Things the manufacturers don't tell you.

[esaj]

2 hours ago, Planemo said:

I don't see why not though. Battery packs can be re-celled and aside from a tyre every now and then and a wheel bearing even less frequently, there's not a lot else to wear out. People keep E-bikes for many years, with 1000's of KM.

With newer wheels, the component failures and software issues seem (at least for the most part) be more of a thing of the past, so it could be possible. Assuming an average of 500 cycles (value chosen by best guess, there's a lot of variables that affect this, from charging/discharging -cycling depth to temperatures and storage conditions ) before the battery begins to degrade, and 50km range, even the battery packs might not need replacement until around 25000km. But many people change wheels every year or two, so very few have racked up considerable mileage on theirs so far. I think the highest number I saw was someone saying they have ridden their wheel for over 8000km, don't remember who, or the make or model though...

[Planemo]

@esajtotally agree with you there, many folk seem to upgrade wheels far before someone with an e-bike might, but the fact remains that there are very few components in an EUC which have a lifespan so it just surprises me that that there are not more around with high km's. Maybe, like you say, the specifications are only now up to a level where the failures are a thing of the past so people will start keeping wheels for longer. Even so, given the consequences of a failure on an EUC can be catastrophic (unlike an E-bike), it would be really nice if manufacturers could implement some detailed diagnostics/telemetry to keep things tickety-boo prior to a failure giving someone a very bad day. On the flip side, maybe the manufacturers feel that adding fancy telemetry would suggest their wheels are prone to failures...

[Heyzeus]

8 hours ago, pico said:

The Wh conspiracy or why the + versions suck(under some circumstances)   Ex V5F+  Ninebot E+

 

480Wh is 480Wh right?

 

Well, it depends what cells you are using, at what temperature and at what current (read also configuration, 2P, 3P etc...).

 

Read the spreadsheet of any battery and you will find that the Wh is specified at a certain current and temperature.

 

Cells for a given form factor and chemistry are optimized for capacity OR current capability(read low internal resistance). 

 

It turns out that if you are looking at super acceleration the + versions usually suck while they will carry you further at a reasonable speed.

I also suspect that many plus versions do more poorly at low temperatures.

 

So a typical conversation may go like this between EEs and management/marketing:

 

MM   They love our wheel but customer complain that there is not enough range!

EE    There are those new 4.200mah cells but their internal resistance suck.

MM   So what? Slap in a new battery pack with those cells and make some more bucks and start charging premium! Just call it the XXX + !

EE     ???

 

Again, please feel free to correct or object, I am still willing to learn...

 

Does that apply to something like these cells which I see are in a variety of the wheels Jason sells (both king Song and gotway use them I believe)

• Panasonic NCR18650PF pack: 2.9Ah (rated capacity) x 3.6V (nominal voltage) x 120 cells = 1252.8Wh
• Sanyo NCR18650GA pack: 3.5Ah (rated capacity) x 3.6V (nominal voltage) x 120 cells = 1512Wh

Both cells (from what little I've read are considered to be good quality) but how would one know whether or not the lower amp hour cell had a lower internal resistance compared to a high quality higher amp hour cell.

[pico]

I would not want to characterize those cells in particular.

My point is really that we need a standardize figure of merit ( for example max current for 10 secs at 25 deg C) to evaluate the quality of the cells that are in the pack.

This is an important factor for us Wheelers because it relates to our capacity to accelerate and avoid over leaning. Even more IMHO than straight Wh capacity.

An old battery pack could therefore be tested to check how far away it is from being unuseable (increasing security).

Bad quality packs from a production run/dubious supplier could also be detected. ( Don't forget that those cells represent the major cost of the wheel)

To be fair: as you increase the number of cells in parallel, the problem is less accute but still exist.

[mrelwood]

11 hours ago, esaj said:

hey have ridden their wheel for over 8000km, don't remember who, or the make or model though...

My 16S is at 9000+ km. And I agree, there is very little that tends to go wrong with these wheels.

[pico]

On 2/27/2019 at 7:33 PM, Heyzeus said:

Does that apply to something like these cells which I see are in a variety of the wheels Jason sells (both king Song and gotway use them I believe)

• Panasonic NCR18650PF pack: 2.9Ah (rated capacity) x 3.6V (nominal voltage) x 120 cells = 1252.8Wh
• Sanyo NCR18650GA pack: 3.5Ah (rated capacity) x 3.6V (nominal voltage) x 120 cells = 1512Wh

Both cells (from what little I've read are considered to be good quality) but how would one know whether or not the lower amp hour cell had a lower internal resistance compared to a high quality higher amp hour cell.

 For a short moment I got out of retirement.

Wh are Wh, right? Think again...

Here are curves of two QUALITY 18650 Cells from the SAME manufacturer. NCR18650GA 3500mAh 10A  vs UR18650NSX 2600mAh 20A 

 

For the sake of argument, I use a 10A discharge rate (about 800Watts or approx 1HP at 80V on a 1P20S configuration. Exaggerated, of course as you put more cells in parallel it will amount to much less).

