E+ powered by A123 20 Ah pack UPDATE

[miro13car]

Yes
If you can weld or have somebody to weld for you
Task I mean very repeatitive
I mean once you get one group of 5 right rest is just repeat
Yes black jumpers are connected from group to group by Allan screws
really all assembly is first class all stainless steel hardware
But can you reuse old jumpers?

 

[miro13car]

I mean can you make new weld over old weld

 

[miro13car]

here are the pictures which should help.
you can see first class assembly , no one zink-plated screw used, all brass and stainless hardware, ALLEN bolts.
I forgot about sense board.
Every group of 5 cells has sense board which you would have to solder to new group.
whatever you do inside front hub you must:
- cover circuit board on component side with masking tape BEFORE loosening any ALLEN bolt /all cicuit board completely covered so no exposed components or traces/
- after finding right size ALLEN key you must completely wrap it with electrical tape except the very end.
failing to do it you risk shortening 36V to componet/trace on board.
But first you have to:
- find new GoldePeak NMH cells.

 

[kaysellters]

thank you. That was very helpful. I will have to talk to someone about this because this seems a little beyond my level of comfort. I don't want to F*** it up on my first try. Thanks again

 

[BVH]

I haven't ridden my E+ since around the last post in this thread prior to this one. This is just data for those of us that still have and use our bikes and more specifically for those who have the Lipo "B" battery. When I last rode, I made sure I ran the Lipo pack down to about 50% charge because I knew I would not ride for a while and 50% is the best storage condition for Lipos. I had the front hub battery full. Needless to say, last week, the front battery was empty. Fortunately, there was enough voltage to continue to power the controller and the battery charged up normally. (Reminder to self to check it more frequently) I charged up the Lipo too. On my 23 mile ride yesterday, the front battery cut off around the time that the fuel gauge turned 50% under a 650 Watt load. This was somewhere around 5 or 6 miles. I then used the Lipo for about 12 miles and it cut off while displaying 60%. I shut the system down and turned it back on to battery "A" which showed about 60% despite having cut off earlier. I was able to ride home the remaining 5 miles on battery "A" so the cutoff was premature. I fully charged both packs overnight and rode again today. I rode 5 miles on battery "A" and purposely changed to battery "B" to see if it would again cut off at 60%. I was able to go 19 miles on it with 30% remaining - no premature cut off.

Does anyone remember any of the details about the controller having a learning function to sync the fuel gauges to the changing capacity of the batteries? I remember reading something on the E+ site and/or in the manual that came with the bike about not estimating your range on the performance of the first few charges when the bike was new because the controller had to learn the capacity of the battery and sync it to the fuel gauge. I'm sure that's what's going on again now and what caused the premature shut down on the first ride.

All in all, though, 19 miles with 30% remaining on a 4 or so year old Lipo is pretty good range compared to when it was new. My typical usage in obtaining this range was between 150 and 350 Watts, with occasional 500 and 800 Watt peaks on some steep hills. On slight down hills and level areas, I'd have about 100 Watts going and be pedaling to maintain about 18 to 20 mph.

 

[arkmundi]

DISASTER WAITING TO HAPPEN
Remember you ride on LITHUM POLYMER=DANGER.
Foolish decision of Electric Motion Systems was to design so dangerous chemistry battery for Eplus bike and sell it to the public.
It was just matter of time before disaster happened .
And it happened, namely fire in bicycle store while charging which triggered recall - very costly operation for EMS .
One thing is when electronic/electrical knowledagble hobbyst build ebike with LiPolymer and run it
another thing is to design and start selling LiPolymer to general public= big NO NO.
GREAT pity.
Eplus is a wonderful, sophisticated system , DSP based with FOC motor control, all kind of protections build in /example you cannot melt Eplus motor-impossible/.

Hey, just reading through this thread for the the first time. Nice build! Thanks for sharing the details. Couldn't agree more about your warning. Can only hope more people, whether professional engineer or hobbyist ebiker, understand the difference between automotive grade LiFePO4 like A123 and LiPo for the RC hobbyist market. After a full & comprehensive R&D on my own build, I too choose an A123 AMP20 12S 36V pack build. Like the ABS plate casing you did and may follow suite, replacing my less adequate casing. Best!

