Difference between a throttle and a PAS system?

[FLBikeDude]

What is the technical difference between a PAS sensor on the crank of a bicycle and the throttle magnet/sensor on the handlebars?
Is there a huge difference or is it that the turning of the crank and the turning of a hand throttle are basically the same thing / mechanism?

 

[amberwolf]

The difference is usually significant in how they physically operate and how htey control the bike.

PAS is Pedal Assist Sensor (or System) and could be anything from a simple on/off of the power to the bike, so that when pedals are turning it also applies power at whatever level the controller is set to (or can output, depending on how it's made), to the complexity of an actual torque-controlled system meaning how hard you pedal controls the amount of power applied by the wheel to the ground.

Some of them use a simple magnet on the cranks or sprockets or it's own ring on the shaft, some use several magnets on such a ring, some use an analog sensor (hall, etc) to detect the magnet(s) and some use a switching type (hall, reed, etc), some use a strain gauge or other analog sensor without magnets to detect torque, etc. Depends on what hte controller itself is designed to do/work with. Then there's external systems lke the Cycle Analyst V3, which can use some of those sensors to make a PAS out of a system that isn't designed for it.


Exactly what any particular PAS system does or works like will depned on how it's designed and setup. You'd have to check with the manufacturer of whichever one you're interested in to find out details if you have a particular one already or one you want to buy/use. If instead you have something specific in mind on how you wnat it to wokr, there is probably already a system out there that does that, but you'd have to be very specific in wha tyou want.



A throttle is pretty simple. It's usually either magnets and analog halls or a resistive potentiometer (with or without gearing in it), though ocasionally it's a set of switches or a special switch that cuts in or out a certain amount of voltage to the controller's speed-control input. usually the more voltage on the throttle input (wihtin it's specified limits) the faster it goes, and at or below a certain voltage it doesn't spin the motor at all.

Again, as with PAS it depndds on the specific systme exactly how it will work, and you could probalby set up a number of variations on the idea (for instance, some people use a cable operated throttle on the bars or elsewhere, and run the cable to the actual throttle somwerhe else on the bike that's enclsoed for waterproofing or other reasons).

 

[amberwolf]

Is there a difference in how the power is conveyed to the motor?

That depends on how the controller responds to the input, and if it's programmable how it is set up, and which kind of system you choose, etc. If it's a simply on/off input, then it'll apply full power to the motor whenever throttle or PAS is engaged, and take it completley away when it's not. If it's a variable control, then it'll vary it's output based on the input in whatever way it is designed and set up to do.

Or, if you could match the indicated output level desired, would the motor see any difference between throttle and PAS?

No difference at all between them (assuming you choose the right system to do what you want it to do, and set it up to do that), except whether you are controlling it via the pedals (or hand cranks) or via a throttle (thumb, grip, or otherwise).


Heck, if you wanted to you could use a PAS sensor setup *as* an actual throttle controled by your fingers, or a throttle sensor as a PAS (for instance, use the throttle with gearing and a spring lever like a "float" that rides on the chain top, so that hte harder you are pedalling, the straighter the top of the chain, and thus the more throttle is given).

So it doesn't matter what the control input is, it matters what you want it to do, and how you want it to respond to your input, and which kind of input you'd prefer to have.

And I suppose, where you live and hwat their laws are (and if you care), so that for those places that forbid throttles and only allow PAS you can still have just as responsive a system as a throttle would be, if you like. (of course, some places may still also require a PAS that simply tells the system that you *are* pedalling, while the throttle still controls the actual speed or power applied).

Any performance or longevity issues?

Longevity: Only if you are using a mechanical sensor with bearings or something in it, that also takes a load of some kind, like the complete-BB sensors (THUN, etc). Eventually the bearings wear out, but most likely if you are not a super-hard-pedaller it's not an issue for the life of the bike.

Performance: that entirely depends on what you want it to do, and whihc one you pick. The most basic kinds of PAS or throttle control (essentially on/off switches of one kind or another) don't "perform" well compared to the more advanced ones, but then again, if you want the more advanced performance then you don't want to get/use the more basic controls.

It would seem PAS would give a more variable input. Or output?

