SWB Vision R-40(E) Recumbent with DD Hub Mid-Drive

[LI-ghtcycle]

Hello all!

Time to make a "new years resolution ...." gonna get a motor that I KNOW will do everything I want, unfortunately, that was NOT the goldenmotor 650W trike motor, it was pretty good, but on serious hills, it just wasn't enough for when I would loose my strength after a long ride, AND I want to be able to haul cargo.

My first impulse is to get another CroMotor, but I will check and see if there isn't something in the middle between the 9C 9x7 which is close but not quite enough, and the 5000W continuous CroMotor.

I will be running a 48 - 74V system (A123 prismatic at 48v 20AH or 74 LiPo 18AH if that isn't able to perform as needed) and I am looking to climb hills with about 450 lbs of cargobike, (that is total weight including the rider, bike and cargo) up a 10% grade for about 2 miles on a regular basis.

I really like the near silent operation of a DD hub motor, and with the NuVinci N171B IGH, I will have tons of gearing to use, with a target of 20 MPH on the flat,(and as close to that as I can get to that speed climbing) while still having reasonable range of around 30 miles.

Anyway, here are some preliminary drawings of my planned build:

(The basic design will use the through holes already on the R-40's main beam, I am hoping probably too optimistically to use the QR skewers which are for the seat mount, but I expect I will end up using larger bolts, but I really want some form of QR fastener for easy chain tensioning.)



(Here is a drawing of how I would design the two side plates to attach constructed out of 1/4" steel, hopefully the local college and or machine shop have a way to bend the steel that thick, that or I will use Aluminum with Steel Inserts at the Hub Axle & Since the NuVinci Freewheels internally, I will be using a track Cog on the Hub, the pedal chain will be on a second complete chain just outboard and will be using a Freewheel at the NuVinci hub in the rear wheel.)

 

[LI-ghtcycle]

Amazon the rescue!

http://www.amazon.com/Axle-Release-Classic-Adapter-3-8-Inch/dp/B004Y97ZLK

I'm going to up-grade the skewers to these 3/8" adapters!

3/8" bolts should be more than enough, AND I get a QR mount!

 

[LI-ghtcycle]

Up-Dated Drawing:

 

[LI-ghtcycle]

I just found a perfect off the shelf item to act as my "Polymer Soft Mount" could only take 2 of these to do the trick, but I would probably use 3 just to spread the load more evenly:

http://hosewarehouse.com/HHD24P-Dixon-Heavy-Duty-Series-Pipe-and-Tube-Clamp-1-1-2-Pipe?search=Dixon%20Heavy%20Duty%20Series%20Pipe%20and%20Tube%20Clamp&page=2

 

[Rassy]

I really enjoy seeing your build of my old Vision. What a great home it found. Good luck with the project.

Here's my thread from 2007 when I first set up the Vision with a pusher trailer. It was configured as a LWB recumbent then.

https://endless-sphere.com/forums/viewtopic.php?f=6&t=2360&hilit=Vision

 

[LI-ghtcycle]

No, my friend, Thank You!

I doubt I would have learned of the amazing difference riding a recliner on 2 wheels Vs the typical DF bike unless I had received your generous gift!

Not to mention, I cannot think of a easier design to add a mid-drive to, and actually, now that I have taken a fresh look at this design and how easily it can be built, I am going to adapt a similar one to the E-Bronco, little to no machining needed (put that design on hold), easily built at home with minimal tools/skills needed, and has an inherently simple & strong design!

Will post more pics soon, and thanks for the link to your push trailer build! I hadn't seen that one before! 8)

 

[LI-ghtcycle]

Well, I have had a realization today!

Not only will this design work for my E-Bronco cruiser, but I can make THIS design much simpler too!

Instead of the fancy bends put into the arms, just have wider spacers and keep them flat!

I am confident that this shouldn't reduce stability as I am only adding 1- 1/2 - 2" total width and it will still be plenty strong, adding in addition spacers in plastic, and if needed, I can space the clamps further apart.

 

[LI-ghtcycle]

Just pretend that I changed this drawing to have the aluminum arms straight and added additional spacers instead of having bent arms.

 

[spinningmagnets]

I will check and see if there isn't something in the middle between the 9C 9x7 which is close but not quite enough, and the 5000W continuous CroMotor

The current value leader for hot rods is the MXUS 3000W V2 (45mm wide stator, thick aluminum stator support, thin laminations)

The common 9C you mention had a 27mm wide stator, thicker laminations, and a thin stamped steel stator that was lighter than thick aluminum, but provided no temperature heat-spike absorption.

