62 year old fat guy

[16bitSprite]

2 hours ago, Jason McNeil said:

Is it possible that a faulty cell could lead to an over-charge in the pack? I'm sending Dave a standard charger tomorrow morning as a first step.

This is the first case of over-charging on the 14C. On the 14B experienced the situation where if it's left into the charger for several hours, it can rise as high as 67.8v. Riding it out for a couple minutes with the smaller 148Wh pack usually gets the voltage below normal range.

 

I can't be sure of anything without testing everything myself.

If he is getting a low power reading directly from the battery, with the charger disconnected, after it has been 'fully charged' this would indicate a bad battery pack. Especially if the charger is outputting the correct 67.2 or slightly higher voltage. 

I'm not an expert on the inner working of the BMS but my line of thinking however goes like this...

In a normal circumstance the charger will keep charging till it gets a reading of 67.2 volts or 4.2 volts from any one cell. At this point the charging light changes from red to green and power from the charger is cut. If however a cell is disconnected or faulty it keeps charging all the remaining cells to the point that one cell hits a safety threshold (4.2v or slightly higher) and the bms tells the charger to cut out before all the remaining cells get seriously overcharged and damaged. This would give the impression of a full charge from the charger while still only having low charge from the battery. If the power reading from the app comes from only one cell it could give an overcharged reading too. You may have enough volts to ride short time but would get a cutout very soon after riding. If you hit a hill, the wheel would fail too not being able to cope with the extra strain on the battery. If you tried to charge after this you would get a very quick charge before getting the false 'battery full' message from the charger due to the BMS safety cutout again.

I can see that like any good mechanic, you test the most common cheapest fix first, in this case replacing the charger. I really really hope for both buyer and seller this will fix it, but from what I've read my money would be on a faulty battery.

I can only speak from my experience and say that the symptoms seem the same as my case was. The only difference is that the wheel I have doesnt't have any Bluetooth connection, therefore i can't confidently speak for how an app will react to the bad /disconnected cell situation. 

Best of luck. I have my fingers crossed the charger solves your problem and you are back riding again soon. 

[Chriull]

7 hours ago, 16bitSprite said:

Sounds to me like your charger is ok but one of your 18650 cells in your battery pack has become disconnected/faulty or a fuse on the bms has gone. I had a similar problem a while back with a bad connection. Managed to fix it too with a bit if work. 

?? I cannot follow you - what are the similarities of @DaveThomasPilotproblem (akku pack a bit overcharchged) and yours and how could this overcharge be caused by disconnected fautly cells ???

7 hours ago, 16bitSprite said:

If you are feeling brave and you are comfortable doing it, I'd be opening up the battery pack and looking for a dodgy cell.

Please don't open the battery pack, at least not before all other possibilities can be excluded. And at least some hints in this direction are existing!

Also, as it seems @Jason McNeilis ready to support him to solve the problem! Tearing "everything" apart will not help the seller to support his customer....

 

1 hour ago, 16bitSprite said:

I'm not an expert on the inner working of the BMS but my line of thinking however goes like this...

In a normal circumstance the charger will keep charging till it gets a reading of 67.2 volts or 4.2 volts from any one cell. At this point the charging light changes from red to green and power from the charger is cut. If however a cell is disconnected or faulty it keeps charging all the remaining cells to the point that one cell hits a safety threshold (4.2v or slightly higher) and the bms tells the charger to cut out before all the remaining cells get seriously overcharged and damaged. This would give the impression of a full charge from the charger while still only having low charge from the battery. If the power reading from the app comes from only one cell it could give an overcharged reading too.

The wheel firmware only measures the voltage of all cells together - not individual cells. If there is a too high voltage send to the app, the overall battery voltage is this bit too high.

With the way the KS14 BMS and mainboard/firmware works there is _no_ way of having a bit too high voltage shown, the wheel running several times up a hill without any signs of cut-outs and having faulty/disconnected battery cells!

