Dial-in your cheap chinese controller&throttle....no more surging forward.

pullin-gs

1 kW
Joined
Jul 31, 2009
Messages
395
This is an old problem discussed in several threads as far back as 2010.
Cheap speed controllers are "speed" throttled, not "torque" throttled, so when starting out, max-amps go into motor until target-RPM is reached.

The specific problem I experience on last build: Bike would "lunge" forward when starting out.
First 15% of throttle movement would not move bike forward. Once 15% threshold was reached, bike would surge forward.
Also bike speed maxxed out at about 90% throttle....last 10% of throttle movement made no difference.
Cycle Analyst is a 3rd-party product which manages this problem. But I needed a fix right away (and did not want to spend $120) so I did some digging instead.

Here is the solution which worked for me (I used 5K pots, not 10K). I cannot take credit (Tommycat wrote it up on another forum), so I will link resource I used.
https://electricbike.com/forum/forum/kits/golden-motor-magic-pie/70584-guide-to-hall-sensor-throttle-operation-testing-and-modification
(C)2018 Tommycat
In a nutshell it is a very simple circuit which goes between your hall type throttle and controller. I spiced it in the middle of the throttle cable run to the controller. It uses two potentiometers to adjust lower minimum throttle (throttle position relative to zero-throttle speed) and upper maximum throttle (throttle position relative to full-throttle speed). I had the parts already, so it took me about an hour to build, insert into cable, and bench-adjust.

An added benefit is that you can adjust it so maximum throttle is much less. Bike too fast? Turn back maximum throttle adjustment pot.
qh91OFR.jpg
(C)2018 Tommycat
 
There is a different problem with a Sensorless setup where if you try to increase the battery current "too much" (by cheating and adding a wire to the existing shunt) the controller loses syncronization and basically rattles and coughs... a kind of misfire.

The cure is to simply lower the battery current by cheating less.

But it would be nice if "somehow" you could hack the controller to make it a softer start while permitting a maximum cheat level at higher motor rpm where the good power is located.

How?

I have absolutely no idea... but if anyone has a great idea feel free to mention it.

One possible idea would be a circuit like this that had a less linear shape and was more parabolic but again I'm just throwing ideas out there and that might not work.

Sensorless really "should" be fixed by literally adding sensors which is a "one of these days" project I need to do.

Also, there are far better controllers than these $35 Chinese products.
 
Back
Top