L-Faster / Dome Motor 450W Mid-Drive Build

richj8990

100 W
Joined
Jan 21, 2020
Messages
277
Finally got everything to work. I wanted to do a test ride up a 12% fire road near my home but decided to instead clear some overgrowth 20 minutes away by riding this offroad about 8 miles total. 50 lbs with front wheel/tire. Will try to trim closer to 45 lbs with different crankset later. Notes below.

image000000.jpg



1. Amazon and their own website is L-Faster. E-Bay is Dome Motor. They are slightly different; the Dome Motor one has more mid-drive specific parts in their kit including a 16T freewheel sprocket connected to the motor with a (flimsy) tensioner. If you are going to buy this motor from one of the three sites, buy the mid-drive one for sure, absolutely do not buy the one that attaches to the left chainstay. Too many alignment problems with that one. Not that this mid-drive was a breeze to install; I still need some major tweaks later that will be mentioned below. The one to buy is the 450-750W 36/48V mid-drive. Amazon sells for $404, E-Bay for $420 not including shipping.

2. Why did I decide to get this instead of, for example a CYC Stealth? Because the accessories are more or less 100% open-source. As in you pick your own battery, controller/display, and possibly even your own crankset (see below for more theoretical details on that). So you can basically set this up with whatever power, throttle/pas, battery life, and number of chainrings that you want to; it's very modular. I would call it a hub drive motor that happens to be placed in the middle of the bike and has a 2nd geared chain. That's very different than a Bafang BSSxx or even a CYC that's their own crankset / controller / display that you have to use.

3. Installation of the motor was...interesting. The 'inner' q-factor, as in the inside of the cranks from crank to crank on the square-taped stuff that came with the kit is 170mm wide. For comparison, most Shimano Hollowtech II cranksets are around 150mm, whether the cranks themselves are curved outwards or straight. The reason I mention this is that the motor's widest point from sprocket to other non-drive side is 140mm. In theory that would allow you to use a hollowtech-like crankset w/o freewheel. In reality, the drive side is offset a lot. Which means to use a different crankset, you'd need to have the equivalent of shims (in this case flat plates or repair plates) extend the motor away from the cranks, up a few inches on the down tube. That creates its own can of worms for a longer drive chain and possibly a more wobbly motor. But I feel it's worth it because this kit crankset is so damned heavy at 7 lbs, complicated with freewheels all over the place, and the chainrings are not even narrow-wide. It's very low-grade, and it's already wobbly so I'm not really losing much by extending the motor out later. I had to invert the motor-driven chainring and now can't use the large 44T chainring; just using the 34T middle one for now. That ring climbs good, details below. Pictures of how I needed to put some shim spacers on the drive side as well. You can see that the motor sprocket, tensioner, and largest chainring are within a couple mm of each other with some mods.

image000003.jpg

image000002.jpg

I've also included an image of what I'll try for extending the motor with...keeping the metal plates that with the kit and adding these to those original plates.


61eS4VhXddL._AC_SL1100_.jpg


4. The stock controller was a complete POS. But that's why this motor could actually be worth it...because like a hub motor, you simply change the controller/display to what you want. KT LCD-5 and generic 36/48V 12 mosfet controller worked. PAS KT V12 however is not working. The controller I'm now using has SM plugs and a male 3-pin for the PAS. I can't find a hollowtech (big hole) PAS with an SM 3-hole female pin to mate to this controller. So...why not simply buy two different PAS systems that are KT-compatible and combine both pieces? I'll see later if it works. The only issue I see is that the chainring magnets have 12, and the 3-pin female SM PAS sensor is built for 8 magnets. But I think it should be fine. Either it works or it doesn't. Pictures of what I'll try later for PAS are below.