1 hour ago, 16bitSprite said:

I can only speak from my experience and say that the symptoms seem the same as my case was. The only difference is that the wheel I have doesnt't have any Bluetooth connection, therefore i can't confidently speak for how an app will react to the bad /disconnected cell situation. 

Your wheel had a bit too high voltage shown after fully charging, you could ride without cut-offs up a hill and the voltage did not get much lower? That's about all we got on information from @DaveThomasPilot

 

5 hours ago, Jason McNeil said:

Is it possible that a faulty cell could lead to an over-charge in the pack?

As written above I could not imagine how a faulty cell and a over-charge could fit together...

5 hours ago, Jason McNeil said:

I'm sending Dave a standard charger tomorrow morning as a first step.

+1!

5 hours ago, Jason McNeil said:

This is the first case of over-charging on the 14C. On the 14B experienced the situation where if it's left into the charger for several hours, it can rise as high as 67.8v. Riding it out for a couple minutes with the smaller 148Wh pack usually gets the voltage below normal range.

That's maybe another point to push the chinese manufacturers to perform better QA. The battery voltage really should not be over 67,2V - especially if a customer leaves the wheel fully charged for a longer time period... Would be interesting, how accurate the internal voltage measurement at the motherboard is and how accurate and "robust" the trim poti adjustment of the chargers are...

For the rest it seems fine - The voltage dropped from about 67,8V to 67,5V after driving 3 times 0,3 miles up a quite steep hill. Depends on the total capacity of the battery pack, but sounds sane.

 

 

[DaveThomasPilot]

As he stated, Jason is sending me a different charger to try.

Summarizing, here's what we know:

1)  The wheel reports voltage above 4.2 volts per cell.  67.65 volts after coming of the charger (and some riding up hill!). 

2)  Both WheelLog and KingSong Apps show same voltage.

3) The voltage as measured by a DVM is also 67.6x volts --higher than the 4.2 volts per cell.  This is consistent with the voltage as reported by the apps.

4)  The wheel consistently beeps three times and rocks back and forth.

5)  The wheel tilts back on even slight down hills and on level when slowing down.

6)  The beeping ceases going up and hill and the voltage as reported by the WheelLog app drops below 67 volts (didn't write down the actual voltage).

7) The charger green light is not on when charginge and the measured voltage is 67.6x volts.

There seems to be a diode in series with the cells, so I'd expect the maximum voltage would be a diode drop higher than 16 x 4.2.  Not sure if it's a schottky or PN diode.

These facts are consistent with an upwards drift in a voltage or current threshold in the charger. 

The only thing that doesn't make sense (to me) is that the battery voltage has not decreased significantly despite a significant amount of watt-hours being consumed.  (1 mile uphill).  In my experience with an overcharge, it takes very few amp-hours drain to drop the voltage below the full charge volts per cell.    That's why I was thinking the voltage reading wasn't accurate. 

Thanks for the replies!

 

 

 

[DaveThomasPilot]

Normally, the charger will transfer from constant current to constant voltage mode when the current drops below a certain threshold.

What happens if it doesn't?  Does it charge at some maximum voltage until the current drops below the threshold?  Maybe that maximum voltage for servoing charge current isn't exactly the same as the final cut-off voltage.

 

Does anyone know the current threshold for transition from constant current to constant voltage?

[Chriull]

7 minutes ago, DaveThomasPilot said:

Normally, the charger will transfer from constant current to constant voltage mode when the current drops below a certain threshold.

No - the charger charges with a constant current (stage 1) until a certain voltage is reached, then he keeps this voltage (stage 2): http://batteryuniversity.com/learn/article/charging_lithium_ion_batteries

[DaveThomasPilot]

Yes, that's right.  So, it's pretty obvious there's a problem with the charger.  Maybe more wrong than that, but voltages that high shouldn't be there. 

