face plant from rolling backward and forward

[tjcooper]

I have a number of threads that describe my "face plants" with my MoHoo system.  But when I did the "shunt" magic, all the problems went away.  I even got a larger 16S battery from China and was using that this weekend.  I had only run the system for like 15 minutes of gentle street cruising.  But the system started to show 1 bar instead of 4 bars (out of 5) which it started with.  I kept riding very gently just to see what would happen.  When I slowed down the system would go back to 4 bars.  No reason why.

So I decided to practice rolling gently forward and backward so I could build up my skills of riding backwards.  A few rolls and the battery indicator said 1 bar.  I then leaned forward gently to go forward again.  No beeps and no messages.  The system just locked hard and sent me for a face-plant.  Lucky I was not going very fast and I had the full protection equipment.

I then very carefully tried riding forward again.  Went back to 4 bars and was behaving normally.  I decided to try the rolling forward and backward to see if it would happen again.  Sure enough, about 3 sets of oscillations and then I started to go forward to ride forward it dumped me off again.  But his time I was prepared for it and just ran off the system.

Has anyone else seen this phenomenon?  Is the drop to 1 bar and the rolling forward and back related?  Does the controller board just think the battery is too low when I start to go forward again and shuts down?  This battery has been tested 3 times running from full charge down to below 1 bar.  I has always done a verbal warning "under voltage" and then lifts up the petals to force me to get off.  This weekends events gave no verbal warning and no petal lifts.  It just shut down hard when I started to go gently forward to make a long run.  Appreciate any comments or other tests I could run.

   tjcooper

[Chriull]

Seems like the burden while "normal" driving is too much for the batteries and they loose quite some voltage (internal voltage sag from overburdening) which recovers after a break.

With rolling backward and forward the burden on the battery is even higher then with normal driving...

So I'd say the batteries are too weak (not enough continous current) for EUCing.

[Slaughthammer]

4 hours ago, tjcooper said:

I even got a larger 16S battery from China and was using that this weekend.

I think this might be your problem. Did you shunt the BMS of that batterypack as well? besides, I don't think it will make much of a difference, if the voltage drops that fast, you just bought a crappy china battery. I guess it doesn't even remotely reach the stated properties in terms of capacity and current delivery.

[US69]

Problem is ....all the amperage draw goes on one 16seriell cell system! So you have an enormous voltage sag....as those batteries are not build for continuous high amp draw....

on "idling"-back and forward that becomes the worst possible...(amp draw wise) also it is normal that the batteries "recover" after a while...

 

like said before: instead of dumping money into a weak system over and over again (mohoo)...you should save this money up for a decent wheel.... 

[Slaughthammer]

Just now, KingSong69 said:

 

like said before: instead of dumping money into a weak system over and over again (mohoo)...you should save this money up for a decent wheel....

I couldn't agree more in this case!

[Cerbera]

2 hours ago, Slaughthammer said:

I couldn't agree more in this case!

Me neither. IMO that MoHoo thing is not fit for purpose, and even less so with a duff battery pack.

[tjcooper]

Not to be contrarian, I have had over 30 long rides on the MoHoo after I did the battery shunt and everything worked fine.  I got the new 16S battery so that I could put the two systems in parallel and have the same riding battery power as my Ninebot 1 +.   I was just trying to verify that the shunt fix I  put on the new battery was working fine (I had not driven the new 16S battery to below 1 bar yet).

The new battery was almost full when I encountered this problem.  A recharge after the problem showed only 10 Watt-hr of depletion.  So everyone thinks that rocking back and forth at slow speeds puts a HUGE load on the battery?   I thought since my unit has regenerative braking that I would have like half the load I put on the battery when I am trying to accelerate going forward for a long run.  I have an inductive DC current reader I have used on the MoHoo in the past.  Maybe I should put it on the new battery and see what peak currents I get when I rock back and forth.  When I only went forward, my average current was only like 1-3 amps (new change).  For accelerating at moderate speed I only got 3-4 amps peak.  That does not seem to be a high load on the 18650 cells.  Any thoughts on this?

The only reason I am "repairing" the MoHoo is so that I have a machine for my son to play with as we ride around the neighborhood.  I always prefer the Ninebot for long riding.  The MoHoo is just for short runs and when we practice doing tight turns.....something we never get from the Ninebot.

   tjcooper

[esaj]

On 20.1.2017 at 9:23 AM, tjcooper said:

So everyone thinks that rocking back and forth at slow speeds puts a HUGE load on the battery?

The motor is more like a short circuit when it isn't turning, applying a large voltage (high PWM-duty cycle) to the motor at this point causes a very large current to run through it until it starts turning. Here's a graph I made starting from standstill and climbing a very steep gravel hill with EUC Extreme's MCM2s:

See the current peak at the start, around the 2.5-3 second mark? The motor drew about as much amperage as in the steepest parts of the hill. I'm not sure if the actual amp-values are correct, it has been said that the Gotway data exaggerates the amps a lot (like 50%), but just comparing the amount of amps drawn, getting going even with my light-weight (<60kg) carcass causes as much current to run through the motor as climbing about 20+ degree hill.

