Throttle Behavior

Discussions related to motors other than hub motors.
This includes R/C motors, botttom bracket, roller and geared drives.
User avatar
fechter   100 GW

100 GW
Posts: 12911
Joined: Dec 31 2006 3:23pm
Location: California Bay Area, USA

Re: Throttle Behavior

Post by fechter » Apr 15 2018 12:59pm

automan25 wrote:
Apr 15 2018 10:04am
Is there a way to disable PAS in our systems, but still have the throttle work? I'm pretty sure there is. Anyway, that would be a way to get around the issue with the throttle not working progressively while pedaling.
Yes. If you open the controller and cut or disconnect the GREY wire going to the PAS sensor, the PAS will be disabled and the throttle will work properly. That's what I did on mine. I suppose you could wire up a switch to make this selectable.
"One test is worth a thousand opinions"

User avatar
Rassy   1 MW

1 MW
Posts: 1979
Joined: Apr 08 2007 10:58am
Location: Eugene, Oregon USA

Re: Throttle Behavior

Post by Rassy » Apr 15 2018 1:16pm

This is becoming discouraging. I have purchased six BBS02s, two on my own trikes, and four on friends and relatives bikes/trikes. I provide most of the maintenance and service for these six units. All have pre-2017 controllers and they all work properly.

The problem is that I currently have no spare parts and I like to keep at least one full kit on the shelf so that I can quickly diagnose any problem by substituting parts and getting the bike running again. Although there haven't been any issues for the last 3 years, my spare kit keeps getting installed on someone else's bike, and then I purchase a new one for the shelf. Now it is impossible to purchase a BBS02 kit that works correctly (pedal/throttle issue), or even a BBS02B controller that works correctly.

While I understand that some don't care for the Bafang simple on/off PAS system that is triggered by pedal rotation, everyone in my group of older recreational/exercise riders is very happy with the system as it operates on the pre-2017 controllers.
-Rassy-
One Tadpole Trike equipped with a Bafang BBS02
https://endless-sphere.com/forums/viewt ... 28&t=69419
One Delta Trike equipped with a Bafang BBS02
https://endless-sphere.com/forums/viewt ... 0#p1291260

User avatar
Rassy   1 MW

1 MW
Posts: 1979
Joined: Apr 08 2007 10:58am
Location: Eugene, Oregon USA

Re: Throttle Behavior

Post by Rassy » Apr 15 2018 1:28pm

fechter wrote:
Yes. If you open the controller and cut or disconnect the GREY wire going to the PAS sensor, the PAS will be disabled and the throttle will work properly. That's what I did on mine. I suppose you could wire up a switch to make this selectable.
Excellent suggestion. If I ever have to use one of those faulty controllers I'll just splice some wires to that grey wire and mount an external switch so I can use the other PAS levels with just full throttle override and use PAS 0 with proper throttle behavior while pedaling. Thanks for the idea.

Is there just a single wire within the grey cable, or are there multiple little wires?
-Rassy-
One Tadpole Trike equipped with a Bafang BBS02
https://endless-sphere.com/forums/viewt ... 28&t=69419
One Delta Trike equipped with a Bafang BBS02
https://endless-sphere.com/forums/viewt ... 0#p1291260

espresso   10 mW

10 mW
Posts: 34
Joined: Mar 08 2017 5:20am

Re: Throttle Behavior

Post by espresso » Apr 16 2018 4:13am

automan25 wrote:
Apr 15 2018 12:16pm
I just got a new controller and it didnt fix the problem. It seems there is little way to know what to expect out of Bafangs firmware. Hence what I’m proposing is figuring out how to get around the PAS component of the firmware.
Are we even sure that the problem is coming from the Controller?
Couldn't it be coming from the PAS board?

User avatar
fechter   100 GW

100 GW
Posts: 12911
Joined: Dec 31 2006 3:23pm
Location: California Bay Area, USA

Re: Throttle Behavior

Post by fechter » Apr 16 2018 8:23am

Rassy wrote:
Apr 15 2018 1:28pm
Is there just a single wire within the grey cable, or are there multiple little wires?
I thought I had a better picture of it, but this is all I could find. There are 4 wires going to the PAS sensor. Snipping the grey one will disable the PAS without causing error messages.
PAS Sensor Signal Wire.jpg
PAS Sensor Signal Wire.jpg (100.05 KiB) Viewed 917 times
"One test is worth a thousand opinions"

