TerraTrike Rover, NuVinci N171B CVP

Just decided to build up another trike, don't ask why. It will be similar to my current TerraTrike Path, but lower geared with a 23 tooth freewheel on the NuVinci instead of an 18 tooth cog. I'll use it locally on our steep hills and when pulling my trailer. Will also have two trikes for use when friends and relatives visit.

This will be a low speed low power trike, and this build will probably only be of interest to someone thinking about adding a mid-drive to a Rover trike.

Below is the 42 pound trike right out of the box. It comes as a single speed for $699 directly from TerraTrike. Utah Trikes doesn't even handle the single speed, but the crank will be replaced with a 48T freewheel and the rear hub will be replaced with a NuVinci N171B auto shift (Developers kit).

Edit: This trike was changed back to a pedal trike with a 7 speed derailleur in 2017 and sold. The 9C motor and Nuvinci auto shift transmission have also been sold.

I really like tadpoles like that. Low enough to slide instead of flip-over, tall enough to be seen by cars. Tons of room under the frame for a large rectangular battery pack (Headways?), room behind the seat for a mid-drive if wanted...

The Rover has about the same dimensions as the Path, once I raised the seat to give enough room to shoehorn the 9C clone motor between the rear tire and the seat. So I briefly considered getting a smaller diameter motor. But I already have another 9C with the slow speed 6 X 10 winding that I wanted to use. So I decided to move the rear wheel back 3 inches. Turned out to be pretty easy to do the way the Rover was constructed. The rear section of the trike is made with 2" square tubing that recieves the slightly smaller square tubing that the boom is made from. There are two set screws that hold these tubes in place and at its longest extension there is a 13" overlap of the two tubes. So I drilled a 3/8" hole for a bolt to tie the two pieces together with just a 10" overlap since the rear set screw, that shows in the picture below, would no longer be useful.

I inserted a short piece of 1/2" round steel tubing inside the inner square tubing to prevent crushing when the bolt is tightened. The tube had to be tapped into place, plus I added some epoxy glue so it wouldn't move around during assembly or when the bolt is removed. Also glued a SS washer to the inner square tubing to take up almost all of the slack between the two tubes and keep the allignment the same as the original grub screw did.

As a bonus the bolt that holds the square tubes together will actually be a piece of SS fine threaded 3/8" rod which will also provide the attachment points for the front motor supports.

1" of the gained space will be used to provide additional seat adjustment since I am just a little tight on the Path. The other 2" will be used to give more room for mounting the motor.

There will still be about 8" fore and aft adjustment to the seat (originally about 11") plus about the same adjustment to the entire front wheel/handlebar assembly. This is all without any chain adjustments, but might require a set of longer seat back supports if the seat is moved way forward for a very short person.

I always thought hub motor mid-drive setups looked more promising than multi-reduction setups. Good luck on your build, I am having a hard time making mine efficient. I'm getting a lot of loss from the tensioner.

Should be a great trike once it's done.

parajared said:

I'm getting a lot of loss from the tensioner

Click to expand...

When I set up my TerraTrike Path as a hub motor mid-drive a couple of years ago I used the single idler tensioner that was original equipment on the Path. For over a year I worked to chase down an occassional chain skip under hard climbing loads. First I thought it was a freewheel, then the Nexus 8 speed I was using, etc. Eventually I put the long return chain in a chain tube to eliminate the long chain slapping around plus a tight clearance issue where the chain would sometimes hit a bolt. At the same time I removed the tensioner. I had already made one front return idler adjustable, so I had some slack take-up available. Anyway, no more chain issues. The tube puts a little tiny bit of pressure on the chain and I have the front idler adjusted so the chain is just tight enough.

One reason the original tensioner caused a problem, in addition to being sort of cheap, was that it applied tension be pushing the chain away from the rear sprocket, instead of increasing the wrap the way a regular two idler deraileur tensioner does.

Yeah, my current setup is knee deep in chain. The chain is so loose a tensioner is absolutely needed to keep the chain on. Since my setup is not adjustable, I bought a half link, and hope to shorten the chain to that perfect length where the tensioner is just a guide rather than something that holds that floppy chain on.

I think your hub build sounds good because single reduction should eliminate the losses you have from tensioners and such in multiple reductions.

Not working on this project very fast. Thought I should post a picture before lots of detail is hidden by the ammo boxes that are coming next week to hold the battery and electronics. The ammo boxes will be mounted just outboard of the motor, one on each side of the trike. No rear rack planned.

I replaced the original 1/8" single speed chain with a 3/32" chain. However, the new chain was purchased at the local hardware store (no local bike shops here) and even without the motor the chain stretched everytime I tested it. After just a few short tests on almost level ground the chain had stretched a full inch. Removed a link, and the chain kept stretching. So I did a little research and sent for some 3/32" single speed chain. The difference is that the outer plates on the single speed chain are thicker and flat. The 3/32" multi-speed chains outer plates had a little curvature to them. Anyway, the new chain does not stretch.

Edit: A few posts ahead AW questioned that the chain would stretch that much. I thought through the process again and AW is correct. See my post below for full explanation.

I really like the look of that.. how does she work out ? Looks like you should have a lot of power there especially for climbing.

Video Please when you have her going.

Are you sure the chain stretched that much? It almost sounds more likely that a mount shifted or bent?

AW wrote:

Are you sure the chain stretched that much? It almost sounds more likely that a mount shifted or bent?

Click to expand...

