Cable Operated Linear Throttle

Raged · Jun 13, 2015

I'm building a cable operated linear throttle for my new build where I will use the front deraileur adjustment on a STI (dropbar) arrangement on a custom 29er setup.

Am I right in thinking this is pretty much how a hall sensor throttle can be set up?

Is there anything more complicated I'm not taking into account?

teklektik · Jun 13, 2015

Or just buy it from gwhy! --- ES For Sale Thread

Raged · Jun 14, 2015

Ahh... someones already done it... your one works out to AUD$46 delivered which I might go with if my design doesnt work. My design is alot smaller and simpler (one spring) in a tube. It will be smaller than a cigarette lighter in length.

Alan B · Jun 14, 2015

I did a spring/magnet in a PVC tube similar to what you are doing. I put the hall sensor in a hole to get it right in front of (actually touching) the magnet to get more range of voltage output, and I had a microswitch that opened when the magnet was fully released since this was for a variable brake instead of a throttle. The one difficulty was getting it adjusted, moving the cable around changes the length of the housing and upsets the mechanical zero.

I also arranged the magnet to move away from the sensor, so the released position was magnet touching sensor, and it pulled away. There are various ways to do it.

Putting the magnet outside the tube was tested but didn't produce the range of voltage I was looking for. I eventually made another with the magnet and hall sensor inside the hood of a brake handle to do what I needed for a variable ebrake, and I use it almost daily on my CroBorg.

Raged · Jun 14, 2015

Alan B said:
I did a spring/magnet in a PVC tube similar to what you are doing. I put the hall sensor in a hole to get it right in front of (actually touching) the magnet to get more range of voltage output, and I had a microswitch that opened when the magnet was fully released since this was for a variable brake instead of a throttle. The one difficulty was getting it adjusted, moving the cable around changes the length of the housing and upsets the mechanical zero.

I also arranged the magnet to move away from the sensor, so the released position was magnet touching sensor, and it pulled away. There are various ways to do it.

Putting the magnet outside the tube was tested but didn't produce the range of voltage I was looking for. I eventually made another with the magnet and hall sensor inside the hood of a brake handle to do what I needed for a variable ebrake, and I use it almost daily on my CroBorg.

The magnet needs to go across from 1 pole to the other pole for full voltage range of the hall sensor? moving away will only give 1/2 range?

Anyways, I ordered one of these : http://www.ebay.com.au/itm/231368510275?_trksid=p2057872.m2749.l2649&ssPageName=STRK%3AMEBIDX%3AIT

The cable hole is a bit large but the idea is to pass a screw through the magnet with a stopper at about the right distance, then use a standard jag wire cable adjustment up front (m5 thread) to do fine adjustments. If you have a long pull brake lever, use a magnet that is 15-20mm long for better control/range. The STI shifter is short pull so I only did 10mm magnet. I might get a longer one if I need more tighter control.

The internal adjustment for big (up to 5mm) changes is done with a cable stop clamp with a longer screw :

Something like this :

But add a longer screw and 4 nuts to hold and adjust the magnet position.

amberwolf · Jun 14, 2015

Have you considered taking a hidden-wire-brake-sensor (HWBS) and replacing the switching hall with a linear/analog one?

It would get you a housing, magnet, wiring, etc.

gwhy!1 · Jun 14, 2015

I played around with this configuration before I finalised my design and it do work ok.. but

... it is fine for short pull cables but if you want to use a MC throttle for example you need around 3-3.5cm of throw/pull plus the gap for the spring which will need to be around another 3-3.5cm, then add around another 1-2cm for cable fixing and end caps so you end up with quite a longish tube.. remember to use a SS spring as a steel spring will upset the range of throw also as will the cable nipple at the end of the cable unless its SS or ali. i have been re-thinking about my attempt of making a tube type sensor and attempting to make it shorter than my first attempts by running the spring down the middle of the magnet so you dont have the compressed spring taking up approx 1-1.5cm in the empty space in the tube at full throttle.

d8veh · Jun 14, 2015

Ebike throttles normally have two magnets spaced apart. Without a magnet present, the sensor gives 50% throttle. The two magnets are polarised opposite, so one pulls the hall down to zero throttle and the other pulls it up to maximum.

gwhy!1 · Jun 14, 2015

d8veh said:
Ebike throttles normally have two magnets spaced apart. Without a magnet present, the sensor gives 50% throttle. The two magnets are polarised opposite, so one pulls the hall down to zero throttle and the other pulls it up to maximum.

but some have 1 curved magnet

, with the sensor starting at one end of the magnet and travels to the other end of the magnet. these are the best types of e-bike throttles but the throw is not very linear.

edit: when I say the best type I mean the most reliable configuration for regular ebike throttles

Raged · Jun 14, 2015

amberwolf said:
Have you considered taking a hidden-wire-brake-sensor (HWBS) and replacing the switching hall with a linear/analog one?

It would get you a housing, magnet, wiring, etc.

