DIY Fix for BBS02/BBSHD Pedal/Throttle Issue

[amberwolf]

Sometimes.

Sometimes not.

Depends on what firmware Bafang put on any particular unit.

See the thread linked in the post above yours for a potential fix if yours does not work.

 

[LowkeeRider]

I ordered a programming cable I could reprogram it to work with 52v battery is that right?

 

[LowkeeRider]

Thanks, I will read that thread.

 

[amberwolf]

I ordered a programming cable I could reprogram it to work with 52v battery is that right?

Maybe.

Depends on what Bafang did to whatever unit you end up with.

Sometimes the updates work, sometimes they don't. :/

 

[maultasche]

My new bafang BBSHD that I ordered a couple of weeks ago has the throttle issue that was described here, really annoying

Firmware update sounds risky. The hardware solution from fechter looks great, but I feel indisposed cutting deep into a new controller.

Is it possible to connect one single wire to the PAS signal wire, lead the wire out of the controller housing and put the fechter transistor circuit somewhere at the thumb throttle and use the ground of the throttle ?

So can i ground the PAS signal also at the throttle ground ? Is that the same ground ?

Sorry maybe thats a stupid question, I am not very familiar in electrical engineering. Thank you in advance.

 

[fechter]

Is it possible to connect one single wire to the PAS signal wire, lead the wire out of the controller housing and put the fechter transistor circuit somewhere at the thumb throttle and use the ground of the throttle ?

So can i ground the PAS signal also at the throttle ground ? Is that the same ground ?

Yes, that should work. They all have the same ground. They make a throttle extension cable you could cut and use if you don't want to hack into the original wiring.

 

[maultasche]

Yes, that should work. They all have the same ground. They make a throttle extension cable you could cut and use if you don't want to hack into the original wiring.

Thanks for your fast reply, I will try it in that way

 

[LowkeeRider]

Edit: I want to make a clarification, this is not a fix to the throttle issue described on this thread but rather just my satisfaction by changing the settings on the controller using Kelper profile. I rode on PAS 5 or 6 mostly, smooth start and fast off the line. This is one possibility for those who don't want to dig into the controller.

I am so thrilled that I don't have to do this mode any more.

I have 2017 BBS02 750 watt
(Stock: no throttle at PAS 0, only has 5 assist levels and no throttle override)

With the programming cable, I was able to reconfigure the controller using Kepler's values. The bike starts smooth no chain grinding.
Throttle works on PAS 0, has 9 levels of assist with throttle override.

I am not tech savvy, took a while to search, download programs and how to assign COM port. Some instructions had to read it three, four times to understand. Along the way, I had help, Google was by my side.

 

[Rassy]

Hi LowKeeRider, glad to hear you are happy with your BBS02 after updating some of the parameters.

However, there is nothing in your post to indicate that you have tested or corrected the problem that was addressed in the first post of this thread. No offense meant, but I would be happy to have you answer a couple of questions so as not to confuse the issue for others:

When using your throttle while in PAS 0 do you get the same range of throttle response whether or not you also pedal?

When pedaling along in PAS 1 or above, can you "gently" increase your speed by applying some throttle or do you get "full throttle" response unless you stop pedaling?

Thanks

 

[LowkeeRider]

I edited the previous post.

I rode over 150 kilometers since I changed settings in the controller.

No full range throttle on PAS 0, it is a bit abrupt starting from a dead stop but once up to speed it is negligible. No need to pedal to get full power. Got up to 50km. (Showed 999 watts) Display limit, could be more.

Throttle does override on all nine PAS levels but comes full power on, whether no pedal, gentle pedal or hard pedal. There was a few instants of throttle delay, happened on PAS 1-3.

My riding is mostly done in PAS 5 or 6, smooth start yet fast to get off the line, I could hit throttle for more boost if needed. Yea, I am satisfy with the new settings.

Changing setting in controller will not fix the throttle issue described on this thread.

 

[LowkeeRider]

Finally fixed the throttle.

Today, I spent a good half a day digging away the potting and soldering the circuit. I now have full throttle range.

Thank you Fechter.

 

[fechter]

Glad to hear it worked out.