Not sure what you mean by "variable". If you mean that you could have more control over the system, then not exactly: Both PAS and Throttle methods at their root *usually* just control a voltage into the controller that tells it what voltage to apply to the motor. Buuuut: Either one can be set up (assuming the controller/etc allows this) to do that in quite a few ways, and exactly how it's done depends on the hardware design and it's programming/customization (if any).

If you mean that one has a wider range of input than the ohter, then no: either one is going to simply send the input the controller itself is designed to respond to. What that response is again depends on the controller itself, and which one you use depends on what you want to do.

Amberwolf mentioned systems specifically designed for PAS? Crap, I thought this was just a sensor that could be ordered with most any kit, like the different throttles. Or do some just not have the necessary connections?

As explained in the previous reply, it's not just connections, it's how it is designed to work, and which KIND of PAS you want / it is designed to use, etc. (you'd have to decide what you want it to do before you'll know what sensor / connections / etc you need).

But in a partial answer: Yes, you *can* just add a Cycle Analyst v3 to any controller, along with one of the PAS sensors it supports (THUN, for example), and get PAS control out of it (with most likely a lot more customizability than with any other PAS system, albeit at a greater cost than most).

Should I be looking for something specific in a kit intended for mostly PAS, other than the sensor itself?

What you look for in the kit entirely depends on how you want it to work for you. If you don't say what that is, we can't tell you what to look for.

It all comes down to what YOU want the bike to do for you. Once you decide that, then it becomes possible to figure out a system that would do that. Most likley an off-the-shelf systme already exists--the cost may vary, but it can probably be done with stuff already designed for the purpose.


If you want some other ideas about what you can do with throttle vs PAS, look around the forum at people's build threads, and the various controller design, modification, or review threads, and at the various Cycle Analyst (CA) threads, including the V3 Beta discussion, and at the various BionX threads, etc. The wiki probably also has some info in it (and links to a number of those kinds of threads).

 

[amberwolf]

What I am trying to get at is this - assuming I have chosen a motor with the correct performance parameters, and I choose the correct type of PAS sensor that the motor and/or controller support, other than the different "mode" of modulating the power supplied, is there any difference to motor, controller, battery which one I use?

It depends on how you want the system to respond to you. Not just during steady-state, but how you want it to apply power in the first place, what do you want it to do when it changes state, etc.

If I pedal the same way, steady-state, and use throttle or PAS to the same speed, is it the same thing? Is one more wearing or damaging on components than the other?

For that specific condition, assuming you have two identical systems except that one is controlled by pedals and one by throttle, then yes. BUT: if the systems are not identical in how they are supposed to respond to the throttle, then they might nto behave the same.

Neither one, if otherwise identical, would be any different than the other for wear and tear, except as stated about loaded bearings. Throttles are more exposed to crash damage, probably, depending on what type and where you put it, but the same otherwise, really.

It seems to me that variable pedaling cadence would be similar to constantly varying the throttle input up and down rather than a constant setting, and this seems like it might be a bad thing.

Depnds on the systme and how it's set up and designed. Poorly designed systems do indeed sometimes do that, but good ones won't.

I've been doing quite a bit of reading and so much of it confuses me, there is so much I am failing to understand.

Then what you want to do first is decide what you want your systme to do, and then find the stuff to do it with.

Since you haven't said how you want it to behave, we still can't help there.

I don't know if I want a low-power assist motor on a road bike or a much bigger motor and battery on a FS MTB. I only just recently learned that a "specialized rock-hopper" was not something that was custom-modified to jump over boulders, and that a "downhill racer" was not just a crazy person. Or a skier. I still call a road bike a "10-speed".

Doesn't matter so much about the terms...we just need to knwo what you want to do with it, before we can help. You can DO almost anything you want to do with a bike, if you don't care about your local laws/etc, with the stuff that already exists, but you have to decide what it is you want to do first.

If you want to go offroad and power around in dirt you're gonna need a totally different system than if you want to lightly pedal along without sweat to a bus stop and then lift the bike one-handed onto a rack and finish your trip that way.

 

[amberwolf]

If you like, I can move our discussion in this thread into a new one for you, and link it here.

I'm afraid if I have a throttle I'll just mash that baby all the time.

