35 posts in this topic

Posted (edited)

Greetings.
Topic: Possibilities for salvage/fix old EUC beyond warranty, used outside specifications (frequently beyond max load and driven in rain, snow/frost, mud, offroad). 

Abstract: Used my Esway Mars Rover ES-E3 daily for over 3 years for transport to and from school, for buying food and joyriding/exploring. 35-45% of those days I rode until battery was empty. Went up and down very steep hills with incline 10-25 degrees paved and gravel. Usually at or above max load capacity. Some days I packed my schoolbooks and skis and rode up a bumpy mountain road that didnt have asfalt so I could study up on the mountain and ride my skis back home. Of course this use has taxed every system of the EUC beyond capacity. Now my beloved transport system suddenly stops when I do a maneuver that requires high power such as rapid acceleration up a steep hill, making very sharp turn at high speed/accelerating or carrying a load near/above max capacity).  Sudden and complete shutdown of all systems (which might very well be my fault for tampering with every single part of the system)

Maintenance: Once a month: Complete disassembly. Inspection, measurements of components and replacement or repair of any component outside specified tolerance. Cleaning and resealing all electronics. On first inspection replaced all wires subjected to stress and high current with extra heavy duty insulated wires soldered in and secured with added deadlength to absorb stress and vibration between solid anchors (metalclamps or heavyduty strips). Cleaning drivebolt/nuts with rubbing alcohol, re-applying threadlock and tighten with torquewrench. Simple loadtest of battery (measure voltage when fully charged, then observe the drop in voltage when applying load). Tirepressure and drivebearings check (low tirepressure makes the motor work harder and decreases battery life. Drivebearings checked after re-applying threadlock and tightening to specified torque using torquewrench (no powertools) by securing the kajigger between your knees or in a vice with thick rubber inlays by the drivebolt when disassembled. Jiggle wheel in caster and camber angle to feel for loose bearing. If jiggling creates movement on an otherwise secures drivebolt the tolerance of the bearing can be checked with a micrometer and be replaced. These bearings are very cheap at most machine shops (around 2USD) and the partnumber is printed on the seal of the bearing.)
Each following month complete disassembly, visual inspection, cleaning and checking torque on drivebolt and how much the battery voltage drops when putting a load on it.
Note: These batteries might seem good when only subjected to slight load. The benchload test using something like lightbulbs is therefor useful for the inspection process of wires, electronics and battery connection. Accelerating sharply in a hill, turning aggressively with the wheel in a very sharp angle or similiar does put a load on your battery because of high poweroutput, but you shouldn't use this method to check if your battery handles load. Retailers selling motorcycle batteries will have the equipment to test your battery if you experience lacking performance.
 
**Note: Warranty void since complete disassembly for cleaning, re-application of thread-lock to drivetrain, batterycheck, re-securing wires and troubleshooting electronics.**.

My initial point to this thread was to get a solution to my specific problem with my specific model (discontinued Esway ES-E3).  The wheel just stops when I ride aggressively and requires a reset which is done by plugging in the charger.
EDIT: loadtest both bench and proper load while riding seems fine. Dang these segment batteries are hard to troubleshoot without cutting open the sealed pack.

Further discussions or ideas here could be similar problems with other EUC and general re-purposing/salvage/overhaul projects.

Edited by Cryptonitor
EDIT: Added my tweaks and routines for optimal longevity and durability.
1

Share this post


Link to post
Share on other sites

Does the shut down probability depend on the battery level?

My initial guess is, the batteries are "spent" and will quickly drop to lower voltage, and the wheel takes this as a clue to shut down.

0

Share this post


Link to post
Share on other sites

No relation with battery level. It can happen when fully  charged or nearly depleted. I suspect it has something to do with the fact that the motor has been run at or above max capacity for years and simply overheats because of degrading of motor windings which causes higher resistance with all the resulting bad things from that. I've ridden for quite a while recently without shutdown. I don't think it's the batteries since they perform well in bench-test. I'm studying engineering. I checked every component individually and they are all fine.

0

Share this post


Link to post
Share on other sites

I'm no expert by any means, maybe you just need a new wheel then if it's just general wear and tear that's happening;)

3 years is pretty good for a noname clone, that's wonderful news for people wondering about longevity of their brand wheels!

 

1

Share this post


Link to post
Share on other sites

I'm not familiar with that wheel, but with @noisycarlos 's Ninebot a similar thing was happening.  It ended up being a cell or two not charging up to spec any more.  I wonder if you can test each cell to see what voltage they are at individually.  He took his pack all apart and measured each individually IIRC.  If you have a spot welder, you might be able to replace the dying cells.  If the pack is getting old it might be wise to replace the whole thing.