 

[eBlur]

So glad I found this thread. I bought a new BBS02 from Luna in June 2019 and it appears to have this symptom of jerky throttle when peddling even in PAS0. The very first post in this thread has an edit from May that suggests a "possible firmware-only fix", but as far as I can tell from reading the linked thread there is no confirmation that the new firmware fixes the peddle-with-throttle issue and it seems more about the over-voltage fix.

The first few rides of getting to know my motor I guess I was using throttle-only off-road so didn't notice there was a problem. As I've gotten more confident - I'm finding I'm wanting to peddle hard with a hit of throttle here and there and now I'm fully aware of it. I even bought a new throttle thinking it got damaged in the rain and was only responsive to full-power.

I'm going to attempt the hardware fix to see if it improves my situation. Quick question for the electronics gurus... I have a handful of BC547B NPN transistors (datasheet: https://www.farnell.com/datasheets/410427.pdf) kicking around from another project. Would these work as a replacement for the KSP13BU? I'd love to try the fix today with what I have on hand, but don't want to kill my controller with the wrong transistor

 

[Rassy]

Glad you found this thread, eBlur. The technical expert is Fechter, and if he doesn't see your post here you might send him a private message.

Good luck with your repair. Even though I have an older controller that works correctly from the factory, I just left it on the shelf and continue to use the one I fixed since it works just the same.

 

[fechter]

You won't kill your controller, but that transistor will turn on at too low of a voltage. You could wire two of them together into a darlington configuration and it should work.

 

[eBlur]

Thanks fechter.

My bike is unexpectedly out of action for some suspension repairs - so I may as well order the proper transistor while I work on that... I have a handful of KSP13BU's on the way from mouser. I like the less-components-small-footprint design as it it

What wire did you use to connect with - I can't tell for sure but the throttle cable is skinny - looks like 26 AWG to me. I was thinking I should use stranded-core 24 AWG for hookups, which I assume would endure vibrations better as a soldered connection than solid-core.

Looking forward to picking though the potting

 

[fechter]

That should be good.

 

[eBlur]

Inspired by BotoXbz’s post I thought I also would try digging into the potting at the opposite side of the controller. The only downside is that the wires are quite taught and you can’t pull up a loop to tap into, so I kept on digging until I got to the PCB connection.



I’m a bit skeptical of simply soldering the new wire to the top of of the existing - especially given long term vibration. I was hoping for a western union style splice where I wrap the tap around the existing wire for a good mechanical connection before soldering. However without getting more slack in the wire, I don't think I could wrap it...

I see three possible ways forward - love to pick the collective wisdom of those that did this before me...

1. Try and apply heat to the upper PCB hole and pull the blue wire out. Top-solder in a new blue lead to the PCB and make nice connection to new mini-board (w/darrington&resistor) This seems a bit dicey not being able to see the bottom layer - and risk of molten solder shorting two holes together. Will probably help to hit use a solder sucker once the the wire is out to clean the hole and possible 'under spill' before putting the new one in.

2. Similar to 1 except top solder wire onto upper hole and not pull old wire. This would mean less heat / time spent in that area but seems worse as it still carry the risks above and will have a crappier connection.

3. Abandon this area and dig up potting on the other side to be able to pull a loop and tap properly.

Which way would you go? Better idea?

<Aside>
In a moment of weakness I considered digging out the whole board to fix it cleanly - but then I found this video and saw the terrifying rails of solder underneath to handle larger current. Now I’m slightly scared of my controller. I'm not sure I'll sleep at night anymore https://youtu.be/z_5RlBvoNoA?t=1740 (at 29:01m) Think I'm going to leave the PCB where it is...
</Aside>

 

[fechter]

Personally, I would suggest just tacking the new wire on top of the blue wire. If you heat the end of the blue wire where it goes through the hole, the insulation will melt back a bit. Let it cool and carefully scrape back the melted insulation so you have enough surface area for a decent solder joint. I would avoid pulling the wire out of the hole because it will be very hard to get it back in without removing the entire board. Just tin both wires and lay the new one on top of the existing one and heat. I wouldn't try to wrap it around, but if you have enough slack it wouldn't hurt. Once you get the new signal wire tacked on, you can cover the joint with silicone or epoxy glue to prevent it from moving from vibration. The signal is a very low current, so your main concern is mechanical strength. The glue blob will secure the joint so only flexible sections of wire are exposed.

