The Golden Goose. 72v Dengfu E55 CYC X1 Pro.

[kaaskopf]

so cool! love how its different!

just a small tip;

did you try your rear triangle without shock ? your rear brake line might bind during full travel

i had the line like that on e10 before and it was pulling alot during bottom out

first ride i broke speedsensor cable and break hose

 

[bram.biesiekierski]

THanks for the tip. Im pretty sure i cycled rhe suspension when i put the lines in. I remeber thinking that it seemed not as good as the e22. I did actually put a layer of dual wall heatshrink on both the brake and shift hoses, as i was concerned with it rubbing at the bend. Anyways, I will definetly re-check now that ypu mentioned it. I have some 3d printed brackets for the speed sensor coming, so when that artives I will check all the cable/lines.

so cool! love how its different!

just a small tip;

did you try your rear triangle without shock ? your rear brake line might bind during full travel

i had the line like that on e10 before and it was pulling alot during bottom out

first ride i broke speedsensor cable and break hose

 

[LAnton]

Got a few little bits and pieces tidied up on the bike today. Swapped in a longer 58mm offset tripple clamp for slightly more steering angle. Finalised the layout for the bars, grips, brake levers etc. Swapped the gold Renthal bars out for black, and swapped grips over to the tan/gum coloured Race Face Gripplers. Everything laid out comfortably, and adjusted. A little bit cramped with shifter, throttle, brake and display on the right. But not really any worse than on the e22. I can get my thumb where it needs to be when I need to.

I put some yellow spiral wrap over the lines/cables, I was hoping for a slightly darker yellow, but thats all I could find on Aliexpress. Tidied up the cable routing into the frame. Bled the brakes with Formula fluid. Sized up the chain. Adjusted the derailer limit screws, and got the shifting all dialed in.

I even went for a little pedal around the yard to feel the bike out. No power ofcourse, but it was nice to feel the geometry, steering, brakes etc. Bike feels good, atleast in the seated posistion (didnt really want to try stand up pedalling on the plastic prototype bracket).

Its high, but its not bad. It doesnt feel like its that much bigger than the e22 considering this has 30mm larger forks, and 20mm taller bars. Fork feels good, think Ive got a pretty good spring rate in both the fork and rear shock. Up front is a 9.7nm / 55 pound coil. Rear is a 571 pound coil. Feels about right for what I like. I do have a couple of different size coils for both front and rear to play with if I need.

The lack of steering angle due to the dual crown isnt as bad as I was worried it would be. I can turn within my driveway, and a little speed and lean actually tightens the radius up further. I think it will be fine for most of the single track switch backs I will encounter. And if not, the bike can just monster truck over whatever I need it to.

Ive placed an order for the metal 3d printed motor bracket. I actually ordered 2, from different suppliers, as I want to check any difference in quality from different suppliers. I ordered from PCBway and Rapid Direct. Both AlSi10Mg aluminium selective laser sintered powder bed fusion process. Price was similar from each supplier, with the Rapid Direct a bit cheaper. The Rapid Direct part will also be anodized. I forgot to spec anodize on the PCBway order. PCBway have stopped accepting credit card payments aswell which made ordering a hassle. I suspect the US has told credit card providers to stop dealing with PCBway for some reason like defense or trade policy???

During quoting, Rapid Direct also said they will follow up and CNC machine the bores for the bolts and bottom bracket holes to dimension aswell. Which is a nice touch. I didnt get any threads specc'd in the order, as its something I can do once landed. The only threads are the 2 x M5s to mount the CYC motor. And then the 6 x M6 grub screws, 3 for each side of the BB to hold the bottom bracket snug. Unfortuantly, due to the Chinese spring festival period, I wont see either part until march.

View attachment 177383

View attachment 177384

View attachment 177385

View attachment 177386

View attachment 177387

View attachment 177388

View attachment 177390

View attachment 177391

View attachment 177392

View attachment 177393

View attachment 177394

Checked the head angle and seat angle out of curiosity. The 200mm fork and angle set were just a guestimate to get it somewhere close to rideable. I didnt really know how it would turn out until after it was all done.

