eMTB - Specialized Enduro with CYC Stealth (aka The e-Duro!)

[handlworks]

Hey all, figured I'd post this most recent build up for interest. Not quite a build thread as I've finished it for the most part, but it's interesting to me and maybe other folks are interested. Here she is on her first ride this past weekend:


Goals of this build:

• A real, real, electric mountain bike: I'm a pretty avid mountain biker here in SoCal, and I want something to shred downhill with me and keep up with my friends who bought eMTBs.

• Big bike: I need decent suspension travel to handle the big hits, and a nice modern geometry for preference's sake.

• Stealth battery: Partly for aesthetics, partly to avoid getting hassled - I want to avoid the battery hanging out in my frame like a big pimple. Needs to be as integrated as possible!

• Power: I'm still looking to get a workout, but alllllsoooooo it'd be cool to have some power on tap when I want it...

• Street cred: It's gotta talk the talk and walk the walk. It's gotta be a respectable build, to e-nerds AND mtb bros. I want to snap necks with any aspect of the build, whether it's the finish or the components or just the cool base bike.

• Limited fab: While I can do a good amount of fabrication if needed, I just finished another fab-heavy project and I want to keep it minimal for simplicity (and the challenge).

What's next: Over the next few posts, I'll zero in on some points of interest on this build with details and pictures. The purpose is to get folks to think about their own builds, feed the stoke, and gather feedback (love, hate, knowledge, ridicule). I want to give each subsequent post enough time and effort (similar to how much has been going on in my head), so I'll take my time with these and post over the next several days.

Finally, more pics: in various states of completion, just things I think are cool to share.

 

[handlworks]

THE FRAME

2020 Specialized Enduro
Size: S4 (Large-ish)
Travel: 170mm front/rear
Material: Carbon fiber
Swag: off the charts

This frame was perfect for many reasons:

SWAT storage: Above all else, the SWAT storage was the reason I chose Specialized in the first place. They were the pioneers of the downtube storage compartment (Storage, Water, Air, Tools = SWAT), and I think they do it best. I said I wanted a stealth build, which means hiding the battery. I think I can hide my battery inside this downtube the best.

Big suspension: I normally ride a bike with 150/160mm travel, so that was the minimum size for this build. That narrows it down to only two bikes inside the Specialized family - the Enduro (170mm travel) and the Stumpjumper EVO (160/150mm travel).

Sloping from the bottom bracket: I knew that adding a mid-drive kit would cause potential clearance issues, with a hunk of metal hanging underneath the bike. The Stumpjumper EVO, although I like its geometry figures and suspension travel better, has a horizontal section forward of the BB. This is a deal breaker, as adding a motor below it is going to hang up on rocks. Enduro it is.

Carbon fiber: I'm putting a battery inside the downtube. JUST IN CASE everything goes wrong and my frame rubs through my insulation or I crash really hard and puncture something inside the pack, at least it's nonconductive. The other benefit of carbon fiber is weight savings - duh.

Swag: I found this frame on Pinkbike for sale at a reasonable price. The previous owner had wrapped it in this custom winter camo frame protector. How sweet is that?? It's borderline garish and undeniably unique, perfectly my style. He even printed his name on the sides of the top tube wrap, which I had to put electrical tape over. Sorry, Tom.

Here's the frame on the day I received it:

 

[handlworks]

THE WHEELS (AND TIRES)

Last mechanical thing. This frame didn't come with wheels, and I took this as a chance to grab exactly the hubs I wanted.

I knew that the CYC motors are pretty loud, and figured I'd match the "Stealth" with a loud freewheel. Maybe I won't even need a bell on the trail

I got a custom wheelset built up by a small shop down in San Diego. i9 hydra hubs (690 points of engagement!) laced to DT Swiss EX511 hubs. He hooked it up with the stronger spokes and brass nipples for the extra strength. This is the loudest freewheel I can think of, and I remember when these were the things that snapped necks out on the trail. Hearing a Hydra coming down the trail strikes fear into your bones and makes your blood run cold. That's the feeling I want.

