wrobinson0413
1 kW
.
Pedal Assist Sensors
- Thread starter wrobinson0413
- Start date
GringoInChina
10 mW
If it's not raining like a mutha tomorrow I'll pull my 'Chingaletti' out of the 'scooter shed' and try to get a number for you.
It's 11:00pm here right now and there aren't any lights worth mentioning out in the drive or the shed.
It's 11:00pm here right now and there aren't any lights worth mentioning out in the drive or the shed.
wrobinson0413
1 kW
.
rkosiorek
100 kW
that is almost like asking how many different species of inseccts are there?
1. pedal sensor that just tells if the pedals are turning. - usually a magnet ring attached to the pedal axle (BB) and a hall or reed sensor attached to the seat tube.
2. sensor that reads chain tension - often looks like a chain tensioner except that it is located on the tension side of the chain and it reads the deflection of the tensioner arm.
3. strain gauges located on the hub motor axle - these read how much pulling force is put on the hubmotor axle when tension is put on the chain.
4. strain gauge mounted between the freewheel and the hub.
i have seen both elastomer and variable reluctance strain gauges used. in both locations.
add numerous variations and combinations of all of the above.
rick
1. pedal sensor that just tells if the pedals are turning. - usually a magnet ring attached to the pedal axle (BB) and a hall or reed sensor attached to the seat tube.
2. sensor that reads chain tension - often looks like a chain tensioner except that it is located on the tension side of the chain and it reads the deflection of the tensioner arm.
3. strain gauges located on the hub motor axle - these read how much pulling force is put on the hubmotor axle when tension is put on the chain.
4. strain gauge mounted between the freewheel and the hub.
i have seen both elastomer and variable reluctance strain gauges used. in both locations.
add numerous variations and combinations of all of the above.
rick
ZapPat
10 kW
Wayne - Have you ever seen this linear pedal torque sensing gadget? I know you aren't necessarily looking for this, but thought it might be of interest to you anyways. I would love to play with this thing to replace the hand throttle and digital pedal sensor. If used with a torque-mode controller it should give the user totaly automatic control over motor torque in proportion to applied pedal torque... and probably give us the impression of being He-Man on a bike! :lol:
Oops, forgot the link!
Pedal torque sensor - Item M25 towards the bottom of the page
BTW, has anyone here ever used any similar linear torque sensors to sense user effort?
Oops, forgot the link!
Pedal torque sensor - Item M25 towards the bottom of the page
BTW, has anyone here ever used any similar linear torque sensors to sense user effort?
GringoInChina
10 mW
WR,
I pulled the top cover off the compartment that houses my chain ring, controller, cables etc. this afternoon.
I couldn't see a part number on the sensor/switch unit but by the looks of it it's a reed switch (and I wasn't smart enough to follow the little cable back to the plug to find out how many conductors it had: 2 probably would be 'reed' and 3 would be 'Hall').
All I can tell you at this point is: it's about 3cm long, 2 screw mount, up next to a magnet ring with 6 magnets on it. The magnet ring was maybe 80mm in diameter.
I'll try to go out with my MM tomorrow, pull the cover off again, to verify that it's a 'reed switch' and let you know if it's N/O or N/C.
I was getting distracted by curious locals watching me wash the 'Yichang' mud, crud and misc road spewed body fluids off the poor bike. The usual cleaning practice (among the locals) with bikes and motorcycles here is to knock the slime off the seat and handlebars while the rest gets caked up and just rots away. So everybody in the parking lot missed their 'soap opera's' for an hour, 'watching the Gringo' use soapy water and 'lemon Pledge spray polish'.
Mark
I pulled the top cover off the compartment that houses my chain ring, controller, cables etc. this afternoon.
I couldn't see a part number on the sensor/switch unit but by the looks of it it's a reed switch (and I wasn't smart enough to follow the little cable back to the plug to find out how many conductors it had: 2 probably would be 'reed' and 3 would be 'Hall').
All I can tell you at this point is: it's about 3cm long, 2 screw mount, up next to a magnet ring with 6 magnets on it. The magnet ring was maybe 80mm in diameter.
I'll try to go out with my MM tomorrow, pull the cover off again, to verify that it's a 'reed switch' and let you know if it's N/O or N/C.
