RC Drive Recumbent Trike Build

[Squilt]

I've been planning my next build. I'll be electrifying this used recumbent trike I bought. At first I thought it was a Performer JC-26X. But that model has direct steering and this has indirect steering. I reached out to someone at Performer Cycles and learned that I have a discontinued model.


Anyway, here are the important specs. It weighs 50 lbs, the top of the seat is 5 inches lower(29") compared to the JC-26X while the bottom seat height is the same, it's got a 9-inch longer wheelbase(54”), and everything else seems the same, with a front track of 29". It has an aluminum frame and CrMo swingarm, with 160mm brake rotors on the front wheels. The weight distribution is nearly 50-50 front and rear. It corners great, but I have to lean a bit. I can lock up the front wheels on flat ground without lifting the rear wheel. I'll try to arrange the weight of batteries to put more weight on the front for better cornering, and hopefully burnouts.


To help decide on a speed to shoot for, I rode it down some hills. I had never ridden something like this before so the first few runs (~35mph) felt sketchy. The trike was sturdy, it was just a me problem. After that, I hauled it out to some more substantial hills where I managed 43mph on a steeper one and averaged about 30mph for over 3 minutes on a longer one. After spending more time at speed I got used to the steering and riding position. It never felt unstable. It didn't do too bad going over bumps and holes in the road. Though I haven't hit anything that substantial at speed. 45 mph should be comfortable after some modifications. I was able to pedal all the way up to about 40, and when I swap the 52t chainring for a 62t, I'll be able to pedal up to 45 with only 100 rpm at the cranks. I'd have to lock out the smallest chainring but even if something fails, I'd still be able to pedal uphill at 6mph with a cadence of 60.

I’m going to use an RC motor driving the left side of the rear wheel because RC drives are fun. Also because I plan to test a 2-speed gearbox I'm thinking up. Don’t want to get into the details yet, but it should allow freewheeling, regen braking, and not have the issue of locking up when rolling backward. I’ll gear it for 35mph in 1st and 50mph in 2nd. It will likely just be a proof of concept since I don't have the facilities(gear hobbing) to make it robust. So I doubt it will be on the final version of this.

For power I figure I'll max out a controller I already have, which would be about 5.7 kW. This number is convenient because building a pack to this spec with the 18650s I have in mind would make it capable of charging at 1 kw at .5C. Which will be compatible with my aspiration to make a series hybrid vehicle. Which won't necessarily be the trike, I want to swap the battery between builds.


Here's a list of components I plan to buy:

(192ct)EVE 29V 18650 2850mAh 8.4A 1.95kWh configured 16s 12p​

I've made some packs with 18650s before so that's why I'm not using pouch cells or just buying a pre-built pack.​

16s 100a DALY Smart BMS​

Have the 60a version on my bike and it's held up fine. The app is handy as well.​

​

Spintend Ubox 75v/100a ESC​

I have this on my bike as well. I haven’t had any problems and I like the VESC Tool app.​

​

Astroflight 3215 Brushless Motor​

Fits the application, but boy are these expensive, if you know of a motor with similar spec that's cheaper let me know. I don't mind losing a bit in efficiency.​

I've concluded that I'll want a steering damper but I'm not sure where to look.

For the wheels & tires I'm thinking between re-lacing the hubs with double wall steel rims and Maxxis Hookworm tires or motorcycle/moped rims and tires.

I'm going to need front suspension, and no such kit exists for this trike. Even if they did, they'd probably be outside my budget. So I'll just have to make it myself. Thinking of copying the style of the kmx trike kit.

 

[chuyskywalker]

Spintend Ubox 75v/100a ESCI have this on my bike as well. I haven’t had any problems and I like the VESC Tool app.

The only annoying thing with VESC for a bike build is lack of a way to integrate a PAS. If you're good with throttle + pedal, it's fine, but I know from my experience with Grin stuff and torque/pas sensors I would really miss it.

 

[amberwolf]

The only annoying thing with VESC for a bike build is lack of a way to integrate a PAS.

I haven't investigated it but there's this thread

Bosch mid drive with vesc controller

Hi All, new project time. Bike: Corratech e-power e28 hybrid, bosch mid drive, alfine igh and hydraulic rim brakes. problem: controller is bricked by wrong one time initialisation code, now gives error 500 slution: gut controller and install enerton vesc foc box controller. I will start...

endless-sphere.com

You can always still use the Cycle Analyst to read all your sensors and produce a throttle signal for the VESC.