Looking at the original advertised head angle of 64°, then fitting a 200mm fork, with a 1.5° steeper angle set has resulted in the frame tilted back approx 2°. The new head angle is pretty close to original now at 63.5°. Once you factor in a little more sag on the larger fork compared to the rear, it will probably settle at around the original 64°.

The seat tube measures 69°, but this cant really be compared to the original advertised seat tube angle of 77° as I am pretty sure that is effective seat tube angle measuring from BB to sest, not just the seat post as I have measured.

Head angle 63.45°.
View attachment 177397

Seat angle 69.25°
View attachment 177398

Can the motor mount be shortened? It can then be hidden in the frame

 

[bram.biesiekierski]

Can the motor mount be shortened? It can then be hidden in the frame

Not really. The space between the cranks and the actual CYC electrical motor are fixed, because the primary chain reduction needs to be that long. At least not with out re-engineering the entire CYC system.

The bottom bracket can be posistioned anywhere really, as my custom bracket geometry defines its location. I have already dropped the BB a short amount to slightly account for having a longer fork. I wouldnt really want to raise it at all, as then your BB drop turns into BB raise, and your standing too high, which would probably feel weird to ride.

I cannot really move the BB rearwards, as the chain ring will start impeding on the chanstays. (And I wouldnt want to anyways as that will shorten the effective chainstay length, and one the best features of these Dengfus is the relatively long chainstays.) And likewise, moving the BB forwards enough to tuck the motor part infront of the down tube would really throw the F/R balance out.

I did toy with the idea of rotating the entire motor up and into the space inside the triangle. Which I could have done quite easy, with little variation to the original BB location. But I was also playing with the idea of building a range extender battery into that space aswell, which I want to keep that option open as I am unsure how long the 20s3p will last with the X1 power. And there would be no benefit of putting it there anyways. It just moves a big chunk of mass higher than it needs to be, and the aesthetics still look a little whack because then you have a big gaping hole down below the down tube.

The ground clearance with the motor in it current location is ample. The motor reachs forwards compared to the original M620, but not any lower. Infact it may even sit higher than the M620. And the frame has an extra 50mm of suspension upfront aswell.

 

[bram.biesiekierski]

Can the motor mount be shortened? It can then be hidden in the frame

If I end up desinging another bracket. One thing I may look at changing is moving the motor control module up into the hollow space where the M620 used to go. In standard CYC config, it sits below the BB, at the lowest part of the arrangement. (Because on a standard X1 conversion, that space is all bike frame.) It probably gets good airflow for cooling, but is slightly exposed. And the wires exit facing up, which is a terrible arrangement for long term water ingress.

It would fit if I redesigned the frame work that supports the forward section of the kit. I could even use the motor controller as a semi structural part of the main bracket. But making it all work would be a design nightmare. Ive spent way too long designing amd optomising the bracket just to get it to this stage. And changind controller lpcation would basically require a full re-design in CAD.

 

[bram.biesiekierski]

I also designed this little plate to block off the standard e55 charging port. I never charge or store my batteries inside the bikes. So the standard charge pocket is redundant.

 

[TQFreak]

I machined the shock mount a little different so there are less custom parts (only 1 actually and 1 part is stock tube trimmed and tapped). It is a little bit lighter, there is no stress concentrator like on the shoulder screw shoulder area and also the axle engages with the carbon area completely

Link with details

 

[TQFreak]

By the way, you can go back to Bafang M620 motor now, the new open source completely integrated VESC based INNOTRACE controller for M620 capable up to 100V and 10 kW has been just finished. That bad boy is certainly will smoke the CYC motor and you do not need to go Frankenstein stile

VESC based INNOTRACE motor controller for Bafang M620 CANbus

The archive with the current versions of the motor controller files (KiCAD version 9.0.8), Gerber files, CAD files and firmware files can be downloaded here (updated on 2026-03-26). https://drive.google.com/file/d/1hCO7RL1AwwIHGNZ7YgvKHUjH4vSG7JOC/view?usp=sharing There are two PCBs in two...

endless-sphere.com

 

[bram.biesiekierski]

Yes I had seen your solution and it was an inspiration for me when designing mine. I actually even purchased some of the same tube. In the end i went with the bolt as i was concerned about bending the tube. Ive bent the e22 shock bolt and shock bolts on previous bikes too. So I wanted to move away from a hollow tube and try and incorporste a grade 10 bolt to give the best chance to not bend. The e55 is quite a wide span and that makkes it easier to bend things inbetween.

