Copenhagen Wheel Battery Replacement

[jkrienert]

Randomly saw a used basically new Copenhagen wheel for about the same price as a non powered wheel. Seller said they thought battery was dead and could not turn on.

After a careful disassembly, I agree. 7v does not look right for a 13s2p lion pack....

Anyone have any recommendations on how to swap in new batteries? I was considering getting a NOS ebike battery with decent cells and salvage repurposing.

 

[docw009]

You can get new 35E's this week at 18650batterystore.com for $2.65 each. Less work than pulling apart an old pack to harvest cells. That's like 70 bucks vs getting 40 old MH1's from batteryclearinghouse.com for 30 bucks. Given all the work you will do to stuff in those 26 cells, might as well do it with new ones.

 

[jkrienert]

Lunch break reply -
Well @docw009, that is optimal. I was pricing those cells out (28, 2 spares ~$90 before shipping), as well as looking at was the Super Cap spot welder at Batteryhookup (~$90) since I don't presently have one.

Now, for $30 I also just purchased a NOS bike battery with 40 35e cells still testing at minimum 3300 mAh each. My plan was to use flush cutters and a diamond wheel to separate the cells in 2p groups and keep as much of the original nickel strips welded on the cells as possible. Then, potentially solder buss wire between those 'tabs', thus avoiding purchase of a spot welder. Yet, I have never done this procedure before. So, I am hopefully going to learn here from the community on the viability of this plan...

 

[docw009]

The Copenhagen looks like it took a lot of manual assembly, There are hand wired series connections between each group? Looks like a lot of soldering to re-cell it. You will have to be soldering wires between every 4 cells anyway. Spot welding only saves you about twelve solder joins,

The problem will whether you can dremel apart your donor battery without cutting nicks in the cells. I used to harvest cells, but the process can result in dented cans, and deep nicks in the ends. My personal rule was take no risks. I never re-used a damaged cell. If you're careful, you can probably isolate 13 pairs of 2P, and have enough nickel tab left to solder them. Just hope they're worth re-using, A NOS 35E still ought to have 3500 mah. not 3300 mah.

I purchased a Liitokala four slot tester to measure cell capacity (time consuming) and a YR1030 AC IR tester. When using harveted cells, I tested capacity and IR. Rejects were rare, like one out of fifty, but that was high enough to justify testing every cell when using second hand cells, It was worth the time for me to buy new cells. as I didn't feel like I had to test every one.

 

[jkrienert]

My personal rule was take no risks

I share that precept with this project, no doubts. So that brings to notion... I don't have a capacity tester. The meter I have has an IR test, not sure yet if that's sufficient. Am I ill posed in presuming that because these cells are matched in a pack, they will have fairly close IR and capacity? To note the quote, I feel a minor risk in this...

 

[jkrienert]

The Copenhagen looks like it took a lot of manual assembly, There are hand wired series connections between each group? Looks like a lot of soldering to re-cell it. You will have to be soldering wires between every 4 cells anyway. Spot welding only saves you about twelve solder joins,

The problem will whether you can dremel apart your donor battery without cutting nicks in the cells. I used to harvest cells, but the process can result in dented cans, and deep nicks in the ends. My personal rule was take no risks. I never re-used a damaged cell. If you're careful, you can probably isolate 13 pairs of 2P, and have enough nickel tab left to solder them. Just hope they're worth re-using, A NOS 35E still ought to have 3500 mah. not 3300 mah.

I purchased a Liitokala four slot tester to measure cell capacity (time consuming) and a YR1030 AC IR tester. When using harveted cells, I tested capacity and IR. Rejects were rare, like one out of fifty, but that was high enough to justify testing every cell when using second hand cells, It was worth the time for me to buy new cells. as I didn't feel like I had to test every one.

Assembly was a bear to figure out, thankfully there is a guide thats semi-public on facebook (no account presently, yet was still able to access). However there are nuances in the process that, make the process challenging. Each P group appears to be hand wired in series.