 

[BVH]

An update on the front hub battery. Did a 29 mile ride today and used the front hub battery until she expired. I was able to ride 25.5 miles and the fuel gauge decreased normally all the way down to the last blinking bar and then off. If I had ridden as I normally do, very conservative throttle, I probably could have eeked out 29 or 30 miles but I was quite liberal with the throttle wanting to be sure I depleted the battery to see if the controls system re-calibrated the fuel gauge.

 

[miro13car]

EPLUS SOC battery indicator is quite precise, I found it very consistent.
not a controller, you mean motor controller?
motor controller has nothing to do with SOC.
No cheap useless 3-light battery "indicator" here.
PIC chip on BMS board does Columb count of charge going to NMH cells and reads voltage at the same time.
Bars on SOC indicator are displayed based on count of charge and voltage.

 

[BVH]

There was a statement in the original owners manual about the "system" having to "learn" by going through a few charge and discharge cycles when brand new in order for the Fuel Gauge to read accurately. I was looking to see if anyone had that exact statement from their manual.

I'm thinking the system shut down on my first ride while 3.5 bars were showing because during the 6 months of non-use, I topped off the Nimh pack a time or two with no discharge by use.

 

[wrandyr]

There was a statement in the original owners manual about the "system" having to "learn" by going through a few charge and discharge cycles when brand new in order for the Fuel Gauge to read accurately. I was looking to see if anyone had that exact statement from their manual.

I'm thinking the system shut down on my first ride while 3.5 bars were showing because during the 6 months of non-use, I topped off the Nimh pack a time or two with no discharge by use.

Here is what I could find:
"The SOC gauge will automatically calibrate itself over time to ensure it is accurate throughout the life of the battery (i.e. throughout the battery life a full SOC gauge will be shown when the battery is fully charged, even though the capacity will gradually become less than when the battery was new)."
and
"Whenever possible it is recommended that you perform one complete 'charge and discharge' cycle to auto-calibrate the fuel gauging system frequently. For definition of complete 'charge and discharge' cycle please refer to the E+ Battery Care Guide."

Both quotes are from the Battery Care Guide, and I could not find a definition of complete 'charge and discharge' cycle. My understanding, which I think I got from Ric, is that this means running the bike until LVCO occurs, then giving it a full charge.

Here is a question I have for otheres: Do you leave the Display/Controller connected when you are charging? I had problems with mine going into errors when I did this, so I got into the habit of disconnecting the display when charging. Now I am having problems with both of 2 batteries getting a full charge. I'm wondering if there is any relationship.

 

[miro13car]

OF course you leave display/controller on while charging to observe progress on charging increasing bars on SOC .
It is normal that used Eplus NMH hub will charge only to 4 bars or even 3 bars.
After long storage NMH BMS electronics needs to be calibrated by full charge followed by full discharge.

 

[BVH]

There was a statement in the original owners manual about the "system" having to "learn" by going through a few charge and discharge cycles when brand new in order for the Fuel Gauge to read accurately. I was looking to see if anyone had that exact statement from their manual.

I'm thinking the system shut down on my first ride while 3.5 bars were showing because during the 6 months of non-use, I topped off the Nimh pack a time or two with no discharge by use.

Here is what I could find:
"The SOC gauge will automatically calibrate itself over time to ensure it is accurate throughout the life of the battery (i.e. throughout the battery life a full SOC gauge will be shown when the battery is fully charged, even though the capacity will gradually become less than when the battery was new)."
and
"Whenever possible it is recommended that you perform one complete 'charge and discharge' cycle to auto-calibrate the fuel gauging system frequently. For definition of complete 'charge and discharge' cycle please refer to the E+ Battery Care Guide."

Both quotes are from the Battery Care Guide, and I could not find a definition of complete 'charge and discharge' cycle. My understanding, which I think I got from Ric, is that this means running the bike until LVCO occurs, then giving it a full charge.

Here is a question I have for otheres: Do you leave the Display/Controller connected when you are charging? I had problems with mine going into errors when I did this, so I got into the habit of disconnecting the display when charging. Now I am having problems with both of 2 batteries getting a full charge. I'm wondering if there is any relationship.

Thanks wrandy, that's what I was looking for. I've never charged either battery without the display connected. Both my batteries are original so they're around 4 years old and both still charge normally all the way up to 5 bars.