My bolt bears on the carbom, aswell as the entire surface of the 2 end pieces,. And the bolt is a tight fit into the end pieces aswell. hopefully spreading the load into the frame ober the largest possible area.

I machined the shock mount a little different so there are less custom parts (only 1 actually and 1 part is stock tube trimmed and tapped). It is a little bit lighter, there is no stress concentrator like on the shoulder screw shoulder area and also the axle engages with the carbon area completely

Link with details



View attachment 177492

View attachment 177491

 

[TQFreak]

I have ordered custom copper bussbar from 2 different vendors. Wellgo and EverBest battery. I went with 2 vendors as that way i could pick the highest quality option.

My Wellgo custom 0.5mm copper bussbars arrived today. One side is perfect. The other side has been manufactured incorrectly, with the weld tabs facing the opposite side. EDIT : Wellgo have agreed to send out replacement bussbars to replace the mirrored ones.View attachment 174394
View attachment 174395

View attachment 174396

How did you specify the shape for those bussbars for them? Did you make it in a CAD program and sent the the 3D models or you made a PDF file with the outlines of these parts?

 

[bram.biesiekierski]

How did you specify the shape for those bussbars for them? Did you make it in a CAD program and sent the the 3D models or you made a PDF file with the outlines of these parts?

I send DXF files. I pulled the DXF directly out of my battery design CAD file. Same with the Isoval 11 insulators.

I think i did share the onshape project and battery cell holder STEP files with one of the sellers to better understand how i wanted the bussbars to fit.

While on topic. The bussbars I got from Wellgo re-seller and EverBest are identical. They even sent the exact same paper work for both orders, and the boxs had the same packaging underneath. The correspondace even had the same last name. So I believe they all came from the exaxt same factory.

 

[bram.biesiekierski]

THat is cool. Had i known about that 6 months ago when i started this project, i probably would have gone with that option. I have had the idea to do the x1 e55 for a long time now, and at the time time any serious attempts to tweak the m620 were all vaporware. The X1, while not perfect, was ready to go and fitted the bill and requirements for my planned build.

Although, in either case of the modded m620, or the X1, i think they are both ultimately going to be battery limited on the e55. 10kw peaks sounds amazing, but you wont get far/long with a battery the size of the e55. A custom battery without a case, or extra range extenders etc would be larger, but then you loose the quick swap functionality. At some point a compromise between power and endurance must be made.

By the way, you can go back to Bafang M620 motor now, the new open source completely integrated VESC based INNOTRACE controller for M620 capable up to 100V and 10 kW has been just finished. That bad boy is certainly will smoke the CYC motor and you do not need to go Frankenstein stile

VESC based INNOTRACE motor controller for Bafang M620 CANbus

The archive with the current versions of the motor controller files (KiCAD version 9.0.8), Gerber files, CAD files and firmware files can be downloaded here (updated on 2026-03-26). https://drive.google.com/file/d/1hCO7RL1AwwIHGNZ7YgvKHUjH4vSG7JOC/view?usp=sharing There are two PCBs in two...

endless-sphere.com

View attachment 177493

 

[bram.biesiekierski]

I Have since purchased an AwithZ P120D spot welder. Toying about the idea of attempting to weld 0.5mm copper directly to the cells. Getting 0.5mm copper laser cut with out any of the spot weld tabs, just some simple spiral current labrinths cut in to direct the weld energy into the weld, and offer some small degree of mechanical flexure.

I have seen that this welder is capable of welding 0.5mm copper directly, with flux and stainless/nickle sandwhich method. So at some point i may build an up-rated battery. Perhaps wait until full tab cells become a little more refined / mainstream.

How did you specify the shape for those bussbars for them? Did you make it in a CAD program and sent the the 3D models or you made a PDF file with the outlines of these parts?

