New to E-Bike Building – Need Advice from the Pros!

[Rhapsody]

Hey guys!
I’m building my own DIY e-bike from scratch — frame, battery, everything — and I’m learning as I go. If you’ve built one before, I’d love to hear your tips or connect for occasional guidance. Just trying to avoid rookie mistakes and build something solid.

Thanks in advance! The bike I want to build is going to be a surron clone. Aesthetically pleasing but not as efficient... I would love to learn under whoever as far as they can help me complete this project.

 

[amberwolf]

We'd need to know complete details about the project, how it's going to be used, and under what riding conditions, in order to provide useful feedback.

Otherwise I could only recommend reading other build threads, troubleshooting threads, etc., that cover whatever it is you're going to do.

 

[E-HP]

Do you weld aluminum?

 

[bananu7]

The general tip would be to fit the project to your fabrication capabilities. Nothing is truly "from scratch" and you will need to rely on the supply chain sooner or later. I suspect you won't be making your own suspension, tyres or battery cells. So it's pretty important to decide what parts you want to make (e.g. frame, seat, battery enclosure) and what you want to buy (literally everything else).

Even when assembling existing components, there's a lot of moving parts (literally!) and getting fitment right is always a challenge. Basically there are two ends of the spectrum here: design everything in CAD and rigorously maintain everything up to spec, replacing and modifying until it all fits, or not design anything and just rig everything in situ. You'll see in the existing builds that something in the middle generally yields acceptable results for one-off builds.

Then there's the decision order. You need to lock into certain choices because they will impact everything else. The thing I personally would recommend would be to start with the exact tyres you want, in the sizes you want, and build everything in sequence from there - rims, spokes, hubs, swingarm and front suspension, and finally the frame itself, with the caveat that you should know at least more or less what the battery will be like. This should minimize backtracking and will ensure that the project has parts that are optimally strong/lightweight.

 

[Rhapsody]

We'd need to know complete details about the project, how it's going to be used, and under what riding conditions, in order to provide useful feedback.

Otherwise I could only recommend reading other build threads, troubleshooting threads, etc., that cover whatever it is you're going to do.

The document attached has a summary of everything I plan to do, problem is I need someone with substantial knowledge in e-motorcycle building and its intricacies to guide me. Where I stay e-motorcycles are quite rare so there are no laws or limits to my creativity. I just need people to mentor and guide me on this project. Thanks in advance!

 

[Rhapsody]

The general tip would be to fit the project to your fabrication capabilities. Nothing is truly "from scratch" and you will need to rely on the supply chain sooner or later. I suspect you won't be making your own suspension, tyres or battery cells. So it's pretty important to decide what parts you want to make (e.g. frame, seat, battery enclosure) and what you want to buy (literally everything else).

Even when assembling existing components, there's a lot of moving parts (literally!) and getting fitment right is always a challenge. Basically there are two ends of the spectrum here: design everything in CAD and rigorously maintain everything up to spec, replacing and modifying until it all fits, or not design anything and just rig everything in situ. You'll see in the existing builds that something in the middle generally yields acceptable results for one-off builds.

Then there's the decision order. You need to lock into certain choices because they will impact everything else. The thing I personally would recommend would be to start with the exact tyres you want, in the sizes you want, and build everything in sequence from there - rims, spokes, hubs, swingarm and front suspension, and finally the frame itself, with the caveat that you should know at least more or less what the battery will be like. This should minimize backtracking and will ensure that the project has parts that are optimally strong/lightweight.

The document attached has a summary of everything I plan to do, problem is I need someone with substantial knowledge in e-motorcycle building and its intricacies to guide me. Where I stay e-motorcycles are quite rare so there are no laws or limits to my creativity. I just need people to mentor and guide me on this project. Thanks in advance!

 

[chuyskywalker]

problem is I need someone with substantial knowledge in e-motorcycle building and its intricacies to guide me.