And I use a Cutoff of 3.0V, realistic for our BMS's.

We find the 2.6Ah cell becomes a 2.4Ah cell.

And SURPRISE (!!!!???) the QUALITY 3.5A cell becomes a 2.6Ah Cell!

Factor in temperature and we may have even BIGGER surprises...

The 10A may seem unrealistic on a 8P configuration at 25 deg C. It can become disastrously realistic on a 2P configuration going full tilt Uphill at -3 deg C...

 

 

 

 

 

[Smoother]

On 2/22/2019 at 10:12 PM, meepmeepmayer said:

I can't recall a single reported mid-ride electronics failure faceplant on any modern wheel where the reason was just "stopped working without excuse".

Here's one for you to recall at a later date.  I'm riding along 12kph, hit a smal bump wheel dies instantly.  Run off, no harm.  Diagnosis, one or more Hall sensor wires is loose at the control board.

[US69]

One word to the Watthour discussion:

Nothing really new that our cells give a lower amount of watthours at a higher current draw :-) 

The cells we use in our wheels are pretty much perfect for our usage. They are the highest available watthour possible for 18650cells(3500mah, lately i have seen a 3600mah cell) with an “average” possible amp draw (all are 10Amp continuous), on a low inner resistance (30-60mOhm).

All mentioned cells (and others used, for example LG 3200mah and 3500mah) rated at a 10Ampere continuous. Meaning they are able todo that current....but doesnt mean their rated watthours/mah are for that 10Amp draw. The announced 2900, 3200 or 3500mah are for one amp draw or even lower.

But that’s not that big of a problem. Our wheels 67/84V “just” consume about 300-600Watt on a “normal cruise” of 20-30kmh.

Please dont nail me on any exact numbers...thats all “about” and the main point is that on average 400Watt are 5Amps on 80Volts.

That means on a 4P or even 8Paralell system the draw from one seriell pack/cell is -just- about 0,5 Amps (8P) or 1,25Amps(4P).

And the maximum available torque of 40-80Amps is only needed when accelerating like crazy :-) ...or pushing against crazy forewinds at 50-60kmh....

 

Btw.: I am checking the status of my battery on every charge by a Chargedoctor, so for every charge i can see what mah/watthours i pump back into the cells. thats the minimum you can do yourself to check your cells and the CD is a tool i highly, highly recommend.

On every “real” maintenance (for me thats 2000 - 3000km) my wheels then go into the hands of an expert, who checks not only the real capability of the batterypacks, but also which innerresistance each pack has....a value from which you can at best get an idea if the pack gots “squashy” or soft.....the latest wheel i have done that whith was my 18S 1680wh....which after 3200km still had a Watthour capacity of about 1620wh...and a perfect innerresistance :-)  So i would recommend such “maintenance” checks, too!

Think about your brandnew car...it also goes through inspection and checks...so i definitly think our beloved EUCs should, too :-)

[pico]

 

Glad you reacted!

Fully in agreement with ALL you said.

I am more concerned about those that ride with those measly 320Wh/480Wh packs under marginal conditions.(like -5 deg C)

Check V5F/ V8 plants threads.

I doubt they are as toughtfull as you are.

[US69]

16 minutes ago, pico said:

 

Glad you reacted!

Fully in agreement with ALL you said.

I am more concerned about those that ride with those measly 320Wh/480Wh packs under marginal conditions.(like -5 deg C)

Check V5F/ V8 plants threads.

I doubt they are as toughtfull as you are.

Yeah, you are totally right!

Over the long run, watching here for a few years, i have seen a -relative- lot of failures/faceplants of 2P wheels when they got older, mostly confirmed cell failure. Such a wheel takes 4 times or double the stress on it cells when accelerating than a 4P or 8P battery wheel.

I personally will not ride a EUC with less than 4P configuration anymore. Perhaps thats over wary , but i am also more the heavy guy :-)

And yip, when your wheel has its place in a cold gar garage or not well temperatured place before you start winter riding, it even gets more dangerous! 

[Rehab1]

12 minutes ago, US69 said:

On every “real” maintenance (for me thats 2000 - 3000km) my wheels then go into the hands of an expert, who checks not only the real capability of the batterypacks

 

 

[Heyzeus]

43 minutes ago, US69 said:

...

On every “real” maintenance (for me thats 2000 - 3000km) my wheels then go into the hands of an expert, who checks not only the real capability of the batterypacks, but also which innerresistance each pack has... So i would recommend such “maintenance” checks, too!

Think about your brandnew car...it also goes through inspection and checks...so i definitly think our beloved EUCs should, too :-)

How feasible is doing that though for those of us who don't have access to someone well-versed in either these wheels or their batteries or who sells them and does repairs on them.

these are the sorts of things that it would be really great if the manufacturers started building in ways for the users to get a quick maintenance report.  Like what @Scatcat and @esaj mentioned previously.