User avatar
fechter   100 GW

100 GW
Posts: 12911
Joined: Dec 31 2006 3:23pm
Location: California Bay Area, USA

Re: Throttle Behavior

Post by fechter » Apr 16 2018 8:25am

espresso wrote:
Apr 16 2018 4:13am

Are we even sure that the problem is coming from the Controller?
Couldn't it be coming from the PAS board?
The PAS board just sends pulses when the pedals are moving. There's not much to go wrong there. This is easy to test too. The problem goes away if the other version of controller is used.
"One test is worth a thousand opinions"

User avatar
Rassy   1 MW

1 MW
Posts: 1979
Joined: Apr 08 2007 10:58am
Location: Eugene, Oregon USA

Re: Throttle Behavior

Post by Rassy » Apr 16 2018 9:52am

espresso wrote:
Are we even sure that the problem is coming from the Controller?
Couldn't it be coming from the PAS board?
fechter wrote:
The PAS board just sends pulses when the pedals are moving. There's not much to go wrong there. This is easy to test too. The problem goes away if the other version of controller is used.
"One test is worth a thousand opinions"
Right, I installed several controllers on the same BBS02B. The original controller and a replacement controller that I purchased, both manufactured in 2017, and both had the pedal/throttle issue. LUNA obtained a used controller for me, manufactured in 2016, and the issue went away.

EDIT: Obviously Bafang could manufacture a replacement controller that does not have this issue, but they won't even respond to questions about it. They could probably also provide a flash update of some sort to our existing controllers but for some reason choose not to. Even dealers such as LUNA and em3ev seem unable to get any support from Bafang on this issue.
-Rassy-
One Tadpole Trike equipped with a Bafang BBS02
https://endless-sphere.com/forums/viewt ... 28&t=69419
One Delta Trike equipped with a Bafang BBS02
https://endless-sphere.com/forums/viewt ... 0#p1291260

User avatar
Alan B   100 GW

100 GW
Posts: 7486
Joined: Sep 11 2010 7:43am
Location: San Francisco Bay Area, USA
Contact:

Re: Throttle Behavior

Post by Alan B » Apr 16 2018 11:10am

Sounds like we need a circuit that shuts off the PAS pulses when the throttle signal comes up above minimum. Can the grey signal wire be grounded instead of cut? Might be able to do it with one or two transistors. Or one of the tiny 8 pin CPUs. Read throttle voltage on one pin, PAS in and out on two others. 5V is available. Adafruit Trinket should do it. Pretty simple firmware. When I get around to firing up my BBS02 Xootr Swift I may have this problem..

User avatar
fechter   100 GW

100 GW
Posts: 12911
Joined: Dec 31 2006 3:23pm
Location: California Bay Area, USA

Re: Throttle Behavior

Post by fechter » Apr 16 2018 2:10pm

Alan B wrote:
Apr 16 2018 11:10am
Sounds like we need a circuit that shuts off the PAS pulses when the throttle signal comes up above minimum. Can the grey signal wire be grounded instead of cut?
I suspect it would work to either ground it or pull it up to 5v. If so, it would be easy to make an analog circuit that does that. Just a NPN transistor with the base going to the throttle signal through a resistor might be all that's needed. This could all fit inside the controller compartment.

Sort of a pain to test, but I will see if I can do it.
"One test is worth a thousand opinions"

User avatar
Alan B   100 GW

100 GW
Posts: 7486
Joined: Sep 11 2010 7:43am
Location: San Francisco Bay Area, USA
Contact:

Re: Throttle Behavior

Post by Alan B » Apr 16 2018 2:16pm

fechter wrote:
Apr 16 2018 2:10pm
Alan B wrote:
Apr 16 2018 11:10am
Sounds like we need a circuit that shuts off the PAS pulses when the throttle signal comes up above minimum. Can the grey signal wire be grounded instead of cut?
I suspect it would work to either ground it or pull it up to 5v. If so, it would be easy to make an analog circuit that does that. Just a NPN transistor with the base going to the throttle signal through a resistor might be all that's needed. This could all fit inside the controller compartment.

Sort of a pain to test, but I will see if I can do it.
Exactly. The min output of the throttle might be too high, but adding a diode to raise the threshold a bit might help.

User avatar
Rassy   1 MW

1 MW
Posts: 1979
Joined: Apr 08 2007 10:58am
Location: Eugene, Oregon USA

Re: Throttle Behavior

Post by Rassy » Apr 16 2018 2:33pm

I like your thoughts on this Alan B. I don't do electronics so my interpretation is installing a normally closed relay in the grey wire and when applying throttle have the relay open. If this could be done with a small electronic module that gets input from the throttle output the new "faulty" controllers would work like the original controllers. And if this module was small enough it could be fitted right inside the controller area.