This is a MULLET setup (29" front, 27.5" rear). Referred to affectionately by "business in the front, party in the back", the mullet gives me the rollover and momentum in the front-end with the light and playful smaller wheel in the back. I'm a mullet convert and figured: I'm building a bike, I'll make it mine!

Tires are:
Front: Maxxis Minion DHR II, 29x2.6, EXO+ casing
Rear: Maxxis Aggressor, 27.5x2.5, DD casing

Because I'm putting a mullet wheelset on a bike intended as a 29er, I'm going to mess up the geometry a bit. The headtube will get much slacker (on an already slack bike), causing my turning to be loose and wide. The BB is going to drop significantly, risking more hang-ups on rocks.

To remedy this, there is only ONE company in the world that sells a product to remedy this. Williams Racing Products in Australia sells a yoke specifically for Enduros to run mullets, and it extends my shock length to restore the original geometry. It's pretty pricey, but if it works then it works.

Getting excited and putting the wheels on before the tires show up:

 

[99t4]

Because I'm putting a mullet wheelset on a bike intended as a 29er, I'm going to mess up the geometry a bit. The headtube will get much slacker (on an already slack bike), causing my turning to be loose and wide. The BB is going to drop significantly, risking more hang-ups on rocks.

I have a question: How much lower does this place the BB compared to stock?

 

[handlworks]

I have a question: How much lower does this place the BB compared to stock?

Just measured, actually! I don't have a reference with the actual 29" rear wheel, so I'm looking at the delta between mullet + stock suspension vs. mullet + WRP link. See below pics, looks like about 1/4" or ~6mm. Measuring to the center of the BB:

Mullet with stock suspension:


Mullet with WRP link installed:

 

[handlworks]

THE MOTOR

I'm using the Stealth Gen1 from CYC. It replaces the crankset and fits into the BB, with a big hunk of a motor hanging off the front via a few brackets.

Replacing the crankset lets the motor system align to the BB, which is great. However, in the spirit of adaptability to any bike, CYC's third axis of rotation is restrained by a generic sheet metal bracket, intended to clamp onto a downtube. I don't want to use that bracket, because it's ugly af and it doesn't fit around my phat bike frame.

Without a bracket:


Where I want it:


Super scientific jig:


I settled on tying the motor offset bracket to one of my big suspension pivots with a piece of sheet metal and some spacers. Had to replace the stock bolts with longer ones, and do some scientific hand fabrication. I had tried to make a sweet aluminum tube bracket at first, but it was too hard to jig without burning the frame, and I ended up melting too much of the small spacers. Steel is real!

Sitting pretty:

 

[handlworks]

THE BATTERY, PT. 1 - FITMENT

Like I mentioned - I want to fit my batteries into the downtube SWAT storage. The ultimate goal here is that while looking at the finished bike, I WANT TO WONDER WHERE THE BATTERY IS. I've never been excited for a packaging project before, let's do it!

I spent a lot of time with a friend's Stumpjumper EVO, which has a similar SWAT compartment. I brought a bunch of 21700's with me, taped the ends, and played with configurations while stuffing them into the downtube. Here's a screenshot of a video I took of the fitment in his frame:


When my frame arrived, I had to find out exactly how much battery would fit. I was committed to being flexible about the construction. I realized pretty quickly that this was going to be like building a ship in a bottle - I'd have to assemble all the different battery parts INSIDE the frame!

One tool that helped immensely is a $30 borescope - I'm able to shove the camera into the downtube to see anywhere with a little light on the end. Here's a view looking down into the SWAT compartment, and finding out that the shock clearance digs into downtube space and ends my storage space prematurely:


After hours of trying different configurations, I decided that I could PROBABLY fit 20 cells into the downtube. I'll talk about the construction in the next posts.

 

[99t4]

I realized pretty quickly that this was going to be like building a ship in a bottle - I'd have to assemble all the different battery parts INSIDE the frame!

With this you've gained my respect. Your dedication and patience capacities are off the map!

 

[handlworks]

With this you've gained my respect. Your dedication and patience capacities are off the map!

