Announcement

Collapse
No announcement yet.

Front Hub Motors Dangerous?

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    #16
    Torque arms are cool and all but on the various social media groups with a lot of hub motor bikes you can find numerous tales and photos of magnesium drop outs that snapped even with good torque arms.

    Comment


    • AltaBrad
      AltaBrad commented
      Editing a comment
      I was thinking about putting a front hub motor on mine until recently. After researching no way, not even with torque arms. On top of that they're gutless compared to mid-drive. Ban them I say.

    #17
    @ynot...
    When you said "there is just no room on the spindle" I thought you meant the spindle isn't long enough. After looking through the pics in the link provided by Tommycat, I think the problem might be that the torque arm won't slide onto the spindle all the way because it bumps up against the shock cylinder before it hits the dropout.

    If so then you might need a custom made torque arm similar to the one in this picture.

    That pic plus a few more comes from a post by Balmorhea on page 8 at the link posted by TommyCat upthread. Notice the innovative way he attached it to the fork cylinder. Notice also Balmorhea states
    This is a one-way torque arm (no regen or electrical braking). It can only hold forward drive torque.
    Depending on your hub motor that might have big safety ramifications for you. Recommend you ponder that idea carefully before you proceed.

    BTW, it sounds like you've been riding that bike up steep hills with no torque arms. That's just plain all balls no brains DUMB. Smarten up and think about what you're doing before you injure yourself and/or someone else too.

    Comment


      #18
      AltaB, the problem is the length of the spindle, especially on the wiring side. As noted I have machined the wheel nut so that the plastic locking engages also machined the wiring spacer, the spindle length is not enough. My current thinking is to weld 19mm sockets onto bars clamped to the forks, not to stop the nuts loosening, but to hold them in place, thus holding, or helping to hold the spindle. Also if the bars are long and strong, the load will be spread. The only alternative that I can see is to make clamping dropouts. very small space for the anchor bolts. I am currently installing hub motors onto a freighter trike, and it will have clamping dropouts.
      The intended use is to continue using the rear hub motor as installed now mostly using PAS, and using the front motor as a boost for starting and for steeper hills, gravel trails and pavement only. The motor is only a 250 watt Bafang, so loads are minimal.

      Comment


        #19
        TC, just now saw your "torque wrenches" Suggest you patent them.
        So far only one objection to my socket nut idea & think that was about stopping the nuts from loosening, it is about stopping the nuts from wandering, thereby stopping the spindle from twisting. Would really like someone to explain to me why it would not stop or at least help.

        Comment


          #20
          You're using the front motor only for starting. Well, hate to be the bearer of news you don't want to hear but when an electric motor starts torque is infinite, doesn't matter if it's 250 watt, 2500 watt or 250,000 watt... it's torque is infinite. So your assumption that the load will be minimal because it's just a 250 watt is wrong, not just a little bit wrong, totally wrong and startup is the worst possible time to use it because at startup torque is maximal (in fact infinite) not minimal. The basic mistake you are making is thinking watts = torque. Nope. Watts is a measure of power not torque. Power and torque are related but they are not equivalent. Torque is equivalent to force. Power is force applied through distance. That's just basic physics.

          IC engines are exactly the opposite... torque is nearly 0 when they start and torque increases as RPM increases. But that's IC not electric. Electric and IC work on very different principles. IC is governed by thermodynamics (specifically volumetric expansion of hot gases) whereas electric motors work on the electromagnetic effect. Thermodynamics and electromagnetism share little in common except for the fact that torque is a force and power = force X distance.

          Note that I am NOT saying electric motors have infinite torque at any rpm. I am saying torque is infinite at startup and that is not just my opinion it's fact they predicted even before the first electric motor was built and it has been verified empirically. In fact torque drops off rather quickly as RPM increases.