So Alan, can you or someone else make a detailed diagram with part numbers that could do this? Also need to identify the throttle wires that are present in the controller area that would need to be tapped into along with the grey wire.

I have a brand new "faulty" controller sitting on the shelf that I would like to "fix" for future use.

More discussion took place while I was typing this. I would be glad to help in any way possible.

EDIT: Just took a look at my BBS02B controller, which is quite different that the BBS02A controller pictured above, with the housing wrapped completely around all the potting material. Might be very difficult to locate or expose the throttle wires. I'll add a picture here soon.

100_0833.JPG
100_0833.JPG (91.87 KiB) Viewed 883 times
Last edited by Rassy on Apr 16 2018 3:33pm, edited 2 times in total.
-Rassy-
One Tadpole Trike equipped with a Bafang BBS02
https://endless-sphere.com/forums/viewt ... 28&t=69419
One Delta Trike equipped with a Bafang BBS02
https://endless-sphere.com/forums/viewt ... 0#p1291260

User avatar
Alan B   100 GW

100 GW
Posts: 7486
Joined: Sep 11 2010 7:43am
Location: San Francisco Bay Area, USA
Contact:

Re: Throttle Behavior

Post by Alan B » Apr 16 2018 3:21pm

I suspect fechter will make this with a few very small parts before I get to doing anything. I'm willing to help but I doubt he needs any. :)

Digging the wires out of the potting may be the biggest issue. For this circuit only 3 connections are needed, and no cutting. As long as the PAS output doesn't mind being pulled down. Just a sample of the throttle signal used to turn on a clamping transistor and stop the PAS pulses from getting through to the controller, which is looking for the edges of those pulses.

The relay is another idea but this should be smaller.

User avatar
Rassy   1 MW

1 MW
Posts: 1979
Joined: Apr 08 2007 10:58am
Location: Eugene, Oregon USA

Re: Throttle Behavior

Post by Rassy » Apr 16 2018 6:50pm

Due to the potting and the limited space within the controller area I might be inclined to do something outside the BBS02 housing. I.e., cut and extend both ends of the grey PAS wire to an appropriate external point. Expose the internal wires within the main harness that runs from the display unit to the BBS02 controller to gain access to the throttle wires. The red wire carries +5V, the blue wire carries the throttle signal, and the black wire is the ground.

So now all I need is some help on selecting the various parts that will break the continuity of the grey wire when the throttle signal reaches something like +2v, assuming that this is really a fairly simple electronic task.

If someone can draw a simple diagram with a parts list for parts I can purchase I will modify my spare 2017 controller and test it on one of my trikes. Thanks for any information, etc.

Edit: Would this 1.5v coil, NO-NC, non latching mini relay do what I want? It seems like it would do the job for an external solution. Using the NC pins I think it should just open the circuit when the throttle signal reaches 1.5v and then close again when the throttle is released.

https://www.mouser.com/ProductDetail/Pa ... Npasv9w%3d
Last edited by Rassy on Apr 16 2018 7:29pm, edited 1 time in total.
-Rassy-
One Tadpole Trike equipped with a Bafang BBS02
https://endless-sphere.com/forums/viewt ... 28&t=69419
One Delta Trike equipped with a Bafang BBS02
https://endless-sphere.com/forums/viewt ... 0#p1291260

User avatar
amberwolf   100 GW

100 GW
Posts: 25128
Joined: Aug 17 2009 6:43am
Location: Phoenix, AZ, USA, Earth, Sol, Local Bubble, Orion Arm, Milky Way, Local Group
Contact:

Re: Throttle Behavior

Post by amberwolf » Apr 16 2018 7:06pm

fechter wrote:
Apr 14 2018 11:05am

I guess another, simpler approach would be to make something that resembles a throttle but is just a switch you can tie into the existing level buttons to make changing levels easier while riding. Imagine a thumb throttle that is centered in the middle and moves both ways to increase or decrease steps. It would be easy enough to add a pulse circuit that would keep incrementing the steps if you hold it in one direction.
I think the old Forrest Mims III circuit books have at least one very simple method of doing that (but I dont have the books anymore).

Regarding pedal torque sensors, theres a few things here that might do what youre after or at least point the right direction
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4481893/
https://stagescycling.com/us/technology/
the latter of which uses a partly hollowed crank arm with the sensor and wireless unit in that space

the search I used to find htem, which might find more without the bluetooth on it
https://www.google.com/search?num=100&n ... zEI_RhfSuA

User avatar
Alan B   100 GW

100 GW
Posts: 7486
Joined: Sep 11 2010 7:43am
Location: San Francisco Bay Area, USA
Contact:

Re: Throttle Behavior

Post by Alan B » Apr 16 2018 8:16pm

The basic idea is to disable the PAS when the throttle voltage rises above 'idle'.