614dFz+EWgL._AC_SX425_.jpg


41MSIRw5rlL._AC_SX425_.jpg


5. Remember I'm trying later to convert from super-heavy square tapered to 24mm hollowtech spindle, so the lower PAS magnet ring will not fit. Only the steel sensor ring with SM plug will fit; that's why I need the upper 12-magnet ring that attaches to the chainring. I'll probably use a triple chainring with the magnets attached to something I most likely will not use (don't have a front derailleur and it's not easy changing the cassette chain on something with a single-speed chain already on the outside ring!). Middle ring will probably be 38T; right now it's 34T and it climbs pretty dang good. Outer ring will be 52T narrow-wide for a 3.25 reduction ratio.

6. This would lead to the pedals automatically moving when the motor turns. No 'safety' freewheel(s). With a 38T chainring, unless I'm in a really low gear on an 11-52T cassette, I don't see a problem with that. If I have to walk the bike up to start out smoothly instead of risking the pedals turning too fast, so be it. Have to walk up on a lot of bikes anyway, e-powered or not if you run out of steam up a steep section. I'm used to it. My area has some super steep hills.

7. I'll update on how this does vs. a Bafang 500W front-hub motor on a 12% offroad incline. My personal record was 6.5 mph avg, normally I casually do about 4-5 mph avg. up that. We'll see how this does. I don't feel much of a torque difference on pavement vs. a hub drive, but this motor does rev up quickly when the throttle is pushed. The rear wheel NEVER, ever spun out today. I was a bit surprised. The bike will stop before the rear wheel spins out. I was in a couple gears too high on a small steep/loose section and the drivetrain just clawed up that with no complaints. That's very different than struggling with a hub drive up something like that.



So I think overall, if I can get the crankset right, this motor is not bad at all with a good controller/display. And really, you 'could' buy the $115 rear wheel chainstay mounted one and then DIY the motor mounts, buy the controller/display separately and still probably be fine if you are good at this kind of stuff. If you are looking for a very modular mid-drive that you can add your own accessories for, this one is pretty good, but you'll need to for sure do some extra tweaks. Isn't that always how it is though?
 
Last edited:
¿Has probado si sube colinas empinadas?how many watts are you running the bike? discharge amps?
 
Last edited:
¿Has probado si sube colinas empinadas?¿A cuantos watts le está dando la moto? amperios de descarga?
 
Have you tested if it goes up steep hills? How many watts is the bike putting out? discharge amps?

Sorry I cannot paste a Spanish translation on here.

Yes, but only on steep pavement for something long. Feels about 700W power, not sure of the torque. If it's throttle only and no pedaling, the power will fade just a bit up steep pavement but not as much as a (geared) hub drive would. Have not done a several-hundred foot hill on dirt yet but I can promise you this thing can climb. It easily climbed a 30-foot 15% incline with loose rocks and ruts in the wrong gear. So there is no doubt. Of course, you don't try that with 26 x 1.95 semi slick tires either! I had on 27.5 x 2.8 Maxxis DHF front, 27.5 x 2.4 Kenda Hellkat rear. I see all these other mid-drive conversions online, and they convert some $300 bike with street tires. What's the point? Hub drive would be a lot simpler. No need for a mid-drive if you have that bike and tires.

You can buy these for $100-150 on Amazon that were designed for rear brake mounting, you just throw out everything in that kit but the motor & chain, buy new controller / display / compatible throttle and PAS, then buy cheap mending plates with holes and attach them to the bottom bracket area yourself. Remember to find a steel plate hole that's around 35mm in diameter, so you can fit that plate with either the square tapered 3-splined screws, or better yet external bb bearing cup screws like Hollowtech, etc.

The whole thing would be under $300 USD and you'd have a motor with around 750W peak power, and I'm assuming 75 Nm peak torque. PAS is not working right now so it's hard to say how it compares with peak hub drive PAS power. Throttle power as mentioned feels around 700W peak. 48V = around 15A peak with a 22A limited KT sine wave controller. Remember THAT controller didn't come with the kit, you should buy the controller/display separately, what came with the kit is crap. It's the motor you want. Because this is one of the few direct drive motors with sprocket that has more than 200W, can actually fit under your down tube, and not stick out too much.





























