GNG motor self starting spontaneously

I have a GNG mid-drive 36v 350w 2014 model kit and in past six months the motor intermittent self starts spontaneously
without touching the throttle.
This is very worrying to the point that I disconnect the battery from the controller
each time I park the bike. This is an intermittent problem but is becoming more frequent of late.

This is a stock kit with no changes to the motor nor the controller and is used for commuter use and has been
very reliable apart from striped sprocket screws.

I have done a search of the forum with different phrasings of the question but have not come across other posting
related to this problem so my apologies if it has been addressed before.

Thank you,
Phil

Check the ground wire in your throttle. Open grounds cause this.

I'll check that out, I did wonder about that, but the intermittent behaviour frustrated were to look.
I'll try a different throttle and see over time how it behaves. I'll also check the current throttle to see
what may have caused this problem and how to avoid it.

Thanks for your time and advise Amberwolf.

A connection problem in the ground wire can be anywhere from inside the throttle all the way back down to inside the controller. Usually it's at the connector.

If you don't find a problem (since it's intermittent) you can just replace the cable (and connector, if any) all the way from controller to throttle.

I was lucky enough to find my intermittent ground at a bad solder connection at the throttle's hall sensor ground leg.
It was like a funny car coming up the the starting line... Very unnerving! You can read about it here...

https://electricbike.com/forum/foru...t-52v-11-5ah-huffy-parkside?p=67316#post67316

Good luck, and good hunting! :thumb:

The Hall connection to the wires appears sound as does the connector at controller end.

The cable continuity is good so perhaps an internal fault with the hall?

When I took the throttle apart it was covered in grey dust which I presume is plastic
debris from rubbing friction on mating surfaces.

This makes me wonder if this debris could attract/hold moisture and cause a fault at the
legs of the Hall were they mate to the wires? Perhaps a dry lubricant such as wax
and also sealing the Hall legs would make a difference.

Anyway for now I have installed a new under bar thumb throttle, not sure
about the ergonomics yet (gets in the way when thumb needs to grip bar on the down hill)
but it will get me back on the road with some assurance.

I have noticed the new throttle doesn’t appear as smooth as the original throttle.
It seems to throttle up in steps with a noticeable thump as acceleration is increased.

The Halls are stamped S49EH 335 on the original throttle and S49EH 546 on the new
one which I presume are copies of Honeywell SS49E.

The closest Hall I can obtain locally is the SS495A compares in most aspects to the S49EH
except in supply voltage (SS49E@ 2.7 →6.5 Vdc) and (ratiometric sourcing)

Honeywell Ratiometric Hall Effect Sensor SS495A
switching current 1.5 mA
supply voltage 4.5 → 10.5 V dc
ratiometric sinking or sourcing

Would 4.5V min. supply be a stretch for the 5V supply line?

Then again I could just purchase another throttle similar to the original for its ergonomics and compatibility.
I just have to get over paying 24$US shipping for a 4$ purchase!

Phil G said:

Would 4.5V min. supply be a stretch for the 5V supply line?



Then again I could just purchase another throttle similar to the original for its ergonomics and compatibility.

Click to expand...

As a data point my Magic Pie V5 controller provides just 4.37 vdc up and running... Here is a plot of it's output with a Honeywell SS49E hall sensor.





I'd be testing and plotting both throttle's outputs to double check if the problem is indeed in the throttle/wiring or not. To understand what I mean, and get a plot chart. Check out my thread on Hall Sensor Throttles...

https://electricbike.com/forum/foru...r-throttle-operation-testing-and-modification

If you feel the output of your second throttle is reliable... Have you thought about swapping the hall sensor itself between the two? (shipping costs sometimes deter me also. )

Thanks TommyCat, changing the Hall from the under bar throttle to the original casing had occurred to me
but I think the problem maybe deeper.

Today while the bike was parked it momentarily jumped forward.

Perhaps the earthing problem suggested by amberwolf could be further down the line and into the controller,
a rabbit hole I hoping to avoid.

Well, we shall see.

Phil G said:

Today while the bike was parked it momentarily jumped forward.



Well, we shall see.

Click to expand...

Rats! But better to know now, as opposed to after changing the halls!

Get after it now... Much to important and potentially dangerous to wait (make sure your brake cut-outs are working)

amberwolf said:

A connection problem in the ground wire can be anywhere from inside the throttle all the way back down to inside the controller. Usually it's at the connector.

If you don't find a problem (since it's intermittent) you can just replace the cable (and connector, if any) all the way from controller to throttle.

Click to expand...

On motorcycles it's very common to see internal breaks in the loom around the headstock due to flexing when steering. I don't recall seeing the same thing mentioned much with ebikes but it's usually the first place to test with motorcycles (usually fairly obvious the fault is somehow connected with the steering and can often be verified simply by shaking the loom around the headstock).

TommyCat said:

Phil G said:

Today while the bike was parked it momentarily jumped forward.

Click to expand...

Much to important and potentially dangerous to wait (make sure your brake cut-outs are working)

Click to expand...

Yes the brake cut off works and normally I tie a cord over the brake lever or just pull the plug when I'm away out of sight
for a length of time.

It's unnerving to say the leased. Even if I replaced the controller I would still unplug or tie/velcro the brake lever
because I can not be assured that such a fault could not happen again.

Phil G said:

It's unnerving to say the leased. Even if I replaced the controller I would still unplug or tie/velcro the brake lever
because I can not be assured that such a fault could not happen again.

Click to expand...

Just a heads up... unplugging the electrical brake cable actually disarms it.

After running over my foot accidently while working on my bike "hot" with a twist throttle. I decided to utilize the unused light switch on my throttle in series with the throttle's hall sensor supply input so I can easily lock out the throttle when desired.
Red latching button lower left...



Bit of an anti-theft device too. :wink: May be handier than tying your bike up.

I unplug the "key" wire that disconnects battery from controller.

And or I engage the brake lever that has the magnetic switch.