The grey wire is reportedly the PAS pulses. So putting a switch or relay in that path will disable the pulses from getting to the controller. So will pulling it to ground, as long as the source of the signal doesn't mind the short.

The trick is detecting when the throttle is advanced above minimum. What is the minimum voltage out of that throttle? Feeding this to the relay coil is not likely to produce the desired result, and it might distort the throttle signal as well. Knowing what that voltage is we can design something to trigger on it without loading it down.

User avatar
fechter   100 GW

100 GW
Posts: 12911
Joined: Dec 31 2006 3:23pm
Location: California Bay Area, USA

Re: Throttle Behavior

Post by fechter » Apr 16 2018 11:27pm

The problem with a mechanical relay is driving the coil. The voltage regulators in the controller don't have much extra capacity and might drop out with the load of a coil. We can make something that works like a relay but takes way less current (and cheaper, smaller). But first we need some measurements on that PAS signal line and take some throttle voltage measurements.
"One test is worth a thousand opinions"

bhaaf   1 mW

1 mW
Posts: 10
Joined: Jul 19 2017 12:43pm

Re: Throttle Behavior

Post by bhaaf » Apr 17 2018 10:09am

I hope you guys can find a fix. I would not purchase another Bafang BBSXX until the vendor would guarantee that it would not have this problem. Question your vendor on this before purchase.

User avatar
fechter   100 GW

100 GW
Posts: 12911
Joined: Dec 31 2006 3:23pm
Location: California Bay Area, USA

Re: Throttle Behavior

Post by fechter » Apr 17 2018 11:35am

I made a little progress using a semi-blown controller board.

Throttle zero voltage = 0.844v Full throttle = 4.18v

PAS sensor signal lines are pulled high (5v) by resistors in the controller. Pullup resistors appear to be 5k.

I need to use a working one to go further.
I'm nearly certain the PAS sensors are something like the motor halls or a SS41, which has an open collector output. This means we can ground the signal line without damaging anything.

So the next test is to try grounding the grey wire to the black wire and see what happens. If we are lucky, it disables PAS the same as disconnecting it. If we are not so lucky, it might show a PAS sensor fault on the display or not disable the PAS.

If the PAS disables when the grey wire is grounded, then it might work to add a super simple circuit like below. When the throttle signal gets above the Vbe for the transistor (around 0.7v), the transistor will start to conduct and pull the signal line down. Since the throttle zero is higher than this, a resistor divider made of R1 and R2 can set the level higher so it comes on around 0.9v. The controller input from the PAS signal is digital, so it might work to leave out R2 and just get the value of R1 right so it comes on where you want it. R1 also needs to be high enough that it doesn't load the throttle signal line.
PAS correction circuit 1.jpg
PAS correction circuit 1.jpg (18.43 KiB) Viewed 832 times
"One test is worth a thousand opinions"

User avatar
Rassy   1 MW

1 MW
Posts: 1979
Joined: Apr 08 2007 10:58am
Location: Eugene, Oregon USA

Re: Throttle Behavior

Post by Rassy » Apr 17 2018 12:10pm

@ fechter and Alan B, Thank you so much for your guidance. In addition to the possibility of causing a problem by over loading the 5v circuit or grounding the PAS sense wire I don't really want to dig into the potting or break into the external wiring harness to gain access to the throttle wires.

So I slept on it and came up with a fairly simple solution that avoids all the above issues. Namely, cut the well documented grey wire found on the four wire PAS connector within the controller. Splice a wire to each end of the cut grey wire ends and run these two wires to the throttle. Install a normally open micro switch to the throttle such that when the throttle is closed the micro switch is pushed to its closed position and the PAS circuit is completed and should work normally. Then when the throttle is opened the micro switch will move to its open position and the PAS circuit will be disabled, allowing normal throttle operation.

I've already checked the BBS02B controller by removing a small amount of potting where the external wires enter and there is enough space to drill a small hole in the grommet for the wires that need to pass through. In my junk box I found a normally open micro switch that was made to turn on a brake light when attached to a normal bicycle brake line. My biggest challenge will be to properly attach the micro switch to the throttle housing. The other possible issue is the length of the wire that will connect the two ends of the cut grey PAS sense wire.

