Bulk charging A123 LifePo4 and Konion Li-Ion packs

[jerrysimon]

Yep the first experience of a self build or any ebike certainly brings a smile to the face

Re voltage of the A123s they tend to suddenly drop off and they pretty much have the same punch to them right to the end. Just checked mine and they read 39V ish having just taken 1.4Ah out of them on the way home. As I said I always charge them at work before returning home to ensure that there is plenty of capacity left. Like I said I didn't for the first year emptying them almost completely daily and they still lasted about a year before I lost a couple of cells.

Regards

Jerry

 

[Joneser]

PS and the batteries just took less than an hour to recharge - Jerry how long do they normally take to charge when fully run down?

 

[jerrysimon]

Most chargers I use charge at around 1.8-1.9Ah so about 1 and 1/4 hours from flat. If you connect your watt meter on the ride and again when charging you will see how many Ah taken out and put in

Jerry

 

[Joneser]

Most chargers I use charge at around 1.8-1.9Ah so about 1 and 1/4 hours from flat. If you connect your watt meter on the ride and again when charging you will see how many Ah taken out and put in

Jerry

I just connected the Watt meter for the first time - it shows about 23 Watts when spinning freely off the ground and goes up to over 500W applying the brakes.

Its quite interesting all the info it gives: Watts, Watt hours, Amps, Amp hours, Volts plus Vm, Wp & Ap (what are these?). A lot of info on a little screen.

Is it silly (in terms of cable efficiency loss) to have the battery and controller in a saddlebag and then run power cables to the handle bars and back so I can display the Watt meter or is the loss insignificant? Or do I just rig this up temporarily to get a feel of performance and then disconnect it?

Sorry, thinking out loud but and suggestions / confirmation of my assumptions would be appreciated.

 

[Deleted member 4366]

Good question about the wattmeter wires. If you use thick wire, losses are low, but it looks ugly. If you use thinner wire, it looks better, but losses go right up. You should keep them as short as possible, and you only need a thin wire for the positive (red) one because all the measurement is done on the negative side. You can therefore run the positive directly from the battery to the controller, and then splice on a thin wire to run up to the wattmeter positive on the input side. You don't need any wire on the wattmeter output positive. You then run a single thick wire from the battery negative to the wattmeter negative input, and back out the negative output to the controller. That way you only have two thick wires up to the handlebars instead of four, which is not so ugly and you have half the losses.

If that's not enough, you can remove the shunt from the meter, and run three thin wires from the shunt, located near the battery, to the wattmeter on the handlebars.

Which wattmeter did you get?

 

[averhamdave]

Vm = max voltage seen
Ap = amps peak
Wp = Watts peak

In my experience the novelty of a watt meter soon wears off. I'd leave it in the saddle bag. After a dozen or so outings you'll understand the characteristics of the bike/battery/motor and won't bother with again - or won't until you have some issues to resolve.

 

[Joneser]

Which wattmeter did you get?

I think the one you recommended. Its a nice looking blue painted aluminium box. Amazing you can buy those things for a few quid.

 

[Deleted member 4366]

That's even easier then because the shunt consists of three 0.003 ohm current sensing resistors soldered on top of each other like this one to makea single .001 ohm one:
BOURNS, CRF2512-FX-R003ELF, RESISTOR, POWER, 2512, R003, 1 | eBay

You can therefore make a separate shunt and leave the original in place. You take a duplicate pair of battery to controller connectors; run the red straight from one to the other; solder three of those resistors stacked on top of each other between the black wires; solder on three thin wires to run up to the meter, one on the red wire and one on either side of the shunt; open up the box and solder the other ends on their equivalent positions on the pcb. QED

 

[Joneser]

That's even easier then because the shunt consists of three 0.003 ohm current sensing resistors soldered on top of each other like this one to makea single .001 ohm one:
BOURNS, CRF2512-FX-R003ELF, RESISTOR, POWER, 2512, R003, 1 | eBay

You can therefore make a separate shunt and leave the original in place. You take a duplicate pair of battery to controller connectors; run the red straight from one to the other; solder three of those resistors stacked on top of each other between the black wires; solder on three thin wires to run up to the meter, one on the red wire and one on either side of the shunt; open up the box and solder the other ends on their equivalent positions on the pcb. QED

I have just been reading about new evidence the universe is a hologram so brain a little weary. I think you are saying the power level is reduced in the watt meter before it does its electrical analysis so you can reduce the power closer to source to reduce cable size and efficiency loss. Would be neat to get a thin 3 core black cable. I have a bag of random old electrical cable. Maybe I'll be pleased I didn't throw it all away.