Have you built a smaller scale ebike before? How's your knowledge and experience in welding? What battery welding equipment do you have? CAD? Etc, etc, etc.

There's a LOT that goes into making something as fast and powerful as you are talking about, especially when you seem to aim to build even the frame yourself. Starting with simpler parts and pieces and working your way up to it would be wise.

Jumping into the deep end of "I'm going to replicate surron's from scratch!" is an extremely ambitious goal... you should temper your expectations.

 

[Rhapsody]

Have you built a smaller scale ebike before? How's your knowledge and experience in welding? What battery welding equipment do you have? CAD? Etc, etc, etc.

There's a LOT that goes into making something as fast and powerful as you are talking about, especially when you seem to aim to build even the frame yourself. Starting with simpler parts and pieces and working your way up to it would be wise.

Jumping into the deep end of "I'm going to replicate surron's from scratch!" is an extremely ambitious goal... you should temper your expectations.

To be honest this is the first ebike I going to be building. I know aiming to replicate a surron seems ambitious... The truth is I want to build an ebike that somewhat looks like a surron with a 48v 20ah battery, motor and controller of equal standards and the ability to get a distance of at least 40km out of it. If you can help I would really appreciate. Also I would be building the battery myself, I have access to a spot welder, nickel strips, 18650 cells, a bms... whatever I will need to complete the task.

 

[chuyskywalker]

Go to youtube, my man, or browse the build threads here. There's AMPLE amounts of how-to's out there already without people needing to hold your specific hand to start this. When you get stuck and research fails you, come back and asks specific questions, demonstrate that you've done some/any leg work on solving it yourself, and people here will be happy to help.

 

[Rhapsody]

Go to youtube, my man, or browse the build threads here. There's AMPLE amounts of how-to's out there already without people needing to hold your specific hand to start this. When you get stuck and research fails you, come back and asks specific questions, demonstrate that you've done some/any leg work on solving it yourself, and people here will be happy to help.

Understood thanks for the help.

 

[e-beach]

Get your motor, your wheels and build your battery. Once you have those components, build your frame around them. It will be easier to build a frame around components you can measure then it is to theorize measurements of components you don't have yet.

 

[TPEHAK]

You need to design whole thing in CAD program first.

Here is how you build the frame

Motor controller for reference design with pedal assist function can be taken here

Archon X1 Programming Thread (questions and experiences)

I played around a little bit and was finally able to connect via VESC Tool to my Innotrace X1! First I made connection using a simple USB to CAN adapter (e.g. canable) and tried to send common VESC CAN commands (VESC CAN-Bus Communication). I got no response and assumed it has been either...

forums.electricbikereview.com

Reference firmware for motor controller can be taken here

vedderb - Overview

vedderb has 38 repositories available. Follow their code on GitHub.

github.com

Here how to make brushless motor

Here is how to build battery

https://www.emtbforums.com/threads/tqfreaks-dengfu-e55-build-thread.32525/post-480116

 

[bananu7]

an ebike that somewhat looks like a surron with a 48v 20ah battery, motor and controller of equal standards

There's zero chance you'll be able to match Surron's quality with those components if you go DIY. Same goes for the frame if you make it from "steel tubing". Your "document" is a wishlist, not a plan.

problem is I need someone with substantial knowledge in e-motorcycle building and its intricacies to guide me

How much are you paying for consulting? Or are you assuming you'll get professional quality consulting by paying substandard prices? The Surron bikes (which you're comparing yourself to) were designed by a large team of people over the course of years, working on every single detail. How much do you think designing it costed?

Now, I don't want to sound too discouraging, but you're heavily underestimating the project right now. That's not uncommon for a first-time builder and happens to all of us, myself included. At least you're using existing wheels, controller and motor. That's a good start and will give you something to build "against". So let's review what we have in there:

* If you want to build on 19" wheels, you'll end up with a teenager-sized vehicle at most (kind of like the Surron). Make sure you're aware of it. I would really recommend going with a 21" up front for a full-sized adult.

