E+ powered by A123 20 Ah pack UPDATE

[BVH]

As I look back thru my posts, I realize that I have not given my intention to continue to use the NMH battery while making the LiMn mod. I saw an earlier post indicating that I have already tried to power up with only the "B" battery connected without success. I repeated that test today. I cannot power up on the "B" battery only. But this does not matter "right now" concerning the LiMn mod.

I am mystified.............Friday, I fully charged both batteries and rode for 22 miles. I used 10% of "A" and finished by using 50% of "B". Without re-charging either battery, today, I went on another 23.5 mile ride and used 60% more of "A" and 0% of "B". So I've gone 45.5 miles and have 30% of "A" and 50% of "B" remaining. In all my 4+ years of riding this bike, I don't remember ever going more than 37 miles before depleting both batteries. Here I am with a combined 40% remaining capacity and already traveled 8 miles more than I ever have. I ride the same on every ride and always have so my style has not changed. I pedal the entire time to maintain a low burn in my legs. I have 3 rides I choose between so that has also been constant. I add power to help with long uphills.

Here's the other thing I just noticed today. Again, after 4+ years of riding this bike, I just noticed that when I apply either or both brakes, the system automatically goes into Regen mode. I always ride with the display showing power (PWR). Imagine my surprise when applying a brake on a long downhill, I saw positive numbers in the display up to about 400 Watts, tapering down to zero as speed decreased and also saw the lightening bolt on the end of the fuel indicator battery graphic. Let off the brake and the numbers go to zero and the lightening bolt disappears. The Regen automatically stops at about 3.5 MPH without letting up on the brake handle. Up to this point, if I wanted to regen, I thought I had to cycle the power down control button all the way down to zero and then into negative numbers. That's how I've always done it. It was a pain in the butt so I rarely used it.

Please, someone tell me if you are also automatically going into Regen mode with brake application? I am pleasantly mystified with both the big increase in range and noticing auto-Regen. I can't believe I have never seen it before.

 

[BVH]

I'm still going on this battery charge. I rode another 8 miles on the "A" battery before it went into LVC at 53 total miles. So I got 38 miles out of it under my normal riding conditions - which is pretty much full-time pedaling with between 125 and 175 Watts on the flats and up to 600 Watts up steep hills. I still have about 45% of battery "B" left so I think I will achieve about 65 miles on the full 2-battery charge. When re-charging battery "A", I inserted a Watt's-Up battery meter between the charger plug and the front hub battery plug so I could see how many Amp hours were put back into it. After the balance phase completed, the Watt's-up displayed 9.42 Ah. I know there are some small losses in heat so the full 9.42 is not the exact number but it's close. So after 4+ years and maybe 100 - 125 charge cycles, I've got better than 90% capacity left. I would guess that the EMS programmed LVC is conservative to extend battery life and if that is the case, I may have close to new capacity left.

EDIT: Well I thought the charge was done but I guess not. It stuffed 9.985 Ah back in. It took over 7.5 hours to charge. 30 minutes after LVC on today's ride, the pack Voltage was 37.25 or 1.24 Volts Per Cell. I thought they might have LVC set for 80% depth of discharge but I'd say it's 95 to 100%. Might be good to routinely ride down to one solid bar instead of LVC.

Miro, if you want me to put these E+ data posts in my own thread, let me know since it's not really on-topic.

 

[miro13car]

BVH
I was away.
Nothing wrong with you posting into this thread ABOUT YOUR TRIPS, it is still about EPLUS .
There are few of us owners.
Your numbers are quite impressive but for those distances you were in low power levels like 3, 4 or even 2?
Beauty of EPLUS that it has those 9 power levels which are not gadgets but real well implemented.
So you are not going ahead with your conversion right now?
The logic tells me that if NMH hub is involved in powering up you must have NMH connected to power your EPLUS on battery B.
So there is sequence.
My conversion gets rid of NMH cells entirely.

 

[BVH]

I always keep the power level at 9 - just in case I have to hit the throttle hard to avoid a traffic situation. But I am easy on the throttle. Correct, I am not going forth with the conversion at the moment because it seems I still have very good capacity left in both batteries. From what I remember, almost the same capacity as when I bought the bike new. I know the electronics in the front hub are absolutely necessary to be powered and I think, show usage and I don't know how I will deal with this when the time comes.

 

[miro13car]

I must correct myself,
you need not all NMH hub but just NMH electronics powered up to power EPLUS on battery B.
It doesn't matter how do you supply 5V to them.
You can do experiment and disconnect front hub while your EPLUS is powered on battery B , what would happen?