 

[TQFreak]

In terms of amperage that 0.3mm copper sandwiched with nickel for the battery should work fine. The copper tracks on the power stage of the motor controller are thinner. The motor will start overheating before the battery will hit the limit.

 

[bram.biesiekierski]

Its not about finding the limit of the copper, its about minimising small losses, which while insignificant on thier own, can compound to be quite large inefficiencies.

Those copper bussbars look great on the surface, but in a 20S battery, you have 21 x bussbar interconnects. Each individual one comprising of atleast 3 seperate pieces. Eg the base copper, plus the previous cell nickle weld tab, plus the following cell nickle weld tab. That is atleast 4 seperate welds / joins per bussbar. That stacks up to like 80-100 joints / welds in series before it even touchs the BMS, let alone exits the battery pack. Each joint, EG each time the copper meets the nickle, or the nickle meets the cell, there is a slight bottle neck at the weld, or in the thin nickle plate itself. Tiny, negligable, difficult to even quantify, but a lpss none the less. At each step, the resistance is >0. So at each step, there is some small degree of I^2R loss. Eliminating 2 of the 3 pieces that comprise each bussbar, and more than half of the welds/joins within the battery pack stacks to a significant improvement in both electrical performace, as well as mechaincal reliability.

Also, having held these copper bussbar in my hand, and destructively testing a universal batch i have purchased previously, i can tell you with 100% certainty, that there are some cold joints in the welds between the copper and nickle weld tabs. Some pulled away with significantly less force than others. Having a cell open circuit because a weld tab joint was cold, could be a serious problem in a series pack like a 20s 3p. Loosing 1 cell open, means the other 2 then get an additional 50% loading. That can easily become a battery fire.

It is actually something i see alot of in my professional career as a HD auto electrical fitter. Whether it be on multi million dollar rail tamping machines, or basic mobile lighting plant. Wiring done to the bare minimum (and in alot of cases these days, below bare minimum) is a major problem in my industry. My company is currently dealing with a + million dollar whoopsie, because some engineer thought he could save his companie a few bob by using insufficiently specc'd wiring.

In terms of amperage that 0.3mm copper sandwiched with nickel for the battery should work fine. The copper tracks on the power stage of the motor controller are thinner. The motor will start overheating before the battery will hit the limit.

 

[TQFreak]

I did not try the strength of the joint between the nickel and the copper of those buss bars, but the strength between the copper and the battery after welding is quite strong. If I pull it hard It tears the copper and the copper remaining stay on the battery terminal after tear test. The nickel tabs on the bussbar is just to provide resistance to generate enough heat to weld the battery and the copper. After welding the joint between the battery and the copper is what important and it should be quite strong.

 

[kaaskopf]

you bent the original shock bolt ?

ow wow
i go bikeparks regurly with my e10 and im not a lightweight , i think it is the same bolt design

 

[bram.biesiekierski]

you bent the original shock bolt ?

ow wow
i go bikeparks regurly with my e10 and im not a lightweight , i think it is the same bolt design

Yes. The original e22 bolt is bent like a banana, well not a banana, but its pretty bent. Its exactly the same as the e55 bolt, even though the e55 frame has a different mounting lug design. I anticipate it will it would be worse with the e55 as there is more weight of battery, motor, suspension etc with that build.

Ive also bent shock bolts in previous bikes aswell like my Trek Remedy.

There is some mega forces going through those parts. For example, some back of the envelope calcs : I typically run like a 550 to 575 pound per inch spring. Which is kinda softish for my weight, but I like that old school lazy coil feel. Common enduro shocks like my e22, e55, Remedy etc are all 230mm x about 60 to 65mm stroke. So at end stroke, you have compressed about 65mm or about 2.5" of spring. So 2.5 x 575lb = nearly 1500 pounds. Thats nearly 700kg pushing on a relatively soft 8mm bolt. If you happen to bottom out hard, I'd estimate it could even be as much 1000kg at peak impact. So its not really surprising to bend such a small piece of metal.