2

Share this post


Link to post
Share on other sites

Hehe I think this was a lucky purchase. Incredible durability. I have taken it completely apart, cleaned everything, done repairs and maintained it a bunch of times because I rode it all year here in the arctic. Everything gets worn from the ice and slush. WOW Hunka Hunka that is brilliant! I tested each and every component but not the individual cells of the battery! If some cells in there are bad it will be cheap and quick to fix. Stresstest, loadtest and visual check on battery was all good but as you say the charging brings a heavier load than just a homemade testbench. Hoped it would be something more interesting so we could cooperate on salvage and DIY. If changing the battery works I might engage in hacking scrap to create open-source replacement parts/upgrades for any EUC if there is interest/need

1

Share this post


Link to post
Share on other sites

Posted (edited)

Aaaah so much goodness xD If the battery is being replaced it has to be AGM(absorbent glass mat. So the battery does not loose function just because some cells are splintered) in makeup and preferably consisting of several indivudual cells with kinetic chargers inbetween to make it charge itself when moving

Edited by Cryptonitor
explanation why AGM
0

Share this post


Link to post
Share on other sites

that is many years of charging.Can you guess hoe many time you have charged the unit? 

I am sure your batteries have reach their end of life.  they only take so many cycles.

0

Share this post


Link to post
Share on other sites

Posted (edited)

Probably 1000-1200 charges. Rode for 30 minutes and still had a bar left yesterday. No steep hills though.

EDIT: maintain-charging has probably increased the max charge/discharge. If my wheel is not used I still plug in the charger regularly to maintain cellvoltage. Some cells always discharge alittle in storage and that makes a voltage difference that the charger cannot compensate for, if left unnatended.

Edited by Cryptonitor
Added note on downtime charging
1

Share this post


Link to post
Share on other sites

Posted (edited)

Replacing battery cells. Made new ones (patent pending. 4xenergy , half weight and 10 minutes to charge 100%) but need a batterypack for reference. Using cheapest possible cells (5.8USD Per cell 3.7V 18650 3.4Ah) from a factory. Pushing everything to the limit. Max discharge for each cell rated at 4A. When using 16 of these can I draw 14.26A cont. and 35.18A peak? Assuming max discharge per cell x pi x 0.7 x number of cells / 4 = ((4Ax 3.14159 x 0.7% x 16pcs)/4) =35.18A. Worried my calculations are mere speculations... 

EDIT: I was wrong about this... Useless batteries for EUC and nonsense calculations. 

EDIT: Opened batterypack and found alot of corrosion on the circuitboards that were sealed in with the old batteries.... All old cells still measured to be good, so it might just be the corrosion on the circuitboards inside the sealed batterypack that caused problems. Now creating a digital mold for 3D-printers that solve the problem of moisture getting into these critical components.  Designed a wireless power-transfer module so battery can be completely independent, easily replaceable and 100% sealed against moisture (parts are really expensive though so might go for springloaded copper connections instead). Just gotta save money to repair my 13 year old 3D printer and finish some other projects and I'll share it like opensource=)

 

 

Edited by Cryptonitor
new design batterypack
0

Share this post


Link to post
Share on other sites
10 hours ago, Cryptonitor said:

Max discharge for each cell rated at 4A.

Use at least 10A rated cells, if you're going for a 16s2p pack (32 cells). If you want to go for a single 16s1p configuration, use at least 15A cells, better 20A.

10 hours ago, Cryptonitor said:

When using 16 of these can I draw 14.26A cont. and 35.18A peak? Assuming max discharge per cell x pi x 0.7 x number of cells / 4 = ((4Ax 3.14159 x 0.7% x 16pcs)/4) =35.18A. Worried my calculations are mere speculations... Anyone know answer?

Your calculation is plain bullshit. If you have 16 cells, each can output 4 A, put tham all in series, they won't du any more amps than a single one. every electron has pass through every cell, as they are all connected in series, so when one cell is delivering 4 amps, each of the other cells is doing the same. For short peaks, they will be able to deliver quite a bit more, but this is payed with battery lifetime.

3

Share this post


Link to post
Share on other sites
11 minutes ago, Slaughthammer said:

Your calculation is plain bullshit.

No need to beat around the bush, @Slaughthammer, why not tell us what you really think!?!?:laughbounce2:

3

Share this post


Link to post
Share on other sites

Posted (edited)

On 13. mai 2017 at 10:12 PM, Slaughthammer said:

Use at least 10A rated cells, if you're going for a 16s2p pack (32 cells). If you want to go for a single 16s1p configuration, use at least 15A cells, better 20A.

Your calculation is plain bullshit. If you have 16 cells, each can output 4 A, put tham all in series, they won't du any more amps than a single one. every electron has pass through every cell, as they are all connected in series, so when one cell is delivering 4 amps, each of the other cells is doing the same. For short peaks, they will be able to deliver quite a bit more, but this is payed with battery lifetime.

