Nazca Pioneer solar recumbent build

[agniusm]

Was going through this thread couple of times and always refrained from commenting but i have to point out that using modules without the lids is not recommended and misses the design purpose. Plastic is pliable and cells make sides bow out. Lid locks the sides to mitigate this effect so cells could sit flush on the contacts therefore have largest contact area.
Second point, it is not recommended to transfer load through a nut as it is least conductive and better to bolt your terminal strait to the module or make terminal that could accept few wires from copper or aluminium and then bolt it strait to the module.

 

[k2orbust]

Third time's the charm for wheel building! Moving the drive-side spokes inside the motor flange creates the clearance needed for a 12 speed cassette and derailleur. Robbie at GRIN Tech asserts this should not be necessary for 11 speed cassettes, as the overall cassette width is less. I never tried putting a spacer behind the cassette because a) I have no such a spacer and b) I worry about clearance from the smallest sprocket to the dropout.






The battery is encased in 4 mm plastic plates and allegedly water sealed. I need to draft up some kind of sheet metal frame to hang it below the seat.


Some soldering and/or crimping is required before I can connect the battery to the [GRIN Baserunner] motor controller and the LED power supply I got for charging.

 

[thundercamel]

That cassette looks awesome! I regularly have clearance issues between the dropout and smallest sprocket on the DNP 11-28 7 speed freewheels I use. Usually have to use two of the flat washers, then the one tabbed washer to get it to clear. Spread the dropouts as needed.

 

[k2orbust]

I've got things together enough to start laying out the cockpit. (Front brake and throttle on the left, shifter and Cycle Analyst on the right.) Aero bars have some distinct advantages, but clear sightlines are not among them...


Regrettably the front brake hose is not long enough, so I'll have to get a longer one and do a full bleed. The rear brake hose and derailleur housing have length to spare and will get chopped.

 

[marka-ee]

How do those bars work out? How do you mount and dismount from bike? I have the type that folds up and away and there is a hand adjustable screw to set the bar height while riding. I think they are the standard pioneer type. I also lift then on slow speed maneuvers now and then, it helps with me balancing.

 

[k2orbust]

by marka-ee » Mar 04 2021 8:57am
How do those bars work out?

A year ago I test rode some different recumbents (Bacchetta and HP Velotechnik) at a local dealer to try different handlebar types, and this was my preference. They are rigid with no pivoting action, allegedly nice to hang onto for hill climbing. Also allegedly nice for pushing up hills after you give up I mount the bike...carefully? With no chain or brakes I haven't done more than trundle Flintstones-style in front of my house yet.

 

[k2orbust]

This Swede has made some great videos about his solar trike build for the Sun Trip race: https://www.youtube.com/watch?v=tN6ykK9Q-CQ

It's a pity there's no third option for MPPT charge controllers in between the generic blue Chinese ones and the Genasun GVB-8. The panels and battery I chose are an awkward match; with the two panels in series their peak voltage is higher than the battery, but in parallel they put out too much current for a single Genasun. If only someone made a controller that could buck and boost...at the rate my build is going one might pop up by the time I actually need it. But I'll probably end up with two of the clunky, hot-running, slow reacting blue ones.

If I was looking for a full on science project I'd see if this could be hacked to increase the output voltage...but that seems like a very lot of work. And I'd still need two to convert 300 W! https://www.digikey.com/en/products...KIT/13404940?s=N4IgTCBcDaIDIBUAcAWAnARhAXQL5A

P.S. From the comments on that video I learn there is a 600 W version of the Chinese one! Then I would have just one hot blue lump to hide instead of two. https://www.ebay.com/itm/MPPT-Boost...-Charger-Controller-48V-72V-Blue/184562363524

 

[marka-ee]

Two 85 watt panels and he gets about 70 watts from both, that's about what to expect. I wonder if you can swap in a better fet in the blue one ? No reason for it to get hot at those power levels.

 

[k2orbust]

It's got a chain, it's got an operational brake lever, it's got a mirror...time for a test ride :-D


Preliminary findings:

• Stable low speed handling
• Stable high speed handling
• Heel strike a definite risk in sharp low speed turns
• Easy to put feet flat on the ground even with suspension fork (but I'm 6' 1")
• Front suspension feels good @ 80 psi
• Rear coil shock feels too stiff, but my buddy gave me an old air shock to try when I get around to it
• Seat pan is slightly too short, which is a pity because it's not adjustable
• Mirror essential to see behind you
• Lowest gear (~27 gear inches) is inadequate for sustained climbing on anything but a gentle grade without the motor. In low juice situations I'll be backing off assistance on the flats to save battery for any hills.
• Pronounced cross-chaining from the drive-side idler to the big cogs on the 12 speed cassette. Lots of idler noise in gears 1 & 2.
• Wide torque sensing bottom bracket (ERider 68 mm) + "Zero Q" pedals = shoes tend to brush against the pedals

I made a CAD model of the triangle battery so I can lay out the mount around it. Thinking to make it from sheet metal, but as that's outside my skill set I'd have to send it out somewhere for fabrication. There are clearance and access issues to navigate around the front tire, the chain, the brake and derailleur lines, the charge port, etc. etc.