I was getting distracted by curious locals watching me wash the 'Yichang' mud, crud and misc road spewed body fluids off the poor bike. The usual cleaning practice (among the locals) with bikes and motorcycles here is to knock the slime off the seat and handlebars while the rest gets caked up and just rots away. So everybody in the parking lot missed their 'soap opera's' for an hour, 'watching the Gringo' use soapy water and 'lemon Pledge spray polish'.
Mark
wrobinson0413
1 kW
.
rkosiorek
100 kW
of course if you really want to do it the hard way you can always do this:
this is a rear crystalyte motor. the motor mounts in a frame with vertical dropouts. the axle has 2 flats gound on it. one front and one rear. 4 thin film strain gauges were mounted. those are the yellow colored plastic things partially covered by the heatshrink. after assembly they would be hidden under the freewheel/gear cluster. 2 go to the front and 2 on the rear. they are connected in a normal wheatstone bridge. the way it works when you put tension on the chain the tension will flex the axle slightly and this will be picked up by the strain gauges. the harder you pedal the more it will flex and the higher the output level.
rick
PS i got a couple of those magnet ring PAS sensor thingies as well. real useless trash they are.
this is a rear crystalyte motor. the motor mounts in a frame with vertical dropouts. the axle has 2 flats gound on it. one front and one rear. 4 thin film strain gauges were mounted. those are the yellow colored plastic things partially covered by the heatshrink. after assembly they would be hidden under the freewheel/gear cluster. 2 go to the front and 2 on the rear. they are connected in a normal wheatstone bridge. the way it works when you put tension on the chain the tension will flex the axle slightly and this will be picked up by the strain gauges. the harder you pedal the more it will flex and the higher the output level.
rick
PS i got a couple of those magnet ring PAS sensor thingies as well. real useless trash they are.
rkosiorek
100 kW
The Sanyo/Rabbittool geared hub motor has a Variable reluctance type torque sensor. details and explanation are shown here:
http://www.endless-sphere.com/forums/viewtopic.php?f=2&t=2744&hilit=sanyo#p38547
rick
http://www.endless-sphere.com/forums/viewtopic.php?f=2&t=2744&hilit=sanyo#p38547
rick
wrobinson0413
1 kW
.
rkosiorek
100 kW
strictly DIY. the gauges were glued in place. the gauges are actually pretty inexpensive. catch is you got to buy them in packs of a dozen or so. so you are looking at $60 or so minimum. also the recommended glue is about $12.00. add to that a few bucks for the low noise amplifier to get a useful signal and it can become quite expensive to implement unless it also is a special case design.
i never did calibrate it for ft/lb or Nm measurements. so all i got was relative values. idea was a PAS type control with the torque in a feedback loop. i do know that the torque the motor could put out on the dropouts was more than i could manage by twisting the stator in both hands.
i agree that the Sanyo is a special case. but it is a nice engineering solution. especially when you think that it is at least a 10 yr old design.
i had thought of using a conductive elastomer as the junction between the crank axle and the chainring as another approach. these elastomers are rubber componds that change their resistance as the rubber is compressed. but i never tried them beyond a sketch and measuring the resistance change as i squeezed a sample in a vise. it would work. i suspect that that is what is used in the linear torque sensor advertised by Island Earth.
rick
i never did calibrate it for ft/lb or Nm measurements. so all i got was relative values. idea was a PAS type control with the torque in a feedback loop. i do know that the torque the motor could put out on the dropouts was more than i could manage by twisting the stator in both hands.
i agree that the Sanyo is a special case. but it is a nice engineering solution. especially when you think that it is at least a 10 yr old design.
i had thought of using a conductive elastomer as the junction between the crank axle and the chainring as another approach. these elastomers are rubber componds that change their resistance as the rubber is compressed. but i never tried them beyond a sketch and measuring the resistance change as i squeezed a sample in a vise. it would work. i suspect that that is what is used in the linear torque sensor advertised by Island Earth.
rick
ZapPat
10 kW
I understand your bikes all having instant start setups, I don't even have a chain on my current ebike for the moment! In my opinion the magnet ring / hall sensor is pretty stupid sounding (and I take it Rick feels this way too), but there are a number of countries where they are nonetheless required by law. I doubt that any places require the linear torque type sensors, but I still have a personnal interest in this type of sensor, specially for off-road driving. As you mention though, it would take a different type of controller to make such a sensor work the way I want it to... but that's the beauty of making our own! Although I don't know what kind of sensitivity/resolution these types of sensors can provide, I would hope that it's sensitive enough to let us configure the assist effort range to a pretty low value if we want it that way.wrobinson0413 said:
I have never seen them elsewhere than at Bret's shop (linked earlier), but I also would love to find one closer and/or cheaper. I would also like to see specs for the thing before deciding anything (I haven't tried emailing Bret about this yet...).