Good point AW. You made me rethink where anything could have shifted a little and you are right, the rear motor mounts are tied to the chain stays with U-bolts and if they spread a bit would have affected the chain length. While waiting for the new chain I added a piece of 1/8" angle aluminum between the front motor mount and rear motor mount on each side to create the base for the ammo boxes. I was surprised that there was a little over 1/8" difference in the length, and loosened everything and forced them even. This is working against the "carefully" filed torque fitting on the motor axle flats for each of the four 1/4" steel motor supports. With the added angle aluminum there is now a very sturdy triangle on each side anchored at the front to the 3/8" SS rod that goes through the frame. The only job the rear U-bolts have now is to hold everything down against the frame, which is where it wants to be anyway, along with a little side to side anchoring.

Thanks AW, I'll go back and update my previous post. I'm only out the $30 plus shipping for the new chain and if or when I need another chain I'll retry the other one.

Rassy said:

So I did a little research and sent for some 3/32" single speed chain. The difference is that the outer plates on the single speed chain are thicker and flat. The 3/32" multi-speed chains outer plates had a little curvature to them. Anyway, the new chain does not stretch...

Click to expand...

Yes I saw the edit about chain stretch. But, chain does stretch some... and it's good to know options regarding chin construction, etc. I appreciate you posting the above, I noticed I bought a few sets of 3/32" eight speed. When that stock runs out I will definitely look into 3/32 single speed, and try to log/objectively measure the wear rate of the two.

The pictures below show how the rover turned out. Only had time to ride about 10 miles so far, but I like it.

First picture shows the motor and chainline before the ammo boxes were added. There is no spring loaded tensioner. The idler above the motor is on a slotted arm to take up chain slack. With the NuVinci plus the rear disk brake I added and the trailer hitch I have the axle clear forward and do not plan to adjust in at all

Second picture shows the inside of the right hand ammo box where the wire/connections/controller/Nuvinci controller/DC to DC all reside. The battery is in the left side ammo box.

Third picture shows the ammo boxes from the rear.

Fourth picture shows the rebadge from THE ROVER to THE ROVER-E

Fifth picture shows how sharp the Rover turns. I was showing it off today to a couple of friends and raised the left front tire about a foot into the air turning too sharp or too fast.

Last picture is the disk brake and bracket I added to the rear wheel, plus the trailer hitch.

Only thing left to do is add fenders to the front wheels. The grips are right over the tires, so even on a wet road your hands, etc. would get wet.

Plus, if I give the 29er with the GNG kit to my sister next week-end I'll need another battery since I'm sharing the 48V 20AH LiNiCoMn which will go with the 29er. I think a 48V 15AH Ping will do the job.

eTrike, thanks for the information on the Rover. I inspected all the welds and can not see any issues. Could you post the pictures that show where the problem is located? Outside of your post I haven't found any other information on this issue.

I was surprised to see that Utah Trikes has dropped the TerraTrike line. I am aware that at under $1000 each my Path and Rover are low end trikes, but I have been happy with both of them.

Thanks for posting the pictures. My trike was purchased early this year and everything in that area still looks solid. Apparently they changed how the front axle assembly is anchored to the main frame square tube. Mine has a grub screw at the bottom left of each end which forces the main square tube to seat against the upper right of the short axle square tube, giving infinite adjustment positions. Except the grub screws mess up the finish. I see I didn't capture the grub screws in this picture.

There have been a number of questions concerning the NuVinci N171B efficiency, or lack there of. First a little background concerning two trikes two of my friends have.

One is a Terra Trike Whizzer tadpole (no longer made) which has an aluminum frame and 20 some gears with a standard front and rear derailleur. He also runs higher pressure tires than I do. I helped him put together a power BOB trailer using a "slow" version Bafang geared front hub motor laced into a 16" wheel. He uses a 48V 10AH Ping LiFePO4 and seldom sees anything much over 10 amps on the Watt meter.

The other is a delta style hand cycle (propelled with arm power), also with 20 some gears with a standard set of derailleurs going to the front wheel. This trike has skinny 26" (or larger) tires and he runs them over 100 PSI). I helped him put together a short one wheel power trailer using an old BD36 (brushed and supposed to be inefficient) laced into a 16" wheel. He uses a pair of 48V 10AH Ping LiFePO4 batteries connected in parallel.

On every ride with either of these guys they only use 1/2 to 1/3 the Watt Hours that I use. I've always assumed that I just didn't work as hard and let my motor help me keep the pace more than they did, even when I actually tried to pedal harder.

So a couple of weeks ago I was on a ride with the guy on the hand cycle and right at the start with our Watt meters set to 0 we rode a mile on a flat road, side be side, at 10MPH without touching the pedals. I.e., powered just by the motors. And the results! He used less than half the Watt hours I used. Less that 9 Watt hours for him and close to 20 Watt hours for me.

Pretty unscientific, but does indicate the NuVinci is sucking up a lot of my energy. The rear wheel that came on my Rover just has a standard hub with an 18 tooth freewheel. One of these days when I feel ambitious I'll program the NuVinci as a single speed to match the same gear ratio and then run the same course with both setups to see how much extra juice the NuVinci really sucks up. In the meantime I really love the NuVinci, so I'll just continue hauling extra battery power!

Yeah, I am giving up on my Nuvinci set up. I think it might make sense for higher power set-ups, but there is just too much loss for my liking. And the weight is pretty crazy too. If you have a trike and a garage you probably don't notice it much, but I have to pick the back of my bike up quite often to put it in the apartment and such. I really notice the weight. These days I have barely been riding mine in favor of my girlfriend's Bike E with a MAC motor. So much lighter and zippier than mine with the Nuvinci and mid-drive, not to mention that I've noticed similar efficiency numbers like you describe here between the two setups. I picked up a 6T MAC kit from the classifieds here a couple of weeks ago and am going to start selling the Nuvinci and other mid-drive components from my Bike E and install the MAC kit. It was an educational project for sure, but for my needs it's just overkill.