 

[marka-ee]

Good to see it on the road! The heel strike takes getting used to, after a while you learn to time the pedal position according to what direction you are turning. The motor makes everything so much easier. On the rear shock I am realizing that my 9c heavy hub motor is not ideal. The suspension sometimes chatters over things like roots raising the asphalt in the road. I don't think there is any damping happening in there really. I like the spring vs air though. On my regular mountain bike I recently changed the fox air shock for a cheapo spring unit I found at a flea market. Much better ride! The reason is much less seal friction letting it move for much smaller bumps than an air shock. My issue with the large unsprung hub motor remains. As you noted with the chain you definitely don't want a traditional mid drive on these bikes. Imagine all that chain whipping around all the time. No thanks. I think the best would be a motor mounted under the seat with a gates belt on a large sprocket on the rear wheel. This would still be quiet while running and the unsprung weight much less allowing the suspension to take more of the bumps. I'm planning on making my own front fork to smooth the front out. The design is based on the 'speedway' style leading axel linkage forks. I've started CADing it up to understand the forces better especially under braking conditions. Should be much more plush than any telescopic fork. I believe if I set the geometry right there will be no front dive during braking. The CAD simulation really helps visualizing the stress and forces.

 

[k2orbust]

I think the best would be a motor mounted under the seat with a gates belt on a large sprocket on the rear wheel.

That's what Warren was talking about on the first page of this thread too. The secret sauce is those sprockets that bolt onto a disk brake rear hub in place of the disk. It's definitely the way to get the most out of a motor, but I went the simple route...

 

[Warren]

I think the best would be a motor mounted under the seat with a gates belt on a large sprocket on the rear wheel.

That's what Warren was talking about on the first page of this thread too. The secret sauce is those sprockets that bolt onto a disk brake rear hub in place of the disk. It's definitely the way to get the most out of a motor, but I went the simple route...

Speaking of Gates belts...

https://electrek.co/2021/03/09/gates-launches-new-belt-drive-system-that-could-lead-to-advances-in-electric-motorcycles-scooters/

 

[k2orbust]

The battery is made of 5 aluminium panels soldered together. I did not want to solder a hinge so I decided to use a sliding technique for opening/closing the box with the 6th panel. Therefore I had two screw 2 L shape sticks along the side of the front battery box to act as a rail/guide for the front panel.

Peterfr12, what thickness of aluminum did you use to make your battery box?

 

[k2orbust]

After that first ride I redrilled the seat pan to push it as far to the rear as possible. This confers several benefits:

• Increases seat back angle, counteracting the lift of the suspension fork
• Lowers seat at the front for easier mount/dismount
• Lets me shorten the pedal boom, giving derailleur more chain slack (because I happened the make the chain too short)
• Moves steerer tube forward relative to knees, letting me move the handlebars down and forward (straighter elbows feels better)
• Higher seat angle and lower handlebars improves forward view

Looking at downsides, it does shift more weight to the back wheel and increase heel strike risk a bit.


For now I have the battery lashed to the rear rack and...it works! Took it for a test ride around the neighborhood and played with the throttle and torque assist settings. It got past 31 mph without any pedaling (@ 55 battery volts). Sometime I'll try the Baserunner's field weakening tricks and see if it gives a turbo boost.

Ergon GC1 grips feel nice. One of these days I'll even get a front brake

 

[k2orbust]

First pass at an underseat battery mount. This is simple enough I could probably fabricate it myself...still some work to do around fasteners, metal thickness, waterproofing, etc.


This orientation is winning because the battery is completely supported by the big, permanently installed piece, and you just take off the little front strap to slide out the battery. Also easier to reach the power button and recharge port.

 

[Peterfr12]

K2orbust,

You made great progress. Congratulation.

I used a thickness of 3 mm for my aluminum sheet (made in UK). The box weights nearly 5 kg on its own. It is extremely robust and I really struggled to drill holes in it. I destroyed several 'cheap Chinese' drill bits.

I cannot believe that Nazca stopped their frame production. Yes their bike are heavy but very robust and with a hub motor it is a 'flying carpet'.

 

[k2orbust]

Playing with the Baserunner field weakening settings is fun! The motor's normal no-load speed is 32.5 mph (which draws 80 W power). Setting field weakening current to 12 A bumps it to 45 mph (and 150 W). I took that out on the road and hit 38 mph throttle only and 40 mph pedaling hard. 36 mph is a nice cadence in top gear so I am content. (At 15 A field weakening the no-load speed jumps to 50 mph...I don't want to fall off going that fast.) The loooooong braking distances at these speeds will take getting used to.

The bike has a front brake now :thumb: Slowly I am taming the tangle of wires, cables, and hoses.