* 1-2kW motor is far too weak for this, unless you're citing the "rated" power. The actual system power you should target for this build should be around 7.5kW +/- 2.5kW. Less than that and it will just struggle to move around. More than that and you'd need to heavily upsize all components making the build impractical compared to an MX frame.

If you want it to go offroad, mid drive is the only option. Hubs are only OK for the pavement or gravel roads. And in that case, you'll probably want QS165 with either the surron reduction mid-hub (you can actually buy them on Aliexpress) or the built-in reduction.

* 60V is unnecessarily low. 72V controllers are cheap enough to go with 20s.

* 200mm fork travel is enough for a kids bike or for use on gravel roads. It's not enough for serious offroading for an adult. At the same time I understand that you'll be fighting for every gram. Realistically, your options are either a 200m FastAce fork (37mm) or a used KTM Freeride front-end (43mm). I know that Surron riders praise the Freeride fork as a massive upgrade; for me it was absolute garbage compared to the full-sized 300mm travel WP Xplor.

* Don't pick the shock spring rate until you know the total weight and weight distribution. Don't pick the shock length until you know what rocker system you want to use and how much progression you will need. Builds without a rocker aren't appropriate for offroad.

* 203mm rotors are fine for a bicycle. Brakes should be picked after the total weight of the vehicle is known.

So even at the component choices you have a few things to think about. You should probably decide on a weight budget ("lightweight" means absolutely nothing) and see how much you can "spend" on what parts.

 

[j bjork]

* 1-2kW motor is far too weak for this, unless you're citing the "rated" power. The actual system power you should target for this build should be around 7.5kW +/- 2.5kW. Less than that and it will just struggle to move around. More than that and you'd need to heavily upsize all components making the build impractical compared to an MX frame.

And in that case, you'll probably want QS165 with either the surron reduction mid-hub (you can actually buy them on Aliexpress) or the built-in reduction.

Don't pick the shock length until you know what rocker system you want to use and how much progression you will need. Builds without a rocker aren't appropriate for offroad.

Good points, but I think 7-10kw is still on the low side, I would want to double that.
There are several other options than QS165, maybe not as narrow though.
I'm not so sure that it is necessary to have a rocker system, KTM hasn't got it but Husqvarna does. (very similar bikes otherwise from the same manufacturer)
Those I have talked to that has had both brands thinks it doesn't matter, not sure what differences there are to the shock though.

I agree about getting someone to guide through a build like this would be difficult, and maybe not even the best option.
I think it is better to learn as much as possible, then ask specific questions.

 

[bananu7]

I would want to double that.

For a full-sized bike, sure, but for a scaled-down Surron clone I personally think that's plenty. Either way, adding power is probably the easiest thing to do in a project like this.

There are several other options than QS165

Certainly. I just think it's going to be hard to beat the price/performance ratio, plus it's well known and easy to set up, but for sure it's not the only one out there.

I'm not so sure that it is necessary to have a rocker system, KTM hasn't got it but Husqvarna does. (very similar bikes otherwise from the same manufacturer)

Yeah, I do own two PDS-equipped KTMs. The thing is that PDS is an exclusive system which is extremely hard to tune to get right. It took KTM a good while to get rid of its flaws and get it working as well as it does in the modern bikes - as evidenced by common mods and improvements that the aftermarket has made for them over the years. And of course it needs a progressive spring. The key part of it is that is different from a regular shock is that the damping force is dependent on the position of the suspension in its travel.

In comparison, a rocker system can use a cheaper, more commonly available linear spring, and the damper is actually remarkably simple in comparison. It naturally lends itself to a wider operational range of compression velocities, and allows even extremely simple designs like emulsion shocks to perform well in wider conditions, resisting bottoming while providing acceptable small bump sensitivity. Historically, when bikes transitioned from twin shocks to a monoshock, everyone knew it as "pro-link" simultaneously because that was the key part that made the suspension work so much better.