Incredibly kind of you, I hope to live up to those words

 

[handlworks]

THE BATTERY, PT.2 - SHIP IN A BOTTLE

Here's a bunch of nebulous and disorganized thoughts that I was working with:

• I found (through hours of taping cells and fitting them inside the downtube) that I can fit up to 8 cells in one segment through the SWAT opening

• I could fit 20 total cells (10s2p), while still allowing room for insulation, wiring, and a BMS

• Each segment of the pack would need its own insulation, power wires, and balance wires - to keep space creep at a minimum, it's in my best interest to limit to the fewest segments possible

• Leaving room for the BMS was going to be tricky, and the most difficult part of building this ship in a bottle

• Every connection between segments is going to need significantly more than the bare minimum length of wiring, because only one segment can go into the SWAT box at a time, while the rest need to be outside (I'm estimating 3x the min lengths will be needed)

The end result is the following battery pack features:

• 10s2p, 36V nominal, 10Ah

• Samsung 50E cells

• 30A BMS

• Segments: 8 cells, 8 cells, 4 cells

To start construction, I positioned 8 cells together:

After welding, the strips were bent so I could rotate the cells into a long block. Then, hot glued into place and power/balance wires soldered on:

When I first constructed the pack, I used waterproof XT60s between the segments for power delivery. I removed them later for continuous wire due to space constraints, but the earlier stage looked like this:

And finally, the first time putting the batteries near the bike:


I dry-fit the battery segments maybe 5 times (slightly differently each time) to get a feel for how space would look without vibration damping. I spent lots of time on the dry fit because I anticipated install and removal would be difficult after vibration damping gets added. After each dry fit, I checked the fit with the borescope to understand how the contact with the frame happened, so I'd know where to add vibration damping. This was a fairly tedious process where most of the work happened inside my head, so I have very little documentation of it.

In the end, I figured out a repeatable installation procedure that worked for this pack. It involves mandatory installation order, precise numbers of wire coils and locations, and an overall muscle memory of how this is going to go.

The next steps are to add heat shrink, use the borescope a ton for vibration damping addition, and do a final install.

 

[99t4]

The end result is the following battery pack features:

• 10s2p, 36V nominal, 10Ah

• Samsung 50E cells

• 30A BMS

Just me, but I would have tested that configuration on a sample outing to see if it could supply the power and range I needed. Before making the ship in a bottle.

 

[handlworks]

THE BATTERY, PT. 3 - FINAL INSTALL

In my head, the final packaging needed to accomplish multiple things:

• General waterproofing (not looking for a submersible battery, but no exposed metal)

• Strain relief for wires

• Vibration dampening (high and low frequencies - riding at speed on singletrack, and large bumps)

• Lock positioning inside downtube (no moving around!)

I used fish paper to eliminate nickel-nickel contact, and to provide wear resistance to wires next to nickel strip. Strain relief was accomplished with strategically-placed kapton tape, before heat shrink. Heat shrink (wrapped in two different directions for overlap) further reinforced the strain relief, while providing general waterproofing.

The part I was most concerned with was vibration damping. I figured that I could provide adequate vibration damping inside the downtube by isolating the battery segments from the frame with neoprene foam sheet. My nervousness comes from whether what I do will be ADEQUATE, and PERMANENT.

This is the part I need to iterate on, over miles of ride testing.

I used 1/8" neoprene foam rubber sheet. I cut one piece at a time, placed it on a segment of battery, installed that segment, checked fitment with the borescope and my fingers, removed the segment, and made adjustments as needed.

THIS WAS SO TEDIOUS. In a few areas, I couldn't even fit the foam sheet and had to settle for cloth tape. Space ran out super quick. In the end, after install/check/uninstall/adjust countless times, I finally was able to cram the entire battery into the downtube.

