Bafang motor with Nine Continent controller

[Russell]

I recently purchased a Nine Continent kit with a 36V/22A controller which can handle 48V (63V caps) so I decided to try it on my Raleigh E-bike equipped with a Bafang front motor. The stock controller that came with the Bafang is a little 36V/15A unit with 50V caps which could very well blow up if I would try running it on a 48V battery pack. To my surprise the 9C controller worked perfectly with the Bafang (well at first, but read on); I didn’t have to swap a single wire I did reterminate the DC input and motor power connectors with Andersons to match my present configuration.

View attachment 4





View attachment 1



I had the original controller mounted on the handlebars with the Watts-Up meter on top which made it convenient to monitor the battery pack however I'm going to use this bike to test out the 9C motor thus having it in back will make that easier due to the short length of the 9C motor cables.

After lengthening the trigger throttle cable and running it and the Bafang motor cables to the rear of the bike and dressing them with a few tie-wraps I was ready for a test ride. I didn't expect any major difference between the controllers and there really wasn't much. The speed was the same unloaded, a tad over 20 mph, and on level ground with a combined weight of 265 lbs it was ~16.5 mph. It did feel a bit stronger accelerating but when you're only going 0-16 mph it's not a big deal.

All was going well until about 5 miles into the ride when the controller started cutting out. It didn't seem to be hitting a current limit since it would accelerate fine but then it would intermittently just cut out. I would release the throttle to reset and squeeze again and I'd be off. At 8 miles I stopped to check the Watts-Up meter which was in the rear bag. It showed the current had peaked at 24A and the minimum voltage was still over 34V. The LVC of the controller is supposed to be 30.5V so THAT should not be the cause of the controller cutting out. Also while the controller was in the rear bag it was cold out (low 40's) and the controller itself wasn't even warm to the touch. I had already used 2.5Ah in just 8 miles when normally I'd go at least 13 miles on that much juice. Anyway since there was obviously a problem I turned toward home early.

At 15 miles the controller was cutting out with much greater frquency. I stopped again to check the meter and still the Vmin was over 33V. I had also used up 3.9Ah. At 17 miles the the thing was cutting out all of the time unless I used ever so little throttle. I had the wind at my back so it wasn't a biggie riding the bike without power.

When I arrived home I had gone 18.75 miles at an average speed of 15.8 mph or pretty much par for the course. I had used 4.239 Ah, 153 Wh or 8.2 Wh/mile with peak watts of 862. Usually I use about 6.5 Wh/mile and the peak watts is about 550. The lesson here is that a higher current limit isn't always necessarily a good thing. I wasn't traveling any faster it's just when accelerating the controller will often max out and in this case that maximum is 60% higher than before. If you want to ride as economically as possible the better choice is the controller with the lower current limit. I didn't do any testing on hills for this first ride and with the controller cutting out it wouldn't be valid anyway so I can't say if climbing was enhanced.

I don't know for sure what was causing the controller to cut out but it appears to me that the LVC might be quite a bit higher than the 30.5V written on the controller. Since the original 31.5 V number is crossed out I wonder if it was modified and they messed it up :?

I put the bike on a stand and used a rubber band to hold the throttle wide open. Even though the pack was nearly depleted the no-load current was insufficient to cause any sag in the battery voltage which was now at 36.1V. With the motor spinning I wiggled every connection with no problem seen.

I left the motor running...and running...and it's still running now.

-R

 

[dnmun]

because of the internal resistance of the cells, the cell voltage when discharging a large current will be much lower than the resting cell voltage. usually called voltage sag. it seems your BMS can detect the sag and shuts down, sounds like you could use a larger battery, or bigger C.

 

[Russell]

because of the internal resistance of the cells, the cell voltage when discharging a large current will be much lower than the resting cell voltage. usually called voltage sag. it seems your BMS can detect the sag and shuts down, sounds like you could use a larger battery, or bigger C.

Yes I'm aware of the voltage/current relationship

I'm using SLA's so no BMS.

The voltage never sagged to the LVC of the controller.

The controller was cutting out seemingly randomly even during a flat cruise when current should be low.

Motor is still running, when it stops the WU meter will display the Vmin and I'll know what the LVC of the controller is for sure.

-R

 

[Russell]

I don't know for sure what was causing the controller to cut out but it appears to me that the LVC might be quite a bit higher than the 30.5V written on the controller. Since the original 31.5 V number is crossed out I wonder if it was modified and they messed it up :?

I put the bike on a stand and used a rubber band to hold the throttle wide open. Even though the pack was nearly depleted the no-load current was insufficient to cause any sag in the battery voltage which was now at 36.1V. With the motor spinning I wiggled every connection with no problem seen.

I left the motor running...and running...and it's still running now.

-R

Well it wasn't cutting out because it was hitting the controller LVC. I verified the LVC is 30.25V and even when it hits that it doesn't cut out rather the power just drops while it tries to maintain the setpoint.

While the motor was running I pulled on every cable, banged on the controller and there was no problem. At one point I even yanked out a motor phase connector and the motor didn't stop it just sounded angry. I even pulled apart the throttle connector while it was wide open and put it back together and the controller didn't mind.