I will also get a few good photos of the process, in case (I or) anyone else needs to upgrade the battery on one of these units. I honestly wished I could fit in LTO or LiFePO4 cells in there, for sake of the greater number of charge cycles possible without having to do this again at somepoint... By then I will maybe upgrade to an external battery and alternative hub. This will be a fun challenge though...

 

[jkrienert]

Upon closer look, the cell pair sequence layout is different than I'd supposed. It starts and begins on the same end of the arc. Interesting.

Note, the cells are not radially balanced within the wheel. This is okay because the pack module is stationary on the stator that is inside the rotating shell rotor and wheel.

It's a cool design from compactness, but the large radial bearing that facilitates that design will be a ____ to replace, if a part can even be sourced... The one on there now seems good, so maybe get at least one more packs life out of er'

 

[docw009]

There were two ways of doing it, first being what they did, stringing thirteen 1S-2P in series, so the positive and negative wind up on the same end, The second would have used a 2S-2P, ending in a 1S-2P for the thirteenth group, and positive would be there.

You didn't describe how your donor battery is constructed. Did you plan to cut out 1S-2P or 2S-2P modules and try to fit them into the Copenhagen? May or may not be easy. I would be breaking down to individual cells, and there will be enough nickel left for soldering after inserting the cells into the Copenhagen cell holders. Every cell pair is already soldered at both ends now anyway,

Might be worth having a cell tester. NOS batteries should test at 100% of their rated capacity, I've got Samsung cells I purchased in 2021 that were from 2017 that still tested at 100%.

I 've used a Liittokala 4 slot tester simililar to this one. but couldn't find them on amazon or aliexpress. Might be obsolete.They're selling units like this now. I've seen a similar LCD display used on Atorch load testers. This would seem superior because it allows one to test between different voltages, but from experiebce, these skeleton devices with open circuit boards are very fragile. Nonetheless, a cell tester is inexpensive and you can qualify all your cells by capacity, I don't know if the IR function works or is just window dressing,


.

 

[offGridDownUnder]

Randomly saw a used basically new Copenhagen wheel for about the same price as a non powered wheel. Seller said they thought battery was dead and could not turn on.

After a careful disassembly, I agree. 7v does not look right for a 13s2p lion pack....

Anyone have any recommendations on how to swap in new batteries? I was considering getting a NOS ebike battery with decent cells and salvage repurposing.

Completely ignorant (of Copenhagen wheel) question:
What is the connection of the battery to the rest of the motor? Is it just a black and red wire?
Can you ditch the faulty battery and just run wires out to a new downtube battery?

 

[jkrienert]

What is the connection of the battery to the rest of the motor? Is it just a black and red wire?
Can you ditch the faulty battery and just run wires out to a new downtube battery?

Excellent questions.

There is a Facebook posting of someone modifying the hub power system to accept an external battery pack.

Copenhagen Wheel Modifications | Paying homage to the genius of JOE BEATTY, for 2 great gifts that he made to me | Facebook

Paying homage to the genius of JOE BEATTY, for 2 great gifts that he made to me. 1. Joe restored the dead battery of my CW (it had been left charging and switched on all winter and was stone-cold...

www.facebook.com

I dont have a facebook account, so accessing that information has been, confounding at best. I ended up taking a series of screen shots before my 'visiting account' got blocked. Those screen shots have been a key reference for me, though they don't discuss the internal pack rebuild with any detail. I'm sharing a few of those screen shots that directly pertain to the internal and external connections modified.

The posting guides the use of three wire inputs from an external pack, to facilitate switching between internal and external source. Clever, but I am not sure exactly how that works with the integrated batteries system. E.g. in their recommended three wire connections, the mount to the positive and negative inputs of the raw cells into the onboard and potted BMS. The BMS has several wires that connect to the hub motor controller, for which I have no clear means of dechipering, other than the black and pink wires shown appear to be the main packs output voltage. As, when I test those individual smaller wires coming out of the BMS in pairs to the controller, they show the same ~7vdc I measured from the cell series direct shown in the first post.