Thanks=) Ofcourse you are right if the pack was coupled in series, and that fomula might very well be bullshit (I wrote it as a result of experience from studies and work. Batteries have evolved since then and characteristics change though). The batteries from the factory will be coupled in parallell and used as reference when making more prototypes of my battery. In that config I think I can draw 14.26A cont and 35.18A max. Might be wrong and wont supply enough voltage for EUC, but the pack will behave in a similar fashion with regards to max chargerate, energy dissipation, heat etc. At least that's the idea. Appreciate any feedback!

EDIT: checked the books and turns out I was totally wrong with regards to batteries at these high discharge currents. Thank you so much for letting me know @Slaughthammer! I guess those 4A max discharge cells will be used as reserve battery on long trips.

 

PS: My shell is beyond repair so I will make a new one. Any thoughts/requests for material? Could do a steampunk shell made from copper/brass, a standard shell from fiberglass/carbon fibre or an aluminium/steel shell.

Edited by Cryptonitor
1

Share this post


Link to post
Share on other sites

How about adding some capacitors from scrapped electronics with limiting diodes so the caps only supply current when the batterypack is subjected to high discharge current? Thinking a few high capacity capacitors wired with microswitches so they take turns being supplier while another is recharging. Batteries rated at 20A max discharge are expensive while those at 4A max discharge are fairly cheap. My electronics skills turns out to be abit shit at this level of engineering but as far as I can tell it should be possible and would provide a possibility for using cells at 1/5 the price. Still on Boylestad Introductory Circuit Analysis and mostly just practical experience from industrial electronc systems

0

Share this post


Link to post
Share on other sites

Posted (edited)

1 hour ago, Cryptonitor said:

How about adding some capacitors from scrapped electronics with limiting diodes so the caps only supply current when the batterypack is subjected to high discharge current? Thinking a few high capacity capacitors wired with microswitches so they take turns being supplier while another is recharging. Batteries rated at 20A max discharge are expensive while those at 4A max discharge are fairly cheap. My electronics skills turns out to be abit shit at this level of engineering but as far as I can tell it should be possible and would provide a possibility for using cells at 1/5 the price. Still on Boylestad Introductory Circuit Analysis and mostly just practical experience from industrial electronc systems

I'm not an expert in electronics, but I do see a few problems with this...

What do you mean by limiting diodes, like the discharge goes only one way, yet it can be charged otherwise through some sort of switch, like a mosfet, with current limiting to not draw too much amperage from the battery? Diodes themselves don't limit the current, they just drop a certain forward voltage over themselves, which changes somewhat with temperature and current. I might misunderstand what kind of circuit you actually mean, but I don't see how adding capacitors would help, surely they can give out very high pulses of current (as long as the ESR is low enough), but if the battery still cannot charge them with higher current than 4A (or not even that, as the motor is also drawing power from there?), and you discharge them at 20A (for example), the discharging cap will be discharged long before the other one has charged. Only part of the stored charge can be used, as the capacitor voltage will drop fast, and once it goes too low, it will either start charging itself by pulling current from the circuit, or if it's behind a diode, it will just stop supplying current once the voltage is less than the circuit voltage on the other side of the diode + diode forward drop.

The discharging capacitor will likely be drained to too low voltage in a matter of milliseconds, of course depending on the total capacitance, for example, a 2200uF cap discharged from 67V to 50V, that is 17 volts, at 20A current will discharge in about 2 milliseconds (1/500th of a second), faster once the battery is no longer full and cannot charge it to as high voltage. If the other one would be charged at 4A simultaneously, it wouldn't be charged by the time the other one is already discharged. Basically the caps would need to have higher charge than discharge amperage to "keep up". If the battery is also connected to the other side of the diode (feeding the mainboard/motor), the cap would not discharge at all, unless the battery voltage drops due to wire inductances or such, which is probably one of the reasons why there are larger aluminum electrolytic caps in the mainboard in the first place, they keep the voltage steadier, filter out transients and act as a very short-term energy reserve while the wire inductance is opposing the change in current or the battery voltage otherwise momentarily drops? But they sit directly between the battery positive and negative (circuit ground) at the mainboard, no limiting resistors or such, and no discharge/charge control.

For the higher voltage-ratings, large capacitance capacitors will be physically large sized, so you might also run into problems with getting them to fit. I used a 2200uF cap as an example above, which isn't really that high capacity, but with 70V+ voltage rating (to leave some headroom when the voltage might go higher than maximum battery voltage during braking), will already be relatively large physically. These are some random Chinese 2200uF / 80V capacitors:

Electronic-original-electrolytic-font-b-

With higher capacitance and/or voltage, they need to be larger, 12000uf / 100V:

chemicon_12000ufa.jpg

 

There are also these tiny "super caps" that can store something like 1 farad (1F = 100000uF) at 5V or something, but their ESR (equivalent series resistance) is so high that they can't give out much current, so trying to put those in series won't work. Seems they'd be mostly useful as a battery replacement for something that uses very little current, like a TV-remote, that could then be charged every now and then from somewhere else.

