new ebike started to pulsate at full throttle

[recumbent]

I just finished assembling my new ebike and took it out for a test ride for the day, after about 30 minites the motor started to pulsate or "skip" when wide open.
I stopped for a drink by the river for 10 minutes and it went away, but returned about 15 minutes of riding later. Nothing bad but it knocks about 5MPH off the top speed also.
Mind you i was using the bike hard and no pedalling, went 33 kms for 1hr and 30 min.
It's a 48 volt LIFEPO4 battery pack and everything is new, but wow what a blast.

Does anyone know if this is the BMS issue?

 

[kbarrett]

What kind of motor is it?

 

[Ypedal]

I would guess either a BMS or a controller lvc cutout ( at low rpm you draw more amps. voltage sags more. etc. )

got a CA ?

 

[lazarus2405]

got a CA ?

He sure does. See the picture.

This is a LVC issue. You're on the second half of the battery's discharge, and voltage sag is dropping your pack voltage below the programmed LVC. When whatever is doing the limiting (either the controller, or the CA, or the battery BMS) detects this, it scales back the PWM to the motor, the current draw decreases, and the pack voltage rises back to its resting voltage, which is above the LVC. When it rises above the LVC, whatever cut it out sees that and allows the juice to flow again. Once it does, the voltage sags again below the LVC. And so on.

That causes the pulsation. Every pulse is it going through that process. The batteries recovered a bit while you sat by the lake, but as you continued to drain the batteries the pulsating returned.

So, the question is, what caused the LVC cutout?

The easiest culprit is the CA. Did you program a LVC into it? If so, try disabling it (by setting it to a uselessly low value). Get your back down to a low state of charge and see if the it still happens.

Next up is the controller. Depending upon where you bought it, it will have a LVC in the 37v range programmed for use with a 48v battery. You can hack that open and change the LVC with some soldering.

And the battery BMS. What kind of battery is it? Who makes it, and what control electronics do they use? This would be the most difficult to fix, since every battery manufacturer likes to do whatever suits the moment. If possible, I'd just remove the LVC from the BMS and use the CA to watch the battery voltage.

If the CA isn't doing the limiting, the short term solution is to take it easy when the pack is nearing the end of the cycle.

 

[recumbent]

What kind of motor is it?

Kbarrett, the motor is a 406 Clyte mounted on a 20" wheel.

I will definately check the Cycle analyst for the LVC cut-out setting and lower it. However the "voltage sag" is only about 3 volts when heavily drawn, 49 amps was the max draw according to CA. But the controler is indeed only a 36-48 volt jobber with 35 amp limit Clyte, bought new two months ago.

This is the first ebike I've owned that can keep up with traffic, 43 km/h against wind no probs, 45km/h (28mph) ideal conditions and 49 km/h freewheeling off the ground. Good strong combination, 30mph is plenty fast for a peddle bike BTW.

Thanks for all the tips so far, I'll phone Justin tomorrow, they were closed today. Here's another pic.

 

[kbarrett]

I'm kinda concerned ... if the LVC for the CA isn't hammering you, then you shouldn't be seeing random cutouts while still halfway through the battery.

Maybe a hall effect sensor or sensor wire defect that only appears when the motor is warm? Test the hall sensor pinout with a multimeter once it starts missing on you.

 

[rkosiorek]

i would suspect that a controller that new is one of the new Crystalyte V2 digital controllers. that is just by looking at it. i assume that it is there mounted to the top of the frame just behind the headset.

this is one of the controllers that only uses 6 output FETs. i suspect that at 48V those 6 little puppies got pretty warm. and the over temp is cutting in and out. sort of a warning to stop and let things cool off.

how warm was the controller case? or am i totally out to lunch?

rick

 

[fechter]

Also check for a loose battery connection. The voltage at the controller could possibly be sagging more than the battery itself.

 

[recumbent]

Okay, I did some checking and here is the story so far, Oh, and thanks for helping me sort out this issue so far.

The "LVC" on the Cycle analyst reads 19 volts, and max Amps is set at 50amps.

I phone Justin but still closed, however he answered my email and said sometimes hall sensors might go bad when hot.

The motor & controler never get hot, nor do any connections, as I monitor them very carfully because I just put it all together yesterday.
Also, Justin said that all the 35amp controlers he sells have 12 mosfets, and yes it is a v-2, at least the CA said something to this effect (could be the CA itself ) haven't figured it all out yet.

One thing is the voltage sag is potentially high for a LIFEPO4 battery.
It reads 52 volts off the charger (the morning after) and seems to hover around 49.5 volts while riding for an hour, but dips to 47 volts when gunning the throttle, that is a 5 volt drop from hot off the charger.
It never "cuts out" unless it's full throttle, which lowers top speed to 40km/h while pulsating but goes 45km/h (28MPH) normally.

 

[Mark_A_W]

Okay, I did some checking and here is the story so far, Oh, and thanks for helping me sort out this issue so far.

The "LVC" on the Cycle analyst reads 19 volts, and max Amps is set at 50amps.

I phone Justin but still closed, however he answered my email and said sometimes hall sensors might go bad when hot.

The motor & controler never get hot, nor do any connections, as I monitor them very carfully because I just put it all together yesterday.
Also, Justin said that all the 35amp controlers he sells have 12 mosfets, and yes it is a v-2, at least the CA said something to this effect (could be the CA itself ) haven't figured it all out yet.

One thing is the voltage sag is potentially high for a LIFEPO4 battery.
It reads 52 volts off the charger (the morning after) and seems to hover around 49.5 volts while riding for an hour, but dips to 47 volts when gunning the throttle, that is a 5 volt drop from hot off the charger.
It never "cuts out" unless it's full throttle, which lowers top speed to 40km/h while pulsating but goes 45km/h (28MPH) normally.

My lithium manganese sulphate pack + lipo booster is sagging from 54v to 42v. Yours is fine IMO.

 

[Ypedal]

Yep.. my 16 cell lifepo4, 12ah square bricks, fully charged 55v, within 10 ft at 40 amps i get a solid 48v and it stays there until 50 % DOD.. :wink:

 

[Mark_A_W]

I don't understand what you mean by "within 10 ft at 40 amps "?

Do you mean after opening full throttle (40amps), after travelling for 10 feet (3 metres really) it's at 48v?

 

[D-Man]

Bike really looks good recumbent. Looks really aerodynamic. I know you haven't driven it much yet but with the fairing and everything are you able to hit 49km/h if you pedal hard on a flat with the motor on? Hope you get the battery issue fixed.

 

[Ypedal]

I don't understand what you mean by "within 10 ft at 40 amps "?

Do you mean after opening full throttle (40amps), after travelling for 10 feet (3 metres really) it's at 48v?

You understood correctly, the voltage drops on my pack to 48v right away at 40 amps.. and then goes up to 50v at cruising speed of 10 amp draw 40 km/h (48v .. X5304 in a 20" )

At the 6ah mark i start to see 45v under load, and 48v cruising.. have not gone over 7ah of the 12 available since i don't have a BMS on this pack ! *( And my butt can't handle more than 20 kms at a time on the BMX !! )

 

[recumbent]

Nobody noticed the "torque wrench" literally, on the front axle.
I painted it black and bent it to follow the fork, painted the strapping same as bike, you can barely see it.

Notice how well i hid all my wires, the battery is behind the seat, the trunk is empty.