Something else to consider with the external pack modification - the owner notes that their wheel eventually bricked, and neither the internal or external battery work anymore. They didn't have a cause for that, and I am left wondering if the complicated connections between the internal battery and controller I just mentioned, might be part of the cause... As in, maybe there is a better way to direct connect?

My take away at this point, given I dont want to invest a ton of time and energy into a ~350w hub motor... Is I will probably replace the internal battery because I have the resources to at this point in time. I will see how that goes. I might at a later date, re-refurbished the internal battery, replace the oversized bearing I'm supposing will need it at that point, and possibly revisit making a connection to external pack. Though, at that point in time, say conservative 400 charge cycles with this new pack, that might be a year or two down the road, and other more economical hub options could surface...

 

[jkrienert]

Here is the donor battery. I will be opening to take a look at the internals soon.

Despite the seller saying new unused, there are scuffs on the aluminum casing, and the little charge indicator button LED shows red (empty).

Honestly, if it was a 48v instead of 36v I would happily switch gears (pun intended) and try for the external mount solution discussed. I don't think I could easily facilitate modifying the donor battery for that purpose though, so disassembly will commence soon

I really hope the cells are not below the LV limit that makes them DOA.

 

[poshtiger]

Hey everyone, I'm trying to do the same thing. I have the 35E's and also some custom nickel plates. Wondering who I should contact to get the work done, or if there are any manuals out there that would facilitate the process. All of the ebike repair shops I've contacted have told me that it's a fools errand to look for someone, but I remain hopeful.

 

[docw009]

If you don't mind throwing away the money for a one time use, most of the 50 dollar spot welders on amazon will last long enough to do the 26 spot welds needed to re-fill this battery, Usually, they;re only good enough to weld the nickel plated samples included with the welders, but that's good enough for this application. Probably too weak to weld anything other than the thinnest pure .10 mm nickel, Most of the detail work in this project is soldering all the series wires.

The problem with the above welders is that they use internal Lipo pouch cells which deteriorate when stored at anything other than storage voltage, AFter you complete the job, the bext time you use it, it won't weld as well, I boiught a Seesil unit. Left a 1 star review,

 

[jkrienert]

It looks like it has quite good factory laser spot welds and they are truly samy 35E, so a spot welder might not be totally needed. The disposability of a device you mention is something if I can afford to, avoid. That's why the capacitor style intrigues me. Those are capable of many load cycles comparitively right? So, triple and some change the price ( $80) of the Lipo style doesn't sound too bad... Again though, I'm striving to avoid that route because this probably isn't a hobby, so far as a one-off repair. Photo drop of the pack internal sans end cap control and mains connections.

Mug is 10.5cm tall for rough scale.

 

[jkrienert]

Again it's a 10S4P pack that I will break down first as 1S4P, then to 2S2P, then lastly 1S2P that I will resolder 13S in series per the original design. I'm supposing there was some engineering basis for that layout (maybe thermal management, or minimizing copper wire length?) that I don't or won't be able to know, so to the KISS principal I will abide.

 

[jkrienert]

Hey everyone, I'm trying to do the same thing. I have the 35E's and also some custom nickel plates. Wondering who I should contact to get the work done, or if there are any manuals out there that would facilitate the process. All of the ebike repair shops I've contacted have told me that it's a fools errand to look for someone, but I remain hopeful.

No clue honestly. That would have been my favored route too, so +2u there. I wonder, or it would be cool if there was a way to find out if there are any local enthusiasts of this upcycling path in the region for meetups...

I also figured this was around the edge of my own envelope , so doable and a good practice in the art of a patient and research driven repair s. On sensitivity to that, I have kept rain checking module breakdown because it's gotta be one of those gigs where all the time I need is there - weekend er something. I'd really like to do contiguous breakdown and reassembly for sake of making sure I get all the BMS conn corr.