This will take me a while to test and install so in the meantime I still believe Bafang should provide the solution, or maybe one of the electronic solutions discussed above will be completed.
-Rassy-
One Tadpole Trike equipped with a Bafang BBS02
https://endless-sphere.com/forums/viewt ... 28&t=69419
One Delta Trike equipped with a Bafang BBS02
https://endless-sphere.com/forums/viewt ... 0#p1291260

User avatar
fechter   100 GW

100 GW
Posts: 12911
Joined: Dec 31 2006 3:23pm
Location: California Bay Area, USA

Re: Throttle Behavior

Post by fechter » Apr 17 2018 12:16pm

A mechanical switch on the throttle will work if you can find a good way to mount it. I'm hoping I can do the transistor circuit all inside the controller compartment. Tapping into the grey and black PAS sensor wires is easy. The throttle signal is burried in the potting but may be fairly easy to access since it's on the edge of the board.
"One test is worth a thousand opinions"

User avatar
Rassy   1 MW

1 MW
Posts: 1979
Joined: Apr 08 2007 10:58am
Location: Eugene, Oregon USA

Re: Throttle Behavior

Post by Rassy » Apr 17 2018 1:44pm

That large harness wire enters the potting near the top on the BBS02B and you might be able to clean off enough potting along it until the blue (I think) throttle signal wire can be exposed and you wouldn't even have to get clear down to the board. I'm hoping your solution works as that would be the cleanest and easiest way to go.

I have my spare controller in the house and I'll see if I can expose the throttle wire.
-Rassy-
One Tadpole Trike equipped with a Bafang BBS02
https://endless-sphere.com/forums/viewt ... 28&t=69419
One Delta Trike equipped with a Bafang BBS02
https://endless-sphere.com/forums/viewt ... 0#p1291260

User avatar
fechter   100 GW

100 GW
Posts: 12911
Joined: Dec 31 2006 3:23pm
Location: California Bay Area, USA

Re: Throttle Behavior

Post by fechter » Apr 17 2018 2:05pm

Yes, on mine the throttle signal wire is blue.
BBSHD Throttle Signal Wire.jpg
BBSHD Throttle Signal Wire.jpg (110.87 KiB) Viewed 823 times
"One test is worth a thousand opinions"

User avatar
teklektik   10 GW

10 GW
Posts: 4020
Joined: Mar 26 2011 1:15pm
Location: CT, USA

Re: Throttle Behavior

Post by teklektik » Apr 17 2018 2:17pm

Rassy wrote:
Apr 17 2018 12:10pm
...cut the well documented grey wire found on the four wire PAS connector within the controller. Splice a wire to each end of the cut grey wire ends and run these two wires to the throttle.
Running a signal through a mechanical switch is workable but the failure mode for intermittent contact or broken wire is that the switched functionality is inoperative (no PAS). I think you would be better off to just tap and run the grey wire to the switch and short it to GND with the throttle switch to deactivate PAS. In this case the failed switch or broken wire failure mode leaves PAS working and the logic signal stays in the controller instead of being routed around the bike.

Just a thought...
Visit Grin Technologies at www.ebikes.ca
Build Thread: 2WD Yuba Mundo V4

User avatar
Rassy   1 MW

1 MW
Posts: 1979
Joined: Apr 08 2007 10:58am
Location: Eugene, Oregon USA

Re: Throttle Behavior

Post by Rassy » Apr 17 2018 3:22pm

EDIT: I was wrong, the main wire harness enters the controller near the bottom. This controller does not have the gear sensor wiring. Sorry about the terrible picture quality. :D

Fechter, here's a couple of pictures with the main harness wires exposed within the controller area:
100_0835.JPG
100_0835.JPG (76.45 KiB) Viewed 819 times
100_0834.JPG
100_0834.JPG (92.07 KiB) Viewed 819 times
-Rassy-
One Tadpole Trike equipped with a Bafang BBS02
https://endless-sphere.com/forums/viewt ... 28&t=69419
One Delta Trike equipped with a Bafang BBS02
https://endless-sphere.com/forums/viewt ... 0#p1291260

User avatar
fechter   100 GW

100 GW
Posts: 12911
Joined: Dec 31 2006 3:23pm
Location: California Bay Area, USA

Re: Throttle Behavior

Post by fechter » Apr 17 2018 4:05pm

I see blue. It should be possible to tap into the wire there.

When I want to tap into a wire like that, I try to keep the copper intact and carefully cut the insulation in the desired spot, then just push the insulation back to expose some of the copper. Solder tap wire and tape.

It might be a while before I can test mine.

Next step would be to put a jumper between the black and grey wires on the PAS sensor cable and see if it disables the PAS. It might work to just jam a jumper into the back side of the connector so you don't need to cut/splice anything. Alternate method is to unplug the connector and wrap a skinny bare wire around the two pins and re-insert the plug. Just be careful not to short the red wire.
"One test is worth a thousand opinions"

Post Reply