[DaveThomasPilot]

14 hours ago, Rehab1 said:

Wow the Cirrus came down hard even with a chute! Compression FXs generally heal well without surgical intervention. I see a ton of kids that land hard on their butts while playing sports that suffer from those type of injuries. Glad you are Ok!!

I had 2 helicopter engine out autorotations, both resulting in tip overs and major damage, one including a fire, but walked away with only a few scratches. I don't know about you but I saved some of the parts from my crashes. I have the main and tail rotor blades hanging on a wall in my barn just to remember those moments in greater detail.?

Wish I could be helpful on your wheel issues but you are in good hands with everyone that has offered advice!

Yes, it's like being dropped from 13' in a chair.  Compression fractures in a CAPS deployment like mine aren't rare.

A bigger chute results in a slower drop speed, but it also takes longer to deploy.  Which means the minimum altitude in which you can successfully pull the chute is lower.  There's some debate on what the best trade-off is. 

But, maybe we should keep this thread about the wheel issues.  Not that I mind, but others may not like wading through more posts in what is already becoming a long thread.

Not wanting to be rude--I appreciate your support Rehab1 (and others)!

[DaveThomasPilot]

Out for more uphill riding.  When it's finally happy with the battery voltage (no tilt back), I'll try tackling the subdivision. 

Is there a battery discharge curve for these batteries (840 Wh, IIRC) ?  (Voltage versus remaining capacity).

[DaveThomasPilot]

Did the full subdivision today.  That's 2.5-3 miles, up and down small to moderate hills and some flat.  It was a blast!

I didn't do it  all in one ride.  I had to get off several times for passing cars (safety precaution) and when my feet/angle bone got too sore.

WheelLog still reports 100% on the fuel gauge and 67.22 volts.    Do I really have this kind of battery capacity on an 840 WH battery?   I would have expected that ride to make a significant dent in the remaining capacity. 

If so, I guess I'll keep riding until the remaining capacity gets below X%.  Suggestions for what X should be for a brand new beginner like me?

The foot pain is getting better, but I was still glad to get off the wheel when I'd done the circuit. 

I figured out the worst part was the ankle bone against the KS pad. That pain was largely mitigated  by simply moving my foot away from the wheel a bit.

That worked fine for my right foot (which is the one I place first), but when I jump to get the left foot on it doesn't end up in the same spot every time.  I"m not good enough yet to move my feet while riding, so I either put up with the pain, or have to get off and on again to try for a better foot position.

So, I "cheated" and used a mailbox to get my feet optimally positioned --this greatly reduced my discomfort, but there was still some soreness.  Much like when ice skating or when hiking with boots that are too tight (but the pain was in different places).  More of an "ache" than a pain.

I still haven't developed an instinct to "twist" into where the wheel is leaning instead of trying to lean the opposite direction.   But, I've learned a "trick" that gets me going every time.

After jumping on with the second foot, I just start "doing the twist".  I just twist back and forth while accelerating.  That way, it seems I never develop enough of a wheel lean in a given direction so that I  end up instinctively leaning my body the other way (which results in me and finally have to jump off the wheel).

Probably looks spastic, but it works every time!

One surprise was the wheel doesn't want to stay "plumb" even when going fairly fast.  It keeps "flopping" against one leg or the other.  I didn't expect to have to hold the wheel vertical with a tight squeeze of my legs, but that seems to be what's required for a stable ride.  Does this sound right?

Or, could this be caused by incorrect tire pressure?

Thanks!

 

 

 

[The Fat Unicyclist]

2 hours ago, DaveThomasPilot said:

Did the full subdivision today.  That's 2.5-3 miles, up and down small to moderate hills and some flat.  It was a blast!

That's done it... I think you may be hooked now! 

2 hours ago, DaveThomasPilot said:

One surprise was the wheel doesn't want to stay "plumb" even when going fairly fast.  It keeps "flopping" against one leg or the other.  I didn't expect to have to hold the wheel vertical with a tight squeeze of my legs, but that seems to be what's required for a stable ride.  Does this sound right?