Other solutions include using an MCU (arduino, etc) to convert a cadence (or torque, or both) sensor output into a throttle signal, though this requires some DIY and programming knowledge, but there are projects to start off from:

Simple PAS to throttle converter using an Arduino Nano - ES 1.0 Graveyard

PAS To Throttle converter using Arduino Nano - ES 1.0 Graveyard

Open Source CA alternative on German pedelecforum available - ES 1.0 Graveyard

GSM crankdrive with VESC and Teensy "ForumsController" - ES 1.0 Graveyard

I never got to coding anything in my TidBits thread

Arduino Nano Cycle Analyst (and other) companion TidBits - ES 1.0 Graveyard

but I may get back to it someday--in the meantime the info and thoughts there may be useful to someone.

 

[chuyskywalker]

Fair, let me amend:

is lack of a native (easy) way to integrate a PAS

There's actually some work going on for it, but....

Pedal assist support (PAS) by nitrousnrg · Pull Request #243 · vedderb/bldc

This commit enables cadence-based pedal assist for ebikes using quadrature-style pedal speed sensors. There are 2 operation modes: PAS only: Motor current is based only on pedal feedback ADC + PAS...

github.com

Feature request: support for one wire PAS · Issue #525 · vedderb/bldc

Currently only quadrature sensors are supported but many (cheaper) PAS sensors use only a single wire. There are 2 signal types for one wire PAS: One pulse every time a magnet passes when pedalling...

github.com

 

[amberwolf]

If they'd basically just take the CA's present (3.2x) various methods of cadence-sensor PAS and implement them in VESC, it'd be great for the majority of people that want this function.

There are some tweaks to the CA's methods that could be had with a few more options in settings that would make it more useful for certain users.

Torque sensor operation would also be implementable with the CA's methods, if needed.

 

[Squilt]

Decided to get the motorcycle rims and moped tires. Here's what I ordered.

• 1x Tusk Racing, Impact Rim - Front 32 Spoke Hole, 19 x 1.4 Silver

• 2x Tusk Racing, Impact Rim - Rear 32 Spoke Hole, 16 x 1.85 Silver

• 1x Heidenau M3 Moped tire - 2.25-19

• 2x Mitas Sava MC2 Moped tire - 2.25-16

I'll work on designing the drivetrain in the meantime. Assuming the rear tire comes out to 23" OD, I'll use two 4:1 stages. First stage using HTD 5M belt and pulleys, 25mm wide. Second stage using #35 sprockets and chain.

Edit: 1:4 to 4:1. I don't plan on going 11,378MPH

 

[ccihon]

I have a similar trike, also built by Performer. It is rigid frame, direct steer. I have fitted a motorcycle steering damper, mainly because in my case the BBSHD mid drive up front shifted the balance forward and caused some oscillation.

I'd be very interested in your plan to craft a front suspension option. I had thought about steering tube approaches etc, as I can't really fabricate anything. For my application, the suspension is not a big deal.

I think mounting a motor to separately drive the wheel from the left side is a great approach for trikes, especially of the "high powered" type. Do you think the RC motor will cool adequately when used for this purpose? The gearbox sound interesting also.

 

[Squilt]

I have a similar trike, also built by Performer. It is rigid frame, direct steer. I have fitted a motorcycle steering damper, mainly because in my case the BBSHD mid drive up front shifted the balance forward and caused some oscillation.

I'd be very interested in your plan to craft a front suspension option. I had thought about steering tube approaches etc, as I can't really fabricate anything. For my application, the suspension is not a big deal.

I think mounting a motor to separately drive the wheel from the left side is a great approach for trikes, especially of the "high powered" type. Do you think the RC motor will cool adequately when used for this purpose? The gearbox sound interesting also.

I considered steer tube suspension but I really don't want to touch the tie rods. The toe-in is already perfect. I'm lucky to have access to CNC mills and lathes at work, so multi-link suspension is an option assuming I can pull it off.

I plan to bolt or clamp the motor to a plate of aluminum big enough to hold the jackshaft and connect to the frame, which should act as a giant heat sink. I might even be able to mill fins into it. All that along with a fan if necessary should potentially let me run the motor well over its rated current.