Looking down onto the first 8-cell segment from inside the headtube: https://youtube.com/shorts/Juvd-2Zv3Wk?feature=share

Looking up on the other end of that segment, from inside the SWAT box: https://youtu.be/shpukSHdDyk

Looking down onto the last 4-cell segment from inside the SWAT box: https://youtube.com/shorts/S2FIhPPN2eA?feature=share

It's been pretty tricky to get any meaningful pictures or videos of this section; this is what I have. Next time I take out the battery to check on how everything is breaking in, I'll take full pics.

The second segment with foam on the corner contact area:


A different angle of that segment and the BMS with liberal fish paper:

 

[handlworks]

Just me, but I would have tested that configuration on a sample outing to see if it could supply the power and range I needed. Before making the ship in a bottle.

I did! Only around the block. It wasn't enough to get good confirmation of the specs. Here's a pic of the taped-in pack:

Reasons I was okay with no real-world testing:

• I'd need to climb mountain bike trails for a good test. Without vibration damping and final install? Hell no, I fall too much!

• I benchmarked against available packs (Specialized Turbo Levo, Levo SL, Trek EXe, etc.) and aimed for the same range

• I relied on CYC's 250W "street-legal" mode as a baseline for power requirements

• CYC's own X-36 battery is exactly this configuration

• My backup plan was to package a second battery inside a water bottle to keep in the bottle cage. This, connected in parallel, will give me more current output and range. This is still in the plan, as an optional range extender!

 

[motomech]

Looks like goat's head thorn country (you should really fill out your profile) and you will be getting flats.
Probably the best way to deal w/ them is flat-resistant tires (kevlar belt built in) and extra thick tubes. Then be prepared to fix them on the trail.

 

[moped-dan]

Looks really cool. Wish I had the skills to do a battery like that

 

[handlworks]

Looks like goat's head thorn country (you should really fill out your profile) and you will be getting flats.
Probably the best way to deal w/ them is flat-resistant tires (kevlar belt built in) and extra thick tubes. Then be prepared to fix them on the trail.

Did the location, thanks! I have the DD casing for the extra weight and gnar, and sealant inside everything. For riding here for several years, no problems even without the DD casing. (I hate tubes)

Looks really cool. Wish I had the skills to do a battery like that

Thanks! Hey, part of the reason I'm posting this is so others might get the idea to do the same

 

[handlworks]

FIRST LOOK

...And here, we catch up to present day!

I installed the battery according to the SOP that I created in my head - it's very regimented now.

I tweaked a few settings in the CYC and ASI apps (battery voltages and thresholds, assist levels, and speed limit so far).

I've been on a total of three rides on this baby, as short as 9mi. and as long as 24mi. SHE RIPS! The 24mi. day was a doozy, I rode on my own in Laguna here and did a relatively big trail with a tough climb (Willow -> TNA for the locals). I WAS ABLE TO DO THIS THREE TIMES IN A ROW. Usually on my pedal bikes, it's part of a long morning ride and we're pooped after TNA. Ebikes are totally cheating, dude.

So far, I've had to make a few tweaks:

• With the mullet wheelset, the bike was just too slacked out to turn well. I installed the WRP link that extends my shock length, and what a godsend! Normal rideable geometry restored.

• I had to reinstall the headset. The Specialized Enduro uses a thin spacer INSIDE the headset, and since mine didn't come with a headset, the Cane Creek I bought didn't work too well. The top dust cap was rubbing on my frame, inhibiting turning, and the headset came loose on the first descent. A 1mm spacer inside the headset fixed things, but I'm pretty peeved that Specialized had to do something silly with their integrated bearing cup...

• Suspension tuning (still ongoing) - man, tuning an ebike's weight is kinda weird! Not a suspension nerd, so I'm still learning. Increase the pressure!

Making a list for next steps to continue with this build (not in this order):

• Range extender + power boost - This is a doozy. This motor makes way more power than I'm using. I'm limited by my battery's power output. To solve this, I've already planned the battery connector to be right next to the bottle cage, so I can create a range extender disguised inside a water bottle. It'll be a 10s1p to just give me that extra 10A continuous for when I want it. Either long days or big sends, both of which I want to ride.