That "flopping" Isn't how I would expect it to be - though I would suggest it is a part of your learning process. Remember of course that the wheel wants to fall one way or another, and it is your balance that keeps it upright. 

A lot of riders hold their calves close to the wheel's body and use pressure to assist in steering. This is particularly useful when mounting, as it helps hold the wheel at that correct angle. 

My suspicion would be that, as you get comfortable and master direction and steering, this problem will go away and you will be using the angle of the wheel to guide where you are going.

So my recommendation would be simple... Practice, practice, practice! 

 

PS. It isn't cheating to hold onto something nearby. While it is ideal to be able to start solo, it is also great to have the perfect foot position. If I'm not comfortable, I often stop at a handy sign / lamppost and jiggle into just the right spot (for maximum comfort / fun).

[DaveThomasPilot]

Another 2.5-3.0 miles and the fuel gauge is finally not "pegged".  92% and voltage is 64.9 volts.

Wow, those are some batteries!

As I mentioned previously, the odometer seems off by a factor of 2 or more.  Someone mentioned this is a known bug in the KingSong.

Is it off by a constant ratio so I could multiply by a correction factor?

The wheel enables a new weight management program.  Whenever hungry, get on the wheel and ride.  The foot pain and fear of injury make you totally forget about how hungry you were!

 

 

[checho]

On 10/16/2016 at 7:03 AM, DaveThomasPilot said:

Rehab1--I lost engine power.  I headed to the nearest airport, After I broke out of the clouds I didn't think I could make it to the airport so I pulled the chute.

Broken back sounds worse than it was.  Compression fracture of T12 vertebra.  Lots of pain for a couple of weeks, but I'm pretty much ok now.  (but 1" shorter!).

I'm anxious to go riding this morning, but the wheel (KS14C) is overcharged.  That happened yesterday, but I figured it was because I left it on the charger too long.  It took over 8 hours for it to discharge down to a voltage that eliminated the three beeps and slow tilt forward/tilt back!

This morning, I took the wheel off the charger when I saw the green LED illuminate.  I might have been on the charger 10 or 20 minutes (at most) after the green LED illuminated. 

Total time on charger was about an hour.  This was longer than I expected, given the short time I rode after a full charge.

I emailed Jason this morning, but it's Sunday and I don't expect to hear back from him until tomorrow. 

I'm beginning to think the voltage reading is inaccurate.  There's a lot of watt-hours being dissipated without much voltage drop, if I believe the WheelLog app. An hour so far with the wheel turned on  Power dissipation varies from 20 to  70 watts, according to the app.

Even at the 20 watts dissipation, 20 watt hours dissipated with the voltage still at 67.69 volts doesn't seem reasonable, does it?

I'm worried that if the voltage isn't accurate, the charge state isn't known.  Maybe if I go riding the batteries could deplete so much I'd get a faceplant.

Or, is the overcharge thing normal and I'll be ok waiting until it finally discharges enough?

Thanks!

Am I safe starting at the bottom of a hill and going up to get the thing to discharge?

The voltage is fairly accurate the reading on the app is off by 1 volt, 1 volt lower, however the watt hours are completly off.

If you want the batteries to last do not overcharge if money is not an issue get the satiator which is the best charger that I have seen otherwise use the charge doctor which tries to do the same differently. With the satiator you have complete control of the charge cycle, with the charge doctor you can create a cutt off based on the current being charged to avoid overcharging, and if desired undercharging. 

The only thing you need to worry on hills is not to overheat the control board, hills draw a lot of power and the control will overheat.

[DaveThomasPilot]

I haven't ridden the wheel since October.  It just hurt my feet too much to be enjoyable.  So, I went on a diet.

Lost 25 pounds so far, so I decided to see if riding the wheel would be more fun now.  But,  even after sitting in the garage for months, it does the three beep, tilt back thing.  And, the King Song app says the voltage is 68 volts.