If only king song built in a way to run a battery diagnostic that covered the important bits (probably easier said than done I know and round back to how much extra cost people want to pay to have such features)

 

[pico]

 

36 minutes ago, Heyzeus said:

How feasible is doing that though for those of us who don't have access to someone well-versed in either these wheels or their batteries or who sells them and does repairs on them.

these are the sorts of things that it would be really great if the manufacturers started building in ways for the users to get a quick maintenance report.  Like what @Scatcat and @esaj mentioned previously.

If only king song built in a way to run a battery diagnostic that covered the important bits (probably easier said than done I know and round back to how much extra cost people want to pay to have such features)

 

As I see it, It would be relatively easy to have readings at the cell level. The BMS checks already the cells for imbalance. The prob. is I don't know if the manufs manufacture their own battery packs(read BMS). The info has to be communicated to the main board. So far all the wheels I have looked at have no communication lines between batteries and main board. Well, our phones have that.

 

2 hours ago, Smoother said:

Here's one for you to recall at a later date.  I'm riding along 12kph, hit a smal bump wheel dies instantly.  Run off, no harm.  Diagnosis, one or more Hall sensor wires is loose at the control board.

On R/C brushless motors we have been sensorless for quite a while.

I don't know if it is feasible on our Wheels.

It could eliminate one more point of failure(if properly implemented). 

[US69]

6 minutes ago, pico said:

far all the wheels I have looked at have no communication lines between batteries and main board. Well, our phones have that.

GW and KS have no board communication (KS until know)...9b Z10 and Ips wheels have....but dont show it that easy to the rider.

Inmotion has a minimal board communication of battery.

Ok, phones have one...or 2 cells max. Our modern wheels have 40 (V8) to 180 (Monster) cells......

[pico]

24 minutes ago, US69 said:

GW and KS have no board communication (KS until know)...9b Z10 and Ips wheels have....but dont show it that easy to the rider.

Inmotion has a minimal board communication of battery.

Ok, phones have one...or 2 cells max. Our modern wheels have 40 (V8) to 180 (Monster) cells......

The only info needed per cell is the voltage under load. The main board already has the total current. 

I²C (Inter-Integrated Circuit), pronounced I- squared-C

Would allow up to 10 bit address space on a few wires...for future Monsters

How a cell "caves in" (voltage drop) under load gives an accurate info of it's internal impedance (read health).

[Scatcat]

11 hours ago, pico said:

The only info needed per cell is the voltage under load. The main board already has the total current. 

I²C (Inter-Integrated Circuit), pronounced I- squared-C

Would allow up to 10 bit address space on a few wires...for future Monsters

How a cell "caves in" (voltage drop) under load gives an accurate info of it's internal impedance (read health).

What we're missing are tools in our apps that makes a battery health analysis automatically, and presents us with a reasonable estimate in percent or remaining cycles.

[pico]

4 hours ago, Scatcat said:

What we're missing are tools in our apps that makes a battery health analysis automatically, and presents us with a reasonable estimate in percent or remaining cycles.

What we are talking about is the ability to have this and more! To tell you which pack is degrading and which cell in the pack. 

My post was more about the nuts and bolts to get there.

EUC'ers like to have a complete picture...

 

 

 

[esaj]

19 hours ago, US69 said:

GW and KS have no board communication (KS until know)...9b Z10 and Ips wheels have....but dont show it that easy to the rider.

What's the 3rd wire on both KS16S packs for, if not communication between the BMS and the mainboard?

If I sometime feel adventurous, I might hook an oscilloscope there to see if there's anything happening, might be simple "on/off" (high/low)-signal or possibly some more complex (one-way?) one-wire -stuff, although if it is, deciphering what it's "saying" would probably be near impossible just by looking at the signal...

Older Firewheels had some very simple warning system, there was an extra wire going to the (single) BMS of the packs, if it was unconnected, the wheel would start repeating "battery failure or bad cell in battery" or something along those lines and wouldn't power the motor. That was in the late 2014 model, later models didn't have the wire anymore for some reason. And it could be bypassed by grounding the pin the wire connected to on the mainboard end, so all it could really tell was that the battery is OK or not.

 

[US69]

8 hours ago, esaj said:

What's the 3rd wire on both KS16S packs for, if not communication between the BMS and the mainboard?

On newer models like S series:

Three main thick wires connected to board, one is ground and the other two: one for power in(charge), one for power out.  The charge port is connected to the board, not the batterys like earlier.

(on earlier models/packs which where daisy chained the Power out/charging  was completly split, 2 wires from charge port directly..directly to battery, 2 wires power out from battery to board.)

Then now on S series there is a small red wire between packs for communication between BMS’s, just to secure the paralleled packs in case one shutsdown.

 

 

 

 

[esaj]

On 3/5/2019 at 2:46 AM, US69 said:

On newer models like S series:

Three main thick wires connected to board, one is ground and the other two: one for power in(charge), one for power out.  The charge port is connected to the board, not the batterys like earlier.

D'oh, that actually makes sense, never considered that the charging goes through the mainboard now