• Program the throttle - I want the option to turn on throttle power when I want it. I'll need to program a mode for it and keep a throttle handy and installable.

• Suspension tuning - Maybe I just suck at it, but it's seriously a huge deal and makes or breaks the riding experience right now.

• Clean the cockpit - I really hate cluttered cockpits. So far, I've done a decent job keeping things off my handlebars. I'd like to consolidate the cables so that they're kept tight with the frame, and won't come loose. Right now, they're just hot glued - I'll probably design and 3D print something applicable for all future builds.

• Hide the battery cable - One of my friends gave me some grief over the battery cable, saying it looks too homebuilt and sketchy. I'll take on that challenge - I need to find a way to hide or disguise it so the whole damn bike looks like it's OEM.

If you've made it this far, here are some glamor shots before the next update:

 

[99t4]

Right now, they're just hot glued...

Interesting. Do you have any closeups showing that?

One of my friends gave me some grief over the battery cable, saying it looks too homebuilt and sketchy. I'll take on that challenge - I need to find a way to hide or disguise it so the whole damn bike looks like it's OEM.

Might be able to camoflage it easy enough by painting it same scheme as the frame camo. Or maybe using (same color as frame camo) sections of tape? Also could paint similar camo pattern on the CYC. Make it blend in better.

 

[handlworks]

Interesting. Do you have any closeups showing that?

Here are some pics I took today:

Two cables, zip-tied together and hot glued


Big harness junction hot glued


Another that already came off


Blurry pic of my speed sensor

 

[handlworks]

MONEY TALKS (how much did I spend?)

This was quite a build - I dug into my credit cards for a bit on this one. Below is a screenshot of the cost tracker that I keep:


Takeaways regarding money:

• For a performance mtb build, the frame is by far the largest cost. I don't think I overpaid - the bike is amazing.

• I splurged on the wheelset. While I didn't overpay, it wasn't necessary to shell out for the highest tier parts that I got. Others who try this could get away with a used, lower-end wheelset for half that cost, easily.

• I saved SIGNIFICANTLY by scavenging a groupset and brakes off a different bike. This saved probably close to $1k vs buying new!

• Even though the WRP yoke was a lifesaver in the rideability of the bike, I still can't believe it cost that much.

I don't have much of an opinion on the costs here. I wanted to share for transparency, in case anybody else is interested. Ask questions!

 

[handlworks]

Update on the cable organization: I found some offerings from some bike companies that I might try instead of designing my own.

Jagwire alloy guides at Jenson: https://www.jensonusa.com/Jagwire-S...2uHs3nY5n6YCPNxyCAnLR5PYsw7M6HTcaAmjdEALw_wcB
Problem Solvers plastic guides at Jenson: https://www.jensonusa.com/Problem-S...F7alw0pidkOo3bu3pDXIrfh6U5Oa1FmkaAk4bEALw_wcB
Plastic guides from State Bicycle Co: https://www.statebicycle.com/produc...PVXzy-BTor39kUv7ZHBoTOus_AqBeKHwaAlLwEALw_wcB

I don't love the alloy guides, and I don't like that the Problem Solvers guide relies on friction instead of the clips used by the others. I think I'll try the ones from State Bicycle.

 

[handlworks]

I rode my build side-by-side with a Specialized Levo SL this weekend. My takeaways were that my bike feels considerably bigger (in suspension travel, frame size, and weight) and that the SL feels like the motor puts out considerably more power than mine.

The second takeaway really bugs me. I can justify the first in that the SL is a considerably smaller bike and frame, but the primary value prop of building this bike myself is that I can play with more power when I want it.

I can think of a handful of ways by which I can affect the perception of motor power:

• Increase power: This one is complicated. I'm limited in physical battery space by the downtube space constraints I placed at the beginning of this project. I can increase the power output from the motor (through programming) or even change the cells that make up the battery pack, but I will be taking a hit on range. I really like where my range is at right now. TBD on this one.

