ah versus 2.2ah), and charge to the normal 4.2v from any lithium charger. 15C is still plenty since even one cell is good for 45 amps, and most EV people will use multiple in parallel. The a123's sound like the better choice for the smaller RC devices that need to pull 60 amps for 5-10 minutes from a single string.