Thoughts from a new convert

Bbbbrass

1 W
Joined
Nov 25, 2022
Messages
62
Location
Charleston, SC
I started down this path a few months ago because I had been harvesting lithium packs from recycling as I was building an 18650 power wall. I didn’t like not using these high current cells for such a plebeian task, and I learned how to fix many of the Ryobi 40V packs. I gave some away, sold some, and wondered what to do since I have 0 Ryobi tools. I decided making my old cheap Diamondback Outlook into a budget commuting machine would be ideal. This Amazon kit (JAUOPAY 36V 750W)https://www.amazon.com/dp/B097YWYK48?ref=ppx_pop_mob_ap_share, a random Pedago rear rack, a couple Terrafirma adaptors, a little Lowes hardware and I was rolling! 1 ride and I was hooked, and suddenly my build that I had under $250 into I was willing to invest in a bit more.

Those adaptors proved to be poor choices for a high-vibration setting and I kept having to adjust how they were mounted, getting bulkier and less rapid for swapping. So I went and built my own pack out of 20700s that I had laying around from Dyson packs. 10S3P, with a (whoops!) Daly BMS and now my rack got the ubiquitous milk crate addition, with the battery under the top tube of my frame. I pulled a headlight off a discarded HiBoy S2 scooter I found next to a dumpster. The controller has a direct pack voltage line from the pack that is switched. Awesome! Wait, that light totally sucked. It made me visible, but was not useful for night riding over 10mph.

I found that I could make my commute (16 miles each way, able to charge) on 2 of the Ryobi 40V 6Ah packs. My DIY pack was 7.5Ah, and good for about 11 miles. I set about designing a pack of recovered 25R cells that was 10S5P, and should be about 12.5Ah. I found a frame bag that would hold this nicely. Then Methods had to go and show how the Daly BMS weren’t good, and down the rabbit hole of opening different BMS I went. Then I decided to crack open the controller and see if the 36/48V sticker was for real, or a hard limit of 48V. Much to my pleasant surprise it was full of 60V and higher rated components! On went a “56V” (14S) Ego pack and the rush for more speed was on.

Being able to hit 30mph on the fresh pack was awesome. With the chainring I have, I can’t contribute above 25mph, and more realistically 20-22mph. My commute is often on busy roads for a short distance between the paths and neighborhoods I can route into. This speed makes for a much safer and more relaxing ride (but drivers here have no idea about bikes; we are rare). I am now rebuilding my 25R pack into 14S4P (9.5Ah), debating making some kind of custom mount to hold more (I have hundreds of 20R cells), and finding a box to fit my controller and wiring. I also put more into a CA and SPL-01 from Grin than I spent on my whole starter kit. Totally worth it to have a real headlight for the dark commuting, and to be able to track my Wh/Mi and speed.

If anyone has insight on figuring out the generic controller wiring to fit into the CA shunt that would be super helpful. I get about 25Wh/Mi on the 52V pack, and 20Wh/Mi on 36V. This is because I can contribute by pedaling all the time on 36V, and acceleration is much more gradual. The controller never pulls more than 20A, so the peak is 1000W on 52V and only 750W on 36V. Debating keeping some 36V packs for distance rides, but it’s hard to not have that speed. I think installing the pedal cadence sensor and getting the CA hooked into the throttle will help with that.

I used to work where I could ride a bike 2 miles to the bus stop, take the bus downtown, and ride a last mile to work. All bike path or college area streets where bikes were normal. Now I have 16 miles on heavy trafficked roads. During rush hour, it only takes me 10 more minutes to bike than to drive. I arrive in a much better mood, am helping the earth, and getting that wonderful smug feeling of building something myself that actually works in a useful way. The issues are when I need to suddenly leave to get a kid, or when the weather turns sour. I will not be able to bike commute full-time for now, but the 2-4 days a week I can are amazing! (Also, I really need to upgrade my brakes, and maybe my saddle, and front fork, and install my fenders....) The only downsides are how heavy the bike is, keeping an extra bike around, and wearing out the crotch in my pants.

Thanks for reading!
 
Bbbbrass said:
If anyone has insight on figuring out the generic controller wiring to fit into the CA shunt that would be super helpful.

Just to confirm, in your current setup, you have the CA standalone shunt wired up, but aren't using the two wires for speed and throttle that are under the shrink wrap?

If this is the case, then it's just a matter of wiring the green throttle wire from the CA to the signal wire of the controller's throttle input (the non-black/red wire), and the yellow speed wire from the CA to any one of the three hall sensor wires (splice into on of the hall wires from the hall sensor connector that goes to the motor).
 
Correct! Issue is, 3 wires on my CA, and the controller has 6 wires (battery level, on/off, throttle). I opened it and thought I had it figured, but no dice yet! I got the CA to register throttle input, but couldn’t get the controller happy with pass through yet.
 
ZeroEm said:
CA shunt mod for Generic Controller
In the manual shows how to wire CA up to the controller.

Makes me happy read your experience. If you have to many cells you can sell them here Must put a price with it.

Thanks! I had been looking and somehow missed this. I’ll try again. Part of issue is throttle is connected with 2 other components.

I’ll sell local, but hesitate on shipping cells…
 
Bbbbrass said:
Correct! Issue is, 3 wires on my CA, and the controller has 6 wires (battery level, on/off, throttle). I opened it and thought I had it figured, but no dice yet! I got the CA to register throttle input, but couldn’t get the controller happy with pass through yet.

Sounds like you were able to get the existing throttle to connect to the CA and working, but need to still connect the CA to the controller throttle input (as I described in my earlier post). Is that the current status?
 