 

[scianiac]

Considering you don't need the absolute highest power to weight ratio motor I would go with a much cheaper outrunner. I would rather have a bit more stator volume, an open can design, and be much cheaper at the cost of a little weight and quality. A big outrunner will require less gearing down and can be forced air cooled for high continuous power output. Although there are some downsides like how open the motor is to the elements but that depends on riding conditions and such but an option to consider, there are many builds around here using both motors to compare.

And if you really didn't care about weight just stick a solid IPM motor like a LR motor which will still be cheaper, capible of more power, have bearings solid enough to actually drive from and have a more flat power profile with MTPA and field weakening.

 

[Squilt]

Considering you don't need the absolute highest power to weight ratio motor I would go with a much cheaper outrunner. I would rather have a bit more stator volume, an open can design, and be much cheaper at the cost of a little weight and quality. A big outrunner will require less gearing down and can be forced air cooled for high continuous power output. Although there are some downsides like how open the motor is to the elements but that depends on riding conditions and such but an option to consider, there are many builds around here using both motors to compare.

And if you really didn't care about weight just stick a solid IPM motor like a LR motor which will still be cheaper, capible of more power, have bearings solid enough to actually drive from and have a more flat power profile with MTPA and field weakening.

I'm already having trouble finding space for the drivetrain in this build. I also want to be able to put the motor in another build I have where space is really tight. It's the tiny motorcycle in my pfp. I could use an outrunner, it would just be harder to cool.

Ive spend a lot of time looking for rc motors. The only way I find anything good is going to the websites of reputable vendors and browsing everything they have. Candidates I've found are the APS8072S and the Maytech MTO6575. I'm a bit suspicious of the quality of the APS motor. The Maytech looks solid, but the lead time is 1-2 months.
Where would you look for a suitable RC motor?

 

[scianiac]

I guess cooling depends on how you do it, putting a fan on one end of an open bell design outrunner can cool it quite effectively and with less space than a heatsink in some cases. freerechobby also has some motors (maybe their 8085) but many of these chinese motors are all made in the same factories. I would keep in mind the speed of the motors when it comes to space, those small inrunners generally like to spin very fast and so you might end up spending more space than you save in gearing them down enough. Another option would be the flat heavy lift drone motors, won't fit in the pocket bike and can be tricky to drive with their low inductance but may be easier to place to the side of the rear wheel on the trike. Also if you're willing to gamble you can get those used on aliexpress for real cheap, I've got one of those sitting around somewhere. And those you can pump serious power into with a fan because they cool well due to their open design and short stator length.

 

[The Toecutter]

I look forward to seeing what front suspension design you come up with for this. You're definitely going to need it for anything over 30 mph.

If I were to use an RC motor, I'd be looking at the larger Astroflight models. They can exceed 95% peak efficiency and put out a massive amount of power in a small package.

Here's a trike that runs 3 of them, able to accelerate from 0-40 mph in about 3 seconds:

Custom Build Gallery, Matt’s 50-MPH Monster Yellow Trike

The videos showing off the acceleration really put things into perspective.

For tires, I highly recommend the Mitas MC2. They're very sturdy, and their rolling resistance is comparable to Schwalbe Marathon Plus. Yeah, they're heavy boat anchors, but not nearly as much so as most DOT motorcycle/scooter/moped tires. Budget-sensitive solar race cars use them for a reason.

 

[Squilt]

I look forward to seeing what front suspension design you come up with for this. You're definitely going to need it for anything over 30 mph.

If I were to use an RC motor, I'd be looking at the larger Astroflight models. They can exceed 95% peak efficiency and put out a massive amount of power in a small package.

Here's a trike that runs 3 of them, able to accelerate from 0-40 mph in about 3 seconds:

Custom Build Gallery, Matt’s 50-MPH Monster Yellow Trike

The videos showing off the acceleration really put things into perspective.

For tires, I highly recommend the Mitas MC2. They're very sturdy, and their rolling resistance is comparable to Schwalbe Marathon Plus. Yeah, they're heavy boat anchors, but not nearly as much so as most DOT motorcycle/scooter/moped tires. Budget-sensitive solar race cars use them for a reason.

I'll start on the suspension after it's pretty much all together. I thought maybe I wouldn't need it, but looking at how the front arms on the trike sweep forward, it's easy to imagine how the force of any bump gets amplified down that long lever, straight to where it bolts to the frame. It's a small miracle there aren't already cracks in the aluminum there. If I had to guess, I'd say that's why it was discontinued. There's no forward sweep on the model that seems to have replaced this one.

