Brushless hub experiment: advanced Hall sensor timing

[Blaze]

OK, I have this idea that I came up with when I found a supplier to get hall sensors to fix my brushless hub motor. After I replace the bad hall sensor in my motor, I will install an entire extra set of hall sensors, only the new set will have advance timing. In theory, I may be able to run the motor using the stock sensors with normal timing, then hit a switch to change over to the second set of sensors with advanced timing, which should make the motor go faster. Low speed, high torque with the stock sensors- high speed, low torque with the advanced sensors.

Total cost should be around $25 or so. Hall sensors can be ordered from Digikey.com (p/n 480-1999-ND) for a couple bucks, the neighborhood MarVac has 5vdc DPDT micro-relays for $8 (you would need 2), and you can get a switch for a couple bucks.

The advanced position I am going to try will only require me to grind one new notch into the motor to install a hall sensor; the other two will fit in with no modification.

Any thoughts on this?

Here is how it gets installed...

 

[gone_fishin]

"superdave! are you alright?"

..."no, you putz, i'm not alright...i've got a set of handlebars wrapped around my ribcage. somebody cut the electrical power, it burns. and call an ambulance!"

 

[Blaze]

I'm actually building it now

Well, I got all my parts and I have the diagram drawn. Time to get this thing built. Last night I ran the extra wire I need to switch speeds. That will probably be the hardest part Photo attached). Everything else will be time consuming and tedious, no doubt, but getting that wire in there sucked. It broke like 5 times when I was trying to thread it through.

I did it by first threading through .032 safety wire. This in itself was difficult. I had to pull the wiring harness out of the axle as far as I could, stick the wire down as far as it would go, then pull the harness back through into the hub with the safety wire inside the harness. If I was lucky, I could see the safety wire sticking out the other end and I could pull it through. If not, I start over.

Once the safety wire was through, I soldered the 24 gauge wire to the outside end of the safety wire. This has to be done in such a manner as to creat a very thin and smooth solder joint. Too thick, and it won't fit through. Not smooth enough, and it will tear the insulation of the other wires.

Lastly, I pulled the safety wire all the way through the axle. When I got to the end of the safety wire, the new 24 gauge wire was pulled through behind it. This also took several tries. The 24 gauge wire broke several times., requiring me to go back to step one and re-thread the safety wire again, then re-solder the wire, etc.

I first tried using .020 safety wire because it is thinner and more flexible, but for some reason the solder would only work on the .032. I don't know why.



Here you can see the safety wire threaded through before the new wire gets soldered to the end of it. Note that the harness is currently pulled out of the axle, exposing the wires. You can also see where I spliced all the wires to remove the damaged section from my wheel spinning out of the forks. Yeah, I insulated them with duct tape. Not very professional, but hey, it's not like anybody's going to be flying it.

 

[Blaze]

the wiring diagram

Here is the wiring diagram in case anybody else wants to do this. Not exactly beginner stuff, but if you have some basic experience with electronics, you can build this (after I see if the advance timing even works, of course).

btw- I decided to switch the ground instead of the voltage because I am working with 5vdc. I can grab a ground from any one of the circuits, but the 5vdc can only be had from a few wires. This way, I can splice into a ground wire at the throttle circuit and just run one wire down to the hub instead of running the 5vdc from the hub to the switch and back to the hub. If I had schematics for the kit's wiring, I might have more options, but I know a ground is the same everywhere on the bike.

 

[Blaze]

Progress...

Removing the hall sensor was pretty much what I expected. I chiselled it out bit by bit.

 

[istbenz]

excellent. i'm following this closely and i'm getting a kick out of your project. keep the updates coming.

 

[Blaze]

sensors installed

I'm glad you like the updates. Here is the next step:

You can see where I have ground out the iron core to make room for the advanced timing sensors. Not only did I have to grind out the notch in the middle of the core for sensor 5, but sensors 4 and 6, which fit in between the cores, also needed grinding work to make them fit. The cores don't have enough room between them to install them at the same level as sensor 5, so I had to widen the gap. Be REALLY careful if you do this. The grinding wheel is right next to the windings. This would be a really easy way to ruin your motor.

In the picture, you will notice that there are two complete new sets of installation notches. One is the advanced position, where I have the new sensors installed. The other set actually retards the timing. Why grind out notches before AND after the original sensors? Because it was a stupid mistake. I realized that I went the wrong direction and had to start over and grind it all out again.

The notches are big and ugly, and I'm not sure how much effect it will have on the magnetic field. Probably not too much just in that one area. I thought about whether or not it was a good idea to grind out the metal, but then I came to this conclusion: f**k it.

The picture isn't the best focus, but here it is. Photo is two original hall sensors, one new sensor to replace the failed one, and three new sensors installed for the experiment. Wiring is next.

 

[Blaze]

it pretty much doesn't work

Well, guys the project is pretty much finished. I wired it up per the wiring diagram and tried it with the front wheel off the ground. In normal mode, it worked great. It even goes about 3mph faster than when I first got it. This is probably because the bearing weren't funlly seated before. I hammered them in pretty good when I reassembled it. I didn't change all 3 original hall sensors, just the one bad sensor. The sensor I got from Digikey worked great.

When I switched over to the advanced timing sensors, however, the motor made a lot of noise and ran pretty slow. It wouldn't start from 0 mph. I had to spinn the wheel before giving it throttle. It ran, but it ran very poorly. Perhaps the timing was advanced too much. Brett White, who does a lot of work with these things, recommended placing the sensors directly next to the old sensors, but it was just too easy to place them in the spots where I did it. I didn't want to grind too much metal in case it didn't work anyway.

Later I will troubleshoot it too make sure everything was doinf what it's supposed to, but for now, the experiment has proven that there is nothing to be gained by placing sensors where I put them. If I ever have to open the hub again, I will move the new sensors to a less advanced position and try again, but for now, everything is working again, so I'm just gonna enjoy riding around on it. All of the mods will stay in place, they will just be disabled.

Anyway, here were the final steps. It may not have worked, but you can see my years of aviation electronics experience is the workmanship. 8)


All sensors installed and wired through the micro relays. Looks really good here.

From this angle you can see the relays and mounts.

Another shot of the sensors. That was a lot of work!!!