          Now here's where we get down to the nitty gritty details and ask ourselves "if the torque is infinite at startup then why does it not snap bicycle dropouts the second the motor starts". The answer is not "torque is not infinite at startup". The answer lies in the elastic nature of metals and the fact that at startup the dropout spreads because it cannot resist an infinite force BUT before the force stretches the dropout to it's breaking point the torque drops off sharply and the drop out springs back. Does it spring back to it's original dimensions? That depends. If it was stretched beyond its yield point then it will not return to original dimension. If it is repeatedly stretched beyond yield point then it will break relatively soon. If not stretched beyond yield point then it will not fail nearly as soon but it will eventually fail due to fatigue, work hardening, notch creep, etc.

          What does all that have to do with your situation? Simple... the fact that it hasn't broken so far doesn't prove it will never break. Depending on the material the dropout is made of you might be stretching the dropout more and more every time you start and every time you hit a hill. You have no way of knowing and, as explained above, the fact that it';s only a 250 watt model means absolutely nothing.

          Some other mistaken assumptions/ideas....

          Originally posted by ynot View Post
          AltaB, the problem is the length of the spindle, especially on the wiring side. As noted I have machined the wheel nut so that the plastic locking engages also machined the wiring spacer, the spindle length is not enough.
          Sounds like they designed that motor before it became obvious they need torque arms. Does it even have flats? Or maybe your fork is too wide for that hub in which case you would have had to add spacers between the hub and the dropouts. Something is not right there.

          Originally posted by ynot View Post
          My current thinking is to weld 19mm sockets onto bars clamped to the forks, not to stop the nuts loosening, but to hold them in place, thus holding, or helping to hold the spindle.
          That will do nothing to counteract the torque on the spindle. If you can't see that simple fact then you lack any understanding of the torque and reaction torque at play. Holding the nuts in place will not stop the reaction torque from turning the spindle into the right-side nut while it simultaneously turns out of the left-side nut. The only way to counter that torque is to hold the spindle itself.

          Take a look at the various solutions in that thread TommyCat linked to and ask yourself why none of those people did what you are considering. Is it because they are all idiots and you're a mechanical genius? Not likely.

          Why not run your idea past them and see what they think? I bet they'll tell you the same thing I'm telling you.... holding the spindle is the only way.

          Seems like your logic goes something like the following.... Dang it, I can't hold the spindle so I'll hold something attached to it. That's faulty logic adopted in desperation and resting on lack of understanding of the cause of the problem. Squirting the nuts with WD-40 and wrapping duct tape over the nuts and around the dropouts would be just as effective and cost you less time and money.

          Originally posted by ynot View Post
          Also if the bars are long and strong, the load will be spread.
          What load? The load on the nuts? There will be no load on the nuts. You are about to spread a load that doesn't exist! Again you don't understand the nature of the problem.

          Originally posted by ynot View Post
          The only alternative that I can see is to make clamping dropouts, very small space for the anchor bolts.
          Can't be sure because I can't see what you see but my initial reaction is small space means the load will not be spread out and will end up still on the dropout just in a different place.

          Comment


            #21
            @ AltaBrad:
            Thank you for the argument/discussion. Just to be clear, I have a lifetime of pulling wrenches, running and building machines, machining metal, sorry only one patent.
            Agreed that the torque load as well as bumps in the road, will put stress on the forks, and if the modulus of elasticity is exceeded something is gonna break.
            The right hand nut will tend to be tightened in my, theoretical situation, So when the spindle tries to tighten itself against the "torque wrench" it will have the effect of squeezing the bikes aluminum drop out, which on a microscopic level will try to press more dropout material against the spindle flats increasing it's holding power. If the nut is tightened to it's specified torque, (whatever that is, will research it) and kept from loosening any torque applied by the motor will be shared by the pressed dropout, by the NM of the tightened nut, by the friction of the nut on the spacer washer and the lug of the washer in the dropout slot, not to mention the wizardly 'torque wrench'. Also the spindle must twist along it's axis because of the torque.

            If the motors torque is not sufficient to break or deform metal either immediately or over time, excess torque will be applied to the spokes and then to the rim tire connection and finally to the connection between the ground and the tire, so tire spin becomes the weak point IMO.
            You are making me think about it, and for that I thank you, will be 79 soon and these bones likely will not think kindly of a close encounter of the thud kind, so to the best of my ability am analyzing risks .