 

[k2orbust]

The Pioneer got its first proper ride today and performed well 43 miles and 3700 feet of climbing:

I don't know the full battery capacity yet, but it did not run out which is the important thing! 57.7 V at the start and 48.6 V at the end. The motor expended 717 W hr but regenerated 91 W hr (14.5% regen ratio) (net 14 W hr/mile), while I contributed 300 W hr. It handled great on packed dirt, though on steep descents the GMAC regen is so strong it can lock up the back wheel.

The left crank keeps coming loose; pedaling it wobbly may have damaged the square taper the very first time I rode it



The road goes ever on, but range anxiety meant I turned around here:

 

[thundercamel]

Great work, and nice scenery! 15.2 Wh/mile is the best I've achieved so far, with no regen.

So the battery you have was the 14s6p EM3EV one you linked earlier? I'm guessing it's the 20.4AH option with Sanyo GA cells and a smart BMS? It's very similar to my 14s6p self made triangle battery I've been using with LG MJ1 cells and a JBD smart BMS.

 

[k2orbust]

I'm guessing it's the 20.4AH option with Sanyo GA cells and a smart BMS?

Yes it's the EM3EV 14s6p, but I cheeped out and got Panasonic PF cells for 17.1 AH nominal. They all have the same smart BMS I think, which matters because when I implement solar it will be recharging thru the discharge wires. https://em3ev.com/shop/50v-14s6p-small-triangle-pack/

EM3EV says the max continuous charging current for this battery config is 12A, well less than the what GMAC motor can output in regeneration. (They also recommended capping regen volts at 57V.) For now I have the regen amps limit set to 15A with the expectation that it will seldom be maxed out for long.

 

[k2orbust]

This afternoon was too beautiful to stay indoors so I gave the GMAC motor a stress test on one of the tougher local climbs. The Chairlift (as I have dubbed the bike) made it to the top with a peak temp reading of 110° C :thumb: My calcs suggest the GMAC is much more efficient that the GRIN All Axle motor on these steep-a$$ hills, but ironically more prone to heat buildup.

After the top tho a funny thing happened where the motor cooled below 100° but all power ceased and I had to pedal with my own dumb legs...until I power cycled the battery and the system magically revived. Has anyone experienced something like that with a Cycle Analyst [v3]? I have it set to start thermal rollback at 90° and cut off completely at 130°. The motor controller has its own thermostat but it was merely warm to the touch at that point.

621 net W hr from the battery, 191 W hr from me, 22.6 net W hr/mile, 7.4% regen. Stats R fun

 

[thundercamel]

EM3EV says the max continuous charging current for this battery config is 12A, well less than the what GMAC motor can output in regeneration. (They also recommended capping regen volts at 57V.) For now I have the regen amps limit set to 15A with the expectation that it will seldom be maxed out for long.

I think that will be just fine. The PF datasheet only mentions a standard charging current of 1.375 amps, while other cells like the MJ1 state a standard charge rate of 1.7 amps and a max charge rate of 3.4 amps. I wouldn't view the regen current as continuous.

After the top tho a funny thing happened where the motor cooled below 100° but all power ceased and I had to pedal with my own dumb legs...until I power cycled the battery and the system magically revived.

Did the cycle analyst lose power as well when the motor cut out? Your wording sounds like that wasn't the case.
When my BMS shuts off for voltage or temperature reasons, my SW900 display and standalone power meter stay on with the minimal voltage being somehow generated by my pedaling. The direct drive motor is still spinning, and the motor controller must somehow convert the brushless AC to a little DC going out the battery feed wires. I can watch the voltage change based on my speed, and it's around 5 volts.

 

[k2orbust]

Did the cycle analyst lose power as well when the motor cut out? Your wording sounds like that wasn't the case.

Not only did the Cycle Analyst remain on, it still saw full battery voltage. Only motor current dropped to 0.0 A and wouldn't budge for anything (throttle, pedal assist, or regen). It was weird because the motor had cooled off from its peak temp when it happened. The Cycle Analyst can't see the Baserunner temperature, and the Baserunner isn't bolted to a heatsink right now, but this doesn't match the thermal rollback behavior described in the Phaserunner manual. (My understanding is that Baserunner and Phaserunner run identical firmware.)

 

[k2orbust]

The GMAC motor is fierce! Riding this way kills the range of course (29.2 W hr/mi avg), but it does shave a few minutes off the commute...

 

[k2orbust]

It's a pity there's no third option for MPPT charge controllers in between the generic blue Chinese ones and the Genasun GVB-8.

A third option emerges! This popped up in Grin Tech's store: https://ebikes.ca/shop/electric-bicycle-parts/solar/adjustable%20400w%20boost%20mppt%20with%20led%20display_cg-mppt_el400.html Given the constraints of my panels it appears one of these could replace two Genasun modules. The tracking speed/accuracy is probably not as good as Genasun, but hard to believe it could be worse than the blue ones...