How and where to mount magnetic reed switch for regen brake?

[electrobent]

Hi:

I ordered the magnetic reed switch from Rick in the cold Ontario a couple of years ago and am finally remaking my bike with a controller that can take advantage of a switch. In the first build, I just made it so that it was ergonomically impossible to squeeze the brake and turn the throttle at the same time.

I want to be able to activate the regen brake without using the mechanical brake and thus I am thinking that the best case scenario is if I can connect it the brake somewhere so that a slight squeeze gives the regen and a big squeeze puts the pads in contact with the wheel.

Rick's kit comes with a stick-on magnet it looks like pretty strong sticky but I don't trust sticky. I also have the magnet from my now defunct bike computer and it can clamp on to a brake cable just as well as a spoke. I am thinking that it is better to do this on the caliper side than the lever side because of the extra cabling and the high probability of rubbing the stick-on magnet off with my hands.

Clamping the computer magnet to brake cable solves one problem, but how to mount the switch itself?

Has anyone done this and been happy with their set-up?

Thanks!

--Eric

 

[amberwolf]

I haven't done this *yet*, but what I would do for mounting is dependent on the brake design you have.

Do you have any pics of the wheel-end of the brakes, to see the cabling and associated fixed framework that might be used for a mounting area?

Without a specific setup in mind, I think that one way to do it is to install a straw-like tube over the end of the open sheath coming out of a brake cable end at the wheel end, and mount the reed switch on that, parallel to the cable inside the tube. Then the magnet is secured (epoxied?) to the cable inside the tube. If the tube is movable or extensible, you can recalibrate it's position whenever you readjust the brakes for any reason, if that changes the cable position within the tube.

 

[electrobent]

Here is a picture:

The situation is complicated by the fact that wheel is still not centered.

I need to spend another couple of hours with a spoke wrench!

 

[amberwolf]

Ok, with that setup I'd put the magnet on the cable as you said, and put a little bar (white rectangle) attached to the "noodle holder" (the black slotted metal "U") so that it holds the reed switch (green oval) out near the magnet (blue oval), *almost* engaging the switch but not quite. It could be made of anything stiff enough to stay in place. Just screw it to the U-bracket with two screws (red circles) , once thru any two of the slots, secured with nuts/lockwashers/nylocks (or loctite).

When you squeeze the brake lever a little, then it would push the switch a little closer to the switch and engage it, before it actually starts mechanically braking.

If the magnet is small, though, it will turn off regen after it passes the switch when you squeeze really hard on the lever for lots of mechanical braking. So you might need either multiple magnets or a long magnet with it's poles in the right direction to keep the reed engaged.

All of this is assuming a normally-open reed, meaning one that is open until you put the magnet next to it.

If it's a normally-closed reed, then you need to put the reed switch at the magnet to keep it engaged until you squeeze the lever a little. At that point, it would then close the switch and keep it closed (keeping regen operating) the entire time the brake lever is being held.

The only bad part about using a normally-closed reed is that if the magnet fell off or became misaligned from vibration/etc, regen would automatically begin happening unexpectedly and stay on until you put the magnet back.

 

[mwkeefer]

What type of brake levers do you have on the bike?

My preference is to locate a place within the body of my existing brake levers (Normally I only do my motor wheel too) for one of those tiny little radioshack neodyum magnets. Once I find a suitable location for both the magnet and the hall sensor I simply epoxy them in place and run the smallest wires I can find (normally really thin speaker wire - 26g or thinner) to the proper eBrake trigger. The same principal would apply to reed switches.

The reason I don't mount them on the brake calipers or pull brake arms is because when I did this, the constant movement and direct transfer of force from the wheel being stopped caused issues with the epoxy mounts and in the end they failed. One other thing I prefer with the lever mounted reed/hall sensor, the power needed is available from the thottle run already (yep you can share VCC and GND) and the wiring can be much cleaner and more stealth (as it should be).