Not an uncommon result.

What I THINK I want is a pedelec road bike or hybrid with a throttle that's like the morphine button I had in the hospital, only works for a brief interval and then you have to wait before you can use it again.

Well, that's relatively easy, but would require a bit of added electronics. A very small and probably very simple "one shot" 555-timer based circuit could pretty easily do that. It'd be "charged up" by any voltage being present on the throttle signal output wire, and then when it reaches the time limit it'd turn off the power to the throttle itself. Another timer would then control the reset of the first timer so you could use it again later.

I don't know any systems that do that by themselves, though, so that part you'd probably have to make yourself (or have made by someone).

I would like to get my cruise up to say 15-18 with a bit less effort, with maybe some assist available to 20 or a touch faster, and able to cruise on motor alone around 15mph, or again just a bit faster. Don't think I want or need to exceed that 18-22 mph ballpark. I've spent some time with the simulator but I'm still kinda undecided. Q-100, MXUS, or Justin's mini seem in the ball park, even maybe a 350W Mac.

Pretty much anything can do that sort of work under the conditions you have listed. The old original 36V Fusin I've got (had, will have again) on DayGlo Avenger could, easily enough.

The battery will be the one thing you really want to ensure you get right for your expected range/power usage/lifespan, though, cuz it's the hardest and/or most expensive thing to deal with problems on later. Controller and/or motor is relatively cheap compared to that.

Most of the geared motors will probably be a little noisier than the direct drive hubs, but probably physically smaller (at least in diameter) than the direct drive for the same startup torque under the same power levels. However, becuase of their "double layer" construction, they don't shed heat as well, so if they are run above their design specs they can get pretty hot (sometimes with no permanent damage, sometimes otherwise). They are also more complex and potentially could fail easier than a direct drive (DD) hub, simply because there are more parts in the path from motor to wheel. The latter is also true of any middrive.

I've used all three, and experienced failures with all of htem, but I'm also abusing them all a fair bit. Most poeple don't see failures at all, with any of them.

Some of the RC-motor friction drives seem to offer something pretty close but the noise is not something I can deal with.

The quietest thing I've used has been the DD hubs, but most of any systme's noise is during accleration or while more current is flowing thru it. But they've all got their little noises, and some are less irritating htan others. Supposedly certain motors like the old TidalForce and it's successor(s) are silent because of the actual motor design, as well as controller type, but they are (AFAIK) only avialable used, and I don't remember anywhere that still makes parts for them if something goes wrong, so you'd have a DIY repair to deal wiht in that case.

There are some controllers called Sinewave (search on "sine*", whcih use a slightly different way of sending hte current thru the same old motors to be a little or a lot quieter, depending on conditons and design/etc. I haven't yet used one to see how much difference that makes.

Pretty sure I want a small geared hubbie, and either buy pre-made or deal with Cell-man or Justin. Then put on the front with the batteries in the triangle. Steel fork with dual TA. I've read enough to know that Cell-man or Justin is a good idea, and front motor may, or may not, be a bad idea.

Remember that in general anything in the back is quieter than in the front just because it *is* behind you. Also, you get more traction out of anything in the back because more weight is on that, except during braking. Under wet conditions it may be easier to use a rear motor because of power in turns, especially with PAS, because sometimes if you have any power on the front wheel it begins to slide and skid and could cause a crash, but if you only apply power to the rear wheel then it's ok. (sometimes leaning forward over the bars will put enough weight on the front to fix it, sometimes it just unbalances you and makes the crash harder). With PAS, then pedalling to apply power in back will also apply power in front, and depending on how your system is designed there could be a significant delay between starting to pedal and getting power, and also a delay in stopping pedalling and having power stop being applied (that's the more dangerous part in that particular situation).

It's harder to service the wheel in back, though, more on some bikes than others.

Batteries to hit 25-40 miles don't seem real feasible with an easily pedal-able bike, but battery plus human input should work.

yeah, even light bateries are heavy. IIRC it's about 20-25 pounds including hte 50-cal ammocan case for my CrazyBike2's 30-ish-mile range pack (flat terrain, little wind, no pedalling, 20mph cruise, 15-18mph avg speed, hard startups). Add about 10-12 pounds for the other 10-15 miles of the "reserve tank" that's usually paralleled with the main one.