When you change the direction back and forth, the motor comes to a standstill in-between. Each time that happens, it's like getting going from still position (or actually it could be even worse, the motor is fighting the roll into the opposite direction, if it's not "just" braking but trying to actively drive the motor in the opposite direction).

 

[Rehab1]

1 hour ago, esaj said:

When you change the direction back and forth, the motor comes to a standstill in-between. Each time that happens, it's like getting going from still position (or actually it could be even worse, the motor is fighting the roll into the opposite direction, if it's not "just" braking but trying to actively drive the motor in the opposite direction).

 

 This gives me pause on practicing that maneuver anymore.

[esaj]

30 minutes ago, Rehab1 said:

 This gives me pause on practicing that maneuver anymore.

Well, I wouldn't say "don't do it", just don't be surprised that it (seemingly) drains your battery (drops voltage) pretty fast and indeed does put a lot of strain on the battery, motor and the mosfets. 

[Mono]

1 hour ago, Rehab1 said:

 This gives me pause on practicing that maneuver anymore.

It is generally true that the largest torques and currents can be produced at zero speed. Yet, you can idle gently or brutally and that will make all the difference. The huge currents only flow if you demand them, that is, if the torque needed to keep you on the wheel is actually large. Current is closely related to torque, so you can basically feel the demands you ask the wheel. I currently often spend 1/3rd of the time on the wheel gently idling and have no issues at all. Like this I have an average power drain of 17Wh/km, including many less gentle accelerations to higher speed.

[UKJ]

I am not sure if this is valid or not but because I love to idle as I do with my pedal unis, I watched the amp draw on a friends phone and noted that if done "gently" meaning delicate change of direction it was much less.

You can feel the EUC working easy as opposed to "forcing" and hear it.

But the problem here is at first when learning it is hard to do this, but with time it has become fun to see just how slow and gentle I can idle!

Any thoughts?

ukj

[Gimlet]

It was idling on my KS that burnt out the mosfets on the first board so be aware it can cause serious damage as well as running your batteries down fast.

[Cerbera]

1 hour ago, UKJ said:

I am not sure if this is valid or not but because I love to idle as I do with my pedal unis, I watched the amp draw on a friends phone and noted that if done "gently" meaning delicate change of direction it was much less.

You can feel the EUC working easy as opposed to "forcing" and hear it.

But the problem here is at first when learning it is hard to do this, but with time it has become fun to see just how slow and gentle I can idle!

Any thoughts?

ukj

That is true. The harsh, erratic, fast idling that beginners do is far more demanding on the machine than the refined, gracefully minimal movements that more experienced riders tend to do when idling.

[Smoother]

I'm no battery expert, but when I read the description of the problem again, the idea that just one of the cells in his new Chinese pack might be bad.  I'm not sure what forms of bad a LiIon cell can have.  I'm sure completely dead is one version.  Are there different types of partly dead cells?, leaking, etc.

i was watching people on YouTube pulling apart "dead" laptop batteries. Inside they would find 80% good cells and 20% bad, for example.  They used the good cells to build power walls.  Wouldn't like to try them in an EUC pack, a power wall can't face plant.

[Joff]

Its the capacitors on the board that provide big current for short burst needs. I wonder if there is something wrong with the ones you have. My weaker euc has 3 220u farad capacitors wired paralell with the battery. My better euc has 4.  Seems like a simple enough mod to add some.

[esaj]

On 24.2.2017 at 5:53 AM, Joff said:

Its the capacitors on the board that provide big current for short burst needs. I wonder if there is something wrong with the ones you have. My weaker euc has 3 220u farad capacitors wired paralell with the battery. My better euc has 4.  Seems like a simple enough mod to add some.

The capacitors hold very little charge when it comes to the amounts of power / energy the motor uses. Probably they're there just for filtering/bypassing any short-lived transients and to keep voltage steady for the electronics (feeding the buck-converter of the electronics), not the motor. As an example, a single 220uF capacitor charged to 67V and discharged to 48V gives out something like 0.066 milliwatthours, if I got it right (might be off-by-decade to one way or the other, won't make much of a difference here). Or roughly 240 milliwattseconds (240 millijoules), so it will be discharged in one second if you pull 0.24W from it. Since the motor will draw a couple of hundred of watts easily, the caps would be discharged in a very small fraction of a second.

[Joff]

Admittedly, I haven't done the math, but I did accidentally short them out when the board was not attached to the battery- made quite a zap!  I agree they are insignificant in terms of batterry capacity spread out over time, but when you need a big current for a fraction of a second that's where capacitors shine,

[tjcooper]

UPDATE of data I should have included in first post:

The new BMS has only top grade Samsung 18650 units in it.  I have done a full test load on battery pack with resistive loads and all cells show balanced output during discharge down to 3.0 volts/cell (measured by hand during discharge).  Done this at least 3 times before riding the new pack.   After the faceplant I described, I immediately pulled apart the battery and measured each cell.  They were all at 4.15 volts +- .1 volts.  There were no weak sisters in the pack.  I really don't think "cheap batteries" is an adequate description of the problem.  This new Samsung pack has also had the SHUNT fix done to it (although it is a different circuit board from first unit).