            Right now am making the dropouts for my freighter trike, and they will be clamping, But will make this promise, if and when I sell the bike, will replace the front wheel with the original unpowered stock wheel.

            Comment


            • AltaBrad
              AltaBrad commented
              Editing a comment
              Thank you for not retreating from our debate/discussion. I'll keep this brief and say you understand better than I thought you did. Your years of experience and broad range of experience has served you well.

              Removing the wheel and replacing it with original would be one (but not the only) responsible thing to do. Explaining to buyer what we have been discussing thereby alerting them to the potentials is equally as responsible. The right price and a proper understanding between buyer and seller is the sweetest deal.

            #22
            Fork lowers are nearly always magnesium. Not aluminum.

            Comment


            • AltaBrad
              AltaBrad commented
              Editing a comment
              I buy nothing but steel lowers. Because I can weld to them with the equipment I have. I'd buy aluminum lowers if I had a nice Tig machine but I don't. I would never buy magnesium lowers, won't even ride 'em. I acquire them from donor bikes and immediately scrap them, don't trust them at all.

            • AltaBrad
              AltaBrad commented
              Editing a comment
              Warning: never weld a shock lower unless you know what you're doing. If done wrong they can and likely will rupture causing severe injury or death

            #23
            @AltaBrad:
            Now you have me up way past my bed time thinking about your comment that the hub motor has infinite torque at start up.
            In theory I guess that is true, as the RPM goes down the torque goes up, so at 0 rpm you could have infinite torque.
            I if that were true, in practice you could not stall the motor, if there was a way to hold the bike in space you could spin the planet.
            The zero rpm only stays that way for a split second, as soon as motor begins to spin the torque is reduced.

            Bafang claims a max torque of 45mn for this motor, and recommended wheel nut torque for a 12mm with flats on shaft, range from 60 to 120mn
            In practice then if you secure the nut from spinning by using an external holder, tightened to 60 mn, you have 15 mn plus, for each nut, in friction on the spindle threads, more than you need for stopping spindle rotation, over and above the holding power of the dropouts etc.
            Please shoot me down if wrong, as it is rather important for safety, and I am in no position to do an engineering study.

            PS That no one has reportedly tried it is not part of the equation, if I felt that way would not have applied for and received my patent.

            Comment


            • AZguy
              AZguy commented
              Editing a comment
              FWIW permanent magnet brushless DC motors don't have infinite torque at startup... not even close from an "ideal" perspective... the torque is limited at all times by the amount of current the controller can supply to the motor windings and so max torque is relatively flat throughout most of the RPM curve and is actually lower at startup (and at higher speeds)

              For super high startup torque a series DC motor (the magnets are replaced by field windings that are in series with the rotor wingdings) is going to win that battle and while it is limited by the current the controller can deliver unlike permanent magnet DC motors where torque vs. winding current is relatively close to proportional across the speed range the series DC motor torque to winding current is greatest at zero speed and drops of as speed increases - this is what they use in electric dragsters and other high startup torque applications like train locomotives... but I digress...

            • AltaBrad
              AltaBrad commented
              Editing a comment
              Excellent rebuttal! I'm Liston on the ropes, you be Clay pummelling the crap outta me :) Time out while I regroup, not easy after 68 trips around the sun, you at 78 (respect), know what I mean. But don't even think I'm reaching for the towel, nope not yet.

            • AZguy
              AZguy commented
              Editing a comment
              LOL - not trying to pummel anyone... just clarifying ;-}

              As another point the permanent magnet brushless DC motor controllers can deliver more current to the motor windings than they draw from the battery so they can provide more torque at lower speeds since as the motor speed gets higher how much more current they can deliver drops an will have a direct relationship with the motor rpm and the maximum motor rpm at full voltage... for a DD hub this isn't likely to happen until you have a really big head of steam going... geared hubs will often be speed limited by the battery voltage and the max torque may drop off more with speed (depends on the controller) but they will still be torque limited by the max output current the controller can deliver at a standstill or very low speeds and most controllers on the geared hubs I've been on were without hall sensors and have a hard time delivering much current at these low speeds for rather anemic torque at a standstill.. And there I go digressing again since a front hub is most likely a DD motor and it will really come down to the controller at this point - some DD controllers can deliver a *lot* of current at stop/low-speed however...