Power-wise, the 35mm wide stators are between those two. The Leaf 1500W has thin lams, but no aluminum stator. The MXUS 1000W has thin lams and an aluminum stator support, but only 28mm wide.

For this application, I'd suggest the Leafbike, and add ferro fluid. It comes in 4 different windings. About the same cost as the MXUS 3000W, but lighter and narrower. If you are leaning towards only needing a little more power than the 9C, get the MXUS 1000W V2 (with FF?)...it can take more continuous and peak amps, generates less heat than the 9C for the same input watts, sheds heat better after its warm...

 

[LI-ghtcycle]

Oh! I forgot to note, in the new design, I am not using the QR from the seat as a mount, instead, I am just using the spacing of the pipe clamp to have bolts outside the main tube, and with two being Quick Release, it will make for quick and easy chain adjustment.

 

[LI-ghtcycle]

I will check and see if there isn't something in the middle between the 9C 9x7 which is close but not quite enough, and the 5000W continuous CroMotor

The current value leader for hot rods is the MXUS 3000W V2 (45mm wide stator, thick aluminum stator support, thin laminations)

The common 9C you mention had a 27mm wide stator, thicker laminations, and a thin stamped steel stator that was lighter than thick aluminum, but provided no temperature heat-spike absorption.

Power-wise, the 35mm wide stators are between those two. The Leaf 1500W has thin lams, but no aluminum stator. The MXUS 1000W has thin lams and an aluminum stator support, but only 28mm wide.

For this application, I'd suggest the Leafbike, and add ferro fluid. It comes in 4 different windings. About the same cost as the MXUS 3000W, but lighter and narrower. If you are leaning towards only needing a little more power than the 9C, get the MXUS 1000W V2 (with FF?)...it can take more continuous and peak amps, generates less heat than the 9C for the same input watts, sheds heat better after its warm...

Interesting!

I'm a bit out of practice here, so the ferro fluid is a liquid that is inside for the purpose of heat transfer?

Also, what would be your opinion of the best over-all quality of the 3 mub motors you suggested?

I'm wanting to be able to soak a lot of heat while still being in the 90% efficiency range, in other words, would you say that the MXUS 3000W has as good or better ability to soak heat from initial starting with a heavy load with next to no pedaling compared to the other two?

How about bearings, and does the FF require special tools or w/e to replace when I need to replace bearings?

Thanks for the input SM!

 

[amberwolf]

FF is demonstrated and tested in the recent posts in the "definitive tests of heating and cooling hubmotors" thread, I think started by Justin_LE.

 

[LI-ghtcycle]

Thanks AW!

I am leaning toward the MXUS 3000 since it already has a temp sensor, and the V2 with aluminum stator sounds about right for me, but I'm having difficulty finding a efficiency rating listed anywhere? *** edit, found an ebay listing saying 85% ? HAS to be true right, it was on Ebay! ***

 

[Warren]

LI-ghtcycle,

The thought of a 3000 watt MXUS running through the gears on a very short wheelbase recumbent is a bit scary. Better run the throttle through the CA to tame it down a lot, or you will find yourself on your back before you can say goddamn!

 

[LI-ghtcycle]

LI-ghtcycle,

The thought of a 3000 watt MXUS running through the gears on a very short wheelbase recumbent is a bit scary. Better run the throttle through the CA to tame it down a lot, or you will find yourself on your back before you can say goddamn!

Well, I could always put on the LWB boom on if it becomes a problem, but ya, my hope is to have more motor that I need so that I will have an nice quiet DD hub motor with a large stater able to soak a lot of heat when I need to climb a long steep hill when I am able to add little to no pedaling with a cargo trailer in tow. (Maximum about 150 lbs, IIRC around 650 lbs total bike, rider and cargo)

Depending on what is most optimal, I can adjust, but I am reading up and it appears that a higher winding count such as the 4T x 16 (still trying to get more info on this, not as many threads as I thought I would find on motor specs, maybe I just haven't found the right one(s)?) since I will be running as mid-drive, and have a lot of flexibility in my choice of gearing.

This motor will also be running to the N171b NuVinci rear hub which should help mellow out any peakiness, but yes, I have a CA, and I want a smooth throttle, so I think your suggestion sounds good, I'm looking to maximize torque and efficiency while climbing a 10% grade, 2 mile hill at around 20 MPH, I wouldn't care if I went any faster than that on the flat.

