CroBorg Super Commuter

Had to finish up the taxes last night, and one thing I noticed is the mileage on my CroBorg for last year was nearly the same as the mileage on my 4Runner. I suppose that is a turning or perhaps a tipping point, when they are equal. It is a significant number of eBike miles (about 4000 of the 6000 total commute annual), so about 2/3 of the commute runs are eBike over the year.

I wonder what the deductible cost per mile is for an eBike? If I include my time and all materials used and consumed, it must be at least a buck a mile. :)

Yesterday afternoon the commute was warmer, I removed the liner from my riding pants to avoid a melt-down. I put them back in for this morning's ride. It is still nippy in the morning, but I'd call it ideal riding weather, ranging from about 50-75 F. Getting to be a lot of other bikes on the road as well.
 
Just checked cell balance, all cells at 3.95V during charging at 12A at the 5AH point of about 13AH total to charge. Not the best test, should check again at end of charge.

Rear tire remaining 0.0320 of 0.375 (inches) so wear is about 92%. Not quite to the wear bars, but close.

Cycle Analyst Odometer shows about 6500 miles, some were lost in CA reset, probably less than 500. This is the second rear tire.

Front tire has 0.108 inches tread depth remaining. Original front tire.
 
Checked the 18S 32AH MultiStar pack on the CroBorg at the end of charge (when charge current was down to 70mA) and it was mostly 4.16V with a few 4.17V readings (on a Battery Medic with resolution of 10mV).

This is excellent balance, so I don't need to worry about that. The most I'm drawing from this pack is about 13 amp hours (out of 32). So it is not a deep cycle. I use about 10 amp hours on the afternoon commute, and 13 on the morning trip.

The electric brake lever is working well, though I'd like more spring tension. I'm collecting parts to try out the cable operated sliding potentiometer. That will have both cable resistance and a stronger spring, so it might feel a bit more natural.
 
20150427_095147.jpg


I installed the Domino throttle yesterday. The Magura throttle was fine, the Domino offers a few improvements. It is slightly larger diameter, softer rubber, and has a 60 degree rotation instead of the Magura's 75 degree. The Domino also has a microswitch that is open when the throttle is released, and closes upon slight rotation, which I'm not using yet but could come in handy.

I've tested the Domino for some brief test runs and half a commute so far. It does seem easier to hold on to with the grippy surface, but wearing gloves hides some of the improvement.

There are reports the Domino is more reliable, but my experience with Magura has been good so I've no failures to compare it with.

The Magura throttle is hard rubber that feels like plastic, so the Domino's patterned soft rubber is nicer. Hopefully no one is riding bare handed though, and with gloves on it is more difficult to feel the difference.

Reducing the angle of rotation is good for reducing wrist strain, but it does require more precise wrist angle. The Domino is so grippy it is very easy to hold onto with less force where the Magura might slip, but the Magura has a lighter spring. So in practice they are both easy to use.

The waterproof plug on the domino is on a short cable. It is a fairly large plug, and the cable is not quite long enough to get the plug into the covers on the Borg. It is a nice Tyco AMP water resistant plug, but it is a bit large for an eBike, and the short cable puts the plug in a less than ideal position.

I added a 560 ohm resistor to the positive side of the 5K Domino pot to match the existing setup and allow open ground detection. The throttle feeds the CAv3 which feeds the Sabvoton. I could whack the big connector off and put the CAv3 connector on directly, the cable is long enough to reach for that. Perhaps later.
 
Rode fast and ebraked hard on this warm commute morning under crystal blue skies with Friday light traffic. Measured the CroMotor temperature after sitting for a few minutes at work and it was up to 139F (60C) near the magnets. Nice and toasty, but not quite hot.