 

[Deleted member 4366]

The wattmeter calculates the current by measuring the voltage drop between each side of the resistors/shunt. In the one you have, the resistance is .001 ohms, so, with 10 amps flowing, the voltage drop from Ohm's law (V=IR) is 10 x .001 = 10mv. You can make that voltage drop over the shunt anywhere that you find convenient, and then take the results to the wattmeter with thin wires.

You don't even need a shunt because you get voltage drop down any wire because all wires have resistance. You could check the maximum current that you get with your wattmeter as a reference point. Then, connect two wires to your negative battery wire about 100mm apart (assuming that you used 14g wire), and one additional wire to the positive. You can use thin wire. Run them up to your Wattmeter and connect the two nearest the battery to the normal input, and the other one to the sense track that runs from the internal shunt. You'd have to cut the track so that it's no longer connected to the shunt. If you open the case, you'll see how it works. Test the maximum current again to see the result. If it's 10% too high, you shorten the distance between the two wires by 10%, or increase it in proportion if it shows the current too low. That way, you don't get any losses, and you only need a thin three-core wire running up to your handlebars.

I made a little spreadsheet that shows how to measure current using any wire. I posted it here:
Endless-sphere.com • View topic - Current Measurement

 

[Joneser]

The wattmeter calculates the current by measuring the voltage drop between each side of the resistors/shunt. In the one you have, the resistance is .001 ohms, so, with 10 amps flowing, the voltage drop from Ohm's law (V=IR) is 10 x .001 = 10mv. You can make that voltage drop over the shunt anywhere that you find convenient, and then take the results to the wattmeter with thin wires.

You don't even need a shunt because you get voltage drop down any wire because all wires have resistance. You could check the maximum current that you get with your wattmeter as a reference point. Then, connect two wires to your negative battery wire about 100mm apart (assuming that you used 14g wire), and one additional wire to the positive. You can use thin wire. Run them up to your Wattmeter and connect the two nearest the battery to the normal input, and the other one to the sense track that runs from the internal shunt. You'd have to cut the track so that it's no longer connected to the shunt. If you open the case, you'll see how it works. Test the maximum current again to see the result. If it's 10% too high, you shorten the distance between the two wires by 10%, or increase it in proportion if it shows the current too low. That way, you don't get any losses, and you only need a thin three-core wire running up to your handlebars.

I made a little spreadsheet that shows how to measure current using any wire. I posted it here:
Endless-sphere.com • View topic - Current Measurement

Awesome Dave. I'll have to try and absorb that another time - having had a few glasses of wine I've just added the rear brake sensor only to find there is no connection onto the KU63. grrhshffs So thats another reason to get the SO6P rigged up (which does have brake sensor connections). I seem to have 2 types of PAS V12 and V7. The number referring to the number of sensor points. I'l try V12 for the SO6P I suppose.

 

[Deleted member 4366]

You can still use the brake sensor with the KU63. You need a 5v supply for it, so splice a wire onto the red wire in the throttle or PAS connectors. The brake sensor has three wires: red 5v, blue (signal) and yellow (0v). The brake connector on the KU63 has two wires: black (0v) and yellow (signal).

So you connect red to red, sensor yellow to controller black and sensor blue to controller yellow, and then it will work.

 

[Joneser]

You can still use the brake sensor with the KU63. You need a 5v supply for it, so splice a wire onto the red wire in the throttle or PAS connectors. The brake sensor has three wires: red 5v, blue (signal) and yellow (0v). The brake connector on the KU63 has two wires: black (0v) and yellow (signal).