For high enough capacitance, you might already be looking at large capacitor banks, but the cost for a bank with large capacity and high maximum voltage is probably more than the high-discharge cells, and you'd run into even more problems with getting things to fit. And still you'd have the problem that the charging bank won't charge fast enough when the other is being discharged at higher current.

In general, it sounds like this wouldn't work at all, but maybe I misunderstood it. If it were this easy to save 80% on the battery cell costs with not much other added costs or complexity, don't you think the manufacturers would have already done it? ;) If you want high discharge, I doubt there's an easy or especially cheap way to get around using actual high discharge current cells...

Edited by esaj
3

Share this post


Link to post
Share on other sites

Need more room to fit large banks of capacitors from scrapped electronics and more cheap batterypacks to keep a charge-discharge cycle going..? Got access to as much scrap electronics as I want from a local recycling station. Gonna try mounting a surfboard on the footboards for fun and possibly for fitting auxilliary power

image.jpeg

0

Share this post


Link to post
Share on other sites

Posted (edited)

1 hour ago, Cryptonitor said:

Need more room to fit large banks of capacitors from scrapped electronics and more cheap batterypacks to keep a charge-discharge cycle going..? Got access to as much scrap electronics as I want from a local recycling station. 

I'm still not sure if I just misunderstand what you're actually trying to achieve.

If you mean to power the wheel from the capacitor banks alone, with some kind of separate circuitry disconnecting and connecting them between the batteries and the mainboard/motor, I doubt it's going to work. You'd need battery packs capable of charging the banks faster than they're being discharged, so that the other bank(s) has/have always charged up before it switches over, and the switch-over needs to happen in a way that at no point in time is the mainboard and motor left without power. Even with relatively large banks, you likely need to switch at least once per second, if not several times per second, for smaller capacitances, hundreds if not thousands of times per second. There's likely going to be all kinds of noise, rush currents and voltage transients in the system, that can play havoc with the mainboard components or the software (it would probably see the battery voltage "all over the place"). If the batteries aren't capable of charging the bank(s) fast enough, when the switch over happens, it'll be connecting only partially charged bank and it will run out of juice before the next switch, or even if the switch is voltage-based, in the next period it will connect a partially charged bank ..? The mainboard will shutdown or the motor will run out of torque. I see this simply as an unnecessary new point of failures introduced into the system.

If you already have battery packs capable of charging the banks faster than what the wheel uses at peak power, why bother with the banks in the first place? Just stick the batteries directly into the wheel and be done with it? Less space needed than with the banks and less points of failures ;) Of course you could put bypass capacitors there, but always keep the batteries directly connected...

But, I could be wrong :P  There's a bunch of actual electronics/electrical engineers, or at least otherwise far more advanced people in this field at this board who probably could shed more light into the issues. @lizardmech ? @Christoph Zens? @electric_vehicle_lover? @zlymex? @DaveThomasPilot? @Slaughthammer? Probably others I don't remember right now...

Edited by esaj
2

Share this post


Link to post
Share on other sites
39 minutes ago, Cryptonitor said:

Dangit this thing is hard to ride...

image.jpg

I WANNA see the video :()

 

1

Share this post


Link to post
Share on other sites

Posted (edited)

On 20/05/2017 at 4:24 PM, Cryptonitor said:

In that config I think I can draw 14.26A cont and 35.18A max.

Oh boy is this a wonderful demonstration of "a little bit of knowledge is dangerous" 4Amp max cells can supply a max of 4Amps - period. There is no complex calculation involved and where the hell would Pi come from in any d.c. Series/ parallel calculation? 300 of them in series would still only be 4 Amps max. In parallel it would be p times 4 Amps ( where p is the number of parallel cells). That is why the good quality reliable packs in good wheels are at least 16s2p and sometimes as much as 16s8p.

8 hours ago, Cryptonitor said:

How about adding some capacitors from scrapped electronics with limiting diodes so the caps only supply current when the batterypack is subjected to high discharge current?

This isn't short sudden peak spikes like I've seen capacitors used in model electric helicopters to prevent receiver brown out due to the receiver voltage sagging, this is spikes that can be sustained for times measured in seconds. Additional cells in parallel or higher discharge cells are, currently the only sensible, cost and weight/volume answer.

P.S. however I do love the idea of your surfboard, I'd also like to see the video of that in use 😄👌

Edited by Keith
1

Share this post


Link to post
Share on other sites
17 minutes ago, Cryptonitor said:

Video here. Laughed so hard and for so long my throat hurts! Dang this is fun to ride!

surfingele.mp4

Cool! How to get started on this thing? Trying to figure out how you would jump on...love it!

0

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now