When I was much younger I saved and researched and read for two years, and finally bought ......... A LAMBERT. The "death-fork" bike. So I'm being a bit anal about gathering as much info as possible, but every new piece of info leads to new decisions and options. At least some of them are starting to make a little bit of sense, but then something new comes along.

Understandable--I remmeber reading on Sheldon Brown's site about that bike. :/ Hopefully yours didnt' actually break on you! FWIW, I had the stem snap on my Columbia 2005 bike, before it became DayGlo Avenger, within the first month of riding it. Almost broke my shoulder...very lucky I was wearing a helmet, too. It's hard to steer when the handlebars are no longer connected to the bike!

Like when you said "specially designed for PAS" I was thinking there might be something different about a "PAS motor" as opposed to a, well, "non-PAS motor", in terms of design or something.

[/quote]
The motor itself usually is the same in either situation, with the exception of some systems like BionX, where the controller and the sensors are actually integrated into the motor itself.

It's usually the controller that would need to be PAS-compatible, and which type it uses and how it works depends on that.

And of course, there's the CA v3 from Grin that can use various types of PAS sensor input and translate that into a regular throttle output for any throttle-compatible controller, so you don't *have* to use a PAS-compatible controller (but as a whole if you don't intend to use a CA then it's cheaper by far to just use a PAS controller).



The nice thing about BionX, and systems like the CA v3 with a THUN sensor, is that they actually sense how much torque you are putting out by pedalling, and can control the motor's output based on that, essentially giving you "bionic legs". That's the way I'd prefer systems to work, and how I originally planned CrazyBike2's powerchair-motor-driven-drivetrain to operate, though I never got that far with it.

Most of the cheaper PAS stuff at best controls the speed of the motor by the speed of pedaling, and at worst just turns motor on at full power when pedaling and off when not.


The only PAS I've actually gotten setup and tested on my own bikes has been the newer Fusin 1000W system, which is reviewed and explained over in my thread for it in the review section (and also linked in my Delta Tripper thread, but I did not install the PAS for it on Delta Tripper as I didn't like how it worked).


All that said: in that situation I'd personally go for a smaller regular throttle system in the 350W range, and a CA and THUN (or other torque-sensing) BB sensor to give me the PAS capability. Then I'd build that timer circuit or something like it to prevent the overuse of throttle. (or alternatley, put a spring-loaded button on the left handlebar that is really hard to press so I cna't hold it down very long, that must be pressed in order to have the throttle on the right side work).

With the CA the response to PAS input (and throttle) can be adjusted to your taste and purposes, and if you like you can change it on the fly when out and about to suit your momentary needs, too.


Battery size, well, that depends on hyour weight limit for that, and the range you can live with, and the money you have for better lighter longer lasting cells. A123 or other LiFePO4 types are nice, but they are significantly heavier than the same amount of Wh in NMC or LiMn, etc. I've got both, and they perform similarly on my setup but I get less range out of the same-sized but notably heavier A123 pack than I do out of hte NMC pack.


Others will have more specific advice about various components.

 

[amberwolf]

I suppose the CA could not be used to program a throttle response similar to what I want?

Not that I know of. But you can look at the CA v3 beta thread, and it has a lot of info on ways ot use the CA, and links to Teklektik's UUG for setting it up, which is very detailed. I blew up my v3 in a wiring mishap and havent' gotten ti sent in for repair yet, so I can't check it's menus directly to see.

Possibly just an awkward-to-use throttle, for KISS purposes. Are any particular throttle types better water-proofed than others?
Or easier to dry out or bypass if wet?

Possibly, but I haven't found one yet. It's possible to seal up the hall sensor and wires in a throttle so that they can't get the moisture on the hall leads, which is often what causes stuck-throttle problems. But I'm not aware of any you can buy off the shelf like that yet. There have been discussions on ES about getting some better ones made, but I don't think anything ever came of them, probably becaues too many different conflicting requirements from different people mean too many versions would need to be made to make it cost effective for small batches.

I have yet to expose my Magura resistive throttle to rain, as it's on the Nishik-E bike that's not yet finished (which is a Kepler friction drive that may not work well in wet itself anyway; have to test that when I get it done).