So you are delicate on throttle on your trips, I know exactely what you mean - because EPLUS throttle is so wonderfully linear.
On many cheap China edrives/ebikes throttle is annoyingly twitchy, non linear when pressing even 1/2 inch gives no effect /on thumb throttle/ and another 1/4 of inch press surges the bike ahead. This contribute to wastage of Wattshours of energy!
To appreciate EPLUS throttle implementation you MUST try other ebikes.

 

[BVH]

OK, I must have forgot that you said that only the electronics must see a 5 Volt signal. That will make it a whole lot easier to accomplish. I was thinking that battery A circuits had to show charge and discharge counts also. IIRC, you moved the controller circuit boards inside your new LiFeP04 enclosure and are feeding them the 5V. With the OEM wiring, could I leave the boards in the front hub and feed them the 5V signal through the original pos. and neg. battery wires? Where exactly, does the 5V signal originate/where is it tapped from?

 

[miro13car]

Sorry
I was away.
To start with,
this is how I understand it....
when EPLUS is powered by NMH battery /battery A displayed/ SOC bars shows state of charge of NMH battery.
when EPLUS is powered by B battery /battery B displayed/ SOC bars on display show state of charge of B battery. It is obvious.
NMH hub electronics are powered on 5V all the time even after EPLUS sits there. NMH battery is connected to NMH electronics all the time.
36V is fed to the motor through discharge MOSFETs /think about them as switches/ which are turned on on the command from NMH electronics.
Even when EPLUS runs on batteryB NMH electronics are active /powered by 5V.
Where is 5V from?
36V is as I understand stepped down to 5V by at least one DC/DC converter which is integrated part of NMH electronics, but I suspect it is not so simple, there is another DC/DC also on the board I suspect. From my experiments with NMH string disconnected - connecting plus 5V and ground to PIC 18 chip /according to data sheet/ does not distribute this 5V to other chips on board, so other chips beside PIC 18 must have seperate source of 5Volts.
OEM power wires pos./neg. carry 36V to the motor. When you remove NMH cells and connect 5V to those power wires it will be stepped down to nothing. So you cannot use hub power wires to provide 5V, you have to go inside hub.
CONCLUSION in any kind of EPLUS Lithium conversion you must dissect front NMH hub.
Even if you find points at which 5V is provided to NMH electronics you still have to carry/charge 5V source.
I moved NMH electronics intomy Lithium but DO NOT feed them any 5V seperately. My Lithium is simply connected exactely where NMH was connected just like factory.

 

[BVH]

OK, everything you said I understand. Thanks for making it simple so I can understand what is needed.

Even though you have not had a Lithium battery "B", do you know how the 36 Volts from Battery "B" get specifically routed to the motor? Are the battery cables for both Battery "A" and "B" connected together somewhere? There is no path for the "B" power to get into the front hub so how is it modulated and how does it get from the battery to the motor?

 

[BVH]

Finally used up the rest of Battery "B" at 66.2 miles so that means 28.2 I got miles out of the "B" pack and a total range of 66.2 miles on both packs down to LVC. I just got thru putting 8.9 Amps back into the 10 ah Lipo. Starting/resting Voltage was 3.67 so just about at 80% dept of discharge. The 8.9 Ah seems a little too much to put back in but I think I know why. And this is a little scary. The charger was steady at a 2.95 Amp rate up to about 7.5 Ah. It then began to back off indicating it had changed to Constant Voltage mode. However, as Amps went down, Voltage continued to climb, which should not happen in a true CCCV charger. And get this, it climbed and terminated at 42.85 pack Voltage or 4.28 Volts Per Cell. This is not good. And "termination" didn't really occur. After the LED indicator on the battery itself said full and the indicator on the charger turned green and the state-of-charge meter on the controller stopped climbing and stabilized, the charger continued to pump in 240 mA and Voltage continued to climb. I am thinking that if the E+ bike was responsible for the bike shop fire, it was probably the charger and not the battery itself. Trickle charging is a definite No-No for Lithium. I have to assume that my charging system has been doing this from day one and the battery is 4 years old and still has nearly full capacity. I really don't know what to think. I measured the Voltage with a near-new Fluke meter that is within 3 millivolts of a 5-Volt reference source. The only unknown is the specialized BMS with communication on-board. I can't think of why that would change termination Voltage but who knows.