Hope it helps!
Mike

 

[rkosiorek]

i also have found that mounting the switches and magnets to the calipers themselves is problematic. just like you i used a screw on spoke magnet. i found that over time the magnet gets loose. tightening it, squashes the cable and makes it tough to adjust. also the cable stretches quite a bit with use and needs to be adjusted often.

the peel and stick magnet uses a 3M foam tape that has the same adhesive as automotive body mouldings do. it is tough. but the surface you stick it too must be very clean and at room temperature for the adhesive to hold properly. it will stick well to most surfaces. it is also meant for those who want the simplest attachment method.

another solution that i have used is to drill a hole in the lever just large enough for the magnet to slip in. then use a bit of glue to hold it in place. as a backup i also used a punch to swage the hole opening to keep the magnet in place. this method is neater and much more secure but requires a few tools to do.

rick

 

[electrobent]

Thanks guys, I'll ponder the brake levers more closely. Maybe I can stick a magnet in the hole for the lead thingy?
Or take that spoke magnet and unscrew the magnet from the rest of it and drill (and tap) a narrow deep hole instead of a short fat one?

Or just get a momentary toggle switch and mount it where I can push against it.

I know, a bite switch!

 

[electrobent]

The attached picture shows what I ended up with--perhaps what Rick intended!

 

[rkosiorek]

yes, exactly right.

how are they working for you?

rick

 

[electrobent]

Works great except for the fact that it is binary: all or nothing. It sure would be nice to have control of how hard it brakes. I have the Golden Motor cruise controller and the brakes slow the bike quickly to the point where the mechanical brakes can stop the bike easily and this was a problem before. But I thought I would be able to to use them to control coasts but I have to pulse them to do this. Golden Motor sells a usb cable to reprogram the controller. If I shell out the $50 or figure out how to mock up my own cable, I will set the amount of regen down so I can use it for coasting down the mountain.

At first it did not want to release right away but that problem seems to have gone away.

Thanks, Rick for a great product and patience.

I did not find your write up until after I installed it.

--Eric

 

[amberwolf]

What I'd like to see in any regen-capable controller is an ebrake input that can take a linear sensor. Maybe some even have that; I'm going to test it with the Fusin I have here as soon as a spare throttle sent by Ianmcnally2 (plus some other stuff) arrives (between times of checking on Hachi).

Basically the operation should be that if you use a standard ebrake lever with just on/off switches, it'd just turn regen on or off. But if you use a linear hall sensor as the ebrake input, it'd use that linear signal to control the amount of braking.

In theory this only requires a software change to *any* programmable controller, but it would require whoever wrote the software for the controller to make the change and then to make the update file available to everyone. Not a likely thing to happen. Except maybe for Keywin / Ecrazyman controllers, as he seems willing to do a lot of stuff to advance the state of technology.

Of course, the controllers are probably designed with input logic that does not read the analog level of the brake line, but only whether it is above or below a certain threshold level, as a simple on/off input. In those cases, there is little likelihood of making regen work the way we'd like in such a simple method as just changing the device it's input with.

However, even with those controllers, if the MCU they use happens to be programmable for what it's inputs do, such as using ADC's as binary inputs, *and* that's how the designers did it for the brake, then in theory (assuming no input filtering logic/etc. outside the MCU between it and the brake input) it is still just a software change, simply a more radical one.

 

[amberwolf]

Just an update--I did test the Fusin controller using a hall throttle (and a pot throttle, and a voltage source input) and it's ebrakes in either regen or non-regen modes are just on/off, with the threshold being when the ebrake signal line drops below 1.4V. Unfortunately the hall throttle I used only goes down from 2.4V (zero throttle) up to 3.5V (full throttle) so it wouldn't do anything at all.

Anyone else want to test their own controllers for this ability?

 

[bart_dood]

Has anyone tried the USB controller software with their golden motor controller?

I have been away from Ebikes but am back on it now; I have a GM controller and noticed this new controller cable and software on their website, I have pretty much figured out how to make my own USB cable but wondered if anyone else has done it?