The most certain way to deal with wet conditions is ot use a cable operated throttle, or old friction-shifter, on your actual bars, and run the cable into a waterproof bag or box that contains the actual throttle. There's at least one thread on that design, but i forgot who made it. It's been used by others on ES too. It might be linked in the wiki.

Regards using the CA plus sensor for PAS - could some combination of simple RPM sensor plus road speed be used to better simulate what the torque sensor does? I've removed a crankset once, and while in my defense it wasn't working very well when I started, it didn't work at all when I put it back in. I do have a very good LBS, they currently stock Stromers, and not only will let me watch but even explain what they're doing, and not too expensive.

Well, you can't really do what the torque sensor does (measure the actual pressure of your pedalling) without one, but you could probably do almost any kind of PAS with the CA v3, depending on how you want it to work. But it may be lots cheaper (if you weren't going to use a CA anyway) to just get the kind of PAS controller you're after and go with that. If it's more important to be able to try differnet PAS methods and variations without changing out controllers and stuff, then the CA v3 with just a regular controller would be better because you get all the other potential benefits of a CA, too.

As for installing a torque sensor BB, well, the good thing is that since it's cartridge bearings it's pretty hard to mess up it's install in a way that keeps it from being used. It *is* possible to damage the wiring, so you have to be careful with that part. If the problem you had with the reinstall of yoru crankset was that the bearings were too tight to spin, and it wasnt' cartridge type, that's also an easy fix, to loosen the cups up again just a hair. If it was some other problem you'd have to describe it for me to suggest a fix. (I realize it probably doesnt' matter at this point, as it could've been years ago for all I know). Also, Sheldon Brown's website is a very good resource for learning about bike-specific things, and there are some very good bike mechanic books out there too; I forget the name of the one I have, but for most stuff it's very handy even if I don't have all the tools it specifies (sometimes I can make them or make do with ones I have).

Also, can using the CA eliminate the PAS delay in cutting off power when the pedals stop? I typically stop pedaling for corners due to pedal strike, powering that front tire thru a curve sounds like something to be avoided. I would assume using somewhat fatter and lower-pressure tires would help somewhat avoiding spin?

If you are using the CA itself for the PAS, and not using a (cheap) PAS-specific controller's PAS input, then yes, it should be able to cut power fast. I'm sure some PAS-specific controllers using their own PAS input from a sensor can also do it quickly, but the one I have here does not, and there are many reports of others that don't. (and ebrakes also tend not to respond instantly to cut power, though often faster than the PAS!).

On moving the thread - no matter to me, I'm going to try to isolate specific issues in particular threads. One problem I'm having is with the many varying opinions I find it hard to sort out the best path for my case. Also I don't even know what questions to ask.

Yeah, that's the problem with experience--everyone's is different, so there will always be different opinions, and sometiems people with zero experience on somethng will still have an opinion about it.

As for which questions to ask...the best thing is to know what you want to do, explain that, and we can probably figure out most of the stuff and ask you the questions, which will eventually lead you to the rest of the questions, most likely.

Your answers have been clear and very helpful. Though I have no clue what a "simple 555-timer circuit" might be, at least I think I know someone local who will, and works cheap.

Sorry--I am used to letting people use google (or other) search to find out what I refer to when I say something like that if they don't know. (it's what I do whenever people say things I don't know about, although it doesn't always find what I am after).

But basically there is an electronic chip, the LM555, that's been around for decades (since I was a kid, actually, whcih was a fairly long time ago) that does several simple functions, and those allow many time-related devices to be builtvery easily with jsut a few wires and little parts added. IF you don't know any electronics, there are still sites that show you exactly what to do and how to do it, but it woudl take a bit more knowledge to figure out how to turn what I described into the circuit you'd need. But most people with basic electronics skills could do it, I think.

 

[amberwolf]

I fix pc's and networks for a living, but to take a circuit, add wires and other parts, that would have to be real simple. I can melt solder, but need a LOT of space. I'm just not very mechanical. PC troubleshooting, I'm great. Many years of experience, I know where the points of no return are, and how to go back and try something else. I just have to remember lots of stuff. Electrical circuitry, I'm lost. A lot of it is I can't get a mental picture of why it goes that way.