 

Last year I bought a GNG Gen1 BB drive kit and mounted it on my Fuji Hybrid Commuter. I was really happy with the kit except from the terrible quality of the frewheeling crank (I think someone here mentioned it was made from butter :)). During the first 6 months I replaced the freewheel 3 times.

Since I anyway don't just drive without pedaling, I figured I'd have an attempt to combine the sprockets from GNG freewheel crank and a standard Shimano Crank in order to eliminate the weak point.

So I found that:

* GNG sprockets have 4 6mm holes with a BCD of 64 mm
* Shimano FCM361 Crank has BCD of 104mm for the two outer sprocket and BCD 64 mm for the inner sprocket

So I ordered a Shimano FCM361, drilled up the holes in the GNG sprocket to 8mm and replaced the inner sprocket of the crank with the GNG sprocket, took off the sprocket in center, and shifted the outer sprocket into the position of the center sprocket.

Result:

* Perfectly aligned outer sprocket with the motor sprocket.
* The inner GNG sprocket used for chain to rear wheel.
* No freewheel.

PS: Remember to not touch the throttle when not pedaling :)

------------------------------------------------------------------------------------------------------------------------------------------------------

This is very similar to what I intend on doing (besides needing to make a new motor to crank bracket), but I don't understand why the above poster needed to keep the GNG sprocket. Does that have a directional freewheel inside or something?

In a way I have an advantage over the above, because with the mending plate brakets and washer/shim/spacers, I can simply move the motor and its 16T sprocket over to get a decent chainline for the outermost (plan on 48 or 52T) chainring. In my case, the middle chainring will drive the rear cassette, probably will be 36-38T. Small 64 BCD chainring will do nothing except hold the magnets for PAS.

Another thing I'm confused about above (not to be picky or anything) is that for most 3x cranksets, the large chainring is a minimum of 42T and is not desigend to be placed in the middle. It has four inner sides that halfway converge (but do not touch each other) to the outside of the crankset with bolts, but it's not circular on the inside like a 30-38T middle chainring. The 4 bolt sides converge together on the inner area of a 30-38T chainring and do not on a 42-52T chainring. That's why you can mount two chainrings on one outside crank spider, because they are not identical in structure; one stacks on the outside of the crankset spider and the other nests inside. Then the really inner 22-28T chainring is 56 bcd on a separate spider. Moving a larger chainring like that to the middle is bad because there is no inner support from the crankset, there are only the bolts holding it in space. Not a good idea for a powerful mid-drive.
 
Si algún día escalas una montaña, publica cómo te fue, pero escalar una montaña, no escalar una simple colina.
 
Si algún día escalas una montaña, publica cómo te fue, pero escalar una montaña, no escalar una simple colina.

Have not climbed a mountain yet. 71 miles on it. Over a dozen motor chain drops. Need to fix that before I proceed to climb mountains lol. But it really does climb better than a geared hub drive. It's amazing, the chain just holds the rear wheel steady up dirt while you can almost randomly select a gear. Even a higher gear! If I produced that much torque naturally on a normal bike, the rear wheel would probably spin out but a mid-drive one just casually chugs up the hill like it was born to do it.
 
That seat looks uncomfortable, however the clamping grips is a nice touch with the color. I'm odd, sometimes I dont use grips.

I see you got holes in your frame, where-abouts the battery is located. You may want to check for cracks around and near the hole in the frame every now and then bruh. Sorry to say your battery case is on the bicycle tube (down tube) with the hole in it. If I was you, I would try to re-locate the battery to hang from the top tube (only if you can, highly doubtful by the looks of things bruh). It is what it is, as long as you aint putting a lot of forces on the battery, dropping the bike on the ground or jumping, you should be fine. If its a generic bicycle, then I would be a little more hesitant.

I hate to be the bummer dude, but your ride is cool and I respect that.

o you can basically set this up with whatever power, throttle/pas, battery life, and number of chainrings that you want to; it's very modular. I would call it a hub drive motor that happens to be placed in the middle of the bike and has a 2nd geared chain. That's very different than a Bafang BSSxx or even a CYC that's their own crankset / controller / display that you have to use.
 