E-HP said:
Bbbbrass said:
Correct! Issue is, 3 wires on my CA, and the controller has 6 wires (battery level, on/off, throttle). I opened it and thought I had it figured, but no dice yet! I got the CA to register throttle input, but couldn’t get the controller happy with pass through yet.

Sounds like you were able to get the existing throttle to connect to the CA and working, but need to still connect the CA to the controller throttle input (as I described in my earlier post). Is that the current status?

Correct. I have hooked throttle back up directly to controller for now to ride, but will look at following the diagram above and wiring through the CA this weekend.

Question: on the diagram from the manual it lists the throttle wire with a resistor added exiting the controller going to a connector. What is that for?
 
CaV3_UUG_3-0e.pdf, Official User Guide page 67
Current monitoring and throttle override are provided by a stock external (SHUNT-CA3) from Grin Tech and a 1K
resistor. (Note that although the resistor power requirement is tiny, choosing a higher wattage part (1/4W – 1/2W)
will give more robust leads for in-line connection – a purely mechanical consideration.)

Have not performed the mod myself.
 
Bbbbrass said:
Question: on the diagram from the manual it lists the throttle wire with a resistor added exiting the controller going to a connector. What is that for?
You won't need it.
"One 1,000 Ohm resistor (only required for version 2.3 Large Screen Cycle Analysts)"
https://endless-sphere.com/forums/viewtopic.php?t=62809#p939063
 
ZeroEm said:
My Bad, don't want to spread disinformation.

It's not well documented. I tried to find everything I could on adding a connector before making one, and that's the only place I saw that information. It isn't even in the unofficial guide, but apparently came directly from the source.

"Thanks to Teklektik for providing corrections and the accurate diagram for this step."

RIP Teklektik
 
Sad info on the people, have always enjoyed the thorough posts.

Appreciate the assistance again!
 
Time for an update, pics later once I've tidied up.
For anyone using the JAUOPAY 36V 750W kit from Amazon and trying to do your own wiring, this will be helpful info.
The brown and yellow wires are the power switch. You can short them together, or attach your own switch.
The white wire is what you want to attach to the CA shunt Green wire.
The black wire is ground, the red is 5V for the display screen, and the Green = max attached battery power. Careful to not short this one! You can cap all the display wires, or use it for a power indicator (will experiment with charging 5V).
IMG_6939.jpeg
I changed my bike from my old 2006 Diamondback Outlook to an older 2005 Specialized Hardrock Sport. The frame is 21 inch (vs 19) so that was the main reason. The 11T cog is nice though, as is the crank that isn't bent. It's originally from Michigan and quite rusty, but I'll keep fixing it a bit at a time.

I finally got my KWeld working, after getting a new battery for my car so I could use my Optima Yellow Top again. I paralleled it with another cranking battery, and kept them connected to a NoCo Genius 5. Changed out my KWeld battery cables to 4AWG, as short as I could make them.

My series connections copper was 0.1mm thickness sheets that I snipped roughly, about 6mm Wide. I used 0.1mm thickness 5mm strips for the parallel. Recycled old tool cell battery 25Rs, 14S5P. Use a 30A 14S BMS from Xiaoxiang (via Battery Hookup). Instead of stressing over getting good solder connections to the cells, I spot welded on some wires with tabs cut off old tool packs. Also put a 25A fuse inline (controller is 22A max). We will discover true capacity as we go, but it should be about 600Wh. I plan on charging to only 56V, and stopping at 49V. Made it the 16miles to work today using 370Wh, averaging 18mph. I have another battery I'm reworking to add another 4-500Wh.

IMG_6969.jpegIMG_6970.jpegIMG_6972.jpeg
 
Well, that didn't last long. After 25 miles, we discover the corrosion on this bike is deeper than it seems, and apparently it may have been too tight of a fit in the dropouts.
 

Attachments

  • IMG_6973.jpeg
    IMG_6973.jpeg
    106.6 KB · Views: 48
Well, that didn't last long. After 25 miles, we discover the corrosion on this bike is deeper than it seems, and apparently it may have been too tight of a fit in the dropouts.
It looks like maybe you skipped the torque arms, which is kind of scary on a front hub bike.
 
Are you sure it has broken due to corrosion and not due to stress on the drop outs. Did you fit a pair of torque arms?
 
Here is a article worth reading about why torque arms are important for hub motors:


Edited: I have just looked at your specs and you are probably peaking 1500 watts on your motor. With torque approaching 130Nm. That’s a lot for a front hub motor fitted to aluminium forks without torque arms. It best to fit one on each side to distribute the load evenly.
 
Last edited:
Well, that didn't last long. After 25 miles, we discover the corrosion on this bike is deeper than it seems, and apparently it may have been too tight of a fit in the dropouts.
img_6973-jpeg.330917


Dude! That's a horrible crash scenario! Are you OK? :oop:
 
img_6973-jpeg.330917


Dude! That's a horrible crash scenario! Are you OK? :oop:
Totally fine. Came off at a stop sign. Lucky.

I will add torque arms, but I’ve realized the dropouts were too tight for the slotted 14mm axle.
 
Last edited:
Totally fine. Came off at a stop sign. Lucky.

I will add torque arms, but I’ve realized the dropouts were too tight for the slotted 14mm axel.
You usually need to deepen the dropout 2mm using a round file (or flat file, then round file) to get the axle to the same position as a 10mm axle would be in. It's more important if you have disc brakes, so the caliper and disc align, but also important so the wheel is better seated when you bolt it down.
 
As a general rule, hub motors and suspension forks don't go together at all. You found out the main reason why, but there are others. The pull of the motor tends to lock out the fork when it's driving. This in turn tends to slop out the fork bushings ahead of schedule. Then there's a radical increase in unsprung weight that diminishes the effectiveness of front suspension.

If you must have a suspension fork, use a rear hub motor.
 
Back
Top