Now, I don't want to sound overly negative, but the chances of a cheap shock being tuned properly for a linear system are near zero. If it was a bike air-cushioned unit, you get a bit of spring progression, but still no real position-dependent damping. And most of the cheap ones people mount to linear-movement frames don't even use a progressive spring! Even if you make the damping extremely progressive (hard shims or even orifice-based), without position-dependent (as opposed to speed-dependent) force you'll be constrained to a really narrow operational range, which in practice limits the bike to either extremely soft for comfort but too soft for jumps/good traction, or too stiff to be comfortable.

Sorry for that unsolicited opinion segment, but I feel that for a novice frame builder/designer, assuming a vehicle meant for actual offroad use, going with a linkage is actually an easier path to get something working well, despite the extra math it requires and fabrication effort involved.

 

[Balgaroth]

For a full-sized bike, sure, but for a scaled-down Surron clone I personally think that's plenty. Either way, adding power is probably the easiest thing to do in a project like this.

Certainly. I just think it's going to be hard to beat the price/performance ratio, plus it's well known and easy to set up, but for sure it's not the only one out there.

Yeah, I do own two PDS-equipped KTMs. The thing is that PDS is an exclusive system which is extremely hard to tune to get right. It took KTM a good while to get rid of its flaws and get it working as well as it does in the modern bikes - as evidenced by common mods and improvements that the aftermarket has made for them over the years. And of course it needs a progressive spring. The key part of it is that is different from a regular shock is that the damping force is dependent on the position of the suspension in its travel.

In comparison, a rocker system can use a cheaper, more commonly available linear spring, and the damper is actually remarkably simple in comparison. It naturally lends itself to a wider operational range of compression velocities, and allows even extremely simple designs like emulsion shocks to perform well in wider conditions, resisting bottoming while providing acceptable small bump sensitivity. Historically, when bikes transitioned from twin shocks to a monoshock, everyone knew it as "pro-link" simultaneously because that was the key part that made the suspension work so much better.

Now, I don't want to sound overly negative, but the chances of a cheap shock being tuned properly for a linear system are near zero. If it was a bike air-cushioned unit, you get a bit of spring progression, but still no real position-dependent damping. And most of the cheap ones people mount to linear-movement frames don't even use a progressive spring! Even if you make the damping extremely progressive (hard shims or even orifice-based), without position-dependent (as opposed to speed-dependent) force you'll be constrained to a really narrow operational range, which in practice limits the bike to either extremely soft for comfort but too soft for jumps/good traction, or too stiff to be comfortable.

Sorry for that unsolicited opinion segment, but I feel that for a novice frame builder/designer, assuming a vehicle meant for actual offroad use, going with a linkage is actually an easier path to get something working well, despite the extra math it requires and fabrication effort involved.

PDS is simply put a shock that has 2 mid valves/piston instead of one like most shocks. It doesn't use or need a progressive coil to work properly as the lack of progression inherent to the mono-pivot design is compensated by the progressive compression damping in the shock thanks to the secondary piston.
The main advantage is that you don't have any linkage down low which is good for logs and tech stuff in general. I have a KTM, a Beta and soon a Suzuki but I can already tell you that I don't really see any real advantage between the KTM and the Beta. One thing to consider is that a suspension with no linkage is harder to make work for hard hitter on proper MX/SX tracks but otherwise it will work for most folks, given that you are using a good shock with contemporary design (basically anything younger than early 2000's for moto suspension, younger than late 2010's for Mtb suspension).
A mono-pivot suspension can be tuned to get some progression, not as much as a linkage design but you can still achieve enough for it to work without issues. If you go full size you can use a KTM PDS shock, if you go small/mid size you want a custom tune and/or a EXT shock with hydraulic bottom-out feature. If you go with a 267mm shock I think you can also go with a Rockshox Superdelux coil which also offers HBO and is significantly cheaper. It should be able to handle between 200-250mm of travel. You could go with more basic shocks but a single pivot design can definitely benefit form HBO. At the bare minimum you will want low and high speed compression adjuster, it should help finding the right shimstack faster and with less revalving.
To design a single-pivot or a linkage suspension you can use linkage, it is a software (cheap) built for Mtb suspension but it will allow you to place the pivot locations and tune the leverage curve easily. Obviously, designing a single pivot will be easier and manufacturing it too. If you want to tune your anti-squat it is also possible. I use it to design my bikes or analyse what is being released on the market but I also used it successfully to analyse some motorbike suspension.