Last edited:
That seat looks uncomfortable, however the clamping grips is a nice touch with the color. I'm odd, sometimes I dont use grips.

I see you got holes in your frame, where-abouts the battery is located. You may want to check for cracks around and near the hole in the frame every now and then bruh. Sorry to say your battery case is on the bicycle tube (down tube) with the hole in it. If I was you, I would try to re-locate the battery to hang from the top tube (only if you can, highly doubtful by the looks of things bruh). It is what it is, as long as you aint putting a lot of forces on the battery, dropping the bike on the ground or jumping, you should be fine. If its a generic bicycle, then I would be a little more hesitant.

I hate to be the bummer dude, but your ride is cool and I respect that.


I had to mount the battery backwards because of the bottle hole placement. Yes, when this battery dies I'll do a top tube one. I've ridden maybe 2000 miles offroad with 3 different frames that had batteries like this (one rightside up, two upside down), never an issue. Of course I don't go through chunk at 20 mph either like the maniacs do. This is definitely, 100% not for enduro riding. I don't even really do that on normal bikes. Some of my downhill times are pretty decent but that's finessing it with switchbacks, that's not going Neanderthal straight down the hill like the unibrow 5th grade education missing front teeth caveman riders do. You know, the ones that are in the hospital every 3 months, and sit around recovering more than they actually ride. Not my style. This is just for climbing and exploring.

I had to redo the drivetrain, switched to Hollowtech superboost and I think it's going to work! Just need to screw around with one side to mount the motor properly, other side has a stock mount attached to the motor and external bottom bracket bearing (in between it and the frame). No PAS, unfortunately.
 
Have to give up on the Hollowtech crankset. No matter how much I reinforce the side w/o the original motor mount, the sprocket side under torque moves over and clips the pedal crank on the other side. If I let off the throttle the motor will literally snap back into proper place about 8mm from the pedal crank. I can't use both stock motor mounts because the motor has to be offset away from the sprocket side too much for both to fit. It was close. Very close. Worth a try. But I've spent a dozen hours on it and need to quit while I'm behind... Will keep this boosted crankset for a non ebike, normally the Deore XT's are over $200 and I got it for $125; can still fit on a normal frame with some spacers. And very very reluctantly buy a square-tapered crankset; already have the elongated bb. That way I can go back to both motor mounts. A lot of respect now to the DIY people who make their own mid-drive motor mounts because this particular motor needs a lot of restraining...
 
what is the spindle length on the superboost crankset? I'm looking for the distance between the inside surface between the cranks. If they were installed on a 73mm BB would there be a gap on the spindle on each side that could accommodate a support bearing for custom FW crank designs?
 
what is the spindle length on the superboost crankset? I'm looking for the distance between the inside surface between the cranks. If they were installed on a 73mm BB would there be a gap on the spindle on each side that could accommodate a support bearing for custom FW crank designs?

Before I answer that, I knew even before I ordered this Chinese one-off that I would probably have a lot of frustration, and also learn a lot at the same time. Both happened. There were times it was so frustrating I had to immediately grab a beer and calm down after a humid sweaty night in the garage trying and failing to get something to work. I'm sure you've had that feeling before with some of these bike components.

OK, the answer: the inside pedal to pedal crank arm measurement is more important than the spindle length. I'll explain why. Superboost (and remember to others reading this we are talking about the middle of the bike's crank axle/spindle width, not the rear dropout width that defines the actual term Standard 135 - 142 / Boost 148 / Superboost 157), has a crank spindle width of about 136mm (bare spindle w/o the pedals or crankset being counted). That's about 10mm more than a standard 135/142 crankset. Yes, if the bearing was 10mm it could fit between a standard frame and the crankset. The kit came with two 5mm external spacers that are 24mm ID and fit outside the bearing cups as opposed to the larger cassette-spacer ones that fit in-between the cups and the frame. So you are getting 10mm more. If you need 5-10mm more, great. I need even more than that because of motor mounting bracket flex.