 

[BVH]

I opened up the battery case again and checked actual Voltage at the battery main output leads. It's 42.32 (4.232 per cell) when the charger LED indicator turns steady green. (Charger output Voltage is 42.92 so there's a drop of .55V through the board) Voltage continues to increase and current is at 240 mAh at the same location and after 10 minutes past LED steady green, actual battery Voltage had risen to 42.37. I then unplugged the charger and after 15 minutes resting, Voltage was at 42.35. While the data is not as drastic as I first thought, it still seems very high to me and more worrying is that it does not shut down completely at least 15 minutes after all indicators show that charging is done.

Since charging Voltage 42.92 resulted in a rested pack Voltage of 42.35, I might just manually terminate the charge at 42.57 which should yield a rested pack Voltage of 42.0 or 4.2 VPC if my math and theory are valid.

Were there Lipo pouch cells manufactured about 4 years ago that could handle a 4.3+ VPC charge?

I opened the charger and there are no trimPOTs or other adjustments.

 

[miro13car]

BVH
I don't know much about LiPolymer charging .
Electric Montion Systems foolish decision to use LIPolymer cells for battery B for sure contributed to bankrupcy of EMS.
On recall they offered EVERY owner of B battery new NMH wheel worth about $1000, imagine cost of it?
If charger was to blame for fire or B battery itself doesnt matter much to me. This battery chemistry is dangerous not the charger, however fault of charger might cause the fire.
Somebody reading this thread and NOT knowledable about EPLUS might form opinion how bad EPLUS was.
Again
quality of battery B has nothing to do with excellent desingn of EPLUS motor and controller .

 

[BVH]

Did you ever have occasion to determine what the 3rd wire is for in the NMH battery charger output cable? Is it to simply power the controller circuit and display?

I would like to figure out a way to use my RC charger to charge the Lipo. The OEM Lipo charger also has a 3-wire cable. In addition to the normal heavy gauge pos and neg wires coming out of it, there is a 3rd wire. It carries a 5 Volt signal and about 550 mA. I thought it might be for balancing current but the "5 Volt" signal seems to rule that out.

 

[miro13car]

Like on every serious NMH charger third wire is for thermistor connected between MINUS and that third wire.
NMH is not easy to charge comparing with LiFePo .
One of the factors to terminate NMH charging is temperature increase hence you need thermistor placed on one of NMH cells inside hub.
I was impressed by level of engineering inside NMH hub when I first opened it and first class assembly, pleasure to work with on my conversion.

 

[BVH]

Well my "B" battery finally died a passive death. One of the cells failed completely but the other 9 were still fine. I removed the PC board and played with the raw battery and noticed it charged and discharged completely normal - as if it was a 9-cell battery but Voltage would not climb beyond 37.9 (3.79 VPC) give or take after putting in about 8 Ah. At 80% charge, cells should have been 4.05 to 4.1 VPC. Now I just need to figure out where to get some new LiMn pouch cells to rebuild it.

 

[arkmundi]

Well my "B" battery finally died a passive death.

Was this "A123 20 Ah pack" and if so, how my cycles did you get out of it?

 

[BVH]

No, this was the OEM E+ "B" 10 Ah Lipo, rear rack battery. I'd estimate I got 200 cycles out of it since new in 09. I'd love to be able to use LiFeP04 but can't without going thru what Miro did to be able to use his. The PCboard is looking for Lipo Voltages.

 

[miro13car]

BVH
Thank you for coming back here.
I understand as I read many times , EPLUS OEM B battery was made of Lithium Polymer pouch cells 10S , ten cells in series for nominal 36V pack.
That is very pack which caught fire in one of the bike shops somewhere in USA, it was not mention in EMS /Electric Motion System/ recall notice which shop.
This event triggered bankruptcy of manufacturer - EMS.
I am thinking how you can reuse your LiPolymer BMS electronics,
I saw pictures of these electronics and like everything EMS - solid /to Western standards/ assembly, seems rather over engineered however we must realize that this BMS communicate/reports to master display/controller over RS485 comm line, for sure it reports low voltage on cells which info is used by motor controller to gradually lower power drawn from those cells.
You said you opened and played with BMS ? Are you sure they are still OK?

 

[miro13car]

I would NOT encourage you to rebuild your B pack with LiPolymer cells of course
UNLESS you take your pack for charging outside into firepit EVERYTIME. UNLESS you never leave pack on your EPLUS.
There is clear evidence out there that even if you follow all charging procedures with shoody quality LiPolymer cells /but chaep/ it still can catch a fire.
- manufacturing defects.
Forgive me but all those precautions with this chemistry cells would make my riding experience horrible
example I roll my EPLUS quite often into shops which would have made me really nervous with LIPolymer battery.
BTW
My converted to A123 cells battery 1000W EPLUS is approaching 8500 kilometers rather fast, plus before conversion I rode at very least 1000 km on it on OEM NMH wheel battery.
It runs like new.
It is no more one-trip-a-day ebike like before due to ridiculously long charging times of NMH , A123 can be charged at 20A-25A easy.