 

[eBlur]

That’s a good point. I guess once’s is re-embedded in clear silicone it should be protected from movement.

 

[eBlur]

Surgery complete. What’s the best way to “bench test” the fix before sealing it up all and reinstalling?

I tried it - and the throttle seems to engage quickly the same if I’m moving crank or not, but motor seems to spin at constant vs variable speed as I apply more throttle. Perhaps this is because I have the speed sensor disconnected and am in 21h error state?

 

[Rassy]

I think you need to be able to "pedal" in order to test the "fix", because the fix makes the throttle dominant instead of pedaling being dominant. As far as how the throttle works on the bench I found it was very hard to tell the difference between a partial throttle setting and a full throttle setting since there is such a small resistance to the motor spinning.

As I started typing this reply, I knew I had bench tested, but now I remember that I had one of the trikes in the house and my bench testing involved just hooking up the controller wires but not actually installing the controller, etc.

So I think you need to hook everything up to the bike to do a meaningful test. I recall when I purchased an extra controller directly from a China Ebay type seller, I thought it worked properly until I actually took a test on the street. That's the controller that I did the fix on and am still using.

Edit: I read your post more carefully and see you were moving the crank. I think your test was successful based on your comment that the throttle engaged quickly whether you were turning the crank or not. Before your fix the throttle would not have engaged as quickly when you were turning the crank. Try this test: In PAS level 0, apply a small amount of throttle, just enough to make the wheel spin, without turning the crank. Then start turning the crank while holding the throttle steady at the "slow" speed. After the fix the motor should continue running while you turn the crank. If you did this test before the fix the motor would have stopped when you started turning the crank.

2nd Edit: If you don't get this sorted, tomorrow I will do the above test on my delta trike with the drive wheel on a jack stand. This will cause an error 21 since my speedometer sensor is on the front wheel and hence will not be working. I'll report my results here after the test.

3rd Edit: Ok, I did the "bench" test, once in PAS 0 and once in PAS 1 with the drive wheel off the ground and the bike in 1st gear. I didn't remove the chain and it was easy to tell that the throttle modulated the motor speed properly. The throttle worked the same whether or not I turned the crank. When anyone plans to do a bench test like this, it would help if you could also do the same tests before you make the fix. I repeated the tests with the chain off, and it was still easy to tell that the throttle was modulating properly. I do change some of the throttle parameters for soft (slow) starts on all my BBS02 installs.

Good luck with your fix.

 

[Alan B]

Another test would be to put a meter or scope on the pedal sensor pulses. The throttle advance should clamp the pedaling pulses off.

Without a load the BBS's torque throttle will rise rapidly to full RPM, so it is hard to judge the situation from motor activity.

I am going to try a PhaseRunner controller instead of the internal Bafang BBSxx controller to get smoother and more programmable control. This fix is a good way to fix this one issue but changing controllers is another option that brings other benefits.

 

[fechter]

A test you can do just to check the circuit is to measure voltage on the output (wire going to the PAS sensor). Measure vs. PAS sensor black wire or battery negative. On standby, you should see 5v on that line. If the pedals happen to be lined up with the PAS sensor, the voltage might be low. In that case slowly turn the crank until you see 5v. Then advance the throttle and watch the voltage. The voltage should drop to less than 0.5v when you apply throttle. If this works, then glue it up and go for a ride.

 

[eBlur]

Hhmm. Something is not right.

So across grey and black on the PAS sensor I get 2.02v on standby with pedals aligned (otherwise 7.3mV). When I apply throttle for the first 10% bite of the throttle I see the voltage start to drop slowly down to 1.8v before any action at the motor. The motor engages and then each step of throttle takes it down until 0.6V at wide open.

I double and triple checked the transistor leads. Using a single KSP13BU with 100k resistor. Note this is bbs02 vs hd.

Maybe I damaged transistor on install. Glad I bought 5 - will cut and try a new one.

Wish I could do a “install and ride” test but my bike frame is getting suspension pivot bearings replaced so unavailable until next week.