 

[bananu7]

It doesn't use or need a progressive coil to work properly

There have been PDS models sold both with linear and progressive coils. All enduro models nowadays go out on progressive springs. It's less important that the position-dependent damping, but still plays a role.

I don't really see any real advantage between the KTM and the Beta.

And that's a testament to the work the WP engineers have done with those shocks.

If you go with a 267mm shock I think you can also go with a Rockshox Superdelux coil which also offers HBO and is significantly cheaper. It should be able to handle between 200-250mm of travel. You could go with more basic shocks but a single pivot design can definitely benefit form HBO.

Yeah, it can certainly be made to work with an appropriate shock. Using a linkage trades the shock complexity for the linkage and geometry complexity. If one feels better at tuning suspension than designing progressive geometry, I suppose it might be easier to do a linear system, but I would say some form of "HBO" is almost a necessity in that case.

 

[BiggKidd]

Hey guys!
I’m building my own DIY e-bike from scratch — frame, battery, everything — and I’m learning as I go. If you’ve built one before, I’d love to hear your tips or connect for occasional guidance. Just trying to avoid rookie mistakes and build something solid.

Thanks in advance! The bike I want to build is going to be a surron clone. Aesthetically pleasing but not as efficient... I would love to learn under whoever as far as they can help me complete this project.

I am in the same boat, sort of. While I have never built a custom bike I have built a lot of custom stuff over 40 years. I started by picking my motor a bit crazy at 24kw. Next I know I'm going more for a street ride than off road but it must be able to handle the two plus miles from home to pavement. I also know I want the ability to travel long distance - 100 miles on a charge at slow speeds, under 28 mph. I'm leaning toward a larger heavier bike with LifePO4 cells although I am not committed at this time. Right now I'm still waiting for my motor I ordered at the beginning of the month to show up. Until it actually gets here I've put a hold on ordering any more parts. Even though several pieces I ordered at the same time as the motor are already here. Personally I'm basically trying to forge a combination of a Schwinn OCC chopper and a late 80's Kawasaki 454LTD. I'm not old or young but I am disabled, old injuries and M.S. among others and have beat on my body hard for many years so comfort for my riding style is the key for my build.

The main point I want to emphasize is failure is just a step in the process of creating what YOU want! Old Edison failed ten thousand times before he got a working light and it still produced 90% heat and 10% light for the power used.

So try and try again until you achieve exactly what you want! Do NOT expect to get it right on the first try just try not to kill yourself or anyone else as you learn!

One thing I am beginning to learn is I will likely need to use a mixture of bike and motorcycle parts. You may need to as well!

Good LUCK!!!

 

[n00b]

Hey guys!
I’m building my own DIY e-bike from scratch — frame, battery, everything — and I’m learning as I go. If you’ve built one before, I’d love to hear your tips or connect for occasional guidance. Just trying to avoid rookie mistakes and build something solid.

Thanks in advance! The bike I want to build is going to be a surron clone. Aesthetically pleasing but not as efficient... I would love to learn under whoever as far as they can help me complete this project.

If you want a nice bike, start with a nice bike. I'm in the "light bicycle that propels itself and goes far enough fast enough" camp.
It has to ride like a bicycle for me, not a motorcycle. I've had both since I was 15.
I pop over curbs n stuff.
I would start with the frame and forks and make sure that fits you and then go from there.

 

[neptronix]

I'm thinking this dude would probably be happy with a cheap stealth bomber clone frame, IE:



OP, are you still reading?