The important thing IMO (and you have more experience in this than me) is the clearance between the motor (sprocket or not) and the crank arms. The outside q factor for this particular Deore XT M8130 is 181mm. Crank arms at the end are about 10mm each. So the inside q-factor if you want to call it that is about 160mm. That's the important measurement for the arms not hitting anything. The frustrating thing is that a $50 standard width Hollowtech crankset was 10mm shorter for the spindle BUT the arms were curved outward more so the inside q-factor was again about 160mm. My motor's width including sprocket is about 140mm. On paper with EITHER crankset I have 10mm on each side for the pedals to clear the motor. So the Superboost didn't help anything, because the crank arm to arm distance was still the same width. In addition, the mounting brackets flex to the point that under high torque the drive crank arm will clip the drive sprocket, just by maybe 1-2mm. Same exact thing happened with both a standard and Superboost crankset installed. Under torque the motor is literally twisting over to the right and making the crank arm clip it! It's hitting enough that this is unreliable. I can't go out 5-8 miles offroad from pavement with this setup! No way. I know when to quit.

So this is probably one of the reasons why the square-tapered spindles on these setups are super long, and the crank arms are curved to the point that they are allowing 20+mm of space between the motor and the arms. For a very, very needed fudge factor. Square-tapered crankset is coming in the mail Thursday. Another week, another waiting for Amazon parts to try again. BTW the stock square-tapered spindle width is 148mm. You may be able to get away with 142mm but maybe not 127mm. Forget Hollowtech unless your entire motor + sprocket length is less than 130mm.
 
OK...everything is good again. New square-tapered crankset came in, and thank God it had a 64 bcd mount on the inside. Because now this 148 spindle is too long. Had to add washers to put the drive chainring several mm on the inside of the 104 bcd spider. And remember I still need to mount the cassette-driving chainring...to...? To the inside 64 bcd mount. How? You can choose from a few different adapters that WERE NOT supposed to be 64 to 104 but they can be anyway. They are direct mount to 104 bcd, but 1/2 way up between the direct mount hole and the 104 holes are additional oblong holes that are maybe just for show...they obviously are not made to do 64 but they work and as long as they work I don't care!

51+DWJAumaL._AC_SX466_.jpg


So you don't use the direct mount hole; that just fits over the square-tapered spindle and does nothing. I ended up joining both chainrings together with washers and this adapter (with long chainring bolts), and then I reinforced everything with 4 more chainring bolts through the adapter's oblong holes to the 64 accepter threads. It works. And trust me, after about 30 hours of working on this, that's all that matters.
 
So you don't use the direct mount hole; that just fits over the square-tapered spindle and does nothing. I ended up joining both chainrings together with washers and this adapter (with long chainring bolts), and then I reinforced everything with 4 more chainring bolts through the adapter's oblong holes to the 64 accepter threads. It works. And trust me, after about 30 hours of working on this, that's all that matters.
Nice, do you have some pics? would like to see how you did it.
 
Wow, this is a real neat thread going on; do you have more pics? And links to some of the specific items used? I never considered mid drives but I could change my tune depending on what's shown here.
 
I had a pretty nice morning! Considering it was 93% humidity. Was able to climb all 4 technical inclines on this one singletrack trail a few miles from my home. Did all 4 with no issues. Now...I have to admit that in the dozen or so times I tried it with a front-hub motor, I also could clean that hill once or twice (I have the video for the hub drive one below). But it's way slower and harder with a 500W hub drive as you'll see; slowing down for embedded rocks is more that I'm running 26" tires on a 27.5 frame with the hub drive than any power issue. I'll make a video later with the mid-drive one on this climb as a comparison to show how it's faster and smoother. And I'm simply not tired at the top with the mid-drive; the bike does almost all the work for me. The hub drive feels way more like a normal bike climb with some electrical assist but it can be exhausting.

Remember this video has a HUB drive, not the mid-drive!




Mid-drive pics below. Notice how the outside drive chain is fairly slack; it has to be that way for some reason. Too tight and the chain will pop off the ring or the sprocket.