I'm looking to make as durable and bullet proof set-up as I can, and eventually use this with a bicycle courier business, however with all the hills around here, I will probably end up with either a 48v 40AH of LiFePo4 Prismatics, or if higher voltage makes more sense, something like 60 - 74V of Li-Ion (18650?) 3C + with 30 ish AH.

Weight is not really a concern as I will have plenty of space on the cargo bike, and let me post a better pic, I posted an older pic of my Vision R-40 so I could design the motor mount easier, here is a pic of the bike currently with paniers and tail box:



Anyhow, off to the shop! Thanks again for any and all advice/info/comments!

 

[spinningmagnets]

The V1 MXUS has the thicker laminations, and efficiency is slightly below 90%. The V2 has the aluminum stator support and the thinner laminations (makes less waste heat per watt applied, plus absorbs existing heat to spread it out), and efficiency slightly above 90%. If you like super hard acceleration and high top speeds, get the 3000W model.

Yes, the ferro fluid (FF) is a new development, and it works. For $25-ish it allows internal stator heat to transfer to the aluminum side-plates much more rapidly, and from there to shed to the outside air...

 

[LI-ghtcycle]

The V1 MXUS has the thicker laminations, and efficiency is slightly below 90%. The V2 has the aluminum stator support and the thinner laminations (makes less waste heat per watt applied, plus absorbs existing heat to spread it out), and efficiency slightly above 90%. If you like super hard acceleration and high top speeds, get the 3000W model.

Yes, the ferro fluid (FF) is a new development, and it works. For $25-ish it allows internal stator heat to transfer to the aluminum side-plates much more rapidly, and from there to shed to the outside air...

Thanks very much for that info, SM!

I will look into the FF, sounds like a great idea, and the V2 sounds like it will fit the bill quite nicely. 8)

 

[Warren]

LI-ghtcycle,

"I'm looking to maximize torque and efficiency while climbing a 10% grade, 2 mile hill at around 20 MPH"

I think you will have trouble running 20 mph, at 650 pounds, up a 10% hill. I am not up on the various MXUS motors, but playing around on the Grin Tech simulator, the best I could do was 17 mph with the 4503, running a pack like my 14s Leaf pack, and a 60 amp controller. At a 290 pound gross. I fly up 10% grades at 25 mph, and do 20 mph up 17% grades.

 

[spinningmagnets]

Since this is using a hub as a non-hub, you can spec a high-Kv/low-turn-count version, then gear it 2:1 or 3:1 to drive the rear wheel, so...the simulator doesn't take that into account. Check with Dave Kaufmann, or Sonny Yalung, they both have longtail cargo bikes with a hub as non-hub...huge performance difference when the motor isn't bogged down at the wheel speed.

 

[Warren]

Actually it does take that into account. You just put in the virtual wheel diameter. You can put in wheels smaller around than the motor, if you like. I tried all the MXUS motors listed, and tried lots of wheel sizes. Using a 54.4 volt, 60 amp pack and a 60 amp controller, like mine. The 4503 did the best for efficiency, and top speed up that hill. It was pulling over 2500 battery watts. The weight is the problem. To get 650 pounds up there at 20 mph will require over 3000 watts. He will need more volts, more amps, or both to do it.

 

[LI-ghtcycle]

Good point Warren, I haven't used this calculator much, but I think I might have made a typo somewhere, the gross weight of my cargo bike while pulling a 10lb trailer (100 lbs of cargo, 190 lb rider, 150 lb bike (it should weigh less, but I want to give myself some wiggle room for the addition of a potentially quite heavy 48v 40 AH LiFePo4 Prismatic battery, 20 lb hub motor, etc. ) totaling 450 lbs.

The closest I see for the simulator is 200 kg (440 lbs) so in the ball-park.

If I am reading this right, that means I would be able to hit about 25 Kph (15 mph) while traveling up a 10% grade with a MXUS 4504 (assuming this represents the V2 16x4T or close?) 48V 40 AH LiFePo4 Prismatic Battery, 60A controller (Lyen 12 Fet Extreme Modder IIRC? bought a couple of the programmable ones from Lyen a few years back, just going off memory) 26" wheel, Full Recumbent, and that puts me at about 1500W + 68% efficiency, and would be putting out 70nm (50 ft lbs, the NuVinci is known to take 95 with np)