Well, I started learning that before computers, as a pre-teen playing with stuff and eventually getting into hamradio and going thru DeVry's tech course (wasted money really as I don't wanna do that for a living nowadays, or computers, which I'm also decent at). But mostly I really "learned" electronics via building blinky-light and sound things for sci fi props for conventions (to wear/carry as costumes, and sometimes to sell); I desinged lots of fairly complex stuff I never built and a few things I did tha tmsotly did what I intended. :lol:

Computers...I did it for so long that ther'es qutie a few things I jsut "know" to do, but sometimes I get stumped quickly, like right now my "new" (very old, really) music computer needs XP activated and it won't talk to the servers, with error 32777, which comes up as a date-time wrong error, but it is right, and I evne set it different and then back again, in bios and in windows and it still won't work. Everything I find on internet about others with the prolbme points to haveing to reinstall from scratch again to fix it, and that means also redoing many dozens of hours (maybe a month to two months of "free" time) reinstaling all the softwrae and configuring it. I can just bypass the activation by resetting the counter every 30 days, but that's a PITA to remember to do and worse to try to do it after I forget and it times out. If I had an illegal copy of XP that'd been hacked, I wouldn't have to worry abou tit, but because I actually own mine I have to deal with problems like this. (this is why software protection is bad, becuase it only punishes the actual owners/users, not the pirates, and doesn't stop anyone that wants to steal it from doing it anyway).

But If it only needs to be reasonably close, and can be made with cardboard, ductape, and glue, I'm your guy!

Tha't smy preference, working with my hands and "sculpting" things to a final shape out of whatever materials.

Like that really bad wire splice you uncovered under the original insulation on the motor that was sent to you. I would have called that excellent workmanship. Now I know it's bad, but no idea why.

That kind of thing is only "bad" if it causes a problem...if water migrated into that cable it would corrode the splice, because it isn't protected by anything. Also, it isn't crimped or soldered so as it oxidizes (which happens even inside the cable) it gets higher resistance, and if it were a current-carrying wire it'd heat up at that point, and drop voltage across it. And because it is inside the insulation, the average person would have no idea the problem was right there, because it woudl *seem* like a connector failure. They'd have to replace teh whole cable (or item with the cable on it) just because of that, to find out that's where the problem was.

I had already planned on avoiding splicing wire if possible, and now have learned I better darn well find out how to do it correctly if I have to. Are wire nuts OK? I can use those, or at least I think so.

Sure, you can use any splicing method you like, as long as you know where your splices are and how you did them so you can go back and find and fix the problems you later encounter because of them. There's a lot of splicng devices made, and all have their good and bad points, so it depends on where you wanna use them. If you'r egonna use plain wirenuts (meant for indoor use in a controlled environment) in a dry desert environment on a bike they'd be fine, but in a humid or wet, potentially salty environment you could have problems due to corrosion. There are wirenuts meant for wet conditions, with a kind of seal and some paste inside to keep water out, and you can emulate that by using dielectric grease (liek for battery terminals on cars) or vaseline, but without the seal on the wirenut, sometimes the stuff migrates out as it gets warm or hot with the biek sitting in the sun (I tried vaseline on mine and all i got was a greasy mess the firs ttime it got hot here, and now all my wires below the connectors are messy and dirty and greasy and yucky).

Anywhere you can avoid a splice, it's better, though. (I have trouble with something I spliced carelessly on CB2 about every month or so, sometimes more often...I really need to go back and rewire the whole thing but it's a fair bit of work and I am tyring to not do new things to it so I can save that time/work/material for it's repalcement).

Now to decide on fast or standard speed, front or rear, 29" or 26", road bike or light MTB, and about 15 other things. I have eliminated SLA batteries, and the ones that catch on fire! Progress.

Well, EVERY battery can catch on fire. So that means you aren't going to use batteries (or any other energy storage device) at all. Really, though, there ARE some that could be more likley to have problems than others, but under the wrong conditions any of them is about as dangerous as any other.

For the rest...that's up to your needs.

 

[Arlo1]

I might play around with this soon. I will have to re-read this thread.