Mid-Drive I.jpgMid-Drive II.jpgMid-Drive III.jpg
 
Nice tidy package there. If you could sneak a sheetmetal bracket under your bottle cage mount and attach it to one of the motor bolts I think it would stop the flexing issues you are having. A little support all the way out in the front of the motor goes a long way. Even a bracket hose clamped to the frame would really help.

I have been keeping an eye on trials bike cranks for a while now. They have lots of evolving crankset and bottom bracket tech that could really help us. Check out these cranks! They use shimano hollowtech 24mm system and fit 68-73mm BB. The drive side is FW threads. Looks like a high quality solution for some of us.
extention-bb24-crankset.jpg

 
Nice tidy package there. If you could sneak a sheetmetal bracket under your bottle cage mount and attach it to one of the motor bolts I think it would stop the flexing issues you are having. A little support all the way out in the front of the motor goes a long way. Even a bracket hose clamped to the frame would really help.

I have been keeping an eye on trials bike cranks for a while now. They have lots of evolving crankset and bottom bracket tech that could really help us. Check out these cranks! They use shimano hollowtech 24mm system and fit 68-73mm BB. The drive side is FW threads. Looks like a high quality solution for some of us.
View attachment 337500


That's pretty cool but how wide are they? How wide is the axle/spindle?

No more flex (at least no more than a couple mm) after I took off the hollowtech crankset and went back to the square-tapered axle (bought a new square-tapered crankset for $40). That allowed me to use the stock motor mount brackets on both sides. Everything works good except the controller/display sometimes cuts out or goes down one or more PAS levels. I assume that's probably because stuff is heating up inside the motor and/or controller. I don't think I'm doing much more than 700W and the limit is 1050W. May need to change the LCD5 to an LCD3 that I have sitting around in the garage. LCD3 can also read watts, the 5 cannot. 5 is only better because it's more compact and fits on the handlebar better.

Other than that, the bike is good to go...need to go from 34T to 38T after the cassette chain finishes wearing out which will not be much longer, maybe a few hundred miles. 34T is sluggish for speed...the bike is asking for more top-end range!

One last thing: you can absolutely pedal with zero problems without any mid-drive freewheel/clutch system. So don't worry about that part if it breaks on your bike or you are thinking of somehow removing it from whatever mid-drive system you have. There is no horrible drag or anything. Battery died last weekend and I was able to pedal a lot better than I thought. Not that I'll get into that habit uphill w/o any electrical assist, but yes you can 100% pedal w/o a freewheel in the crankset with also no electrical power.

Only issue with a fixed crankset and mid-drive is that when I'm approaching a chunky, rocky area on the trail, my natural instinct is to hold the pedals at 3 and 9 o'clock (flat) and coast through the rocks. But you can't do that if electric power is still going through the system, the bike is still pedaling! You have to cut the power off and then try to avoid a pedal strike. So one extra step. Oh well. Not a huge issue. If it's slightly uphill and you need e-power, you just have to plow through the rocks slowly and hope there are no pedal strikes, most likely there will be none anyway but if there are, the strike is deflecting the direction of a 50 lb bike under power off the trail and you need some immediate skills to save the bike from an awkward crash. That's probably the largest 'potential' issue of a fixed mid-drive crankset on singletrack. Pavement/gravel, no worries at all with a fixed crankset.
 
Just found out last night that this motor (which is pretty common) is originally a 2013 GNG 450W mid-drive. Spinning Magnets did a writeup on this motor a long time ago. It has since "percolated" through the Chinese / Taiwanese monolith of distributors. Don't believe the 'brushed' term in the link, I think that was a typo. I'm pretty sure it's brushless.