I understand you here, as I am also finishing up my latest PCB and had to choose what I/O's to include. As it is I could still add two extra digital I/O's by using the ICD data and clock pins, but I'll wait for my next minor revision if I think it will be usefull to add any more. As it is I have most of my PIC's pins used already, except that I'm keeping 7 lines for a possible integrated LCD in the future. I have included two analog throttle inputs, so will be able to experiment with this type of pedal torque sensor or use it for a seperate analog regen brake lever.
wrobinson0413
1 kW
.
rkosiorek
100 kW
just in case Brett (solarbbq) has not been following this thread, i pm'd him asking him to toss in his 2 cents worth about that pedal torque sensor.
after all it is listed on his site.
hope he chimes in soon.
another suggestion. for those PAS boys. how about this senario - you pedal up to speed. once you reach the desired speed you touch a button and a cruise control maintains that speed until you tap the brakes. would that be legal in the EU?
rick.
PS. you right. that magnet ring thingie is a waste of time. i don't like the way they work at all.
after all it is listed on his site.
hope he chimes in soon.
another suggestion. for those PAS boys. how about this senario - you pedal up to speed. once you reach the desired speed you touch a button and a cruise control maintains that speed until you tap the brakes. would that be legal in the EU?
rick.
PS. you right. that magnet ring thingie is a waste of time. i don't like the way they work at all.
solarbbq2003
10 kW
- Joined
- Apr 7, 2007
- Messages
- 500
the torque sensors on my site work via how much effort put in determines output voltage, same as throttle output voltage 1v to 4v approx, uses 5v input, you use via the throttle wires or in conjunction with the throttle wires so can use throttle at the same time. The output is 2:1 very approx, there is another brand available where you can vary the output ratio via small pad on the handlebars. I've not use the torque pedal sensor though I've sent a couple out to customers, I can ask them if you like how they went. attached pic of the torque sensor with variable output ratio, the one on website is the fixed output ratio type, they make the fixed ones in 1:1 or 1:2, 1:2 seems to be the preferred ratio.
I'm not sure whats inside the torque sensors have never taken one apart, so no idea how the output is determined, guessing it a strong spring and linear hall sensor with small magnet arrangement but no idea really.
I'm not sure whats inside the torque sensors have never taken one apart, so no idea how the output is determined, guessing it a strong spring and linear hall sensor with small magnet arrangement but no idea really.
Attachments
wrobinson0413
1 kW
.
Tiberius
10 kW
Guys,
The law in Europe requires pedal sensors. The motor is not allowed to supply power unless the rider is pedalling. This rules out throttle only operation. There is an exception in the UK, so we can use throttles, but the exception may get phased out one day.
The normal method is a hall and magnet system on the crank that just detects whether the pedals are turning. Some of them don't discriminate which direction, so you can get the motor to come on by pedalling backwards.
There are other systems that use a strain gauge on the crank, and then apply motor power in proportion to the rider effort. This is what is done in the complete systems with crank motors made by Panasonic and Yamaha. As you surmise, these schemes can get very high range out of a battery.
Nick
The law in Europe requires pedal sensors. The motor is not allowed to supply power unless the rider is pedalling. This rules out throttle only operation. There is an exception in the UK, so we can use throttles, but the exception may get phased out one day.
The normal method is a hall and magnet system on the crank that just detects whether the pedals are turning. Some of them don't discriminate which direction, so you can get the motor to come on by pedalling backwards.
There are other systems that use a strain gauge on the crank, and then apply motor power in proportion to the rider effort. This is what is done in the complete systems with crank motors made by Panasonic and Yamaha. As you surmise, these schemes can get very high range out of a battery.
Nick
wrobinson0413
1 kW
.
317537
10 kW
- Joined
- Oct 19, 2008
- Messages
- 939
Tiberius said:
True! Maybe.
But maybe we feel like you have this law why dont they. STOP!
Forgive your opinion Nick, and youre not telling lies either, but allow me to run with your post objectivly and not against you personaly.
Why?
Give me one "real" reason why people have to make the pedal movements to satisfy the community. Because that’s all it is, making out you’re are pedalling. Like a variety style show lip syncing to when the saints go marching in.
One word!