Had a good test of the rear eBraking twice on this morning's commute. First one was on a right turn at a green light, I slowed and started the turn but a lycra clad road biker was making a U turn at the mouth of the street I was turning into. I wasn't going very fast, and as the eBrake came on quickly the rear tire skidded slightly, but kept turning and grabbing pavement so control was easily maintained. I was starting to clamp on the front dual disc hydraulics, but the cyclist stopped mid street so I didn't have to execute a full emergency stop in the middle of the turn.

Near the end of the commute run there is a hard right turn in the midst of a steep 15% downgrade, I was up to 37 when I eBraked hard (saw 3.5kW on the CA) down to 15 for this reverse-banked turn. More hard eBraking in the driveway and these descents are likely much of the motor heating's cause. Of course humming along through the park warms up the motor too, so it wasn't cold when the heavy eBraking started. :)

Another fine eBiking day on an excellent commuting machine. 29 minutes. :p
 
Periodically I re-check the tension on the CroMotor lug nuts and NordLock washers, and after this morning's eBraking events it seemed to be an appropriate moment to review their condition.

I use a combination wrench about a foot long for this as it affords both a good grip on the nut and good leverage, and is light enough to carry in the trunk.

Neither nut would move at all. So they are staying fully locked tight despite frequent heavy acceleration and braking cycles. at the 4-6 kW level.

Excellent. 8)
 
Some months ago I upgraded the front brakes from dual 160mm four pot Gatorbrakes to dual 203mm Tektro Aurigas. Besides the larger and thicker discs the Aurigas offer an eBraking switch and a parking lock.

Today I noticed that the braking power of these brakes was not very good, I've been using primarily the rear eBrake for nearly all braking so I hadn't been using the front discs much. So I exercised the dual discs thoroughly and the braking power came way up. I'm not sure if I never had broken them in properly (I did try), or if they were glazing over due to some damp weather and all the light braking.

In either case, the brakes are back. :twisted:

I will remember this in the future, don't let those disc brakes relax too much and "go to sleep", they need a bit of periodic exercise.
 
20150521_090216.jpg


I picked up an ebikes.ca Satiator charger from Thunderstruck Motors. They shipped quickly and the charger arrived the next day, nice service!

So far I've only run a charging test on a 10S Lipo pack. The Satiator is a very sophisticated charger, the present model has a voltage limit of 60VDC. So it is not sufficient for my 75V 18S pack by itself. I have a couple of bikes the Satiator will work for, but I'm also considering adding a series DC LED supply to allow it to handle the 18S pack.

The 240 Watt Meanwell HLG LED supply is very close to the right size to bolt to the back of the Satiator. The Satiator is just a bit shorter than the Meanwell (and a bit wider and thicker), and the mounting holes nearly line up when placed back to back. An adapter plate would not be difficult. The weights of the Meanwell and Satiator are similar, the specifications indicate a bit over 1kg each. Both the Satiator and the Meanwell HLG units are rated for vibration and water resistance.

One interesting feature of adding a supply like this is the increase in available charging power. The higher voltage Satiator (when it becomes available) will be limited to about 4-5 amps at 75V whereas adding the 24V Meanwell allows the standard unit to reach 7-8A. Using a 30V Meanwell would allow the full 8A. Adding a second supply does increase weight and complexity, so it is, as usual, a tradeoff, especially if carrying this charger onboard.
 
Here is the Domino Throttle to Cycle Analyst v3 cable that I made earlier and have been using. Last night I cut the throttle cable at the connector and made a direct to CAv3 cable for the Domino to clean up the cabling on the CroBorg. The cable pictured below worked just fine, but I didn't care for the large water resistant connectors on the bike. I'm sure they are good connectors, but I decided to clean up the wiring a bit.

20150521_074456.jpg


20150523_103343_001.jpg


This cable contains a 560 ohm resistor from the top of the throttle pot to the +5V so the CA can detect and avoid a runaway throttle on a broken ground wire. The Domino switch is brought out to the two unterminated wires near the CAv3 connector on the right.
 