I'd charged it  back in October with the replacement fast charger Jason sent me, but didn't ride it.  So, it had a full charge.  Still, should it be that high?

It's not easy to get the voltage back down to where I can ride it down hills without the tilt-back thing.  I can't get on it at all right now, so I have it sitting powered to see if I can get it to bleed down some more.

I'm back to suspecting something besides the charger is wrong.  I've never seen the battery voltage below 67.69 volts, even after miles of riding. 

Any suggestions on what I should try?

 

[UKJ]

Charge Doctor set to auto stop at 90% should do the trick, sure like and trust mine.

U K J

[DaveThomasPilot]

So, I have a Charge Doctor.  But, must not be the same as what you are referring to.

Mine is a USB gadget that displays the USB current and voltage.

Can you post a link to what you are referring?

Thanks!

[DaveThomasPilot]

I'm leaving it on and recording the discharge voltage versus time. Both the tail light and headlight are on.

Power dissipation as reported by the KingSong app is .3 amps with power dissipation between 18 and 19 watts.

It's been on for about 15 hours and the voltage is 67.4 volts.  So, 270 watt hours and the voltage is only down to 67.4.  Does that sound reasonable?

I'm going to keep it on and record voltage until the battery voltage gets down to something like 3 volts per cell (or something cuts off).

I bet not too many riders complain about the battery not discharging as fast as expected!

[DaveThomasPilot]

Found what's probably the Charger Doctor referred to:

 

http://hobby16.neowp.fr/buy/

I'm going to order one. 

But, I'm impatient-- I bought some connectors for the KingSong last summer figuring I'd "roll my own" at some point using a raspberry pi as the computer that does the A/D conversion and data logging.  If it takes too long to ship and transit, I'll start on that project.

 

[meepmeepmayer]

Usually charge doctor refers to this, not sure if there are similar alternatives: http://hobby16.neowp.fr/

It allows you to see voltage, current, Wh charged. You can set threshold current/voltage (for notr charging 100%). Some allow to use 2 chargers at once.  Also there's a USB output for data on PC. Some other stuff I guess.

edit: oops to late That guy ships quite fast (from France), had mine after a few days.

 

[DaveThomasPilot]

Found what's probably the Charger Doctor referred to:

 

http://hobby16.neowp.fr/buy/

I bought some connectors for the KingSong. 

[DaveThomasPilot]

After 48 yours of 18+ watts power dissipation, voltage is 65.8 volts.

That's 48 hours * 18 watts = 864 watt hours with voltage indicating the battery is still nearly full.

I don't think that's reasonable--I'd love to hear other opinions.

I don't have confidence in the voltage reading, so I feel it's unsafe to ride.  I figure when the voltage decays to some minimum value, the wheel could cut-off and cause an accident.

So, until the voltage decays to something signficantly lower, I'll keep it sitting with the lights on until I see a full discharge voltage curve.  And, when I get the Charge Doctor, I can record the charge curve.  Then I'll either have confidence in the battery monitoring circuitry or need to get it fixed.

Suggestions on long to keep it going? 

[steve454]

On 4/3/2017 at 6:46 AM, DaveThomasPilot said:

 

I'm leaving it on and recording the discharge voltage versus time. Both the tail light and headlight are on.

 

So you are saying that the Kingsong will stay on and never shut off automatically?  That's interesting, the Ninebot shuts off after 5 minutes of not moving.  It seems a little dangerous to leave it on for so long, I don't know why, but after watching the Ninebot turn itself off many times, just thought most good wheels did that.  I wonder if Gotway has auto shutoff?  But then again, the Ninebot has no tail light or headlight, maybe it is a safety feature in Kingsong in case of needing a light in the dark or finding it in the dark.