GNG 450W Mid-drive Kit

GNG is doing the side-mounted options with it as well. L-Faster is either ripping them off or in collaboration with GNG, lord knows which. The side-mounted motor is ridiculous because it removes the rear disk brake in order to work. Losing part of your braking on an electric bike does not sound very wise. Will stick with the mid-drive motor. Some reviewers have said they got the motor up to 2000W. On the Grin simulator, peak torque (46 chainring / 52 granny cog) is 150 Nm with a 45A max controller at about 600W (most simulations put the peak torque of this between 450-600W). Take that BSSHD! 110 Nm peak with a 30A max controller. Not bad for a motor that by itself costs $100-150 to purchase including shipping. Don't ignore the older stuff out there, just modify it for your needs.
 
SW900 and generic 30A controller did not work well today. Wayyyy too much lag between shifts. Could hear the drive sprocket try to engage the drive chain for 2-3 seconds and then after it did, a sudden lurch forward in gear. PAS 3 or higher was a large lurch! Not cool. Happened every time I shifted. Every time. Power was smooth after the shift. But this controller/display pair has failed the test. The lag (especially in a lower gear before an incline) happens sometimes with KT but not nearly as laggy or jerky. And this KT controller is square wave, go figure. But the generic controller paired with the SW900 is even worse. And that's OK, I'll return it. It was an experiment, and now I know the results. Have to stick to KT for better or worse on this motor.

Waiting on KT sine wave 45A controller to come in.
 
I'm not surprised to hear a generic controller was too abrupt. A lot of them are still speed based which means they throw all of the current at the motor until the commanded speed is reached. They are notoriously violent on ramp up. Using a device like a cycle analyst to PWM the throttle signal to the controller can temper this behavior to some extent but that's a bit dated approach from the old days when there just werent advanced controller options.

To get smooth non jerky power application I think you are going to need something with advanced control of current/torque. Some of the advanced controllers allow you to do a virtual freewheel or adjust a minimum torque to keep the backlash in the drivetrain loaded all the time. I think it's the zero resistance to torque at the initial throttle that allows the cheap controllers to violently ramp up.

Something like a nucular 6fet would be a good fit. Nucular has a "clutch" feature specifically for preventing the abrupt lurch when using PAS system. The KT with open source firmware will likely be much better than generic ebike controllers. A VESC variant might also provide you with the precise control of the ramp up speed but last I saw the PAS integration was not very common in that world. That may have changed.
 
I'm not surprised to hear a generic controller was too abrupt. A lot of them are still speed based which means they throw all of the current at the motor until the commanded speed is reached. They are notoriously violent on ramp up. Using a device like a cycle analyst to PWM the throttle signal to the controller can temper this behavior to some extent but that's a bit dated approach from the old days when there just werent advanced controller options.

To get smooth non jerky power application I think you are going to need something with advanced control of current/torque. Some of the advanced controllers allow you to do a virtual freewheel or adjust a minimum torque to keep the backlash in the drivetrain loaded all the time. I think it's the zero resistance to torque at the initial throttle that allows the cheap controllers to violently ramp up.

Something like a nucular 6fet would be a good fit. Nucular has a "clutch" feature specifically for preventing the abrupt lurch when using PAS system. The KT with open source firmware will likely be much better than generic ebike controllers. A VESC variant might also provide you with the precise control of the ramp up speed but last I saw the PAS integration was not very common in that world. That may have changed.

Changed to KT LCD3 (again, had it on another bike), problem for now is fixed. For now. If the throttle is squeezed right, power is fairly smooth. And actually get to see watts again. FYI only the LCD3 and maybe 1-2 other KT LCD's can read watts, which I find extremely bizzarre. Watts are extremely important to monitor especially if you are troubleshooting. I could not figure out a different problem in the system until I read the watt output with this display.

Cannot use PAS; my crankset is fixed and directly tied to the motor (no freewheels/clutches). Well I could do PAS if I was feeling suicidal lol.

I'm going to test out 1/2 dozen different throttles one at a time to see which one has the least deadband at the beginning. This one has at least 30% of the beginning travel as deadband, no power. Will return the generic controller. I'm going to stick with KT stuff for a while on this motor. KT is not perfect, there are several things I would change in the display for options, but it does what I need it to do so will stick with it for now.
 
Back
Top