So you connect red to red, sensor yellow to controller black and sensor blue to controller yellow, and then it will work.

Oh dear. I've solder connected the rear brake sensor and also soldered throttle control on (as I had been using thumb control). The brake sensor light comes on and initially went off when applying the brake, although now it is not. The main point is the motor us not working with the throttle. Stupid because I should have tested the brake sensor with the thumb control so I knew what was happening. Now I've changed 2 things at once I don't know what to do.

Update. 1st thing i decided to do is disconnect the brake sensor red wires. Bingo the throttle works so I know I've done something wrong with the brake sensor. I'm pretty sure I've done what dave suggested. Will post a photo later.

 

[Joneser]

here it is.

 

[Deleted member 4366]

The brake sensor is the wrong way round on the cable. The LED should e off and goes on when you apply the eake. If you have it at the lever end of the cable, put it at the caliper end, or vice versa.

 

[Joneser]

The brake sensor is the wrong way round on the cable. The LED should e off and goes on when you apply the eake. If you have it at the lever end of the cable, put it at the caliper end, or vice versa.

Thanks dave (he says slightly embarrassed)
Switching it was easier said than done as the brake cable started to unravel and its a miracle it eventually fed through, albeit with several strands missing. Hopefully the cable is still man enough.
But it works. Thanks again.

 

[Joneser]

Most chargers I use charge at around 1.8-1.9Ah so about 1 and 1/4 hours from flat. If you connect your watt meter on the ride and again when charging you will see how many Ah taken out and put in

Jerry

Morning!

Connected my watt meter for the commute for the first time this morning.

Christmas party so stonking hangover so may not have peddled as hard up the hills as normal.

The commute in is the more down hill stretch so was particularly surprised when the motor stopped working after 8 miles.

Checked the watt meter which reported 1.5Ah but only 29V!

What does this mean? I'm sure they were (nearly) fully charged.

They have never showed signs of running out even on the way home (more uphill) previously.

 

[Deleted member 4366]

I think it means that you need a bigger battery!

Did you balance the cells before setting off, and what are the individual cell voltages now? It's very important to check them before charging. You might find that one cell has gone right down to near zero, while the others are above 3v. Which battery was it?

Youhave to think seriously about the benefits of using such low capacity batteries. When they're working properly, the voltage will decrease as you use them, so that your average voltage will be much lower than witha larger capacity pack. Power is directly related to voltage, so you're giving up as much as 20% of the power you'd have with a bigger battery.

As an example using a 10aH bottle battery:
Start voltage: 42v
Max current: 12 amps
Charge consumed: 2aH
Finish voltage: 38.5v
Average voltage: 40.25v
Average power available: 483W

For a 2.3 aH LiFePO4 battery
Start voltage: 42v (drops rapidly to 39.6v)
Max current: 12 amps
Charge consumed: 2aH
Finish voltage: 33v
Average voltage: 36.3v
Average power available: 435W, which is 10% lower

 

[Joneser]

I think it means that you need a bigger battery!

Did you balance the cells before setting off, and what are the individual cell voltages now? It's very important to check them before charging. You might find that one cell has gone right down to near zero, while the others are above 3v. Which battery was it?

Youhave to think seriously about the benefits of using such low capacity batteries. When they're working properly, the voltage will decrease as you use them, so that your average voltage will be much lower than witha larger capacity pack. Power is directly related to voltage, so you're giving up as much as 20% of the power you'd have with a bigger battery.

They are the 12C A123's. No I didn't balance them for this journey (but did for the one before that) - every time I have balanced them so far they are been virtually balanced already.

Until today they have been perfect for my commute so I'm not giving up on them yet Monitoring to continue...

 

[jerrysimon]

Looks like they were not fully charged before leaving. I regularly pull 1.5Ah out of mine (that leaves 0.9Ah for the 2.4Ah pack) and they still read around 2.8V so as I said they were probably not fully charged.

29/12 = 2.4V which is well within their operating range anyway. According to specs though they are pretty much empty at 2.5V. Probably best not to go below 2V.

Jerry