• Mechanical advantage: This is really just gear ratios. From the first 50mi I've ridden, I feel that the motor is engaging at a low speed, where it doesn't have as much power. Maybe gear ratios can match up my pedaling cadence to a better motor power band. I can reduce the motor input gear teeth, increase the motor output teeth, or maybe reduce the size of my crank input chainring. Not sure yet...

I have a set of questions I need to answer in order to inform the next steps:

• How much of my max constant current am I using? My 2P 50E's should be able to output 20A total, netting 840W from a fully charged pack.

• How much room do I have to increase the motor power? The display shows that my highest assist mode is using ~200W max. Not sure exactly what that number is, or how accurate it is.

• What's the best way to match up my leg + motor power bands? Which gear is best to change, and what is available without custom machining? CYC uses a 219H chain for the motor reduction, and the typical 3/32 Narrow/Wide chain for the crank connection.

 

[scianiac]

CYC offers two different ratio sets for the 219 reduction, the other set I think would increase low end torque at the cost of speed, but considering the max rpm of the motor (depending on voltage of course) is may be higher than you can pedal that may not be an issue. I know it's way way higher on 72V, maybe just above normal cadence on 36V?

I don't think however you need that, I built a similar bike for a friend (lightweight bike, CYC stealth, small battery) and it has way more power than can be controlled with the torque sensor and well enough ride around in baby dirtbike mode no problem. Sounds like you just need to turn the current up to the motor. As for the batteries to provide that power it might be possible to fit more battery in the space available by using prismatic RC lipos, but really only if you have a reasonable amount of free space left. As in the energy and power density is worse for RC lipos but have better packing efficiency and are thin enough that you can fill the space more efficiently. Also would allow you to run a higher voltage pack which may help. Could also build a small pack that sits just above the motor which would blend in pretty nice.

 

[handlworks]

CYC offers two different ratio sets for the 219 reduction, the other set I think would increase low end torque at the cost of speed, but considering the max rpm of the motor (depending on voltage of course) is may be higher than you can pedal that may not be an issue. I know it's way way higher on 72V, maybe just above normal cadence on 36V?

I don't think however you need that, I built a similar bike for a friend (lightweight bike, CYC stealth, small battery) and it has way more power than can be controlled with the torque sensor and well enough ride around in baby dirtbike mode no problem. Sounds like you just need to turn the current up to the motor. As for the batteries to provide that power it might be possible to fit more battery in the space available by using prismatic RC lipos, but really only if you have a reasonable amount of free space left. As in the energy and power density is worse for RC lipos but have better packing efficiency and are thin enough that you can fill the space more efficiently. Also would allow you to run a higher voltage pack which may help. Could also build a small pack that sits just above the motor which would blend in pretty nice.

You're totally right. My last post was a brain-dump riddled with logical holes that I hadn't worked through yet.

I have the 11/53 gearing, which is the lower torque / higher speed gearing. Shame on me. I haven't found a 10t 219H B-type sprocket that would fit my motor shaft, and I'm not trying to have a one-off made. I'm willing to bet the motor speed won't be a problem with pedaling cadence on the high-end, just need to match them up!

I totally mixed up my thoughts on the power, and forgot that I was still mostly on the stock tune power limit! I've upped Street Mode from 250W to 500W, will have to ride a trail to evaluate performance.

Not looking to try lipos, as I have no experience with them, and I crash pretty often haha! I think my next steps are:

• Increase power through firmware, while watching out for range cannibalization (I really like my range where it's at right now)

• Play with gearing, in this order: crank input chainring (easiest to get, probably least effective) > motor input sprocket (if I can find a 10t 219H) > motor output chainring (seems to be very custom, with the 219H spec but drilled for 104BCD)

And, of course, re-evaluate at every step!

 

[handlworks]

Okay I upped my Street Mode to 500W instead of 250W and it completely changed the ride. Now my lowest assist setting is too powerful to not ghost pedal!

I think I'll set my lowest assist setting back to the original setting and give myself the option of upping all the way to high power on demand.

Then (and this is completely uninformed by feasibility), I'd like to set Race Mode at a ridiculously high power and let that be my "throttle connected" setting where I can use this as a motorcycle.