 

[jkrienert]

4p groups on the recycled donor battery measure ~ 1.24v. They have been that low for at least the 4 days shipping and likely longer. This seems like too high a fire risk for me given my living situation.

Anyone with experience have any comments on (not) using these cells?

 

[jkrienert]

Closer inspection of the cells last night revealed the nickel tabs have several open dedicated soldering pads that would have been more than adequate for my plans to recycle without spot welding.

Though, maybe 26 new cells and spot welder might be the only way I can really salvage this hub motor. If I'll be going that route, I kinda wonder if I can fit Samy 58E (21700) cells in the same housing, or if that's even a reasonable idea? They are about 50% the cost per cell new compared to the 35E. They also have a lower thermal threshold for runaway, so maybe not the best option for this closed housing where the warm motor and cells both reside.

 

[docw009]

4p groups on the recycled donor battery measure ~ 1.24v. Anyone with experience have any comments on (not) using these cells?

The physics says copper dendrites can form when cells are overcharged or discharged too low. These can cause an internal short. The process takes time and is not easily predicted. Who knows. Might never happen. I'm sure lots of people recharged (there's a process for it) from 1.5V and had no issues. My comment is why take the risk?

 

[jkrienert]

Some risks are worth it, and I'm with you, this probably isn't one.

Being unfamiliar with nmc cells, what's the proper recycling method for these?

I know the recycling center here will not take any lithium chemistry batteries. No fault to them. I know big box hardware stores have power tool drop offs but... That seems a bit of a stretch for a 10s4p 36v nom. pack.

 

[docw009]

I was at Lowes and saw they have Li-ion drop off by the returns counter, for camera and small tool batteries, but it's not a place for a dodgy ebike battery. The guy at the counter did say he had a locker in the back of the building for bigger batteries.

 

[jkrienert]

... The guy at the counter did say he had a locker in the back of the building for bigger batteries.

Well that's cool. I will probably start there with recycling the donor battery that I safely reinstalled into its original housing.

I also carefully recorded orientations and color orders of the original pack harness, then removed, and disassembled the modules down to the single cell level, so as to also preserve the plastic holders which I hope to reuse.

The general method of disassembly was:
1. Cut all series connections, one at a time.
2. Unplug all BMS monitoring wires and thermistors (one in each module!).
3. Diamond wheel slice each 1S2P bus.
4. Pry back each side of the slice to cells can easily slide out of the plastic housings.
5. Flat press ends of each cell, and cover with tape for safe recycling.

 

[jkrienert]

Few more views...

 

[jkrienert]

...and harness wire map for the BMS taps and series bus connections. roughly annotated.

 

[jkrienert]

New 35E cells made it safe. I added insulation rings to prep for pack assembly. From all learned, rings are normally just on the positive end to prevent short to case negative... Though, I had enough rings leftover to do both ends, and I figured it will provide some protection for cells heat shrink, e.g. keep the negative ends shrink edge from wearing and tearing. Are there any downsides to adding on the neg end? Some possible problems I ruled out include measuring the cell module casings has plenty of room. There also isn't anything above the neg end which would potentially contact the nickel from the additional <1mm rise (thickness of ring).

Spot welder also made it, and I immediately added some type2 silicone to a few components that were dangling in the wind. I tested quite a few welds to tune setting for these cells and the .1x7mm 98% nickel ribbon I got. Test base was stainless steel sheet of same mill the cells casing and + cap.

Weld testing for this super cap design (batteryhookup) converged on 2ms initial, 2ms pause, 3ms final weld. Initially I was using the factory set default of 8ms and no second tack, and that was yielding some dark halos and burn through.

I actually asked the cell distributor some advice about welder setting based on my tests and one of their reps provided some outstanding emails to help me in less than 24hr. They (diy500amp) were not even the ones that sold the spot welder. Pretty cool.

Anyone have any guidance on testing the nickel ribbon to make sure it's pure and not just plated steel? I'm mainly concerned about resistance and heat.