 

[bram.biesiekierski]

Yes, look I have no problem using these bussbars on my first batch of batteries. It is ample for my needs. And it is simple for producing reliable welds with a wide variety of consumer weld devices. All i am saying is from a purely electrical stand point, it could be alot better. And the technology to make it better is available. Although, that said. Making reliable welds with 0.5mm copper directly to a cell is pretty advanced level skill, (more advanced than what I can currently do.) And requires a VERY powerful machine and alot of patience to dial in the perfect recipe. So I am not there yet, but it is on my mind.

With regards to the bussbars. Another way to look at it is this. Those nickle weld tabs are 0.15mm thick, x 8mm. (With a slot cut in the middle aswell). Each one is dropped into a hole in the copper about 8mm across. So about 10mm from weld to weld. Or about 5mm from center of the cell to where it meets the copper sheet.

On a 20s3p battery, ALL of the current must pass through just 3 of these tiny nickle weld tabs in parallel. And it must do it 40 times in a row. (20 series cells, each with + & - connections = 40 times).

Now imagine getting a piece of 0.15mm x 8mm nickle strip, about 400mm in length (40 weld tabs x 10mm = 400mm) and then passing 25a to 30a through it. Or getting 3 x of those pieces in parallel, and then passing 80a through it. Because that is exactly what will be happening in this battery pack. When you think about it in this way, then you very quickly realise why I say it could be better. For those that dont know, A 400mm long, 0.15mm thick x 8mm wide nickle strip, carrying 25-30a is basically a giant resistive heating element!!!

The 0.5mm copper sheet afcourse is overkill, but there will be losses in the nickle in-between. And ofcoarse 80a peaks is very different to 80a constant. The effective duty cycle at those 80a peaks will be relatively low. So this is why I am comfortable to use the copper bussbars. They are already an order of magnitude better than the old way of joining hubdreds of nickle strips together. And way less welding aswell.

I did not try the strength of the joint between the nickel and the copper of those buss bars, but the strength between the copper and the battery after welding is quite strong. If I pull it hard It tears the copper and the copper remaining stay on the battery terminal after tear test. The nickel tabs on the bussbar is just to provide resistance to generate enough heat to weld the battery and the copper. After welding the joint between the battery and the copper is what important and it should be quite strong.

 

[TQFreak]

But those nickel strips are on top of the copper tabs and the copper tabs make contact with the battery cells terminals which is low resistance. So the current from a battery cell flows through the copper toab and not through the nickel tab on top of the copper tab. Nickel here does not handle the current from the battery cell.

There are copper tabs directly under the nickel tabs, you can see it on your pictures the bottom face of the tabs is orange copper color. When you weld your battery the nickel should be on top of the copper, not on the bottom and the copper is what makes direct contact with the battery cell terminal.

 

[bram.biesiekierski]

I Wouldnt call that a copper tab. Its nickel with a very fine electro plated (or some other plating prcoess) layer of copper on one side, maybe a few microns thick. Those tabs are magnetic, i can lift an entire 6 tab bussbar with a small fridge magnet stuck to one tab. Nickel is a magnetic metal, copper isnt.

But hypothetically, even if it was the opposite, say made of copper with a few micron thick layer of nickel plating on the backside, its still 0.15mm x 8mm. That is like 1.2mm CSA. Still not great for upto 25-30a. And it is still attached to the larger copper plate by 4 small weld dots. And you still do that 40 times in a row in a 20s series string.

But those nickel strips are on top of the copper tabs and the copper tabs make contact with the battery cells terminals which is low resistance. So the current from a battery cell flows through the copper toab and not through the nickel tab on top of the copper tab. Nickel here does not handle the current from the battery cell.

There are copper tabs directly under the nickel tabs, you can see it on your pictures the bottom face of the tabs is orange copper color. When you weld your battery the nickel should be on top of the copper, not on the bottom and the copper is what makes direct contact with the battery cell terminal.

 

[bram.biesiekierski]

Placed an order for 120 x 21700s to make 2 batteries initially. I opted for Samsung 50s v2. I know they arent the bleeding edge with battery tech. But after reading up on some reviews I believe they are suitable for my needs. Looking on Moochs spreadheet of batteries, they still actually score quite well at around my anticipated peak current draw needs.