Finally got to the motorcycle shop yesterday afternoon and had the new rear tire mounted. I can do these myself but it takes six hands and is not a fun job with these Moped tires and rims. Last time I did it the tire started out right but ended up, after all the problems, rotating the wrong direction. Not that it makes a huge difference, but I try to follow the recommended setup. I'll have to get the odometer value, but this one lasted about 3500-4000 miles. I also swapped in a new freewheel, though the last one is still working fine since I started lubing it monthly or so. Cost $25 to mount plus a $2 tire disposal fee. Used a new tube with it as they recommended replacing both. Tire was $26 and tube was $11 and the freewheel was about $7. Reinstalling the motor wheel and reconnecting the wiring is a bit of work, but it went well and everything worked on the test ride.

I wrote up the procedure for tire changing, it is a bit complicated. I'll post it later.

I also added a schematic to the Domino Throttle cable posting above. This adapts the Domino to the CAv3 Throttle input.
 
20150524_093429.jpg


On a brand new rear tire.

This inexpensive dirt bike rack works well for a hefty ebike like the CroBorg.

Not a lot to report on the ride. Nice weather, nice bike paths most of the way. Also visited my Dad and showed him the bike.

Tried out a handlebar phone mount I got some time ago, but the newer phone I have doesn't fit. Very few if any water resistant mounts large enough to enclose a Samsung Note 4.
 
I picked up a keyswitch awhile back and was looking at different ways to mount it on the CroBorg. I noticed that it would not quite fit into 1/2" size PVC, and that same PVC was slightly too large to fit into the large open holes in the triple clamp on the front fork.

So I chucked a length of the PVC into the lathe and turned out the ID on one end to fit the keyswitch, and the OD on the other end to fit into the forks.

Then I wired up the switch, slid it into the pipe and filled the wiring end with epoxy which takes care of insulating and strain relieving the cable.

20150525_172143.jpg


A 1/2" PVC cap can be used to cover the keyswitch and keep things out of it when parked. I could embed the key in epoxy in an endcap and make a waterproof key..
 
No rain this morning, but the wet pavement from some overnight drizzle in the hills made the last half of my ride the wettest I've ever ridden (it was dry at home, if it had been this wet I'd not have ridden). Hard to see with all the mist on the visor. It will be nice on the way home though, and I did pass a LOT of slow traffic on the way in.

The rear wheel lug nuts are still tight. After changing the tire I need to keep checking them daily for awhile, just to be certain. Odometer showed 6854 miles (some are missing) at tire/tube/freewheel change.

Here's a draft of the CroBorg Rear Tire Service Procedure:

The CroBorg rear wheel has a hubmotor and requires some special considerations when changing the tube, tire or freewheel to insure safety against electrical shock, physical injury and damage. The front wheel is not covered by this procedure.

0) Prepare for a rear tire, tube and/or freewheel change by obtaining the parts and tools ahead of time. Use: 17x3.0 Michelin Gazelle tire, compatible tube (best to change tube with new tire), 16 tooth Dicta freewheel with English threads and modified freewheel removal tool (to clear hubmotor’s extended freewheel threads). It is recommended that this cheap Dicta freewheel be replaced at each tire change and periodically lubricated. (Eventually - turn the threads shorter to clear and be compatible with and change to higher quality and serviceable White Bros freewheel).

1) The motor cabling connects directly to the motor controller via ring lugs. For safety the power to the motor controller must be removed before contact with any of the electrical connections. De-energization is accomplished by:
a) Turning the main circuit breaker off and unplug the main power connector. Insure this connector does not get reconnected prematurely, review it each time you return to the bike to continue the work.

e) The lower covers may be removed and the motor phase wires removed from the controller, keeping track of screws and washers.
f) Pre-loosen the axle nuts (since they may be quite tight), but leave them with some tension. They will be fully loosened later.

2) The bike must be supported at the handlebars and under the front of the swingarm to lift the rear tire clear of the floor.