            #24
            One other thing we have not evaluated is the force of braking on the spindle, in theory (that word again) if it takes 50mn to get you and the bike up to speed, it is going to take the same amount of energy passed through the spindle to stop

            Comment


            • AZguy
              AZguy commented
              Editing a comment
              There's a big difference between accelerating and braking with a hub motor in the forces on the axle when that energy is transferred to/from the bike

              During acceleration there will be a force applied as a torque to the axle and that torque will also create a resultant lateral force in the direction of travel also applied to the axle... the torque to lateral force will be proportional to the wheel diameter, larger wheels more torque for the same lateral force - the lateral force is what accelerates the bike

              During braking there will be some torque from regen if the motor has it and the rest will be a lateral force opposite the direction of travel - if no regen then no torque

            #25
            My misteak <sic> the calipers take the load not the spindle.

            In the absence of regen.
            Last edited by ynot; 03-15-2022, 12:39 PM.

            Comment


              #26
              Just from my experience with keeping the motor axle in place...

              Rear mounted, direct drive, 1000w (derated to 750 watts) hub motor in a steel frame, and at the time using excessive regeneration braking. Not enough to throw you over the handlebars, but enough that you would want to brace for it. Turned it down to less than half now.


              Having nice axle shoulder nuts with the serrated flat side that goes toward the dropout locking the nut in place. Having them securely torqued down at 50 ft/lbs, which seemed the maximum I wanted to go. What with the percentage of threads removed for the flats... And with the anti-rotation washers with the tab that sits in the dropout. I felt confident that all would be well. About a week later I removed the motor assembly for other reasons and spied this.




              An inboard spacing washer that showed signs of the axle rocking back and forth. :-( Now the nuts were still tight to specs, and that axle was STILL moving!

              Now the axle flats seemed to fit nice and tight...




              So at first I tried a store bought type.



              But with it's gear type design that allowed for different axle orientations, which was sloppy. And something about a worm gear clamp around a frame member just turns me off.

              So the "Prowler" TA was manufactured and installed. ;-) And added jam nuts for fun.


              Bottom line point. With my axle nuts securely held in place, the axle was still rotating. Enough to cause damage in the future...? Perhaps not, but I feel better about it now.
              Could one have increased the nut torque? Perhaps, but the last thing I wanted to do was strip the axle or nut threads...

              Now I know we are not "apples to apples", but just wanted you to have another data point.


              Difficult to offer possible helpful suggestions without "seeing" your installation. Such as flattening the axle on the inboard side and putting a TA there (no room?), or perhaps "clamping" the existing dropout tight to the axle...?

              See my completed Magic Pie V5 rear hub motor E-Bike build HERE.

              Comment


                #27
                Thanks for that TC, pleased that you are getting into this, So you tightened to 70nm +/- bet your motor at 750 watts is putting out more than 70 mn,

                An O/T, Favorite wife was out yesterday on her trike, using the help you gave she is starting to tune the parameters and is closing in on getting to where she is most comfortable, and is enjoying herself. Looking fwd to taking a few rides together.
                Cheers,

                Comment


                • Tommycat
                  Tommycat commented
                  Editing a comment
                  Don't have specific torque specs, but very pleased with its performance. Gets me where I want to go!
                  Wish I could join in on the hill climbing trials. ;-) Great to hear that your wife is having fun, and your weather is getting warm enough to ride!
                  Cheers.

                #28
                Spent yesterday morning machining the dropouts for the freighter trike, machined them in a stack so all four are identical, in a fit of utter stupidity made them too sloppy. Oh well they only took me about 4 hours to make & I have four more I can cut today.

                Comment

                Working...
                X