Not ideal numbers, but considering I have a lot of gearing ratio choices with for my initial reduction from motor to to rear hub (9 KV for the MXUS 16x4T putting the RPM near 400 under load) I can use 1/2 single speed chain and get my equivalent "rear wheel size" quite low, maybe I will have to take a speed reduction for greater efficiency, that is my real concern, even if I were to have to slow to 10 mph and have things running efficient, that is a "win" as this represents a "worst case" senario, other than the one hill that I live on the top of, there aren't many around here that long with that steep of a grade (average, it gets even steeper for a few blocks, not fun in the freezing rain! )

So, If I can figure a way to see what the "simulated" gearing would be with both my initial reduction, and the range of the NuVinci, (total of 3.5 to 1), I'd be happy to get say the 1200W & 78 - 80% efficiency with a simulated 16" rear wheel, traveling at 13 mph while climbing that hill, even if that meant that unloaded, my gearing would be limited to 20 MPH on the flat.

Feel free to correct my calculations (I am blond after all :wink: )

 

[LI-ghtcycle]

Ok, call me silly for not doing it b4, but as much as 48v is what I have used in the past, WOW is 74v soooo much more efficient! (*** edit, forgot to mention 74v LifePo4 and 40A from controller)

Just switching to 74V, 25 AH I am gaining quite a lot more speed and efficiency! (same total wh's or there about)

Not to mention that I now exceed my goal of 20 MPH (well, in this very simulated example, not counting that cross wind, stopping in the middle of the hill for the cross-walk, my hair-brained calculations .. etc.

Anyhow, this gives me something to think about! Am I reading something wrong, or is it true to say that sometimes higher gearing makes the motor happier even with increased load?

This seems counter-intuitive to me, with the simulated 16 Wheel, I am getting more speed, but also pulling more watts and I actually gain some torque and use less watts with simulated 26" wheel? :?

 

[Rassy]

Li-ghtcycle, I've been meaning to share the following concerning my NuVinci N171 experiences. I think when you rode my TerraTrike Path several years ago it still had the Nexus 8, along with the mid-drive hub motor. After replacing the Nexus 8 with the NuVinci developer's kit I really loved the result, so much that I set up another trike (TerraTrike Rover) with the same drive train to use as a 'visitor' trike.

When I had a WE BD36 hub motor on the front wheel of the Vision and a 48V 20AH LiFePO4 on the rack I rode several metric centuries of 63 to 68 miles without running out of juice. With the trike and the NuVinci, using a 48V 15AH LiFePO4 I always ran out of juice several times between 30 and 40 miles. After setting up a couple of friends with pusher trailers behind their trikes, one with a little geared Bafang and the other with a WE BD36, both in 16" wheels, I was always seeing about twice the Watt Hours used as either one of them.

So one day, on a nice flat road, my buddy and I both set our Watt Meters to 0 and rode side by side for a mile at 10 MPH without touching our pedals. I don't remember the numbers, but I used almost exactly twice the Watt Hours.

Another issue with my trike is that I can not pedal it for more than a few hundred feet without using some assist. So early last year I decided to set up another trike with the goal of not only being more efficient, but also being lighter and being able to pedal with no assist to eliminate range anxiety. All goals were met using a 48V 500W Bafang crank drive and a conventional 9 speed derailuer. Even though I still love the NuVinci trikes, the new one is now my favorite. One of the NuVinci trikes has found a new home with my sister's boy friend.

So, in wrapping up here, I know you have used your NuVinci on the pedal side for some time so that you could match your cadence to your motor speed. But once you also put the motor power through the NuVinci your actual results may not be quite what you are expecting. It will be interesting to see how it all works out.

 

[LI-ghtcycle]

Interesting, ya I am sure there are some inherent losses with the NuVinci, probably amounts to the same resistance to pedaling as a small DD hub motor, however, I do "cheat" a bit. :wink:

I still have 3 rings & derailleur up front for pedaling, so yes, you're correct in that the NuVinci's cost for efficiency especially with a lower weight/powered system is probably not ideal, and I am sure that if I go with this next build using 72V pedaling along with the motor isn't really going to happen unless I am helping it out going up hill.

I think it should pay for it's self with the durability and I won't probably be able to pedal along as much, but the biggest danger is I might start liking the faster speed a bit too much and get pulled over by someone .. I can just see it now ... mug-shot down at the PD with a 40 MPH sized E-Bike grin!

(good thing I can limit top speed easily enough with the CA! :lol: )

 

[LI-ghtcycle]

Oh, and I am right there with you on the Bafang BBSo2!

I have been really impressed by how clean it looks, and if I just wanted to convert the average bike, that would probably by my first choice!

After seeing some of Bruno's Video's I might have to get one for my 700c Hybrid too! 8)