[DaveThomasPilot]

23 hours ago, steve454 said:

So you are saying that the Kingsong will stay on and never shut off automatically?  That's interesting, the Ninebot shuts off after 5 minutes of not moving.  It seems a little dangerous to leave it on for so long, I don't know why, but after watching the Ninebot turn itself off many times, just thought most good wheels did that.  I wonder if Gotway has auto shutoff?  But then again, the Ninebot has no tail light or headlight, maybe it is a safety feature in Kingsong in case of needing a light in the dark or finding it in the dark.

 

No, I'm worried the voltage measurement is inaccurate so that the battery will eventually discharge completely without the pedal tiltback warning.  That would cause a face plant or worse!

So, I want to see believable discharge and charge voltage curves before I do more riding.

After another 30 hours of discharge, the voltage is at 63.2 volts, or 3.95 volts per cell.

78 hours * 18 watts = 1404 Watt hours.  I don't think the battery has that much capacity, yet 3.95 volts per cell would indicate the battery is still nearly full.

When the charger doctor gets here, I'll put it on and see what voltage it reads, then record the charge voltage and current as it charges. 

[DaveThomasPilot]

On 4/4/2017 at 7:14 PM, steve454 said:

So you are saying that the Kingsong will stay on and never shut off automatically?  That's interesting, the Ninebot shuts off after 5 minutes of not moving.  It seems a little dangerous to leave it on for so long, I don't know why, but after watching the Ninebot turn itself off many times, just thought most good wheels did that.  I wonder if Gotway has auto shutoff?  But then again, the Ninebot has no tail light or headlight, maybe it is a safety feature in Kingsong in case of needing a light in the dark or finding it in the dark.

 

No, I'm worried the voltage measurement is inaccurate so that the battery will eventually discharge completely without the pedal tiltback warning.  That would cause a face plant or worse!

So, I want to see believable discharge and charge voltage curves before I do more riding.

After another 30 hours of discharge, the voltage is at 63.2 volts, or 3.95 volts per cell.

78 hours * 18 watts = 1404 Watt hours.  I don't think the battery has that much capacity.

break...

Reread your post more carefully.  No, the wheel does not shut off automatically when it's not moving.  At least, the lights are is always on when I check and the power dissipation is about 18 watts.

 

[steve454]

19 minutes ago, DaveThomasPilot said:

the wheel does not shut off automatically when it's not moving.  At least, the lights are is always on when I check and the power dissipation is about 18 watts.

OK, I am not an electrical engineer, I have no idea what you are doing here.  At least the batteries seem very good, no?

[Christoph Zens]

9 hours ago, DaveThomasPilot said:

After another 30 hours of discharge, the voltage is at 63.2 volts, or 3.95 volts per cell.

78 hours * 18 watts = 1404 Watt hours.  I don't think the battery has that much capacity, yet 3.95 volts per cell would indicate the battery is still nearly full.

When the charger doctor gets here, I'll put it on and see what voltage it reads, then record the charge voltage and current as it charges. 

Maybe it draws 18W while you are looking at it with the app, but enters a low power state when sitting ilde. Like a friend of mine who wondered why his washing machine would not finish. Every time he looked after it, it was running - yet, the program didn't finish even after hours. Turned out that the outlet to which the machine was connected was only powered when the lights were on. So when he left the room and turned off the lights, the machine stopped. When he came back later and turned the lights back on, the machine was running... crazy things happen :-)

Another likely possibility is that the 300mA reading is just very inaccurate. Power measurements done by the controller board are designed to protect electronics and battery by warning the user about over power conditions. These conditions imply currents of around 50 Amp, which the board needs to handle and measure. To allow for some safety margin, the circuit will be designed to handle even higher currents than that. Now if you expect the board to be able to accurately measure something like 100mA up to 50A, that would be a 1:500 range. Considering that the controller board is not (and does not need to be) a general purpose power measurement equipment, I would be very surprised if it could actually measure power from 10W to 3kW with good accuracy.

What happens to the 0.3A reading if you turn the lights on and off? Does it change in a way that makes sense?