Hmmm...more testing tomorrow :?

-R

 

[Russell]

Weather kept me indoors for the past few days but I finally got out today to further test the 9C controller w/Bafang motor combination.

Set-up:

36V/22A controller from a Nine Continent kit was connected to a 36V Bafang motor with NO wiring issues; plug 'n' play as they say Teamed with a 36V/9Ah SLA pack which I have been using with the original controller with absolutely no problem.


Problem:

5 miles into the initial ride the controller started cutting out randomly. Problem got worse until at 17 miles it was useless to try anything but minimal throttle.

According to the Watt's-Up meter the battery pack never reached the LVC of the controller which is spec'd at 30.5V.


Tested to find true LVC:

When I got home I let the motor run on the bike stand. The starting voltage was 36.1V. After a few hours of running under these no-load conditions the pack voltage hit the LVC which was found to be 30.25V.


Reconfigured bike:

To see if the problem was purely one of voltage I installed a 6V SLA in series with the 36V SLA pack.

I had installed the controller in my rear trunk with the FET side DOWN and while it was chilly out (low 40's) I thought the problem may be heat so I moved the controller to the top compartment of the trunk bag. It still is enclosed but it would be lying flat for the next test with a bit more breathing room.

I had my original controller mounted on the handlebars but when I switched to the new controller I lengthened the throttle cable and ran it and the motor cable down the side of the top tube right next to one another. Since the problem seemed intermittent I thought perhaps current from the motor cable was inducing noise onto the throttle line. I also noticed that each time the cut-out happened my wireless speedometer would briefly register a higher MPH. I don't have a clue why it might get worse during the course of the ride. Anyway to be safe I ran the cables down opposite sides of the top tube for the next ride.


The ride:

Sunny and cool in the upper 40's.

I was very happy to see absolutely NO cut-outs, glitches, or burps from the controller today. I traveled one of my standard routes of 24.17 miles. My average time was the best yet at 17.7 mph due of course to running on 42V. Average Wh/mi was 8.6; higher than my normal 6.5 but this was almost 2 mph faster than my last ride, though it was warmer with less wind today.

when I got home the batteries were pretty much fully drained. I disconneted the 6V pack and went out for a short ride on just the 36V batteries using the motor quite a bit. The Vmin was 30.99V and still no problem from the controller.


Conclusion:

Problem is resolved but I haven't found the root cause...yet. I know the problem wasn't about hitting the LVC or voltage sag or even voltage per se. That leaves the throttle cable routing and the controller placement. For the next ride I'll put the controller back to the original configuration next to the 36V battery pack and see if heat caused the cut-outs.

-R

 

[Russell]

Today I placed the controller back in the main compartment next to the 36V SLA pack just as it had been during the initial ride. Therefore the only difference between the ride today and the one a few days ago when the controller kept cutting out is the routing of the throttle cable. It was in the mid 30's today so about 6 degrees cooler than the initial ride but if heat was causing a problem at 42F it would probably show up at 36F too.

I set out on essentially the same route as the first ride. At no point in the 18.44 miles did the controller exhibit any problems. Without anything to worry about I was free to concentrate on pedaling and that combined with using a bit more throttle due to the shorter than normal course resulted in my best trip average speed yet on 36V of 16.7 mph using 3.763Ah/136.9Wh or 7.4Wh/mi. When I got home I checked the controller and it was stone cold. Peak amps was 22.23 compared to the spec'd 22A.

So this leaves the throttle cable run along side the motor phase power cable as the last possible cause of the original problem. It seems improbable but one last test could prove it. I am curious :wink:

The 9C controller has proven to be a good companion for the Bafang motor. The higher current limit compared to my little stock controller is most noticeable when scooting across intersections. It's also a simple swap with no need to sort out the connections, just plug it in and it works 8)

-R

 

[morph999]

Russell, that's exactly what happens to me everytime I go out riding on my Forsen hub motor. Let me know if you find an easy fix.

 

[morph999]

what kind of throttle do you use? Mine says ORO on the top of it. I think Cyclone-usa also sells them with their kits.

 

[Russell]

Russell, that's exactly what happens to me everytime I go out riding on my Forsen hub motor. Let me know if you find an easy fix.

It has been fixed since the second ride, now I'm seeing if I can reproduce the original failure. The last thing I need to do is run the throttle cable the length of the bike tie-wrapped to the motor phase power cable to see if this was the root cause of the failure. I've already verified it wasn't a battery/voltage/LVC or controller heat problem and the cable routing is the only thing left that I changed after the initial ride to fix the problem.

I am using a modified scooter trigger throttle ( http://www.endless-sphere.com/forums/viewtopic.php?f=2&t=8026 ). Until I installed the 9C controller in the rear bag I had the original controller mounted on the handlebars with just a very short lead to the throttle. The excess cable from the front motor was coiled up on the right side of the forks then routed up to the controller. In this configuration the two cables did not come close to one another.

It's probably a long shot but If you have your throttle cable running in close proximity to the motor 3-phase power cable you could seperate them to see if it relieves your problem.

-R