 

[miro13car]

BVH,
HERE is place which sells Nissan Leaf cells. They are LiMn types, in voltages close to LiPolymer.
BVH , can you post pictures of cells used in EMS B battery?

 

[BVH]

I bought 12 of those Nissan Leaf modules and was going to use them for powering my military searchlights (1.6KW) but the Voltage sag and curve soon yielded Voltages too low for normal operation. I could only get about 1/3 of the capacity out of them before the Voltage was below the lights specs. They were fairly heavy, too. I sold them.

http://www.candlepowerforums.com/vb/showthread.php?375296-Using-7-cells-of-an-8S-pack-Nissan-Leaf-LiMn-Battery-Modules-for-NightSun-Ops&p=4325017&viewfull=1#post4325017

This is what I built instead:

http://www.candlepowerforums.com/vb/showthread.php?366273-My-Holy-Grail-Has-Been-Found-Parts-of-One-Anyway-SX-16-NightSun&p=4341912&viewfull=1#post4341912


Batteryspace sells the exact E+ "B" battery cell here. Same part number right on the pouch.

http://www.batteryspace.com/high-power-polymer-li-ion-cell-3-7v-10-ah-9759156-5c-37wh-50a-rate-----ul-listed-un38-3-passed-3-0.aspx

These cells would most likely fit - they are a bit larger but I don't know if they are really any safer than the OEM Lipo's? I know LiMn cells like the Leaf cell are but am unsure of LiNiMnCo are? Do you know Miro? They are good for 4C - 40 Amp continuous which would be fine for my use.

http://www.batteryspace.com/high-power-lithium-linimnco-polymer-cell-3-7v-10ah-10059156-5c-37wh-100a-rate----un38-3-passed.aspx

If I were to go with these, I would still be looking for someone to solder them together (including the balance tap wire) as I am not comfortable with doing this specific part of assembly. I would do everything else. E+ used tiny micro temp sensors on one side of each cell (total of 10) which were combined with the balance tap wire from each cell into a 3-pin plug that plugged into the main BMS. I don't know if the sensors are simple "make and break" or if they are more like thermistors that vary the thruput.

I think this is the company that manufactured the battery:

http://www.gbp-battery.com/

 

[miro13car]

thank you for really quality pictures of EMS B battery.
LiNiMnCo Polymer versus LiPolymer .
I am not sure if they are safer than "regular" LiPolymer.
I tried to dig more info on Google but it is hard to find any info on LiNiMnCo , let me see, Kokam makes them?

 

[BVH]

Well, I've decided to try to build the pack with the somewhat safer LiNiMnCo pouches. I ordered 2 just to try. I'll test my soldering skills and verify if they will fit - being just a bit longer and thicker. If I'm happy, I'll order the remaining 8 pouches. Other than the Kapton & monofilament tape and the red outer surround paper, I can re-use everything else. I should mention that I do lots of good soldering - it's just I've never attempted to solder these pouches - being a little concerned about starting a fire. The BatterySpace pic shows them with Copper tab surfaces, not Al so it really should not be that big of a deal.

By the way, Cell number 1 tested as failed It showed 0.00 Volts whereas the other 9 cells held some residual Voltage after a complete discharge down to 2 Volts (pack Voltage)

 

[miro13car]

It would be much, much easier for you to solder if both terminals are copper.
About EMS BMS you can see in the pictures,
I now understand what those 10 2-pin connectors are for. Thermistor on every cell.
You can clearly see discharge MOSFET and charge MOSFET.
How is your front NMH battery doing? It is pity that you cannot run your EPLUS without NMH battery and you need it only to initialize/turn on battery B .

 

[BVH]

The front hub battery is doing surprisingly well. I can still get 15 to 18 miles out of it the same way I used to get about 20 miles out of it when new - peddling almost full-time but using power to ease the hills and overcome the natural motor drag and 90 lb bike weight. The ten connectors are actually 3-wire - the third wire being the normal balance tap wire. They just ran all three into one connector header.

 

[BVH]

Got my new cells Monday and finally got them all assembled and the copper tabs soldered. Just need to get some double sided foam tape to mount the board to the sticky paper covering the cells, make the final connections, give it a charge and re-install it on the bike.