3) Remove the rear wheel by:
a) Cut the zip ties that hold the motor cable to the frame carefully to avoid damaging the finish.
b) Disengage the chain and fasten it out of the way with zip ties.
c) Loosen the axle nuts, taking care not to disturb the supported bike.
d) Pull the wheel rearward taking care to avoid getting fingers near the dropouts which will spring in slightly when the axle exits to the rear.

4) Service the wheel - change the tube, tire and/or freewheel as needed:
a) Zip tie the cabling to the spokes to keep it safe and out of the way.
b) Take the wheel, new tube and tire to a motorcycle shop and have them installed. Verify the direction of rotation vs the tire markings, hubmotor cables exit the axle on the left side. Verify tire pressure. Verify the tire is properly seated fully and symetrically on the rim.
c) Remove the freewheel and install a new one using the modified tool, using no-sieze on the threads.

5) Reinstall the rear wheel
a) Verify the main electrical connector is still disconnected and the bike is safely supported.
b) Use a pipe and web strap to separate and hold the dropouts to the proper width. This is necessary since they are slightly closer together than they should be, eventually some careful bending can fix that and eliminate this step.
c) Insure that the washers are in the proper order (NordLock pair on the outside of the dropouts, all others inside) and that the axle is fully forward in the dropouts. Retorque the axle nuts (and recheck them frequently in the coming days).
d) Reinstall the motor wiring connections, power up and test with rear wheel in air (insure that the bike is well supported during this test). Power down.
e) Remove swingarm support, lower rear wheel to the ground.
f) Reroute the wiring, install zip ties and replace all covers, avoiding contact with the now-energized electrical connections. (if contact is necessary then de-energize before proceeding).
g) Reinstall chain.

6) Final Inspection
a) Check wheel nut torque.
b) Inspect wiring, zip ties, rack clearance, pedal chain.
c) Insure all tools removed, all covers replaced and fastened.
d) Insure power is on and the on-off controls function normally.
e) Test ride, use gentle acceleration and braking and keeping speed low. Verify pedaling as well as motor operation.
f) Check wheel nut torque, inspect chain.

7) Ongoing Monitoring and Maintenance
a) Verify wheel nut torque daily for 2 weeks, then weekly.
b) Verify tire pressure weekly.
c) Relubricate the freewheel monthly.

Edit - Removed the zero voltage verification since the battery power connector isolates the only hazardous power source.
 
20150528_190834.jpg


The 100A circuit breaker suggested earlier in this thread arrived, and today it was installed. The previous 63A breaker tripped on a long hill climb with the controller set to 80A battery, this one should hold. It works out better in that the input lead is on top which is the natural direction for this battery layout whereas the other breaker was reversed and the existing wiring was a bit short for that. This one is also smaller but thicker which fits the space nicely and gets a bit of support from the aluminum bracket. It is much harder to turn on, however. The force required is surprisingly high. The wire entries on this breaker are huge, making the #8 AWG wire look too small, but it holds well so it will probably be fine.

I have not been using a precharge circuit, some say they are not needed. We'll see if the breaker fails from the closing arcing. They are certainly designed to open under load, but are they really designed to close into a discharged capacitor?
 
I've seen you a couple of times on the road, I was on my EVG. There is also a guy with recumbent etrike in the area. We should collaborate :)
 
Lost one of the cap nuts on the ATA Speed drive recently, so today I took one to the hardware store and measured it to be 5mmx0.8 thread. I picked up some zinc steel nuts with nylon lockrings to use instead so they won't come loose or spin off if they are loose. Stainless would have been better, but with the lube coming from the planetary gears inside these are unlikely to rust.

Speaking of lube, they recommend re-greasing the ATA Speed Drive every 2-3K pedal miles, or using an oil even more often. I should do that soon, though my pedal miles are probably not there yet.
 
Yesterday's homeward commute was windy and since I was late there was less traffic. Established a new record for that leg, just under 27.5 minutes. Maybe that was the tailwind helping out, who knows. I had to manually trigger one traffic light and it took awhile so it could have been even faster.