 

[miro13car]

BVH
so 4 years NMH hub and they still charge to 5 bars.
When I starting my conversion to Lithium my hub had around roughly 1,000km and of course still charging to 5 bars.
Can you tell us what approximately millage on your NMH hub right now?
When I first opened front hub and saw electronics and all mechanical assembly I was impressed.
A lot of first class engineering went into this , they made sure NMH cells are protected well.
It was ingenious to use solid billet centre aluminum hub as a heatsink for battery charging and discharging MOSFETs.

 

[BVH]

I am guessing that the front pack has approximately 50 to 75 charges on it. Let's estimate about 20 miles per charge so about 1,250 miles. You can see that I don't ride a tremendous amount.

 

[wrandyr]

I have two chargers and two hubs. I have noticed that the number of bars I get on either hub depends on which charger I use. One seems to give me five bars on both hubs, the other never does.

 

[BVH]

It would be interesting to rig a temp set of sampling wires to read Voltage at termination from both chargers used on the same hub. It would also be interesting to rig those wires such that one could insert a Watts-up meter in-line to read Amp Hours pushed into the packs based on each of the packs having been run all the way down to LVC at the same rate of draw. I wonder if the chargers have an adjustable POT inside to set termination Voltage?

 

[BVH]

Both my "A" and "B" batteries are still going strong but I frequently think about the day when the Lipo will need to be replaced. My future plan is to rebuild the pack with the somewhat safer LiMn pouches and re-use the OEM communicating BMS since Lipo and LiMn produce the same Voltage. It should be a fairly simple swap out. Is anyone aware of 10 - 15 Ah pouches that are similar in size to the Lipo pouches? Even if they are a bit bigger, I could put the finished pack into a slightly larger box as I have 3" in length, 2" in width and some height left in my rack battery bag.

Miro, I remember something you said (I think) about the controls system "counting" discharge points and re-charge points and that the numbers have to somewhat match, otherwise there is a system error produced? If this is correct, as long as the counts match, do you think the system would allow a 15 Ah pack versus the OEM 10 Ah pack? Can it deal with a higher count number?

 

[miro13car]

Just noticed your posts
I presume you want to rebuild EMS LIPolymer battery B pack, right?
Just to remind,
The reason EMS /EMS was manufacturer of EPLUS/ recalled those LiPolymer B batteries was a fire of such battery while charging at bicycle shop. This triggered total recall.
I never saw battery B and don't know what BMS elecronics it contains, obviously completely different than NMH front hub electronics.
i AM NOT SURE if LiMn and LiPolymer can use the same BMS??? But if cell voltages are the same and LVCut-off and HVCut-off are the same AND capacity is the same/so SOC meter would work approximately at least/ ??
But can you charge LiMn cells through LiPolymer BMS ? Or better charge them outside this BMS and observe voltages on celllog during charging.
Before starting EMS batt B conversion I would absolutely make sure that you do NOT have any batt B power up problems on your EPLUS!!!!
Did you try to power up your EPLUS by batt B with front NMH hub removed? - that's the first thing you must check. Next you should be able to power EPLUS on battery B by just pushing power button on console, if it does not happen you are not ready for conversion. Unplugging , plugging back connectors, resetting is NOT the way to go.
I converted Eplus to LiFePo chemistry pack.

 

[BVH]

I'm not going to rebuild now because both batteries are working fine and yes, I am aware that allegedly, the EMS "B" caught fire during charging and as a result, a shop burned down. LiMn battery charging algorythyms and Max Voltage and LVC are all identical to LiPo. The big advantage is that LiMn won't rapidly vent with flame as will Lipo's. I see no reason that I just cannot transplant the BMS onto a 10 pouch LiMn pack. My main question was if I went with a larger pack - 15 Ah, would the EMS controls system with it's internal counter still function? If the counter function is responsible for the fuel gauge, then as long as it can handle a higher number "in" and "out", then it should work and still be accurate, right?



Large 3-wire connector is from charger and smaller 4-wire is for a multi-color LED that shows state of charge right ont he outside of the battery next to the charging plug.