Pathetic
Because they want to see people pedal as fast as the motor can push them. People really cant get passed the bike moving without pedals. I just doesn’t look right does it?
Stupid people running the world I tell you.
Leslie,
The directive was designed to frame a primarily human powered vehicle with power-assistance, and exempt it from motor vehicle regulations. The requirement to pedal in order to obtain assistance is the most effective way to ensure design compliance, and also implement a simple safety measure.
The directive was designed to frame a primarily human powered vehicle with power-assistance, and exempt it from motor vehicle regulations. The requirement to pedal in order to obtain assistance is the most effective way to ensure design compliance, and also implement a simple safety measure.
EU Directives are about as insane as any government regulation / mandate.
Here is how to disable it on the Schwinn I Zip with a Hall Effect Sensor.
http://endless-sphere.com/forums/viewtopic.php?f=2&t=10554&p=161886&hilit=pedelec#p161886
Other types of sensors (torque, etc.) can be just as readily bypassed.
My real objection to the Pedelec function is that it is a major safety hazard for e Bikes.
If you do not put yourself into low gear as you come to a stop (like a motorcycle), it is very difficult to restart the bike in high gear because of the weight of the system.
There can be a much longer interval where you are not going fast enough, and maneuvering to maintain balance.
Now, there is an argument that without the current draw from starting the bike, batteries last much longer, but the other argument is that the motor starting the bike (getting it going, accelerating it to speed) is the perfect application of power for a e bike.
What the EU regulations need to be rewritten for is for PEDAL assist bike, meaning the motor is allowed to be a primary source of power (without it tripping the regulations that define it as a motor vehicle).
EU diktats have a long history of being ignored or not enforced when it becomes ridiculous.
Go to places like Spain and Italy, and everyone and their brother is driving up on tricked up 50cc "mopeds" that can outrun many cars.
Same is going to happen with Pedelec functions in most places in Europe.
Here is how to disable it on the Schwinn I Zip with a Hall Effect Sensor.
http://endless-sphere.com/forums/viewtopic.php?f=2&t=10554&p=161886&hilit=pedelec#p161886
Other types of sensors (torque, etc.) can be just as readily bypassed.
My real objection to the Pedelec function is that it is a major safety hazard for e Bikes.
If you do not put yourself into low gear as you come to a stop (like a motorcycle), it is very difficult to restart the bike in high gear because of the weight of the system.
There can be a much longer interval where you are not going fast enough, and maneuvering to maintain balance.
Now, there is an argument that without the current draw from starting the bike, batteries last much longer, but the other argument is that the motor starting the bike (getting it going, accelerating it to speed) is the perfect application of power for a e bike.
What the EU regulations need to be rewritten for is for PEDAL assist bike, meaning the motor is allowed to be a primary source of power (without it tripping the regulations that define it as a motor vehicle).
EU diktats have a long history of being ignored or not enforced when it becomes ridiculous.
Go to places like Spain and Italy, and everyone and their brother is driving up on tricked up 50cc "mopeds" that can outrun many cars.
Same is going to happen with Pedelec functions in most places in Europe.
ZapPat
10 kW
Good arguments to think about, GTA1! Forcing pedal-first operation for "safety" reasons doesn't seem to make sense to me either. This kind of law was probably made by [old] farts who would likely never even use a bike, let alone try to replace their car with it... 
Well, regulations aside, I actually like the idea of pedal controlled assistance. Of course, it needs to be more sophisticated than the garbage out there, at the moment.
A under-powered motorbike that needs you to pedal to get up hills, seems more pathetic, to me.......
A under-powered motorbike that needs you to pedal to get up hills, seems more pathetic, to me.......
ZapPat
10 kW
If "pedal controlled assistance" means motor torque being controlled by applied pedal torque, then I agree that this might be a very interesting control option. I just can't see how the ON/OFF type pedal sensors could be of any use at all, nor how this might improve safety.Miles said:Well, regulations aside, I actually like the idea of pedal controlled assistance. Of course, it needs to be more sophisticated than the garbage out there, at the moment.
A low powered motorbike that needs you to pedal to get up hills, seems more pathetic, to me.......
TylerDurden
100 GW
I'm confident that many instant-start EV/Ebike owners have seen a neophyte take test-ride/drive; where the nooB didn't understand the system was active, even though they were advised.ZapPat said:
:lol: Better to have a nooB forget to downshift, than to have a runaway (from a policymaker's perspective), since a cyclist should know that already.
Similar threads