The blustery winds that occur along this route are very turbulent. Kind of fun, but it keeps you on your toes and made me glad I'm wearing motorcycle armor both for the protection and for the insulation.
 
In the last few days I've broken my time records for both commute directions. Not that I'm aiming to do that, but since the GoPro reports the length of camera time it is easy to see as I shut the cam off. Slightly under 27.5 minutes.

Some of this is due to school letting out and the traffic improving, and lately it has been windy, a tailwind actually helps a little. :)
 
Beautiful warm morning, almost too warm for the riding suit.

After sitting for 15 min or so the motor is 140F 60C on the outside. Quite toasty. This motor is just right for this job. Any less would be really stressed. I use the torque both for climbing hills and for ebraking down the steep stuff. 28 minutes today, though I've had some 27 minute commutes in the previous week. School is mostly out, vacations have started, traffic is reduced. Warmer weather helps the battery voltage a bit, gain a little speed there too.

Excellent commuting machine, living up to its name.
 
I forgot to charge last night so I made a round trip on yesterday's charge, 26 miles on 22 amp hours. The pack capacity is now 32 amp hours, so this was not a problem, the old pack was 20 amp hours and a round trip involved slowing down and probably pushing the bike up some hills (been there, done that). :shock:

I like this outcome much, much better! :D

Edit - checked the battery pack cell balance at the end of this charge, still excellent!

For those who want to see an example of my commute in video, I have one section online now. This is a RAW GoPro MP4 file, which is only slightly compressed, so 1.9GB for a 21 minute section that covers most of a homeward commute. It starts near work, climbs up the steep 10-15% hill, traverses through the winding scenic roads in the park, and then along the bike lane by the highway adjacent to the lake, ending mid lake.

This video is HD, sharp and wide angle, taken from the front fork facing forward and including the front tire (so you can see the suspension working). All the pixels are moving radially out from the center, and compression algorithms don't get much traction on that. Compressing it farther tends to detract significantly from the appearance.

Only download this if you have space for a 1.9GB GoPro file. One viewer said it took 6 minutes to download, depending on your connection speed it could take a lot longer. This video won't be online for long. Download it and then play it, don't try to play directly from the link.

The speed appears fast, the camera position is slightly above the 24" front tire and the low perspective makes it look faster than it does from the rider's viewpoint. On the downhills the speed is below 35, the rest is even less.

https://www.amazon.com/clouddrive/share/aPU2_TOkXoPeOXm1Q-lCiwLzrMP70u79UbQ4zdxBrlg?ref_=cd_share_link_copy (right click link and select download)
 
Climbed the entire steep section of the hill yesterday in one continuous pull, last time I tried this the 63 amp circuit breaker tripped near the top. This time with the 100 amp circuit breaker the power stayed at 5.6 kW most of the hill which is 80 amps, and no trips. This tells us that it is better not to depend on the temporary overcurrent spec for the breaker, it works most of the time but not all the time.

Looking good!

20150528_190834.jpg
 
On the way home I heard a distinct POP that sounded like it could be a lipo cell popping as I rode along. I pulled over and opened the covers to see if anything looked awry with batteries or anything else. I didn't see any sign of a problem, and the pack voltage was fine under load so I continued home, keeping an extra watch on the pack voltage. Later I reviewed the video and found the sound, it is much weaker at the camera's location than it was to my ears. I'm going to not charge the pack tonite and continue to check it closely as I charge it later this weekend, and do some extra balance voltage measurements. It was probably a noise from the golf course nearby, though I pass it twice daily and haven't heard anything like that before, and I was a long way from the tee locations. If the noise had come from the pack it would have been close to the camera's rear open side where it is more sensitive to sound, so perhaps the weakness of the sound indicates it was coming from a higher level. The helmet makes sounds appear to come from below since that's where it is open.
 
Back
Top