 

[miro13car]

EMS battery B electronics - WOW , that is one complex, crowded board.
Looks like that BMS dosnt use PIC 18 chip like hub NMH BMS but other processor. More electronics experienced ones can chime in what kind of processor is it.
You need these electronics mainly to maintain communication with display/controller.
I will tell you later what happened when I first run my EPLUS on A123 20Ah.
To start with,
NMH electronics /to which my A123 20Ah is connected/ measure charge going into cells and discharge by performing Columb count.
I am sure NMH electronics also continously monitor voltages of cells.
To unswer your question how accurate EPLUS SOC on the disply/controller would display SOC of your LiMn cells we have to know:
based on what EPLUS SOC displays bars?
Based for sure on charge count compared with programmed/registered max capacity of NMH pack.
So when the NMH cells are fully charged I think PIC 18 chip registers full charge count and displys 5 bars.
I often heard about need to "calibrate" EPLUS SOC. I understand NMH BMS must know what is the actual full charge count for accurate calibration.
The question is if NMH BMS SOC is factory programmed for full charge value of say 9Ah , so it can compare this with charge count gone into NMH everytime is charged?
I think it is sort of factory programmed , how else would you explain that with weak/old NMH hub SOC will display 4 or even 3 bars with full charge?
All above I wrote is about NMH BMS.
I think EMS LiPolimer BMS works in similar way , it compares factory programmed capacity value with actual charge. IF LIpOLYMER is 10Ah and you want to use 15Ah so what does it tell me?
EPLUS SOC might not be accurate if you use 15Ah pack connected to LiPolymer BMS.
BUT again before strting any conversion your EPLUS must be able to power up normally by pressing power button with no NMH hub.

 

[miro13car]

IF I were you I would disconnect LiMn for charging and charge it outside LiPolymer BMS with some kind of cells monitoring like celllogs,

 

[BVH]

I will test the ability to power up on "B" battery only with NMH NOT connected in circuit. I will unplug the "battery" connector but I assume not the communications connector.

I guess there's really no way to tell if a larger ah battery would result in an accurate SOC without actually making the modification. If it works both on Voltage and counts then maybe there's a chance. Beyond that, if the SOC is not accurate after the mod, then would it still allow full discharge of the larger battery or would it cut off early - after a certain number of counts? Again, we'll probably not know until it is tried.

Still curious why you recommend charging LiMn with BMS not in the circuit? All electrical aspects of LiMn are same as Lipo so BMS should perform same functions.

In any case, I think it's going to be fun and exciting trying the mod when my "B" battery goes South.

 

[miro13car]

Before I continue,
to start with..

Your EMS LiPolymer battery ,
EMS designed it as part of EPLUS fully integrated, it is called B battery, there is even "B" logo on the display
BUT
I honestly do not know how you even turn on that EMS battery, tell me about it, there is nothing about it in my EPLUS manual.
One thing is certain software is designed so master controller /inside display/ would enquire/handshake with all components including batt. A NMH and B battery.
So how you turn on B pack normally by instruction? How you switch between A NMH and B LiPolymer?

 

[BVH]

I have no official instructions, just what my dealer demonstrated to me when he installed the battery into the system (just plugged it into the two connectors) I should also mention that a new display came with the battery so I assume it had some upgraded software in it to recognize the "B" battery. I had to send the old display back.

There is no independent control of either battery. My only option is: Press and hold On/Off switch on Display to turn system on. It will always default to Battery "A" (an "A" shows under the fuel gauge) and show it's SOC. You can use Battery "A" at this point if you want or you can go to Battery "B" by again, pressing and holding the Display power button for about 4-5 seconds until the Display shows Battery "B" in use (A "B" shows under the SOC) and its' SOC.

You cannot go back to Battery "A" at this point. There is no ability to toggle between the two batteries. You must turn the system off and then back on again if you want access to Battery "A".

 

[miro13car]

Now I see.
So there is sequence ,
master controller in display makes handshake with NMH electronics first before you are allowed to press power button second time and hold for accessing batteryB .
Does it mean master controller does not make handshake with batt B electronics unless you press second time?
All this tell me that you cannot get rid of NMH battery with your conversion. You must have NMH and LiMn on your Eplus.
Are you comfortable with it , before starting your conversion? Consider NMH cells will deteriorate and die no matter what.
But do your tests first with batt B connected of course,
disconnect power wires to NMH and try to power up,
disconnect power wires AND comm.cable to NMH and try to power up.