NEW PROBLEM: testing out pack a few weeks ago to ride gently on street.  I had two good rides and was starting a third.  All of a sudden the cycle will not keep petals level.  It just rolls around like the wheel has no current in either direction.  Tested 5 different times.  Sometimes I could get 3-4 feet but then loss of all power and unit goes out from underneath me.  I figured I had burned a fuse or something on the master control board.  Pulled unit completely apart looking for burned wire or broken fuse.  Pulled every cable to check for loose connectors or solder joints.  There is one black pushbutton on motherboard but no fuses and all chips and resistors and capacitors looked new.  Nothing to indicate a problem.  So I re-assembled every cable and put it all back together.

Guess what?  It now worked perfectly with no faults.  Done about 6 full loads of battery now.  No problems.  Son came home from graduate school so we decided to practice going forward-stop-backward.   He was able to do many oscilations of forward then back.  We even got up to 12 seconds of backward riding.  No hint of a problem.

Could it really be as simple as loose wiring to the motherboard?    Does anyone know what that black pushbutton on the motherboard is for?  Hitting it does not seem to change anything on the unit.  Much thanks for all the suggestions.

   tjcooper

 

 

[Carlos E Rodriguez]

1 hour ago, tjcooper said:

UPDATE of data I should have included in first post:

The new BMS has only top grade Samsung 18650 units in it.  I have done a full test load on battery pack with resistive loads and all cells show balanced output during discharge down to 3.0 volts/cell (measured by hand during discharge).  Done this at least 3 times before riding the new pack.   After the faceplant I described, I immediately pulled apart the battery and measured each cell.  They were all at 4.15 volts +- .1 volts.  There were no weak sisters in the pack.  I really don't think "cheap batteries" is an adequate description of the problem.  This new Samsung pack has also had the SHUNT fix done to it (although it is a different circuit board from first unit).

NEW PROBLEM: testing out pack a few weeks ago to ride gently on street.  I had two good rides and was starting a third.  All of a sudden the cycle will not keep petals level.  It just rolls around like the wheel has no current in either direction.  Tested 5 different times.  Sometimes I could get 3-4 feet but then loss of all power and unit goes out from underneath me.  I figured I had burned a fuse or something on the master control board.  Pulled unit completely apart looking for burned wire or broken fuse.  Pulled every cable to check for loose connectors or solder joints.  There is one black pushbutton on motherboard but no fuses and all chips and resistors and capacitors looked new.  Nothing to indicate a problem.  So I re-assembled every cable and put it all back together.

Guess what?  It now worked perfectly with no faults.  Done about 6 full loads of battery now.  No problems.  Son came home from graduate school so we decided to practice going forward-stop-backward.   He was able to do many oscilations of forward then back.  We even got up to 12 seconds of backward riding.  No hint of a problem.

Could it really be as simple as loose wiring to the motherboard?    Does anyone know what that black pushbutton on the motherboard is for?  Hitting it does not seem to change anything on the unit.  Much thanks for all the suggestions.

   tjcooper

 

 

The black button is the super secret fix any problem button.  Just kidding. 

It sound you had a loose connection. 

Hopefully is not a cold solder joint on a connector pin because it might come back again. 

Glad is working. 

[alfu]

I'm having a similar problem with a generic Chinese wheel (14" Step-n-Roll) described in another thread: have been thrown from the wheel twice and the rotor is now electrically locked.  I suspected bad wiring.

Edit 170527: No, not bad wiring. The motor runs forward and backward then stops as advertised during calibration, so the windings, mosfets and hall sensors must be all good and wired correctly.

[alfu]

On 1/20/2017 at 2:23 AM, tjcooper said:

...  So everyone thinks that rocking back and forth at slow speeds puts a HUGE load on the battery? ...

I certainly don't. See my reply at the end of this thread.

[alfu]

On 2/13/2017 at 1:33 PM, esaj said:

The motor is more like a short circuit when it isn't turning, applying a large voltage (high PWM-duty cycle) to the motor at this point causes a very large current to run through it until it starts turning.

Nice graph. Data is data, but I am unconvinced that gently rolling forward and backward through the wheel's resting point draws a large current (at least no more than overcoming the momentum from accelerating). My understanding is that at rest all three leads feeding into the delta winding of the motor are fed with a 50% duty cycle square wave, and are at the same phase, hence no current flows from one of the phases to another. Rocking back and forward would result in slight deformation of the duty cycles away from square, hence small currents would flow in the coils. At speed, the duty cycles on opposing phases distort to something like 25% to 75%.

Jes' sayin' ...