The few full tab / tabless cells that there are available, it sounds like there may still be some teething issues. And the supply chains are certainly alot more muddy with the likes of Relience, EVE, Tenpower, Ampace etc. Which presents risks of counterfeits, pre-productions etc etc. Id rather take a step down in performance in this area, than risk a problem of that sort.

 

[bram.biesiekierski]

I Put a little chain slap protection on the chain stay. No idea how long it will last. Suspicion is not that long. Between the heavy chain, low engagement HD hub, and fairly weak / sloppy derailer, there is alot of chain slackness.

I can place one finger on top of the chain and deflect it until the chain touchs the chainstay, and the freehub doesnt even engage the next pawl detent. And the derailer arm has so much backlash that it just flexes to let out the chain. No resistance, no clutch engagement, no spring force. The chain can slap before it even engages derailer spring. Really concerned with the quality of the Microshift derailer.

 

[kaaskopf]

it does leak neat

i always wrap an inner tube with tape or tywraps around the stays
does not look nice but stays functional for long time

 

[bram.biesiekierski]

it does leak neat

i always wrap an inner tube with tape or tywraps around the stays
does not look nice but stays functional for long time

Yes. Ive used self amalgamating tape before. Works pretty good for a while.

 

[bram.biesiekierski]

Been playing around with the AwithZ P120D today. Using some salvage 18650s and spare bussbars to try and dial in the weld process.

This welder is a freaking beast. To do the copper bussbar weld tabs, it likes to use about level 5 of 999.

It will weld 0.3mm copper directly to the cell (with flux) with about level 100 of 999.
I was able to weld 0.5mm (with flux) directly to the cells at around level 400/500 of 999. But the heat was so much that the cell wrap melted. So i dont think thats a viable option without alot of refinement to the technique. The leads have brutal kick from the lorentz force. This welder scares me a little haha

This is a pretty good setting for the weld tabs.

Specs sound rediculous, but I think its preey legit.

This is 0.5mm copper fused directly onto the cell.

 

[bram.biesiekierski]

0.3mm copper with flux, welded directly to the cell.

 

[bram.biesiekierski]

Experimenting with doubling up the nickle to join between the cell and the copper bussbar. This is looking like potentially a good way to really beef up the connection. Effectively doubling the cross sectional area and spot weld surface area between each cell to the common copper bussbar. But a massive amount more welding involved. One of the key advantages of using prefabricated bussbars is to get away from having to do so much welding.

 

[bram.biesiekierski]

Batterie cells arrived today.

 

[bram.biesiekierski]

Played around with the battery build a bit. My V1.0 Delrin cell scaffold is too tight for the case. I already knew this which is why I had the V2.0 items printed.

The V2.0 are a better fit, as I had slightly increased the width and depth the cells could insert, which made the overall assembly slightly thinner. But I had overlooked one critical aspect. When I had the V2.0 3D printed, I opted for a resin sterolithography print process, and selected the basic white resin material. This material is great for 3D printing as it is cheap, has relatively good mechanical properties and makes smooth prints with great detail, and fine features. But it has relatively low thermal characteristics. It softens, deforms and potentially melts at quite low temperatures. The softening temperature is below what could be expected inside the battery case at high demands.

Thankfully I caught this as having the cell scaffold deform with temperatures could result in an internal short circuit and possibly even battery fire.

This has put a slight delay in the battery build, while I wait for some V3.0 prints to arrive. I have opted for a high temperature resin material which should remain stable at potential expected battery temps. Its otherwise looks to be a relatively similar in its mechanical properties, and hopefully prints just aswell. It is slightly translucent red aswell, which should look pretty cool.

In the mean time I have experimented with a vinyl wrap for the battery case. Because I ordered empty battery cases, the lids are unpainted and wont match the luscious red paint of the Degfu frame. I fpund this red which is pretty close colour wise, but it lacks the metallic fleck in the frame paint. It was a few bucks for enough to wrap both batteries. So Im not too concerned with it not being perfect. It is on the under side of the front triangle and the bike wont stay clean and pretty for long once riding anyways.

I might see if i can find another vinyl that is a better match. It is relatively inexpensive.

Tried gold aswell