The kv of the larger 3220 would be convenient, but it is unfortunately sold-out at the moment. The efficiency is appealing for easy cooling, though the Maytech MTO6575 gets that from built-in water cooling while still being quite efficient and smaller. Even if some specs are bogus, at less than half the price of a 3220, I'm willing to accept that. I messaged Maytech, asking if they could make one with hall sensors and a sealed can. I'm waiting on a response.

The rims have arrived, and the tires should arrive today. I'm excited to see how they look on the trike versus the comparatively puny stock wheels.

 

[kylerlaird]

The only annoying thing with VESC for a bike build is lack of a way to integrate a PAS. If you're good with throttle + pedal, it's fine, but I know from my experience with Grin stuff and torque/pas sensors I would really miss it.

I'm using the VESC ADC app for my torque sensor bikes. Grin's solutions are enticing, but I'd miss the ability to customize. And the price.

 

[DanGT86]

What about the CYC motor? Seems almost like a large RC motor.

Some users on here have had good luck with neumotors as well.

NeuMotors 8000 Series BLDC Motors 7,500-45,000 watts - NeuMotors Brushless Motors

neumotors.com

 

[Squilt]

I was busy presenting a project at an expo in San Francisco. Didn't make any real progress until the last few weeks.

But here it is. The wheels are done.

Pretty slick huh?

I took it for a quick ride around and it feels great. These massive gyroscopes make it super stable at higher speeds, a huge difference from before. It expectedly handles bumps pretty well too. The best thing is that next is the fun part, the power. Though I will be making a different thread before I post again here. I'm building a cheap e-bike for a friend so he can get to work. It shouldn't take too long.

 

[The Toecutter]

This thing is awesome. Suspension is really going to improve its ride quality. Glad to see you took my advice on the front tires. They're rated for 62 mph and won't disappoint.

 

[Ecyclist]

You might want to change the rear shock. I will be building a trike myself. I bought three different and I will use one as a donor for parts.
Instead of Astro maybe you can use Alien outrunner. It's much cheaper and you can get it with low KV.

Brushless motors

Brushless motors

alienpowersystem.com

 

[Squilt]

You might want to change the rear shock. I will be building a trike myself. I bought three different and I will use one as a donor for parts.
Instead of Astro maybe you can use Alien outrunner. It's much cheaper and you can get it with low KV.

Brushless motors

Brushless motors

alienpowersystem.com

The rear shock feels pretty good to me, but I don't know much about it.

Most of the options on APS' site look to be on backorder so it'll take up to 6 weeks until it ships. Though I don't know what Astro's lead times are like. And they don't provide important info like no load current and resistance. I'm still looking around. The 3215 has efficiency and the convenience of it's rated voltage and current being right where I want it. It's form factor, lack of hall sensors, and use of an imperial shaft size, and of course cost, are things other motors have over it, but I'm struggling to find quality options that I can just drop in an online shopping cart and order. Opposed to calling a salesperson only to be informed of long lead times and minimum order quantities. I don't think I'll find a better option than the 3215 before it's time to order a motor.

 

[Ecyclist]

Your shock is only a spring and has no rebound control. Maybe I'm wrong.
I used APS 6373 on my build and it works good for four years.
I'll use 80mm on the trike. Anyways, good luck.

 

[Squilt]

Your shock is only a spring and has no rebound control. Maybe I'm wrong.
I used APS 6373 on my build and it works good for four years.
I'll use 80mm on the trike. Anyways, good luck.

I sent aps a message about their 83100SW. The ratings they have for torque, current, and voltage are almost the same across all of the different kv motors. I'll likely pick a higher kv motor for more rpms and power. Though I doubt their 200kv motor can handle 15k rpm like the voltage rating would suggest. I was planning on a 2 stage reduction anyway. The watercooling is a plus. If the efficiency is good enough I might get it.

 

[Ecyclist]

Just cut all ratings in half and you will be OK.
My plan is to have a single stage by replacing a brake rotor with a custom made sprocket.
I'm not that ambition and will be happy with 5 maybe 7kW.

 

[Ecyclist]

Is this project caput?

 

[marka-ee]

I was watching with interest also.

 

[Squilt]

I was going to wait until I had something more substantial to show. I haven't had much time but progress is picking back up. I think I know what the drivetrain is going to look like. I'm 3D printing test pieces now.



Choosing a belt for the first stage reduction would have made the design way easier but I couldn't shake the appeal of using steel spur gears. Clearances are tight but it's looking good so far.