Good idea! I do have a pair of magnetic brake sensors arriving with my BBS02! Will open her up and see what's inside =) Thanks AW! The only issue is it would still need to be connected to a dud deraileur for the cable tension unless you jimmy up a sealed spring (in which case you're back at square one).

gwhy! said:
I played around with this configuration before I finalised my design and it do work ok.. but ... it is fine for short pull cables but if you want to use a MC throttle for example you need around 3-3.5cm of throw/pull plus the gap for the spring which will need to be around another 3-3.5cm, then add around another 1-2cm for cable fixing and end caps so you end up with quite a longish tube.. remember to use a SS spring as a steel spring will upset the range of throw also as will the cable nipple at the end of the cable unless its SS or ali. i have been re-thinking about my attempt of making a tube type sensor and attempting to make it shorter than my first attempts by running the spring down the middle of the magnet so you dont have the compressed spring taking up approx 1-1.5cm in the empty space in the tube at full throttle.

Yeah for a MC throttle where you have a long twist range my design would be way too long =) For the STI 9x2 setup, The front deraileur pull of 10mm should be fine with a bit of dead zone up front. I calculated about 6-7cm long and 25mm wide for my setup which makes it easy to hide on the handlebars.

d8veh said:
Ebike throttles normally have two magnets spaced apart. Without a magnet present, the sensor gives 50% throttle. The two magnets are polarised opposite, so one pulls the hall down to zero throttle and the other pulls it up to maximum.

All mine are curved magnet?

1x from em3ev for my MAC10T is curved
3x from ebay are also single curved magnet... These have stiff springs, so they dont feel as good to use... Maybe just needs more wear in period.

amberwolf · Jun 14, 2015

Raged said:
The only issue is it would still need to be connected to a dud deraileur for the cable tension unless you jimmy up a sealed spring (in which case you're back at square one).

You know those cone-shaped springs on QR skewers? Maybe if you put a plate on thewide end of one of those and clamp it on the cable end stickng out, it would provide enough travel and tension to work?

If not, there might be another type of spring like that from a ball point pen that could fit within the HWBS housing and provide the tension, against a ball on the end of the cable?

Raged · Jun 15, 2015

FYI.

em3ev throttle and ebay throttle. Single magnet. The em3ev spring is lighter which has a better feel compared to the ebay unit. I tried to swap the springs, but they align at different places =(

The QR skewer spring is too weak =( Get this... My crash, the fork is tacoed... front 180mm disc is taco, QR skewer is taco... but wheel spins true. How the hell does a wheel survive when everything else is bent to shit??

d8veh · Jun 15, 2015

Whether a single curved magnet or with two separate magnets, the result is the same. Measure the output of the hall sensor with a 5v supply and you will find about 3v with no magnet present. Bring a south facing magnet to the hall sensor and you will see the output go down. Bring a north facing one to see the output go up.

Are those curved magnets not just a piece of soft iron with a magnet at each end?

A HWBS won't work. It only gives on/off.

amberwolf · Jun 15, 2015

d8veh said:
A HWBS won't work. It only gives on/off.

Guessing you didn't read the whole post about using one.

Raged · Jun 15, 2015

the throttles look like a full curved magnet? pole on each end and it adjusts voltage across the magnet.

gwhy!1 · Jun 15, 2015

d8veh said:
Whether a single curved magnet or with two separate magnets, the result is the same. Measure the output of the hall sensor with a 5v supply and you will find about 3v with no magnet present. Bring a south facing magnet to the hall sensor and you will see the output go down. Bring a north facing one to see the output go up.

Are those curved magnets not just a piece of soft iron with a magnet at each end?

A HWBS won't work. It only gives on/off.

normally the throttle with the individual mags has the hall sensor in a upright postion so one magnet uses the front of the sensor and the other mag uses the rear of the sensor. On the curved mags..( all the ones i have seen are a curved magnets but would work just as well if a mag was on each end of a soft steel curved bar ) the sensor uses just the one face. The curved mag arrangement is a better design over the 2 seperate mags because mechanical failures are more common, but both the arrangements introduces more non-linearity due to the curvature of rotation.

Raged · Jun 17, 2015

My STI shifters came in with the drop bar. I was right in that theres about 10mm pull. The spring is from a ball point pen =)

[youtube]GcKfl0ZPffA[/youtube]

Now... how to mod the shifter so that it doesnt ratchet to 2nd ? It would be an ok cruise conrtol but potentially dangerous if you hit something and the lever was depressed.

Update - 10mm is too long for the amount of cable pull. (felt like jumping from 1st to 3rd chainring to put it on full throttle). I've ordered a 5mm magnet to reduce the lever pull further and see how it goes.

4r3st3r · Sep 15, 2021

Does anyone know where I can get something like this? Need to convert the original throttle cable!
Cheers

amberwolf · Oct 29, 2021

This uses one for a proportional regen brake lever, but I also use one for my throttle.

https://endless-sphere.com/forums/viewtopic.php?f=2&t=105460

This has some links to parts
https://endless-sphere.com/forums/viewtopic.php?f=2&t=107194&p=1570075&hilit=cable+operated+throttle#p1570075

Cable Operated Linear Throttle

Raged

1 kW

teklektik

100 MW

Raged

1 kW

Alan B

100 GW

Raged

1 kW

amberwolf

Administrator

gwhy!1

100 kW

d8veh

1 GW

gwhy!1

100 kW

Raged

1 kW

amberwolf

Administrator

Raged

1 kW

d8veh

1 GW

amberwolf

Administrator

Raged

1 kW

gwhy!1

100 kW

Raged

1 kW

4r3st3r

1 mW

amberwolf

Administrator

Similar threads