Electric Roadster

[amberwolf]

If you are using platsil (platinum cure silicone) or tin cure, and something like the silc-pig pigments SmoothOn sells, you only need a *teensy tiny itsy bitsy* amount of pigment to make a very solidly colored item.

If you use more than you need to...you get pigment leakage at the surfaces.

However...under weathering conditions, even with "just right", some of it (along with silicone oils that are in the silicone itself to make it soft) can slowly "leak" out, and may end up on your clothing if you don't cover the silicone with some sort of cloth, etc. I would recommend a simple top-wrap of spandex that tucks down into the seat frame under the silicone pad, since it will be able to be stretched over the pad to form fit it.


Also, to be sure you are thoroughly mixing everything, first put just a teensy bit (I mean, really teensy!) of your pigment into part A (or b, but only one!). Then stir that up thoroughly by itself to be sure it's the kind of solid color you really want making sure you scrape the sides and bottom of container. If it's not solid enough, add a *teensy bit* more, and restir, making sure you scrape the sides and bottom of container. Then repeat until it's as solid as you want it to be.

Only when that is done, and it is consistent in color, then mix that with your other silicone part b (or a), and mix until it all looks the same color and solidity, again making sure you scrape the sides and bottom of container.. Then you know it's all mixed thorougly.

This works better if you have an end color that is made by mixing two other colors, because then you can put the right amount of each into part a or b respectively, and only when they are correctly mixed at the end will the right color show up--if it's not the right color it's not mixed well enough. But this takes experimentation and good measuring of color amounts vs material amounts, or experience with color mixing, etc., to get the exact shade you want if that matters for the application.

But it works ok even with just one color in one of the two parts.

 

[amberwolf]

Oh, also, do you plan to only park on solid pavement?

I ask because non-footed kickstands I've seen (and used) never seem to work well with anything that's not solid (even hardpacked dirt&gravel, if wet from rain, etc., lets the "pointy end" stands just sink right in and the bike sometimes to fall over if it goes down far enough.

So I've always either added a "wide" flat foot on the end of my storebought kickstands or built them with them on there in the first place when I made them myself, for anything heavier than just a plain bicycle (with no way to put anything on it except when riding).

 

[Dui ni shuo de dui]

If you are using platsil (platinum cure silicone) or tin cure, and something like the silc-pig pigments SmoothOn sells, you only need a *teensy tiny itsy bitsy* amount of pigment to make a very solidly colored item.

Thanks for the tips
I'll use these for the next saddle design.

Oh, also, do you plan to only park on solid pavement?

Yep. I live in shanghai, so the chances of parking into mud or gravel ar close to zero.
And if really this situation happens, you can always find some flat stone somewhere to put under.
I thought about making the shoe larger but there would be no point in doing it in my particular case, plus I prefer this look.

But I can 3D print any shape I want, so I could replace the current one in a few minutes if needs be.

 

[JimVonBaden]

That is looking outstanding. Happy it works now.

My next step is calibrating the controller, and frankly I am nervous. Did you use full battery power to power the controller for calibration? I wish there were a video showing the process, but I can't find one, just the manual to do it.

 

[Dui ni shuo de dui]

That is looking outstanding. Happy it works now.

My next step is calibrating the controller, and frankly I am nervous. Did you use full battery power to power the controller for calibration? I wish there were a video showing the process, but I can't find one, just the manual to do it.

Yeah, the manual explains everything but you have to read it very, very carefully. And being fluent in chinglish really helps too

To get the wheel spinning, here's what I've done:
IMPORTANT: I'm using a KLS72701 controller, so it is possible that the procedure is slightly different on yours. Read the manual for your controller carefully, if it says anything differently than what I just wrote, then refer on your manual_

Connected the following things on the controller:
-throttle
-hall sensor
-motor phase wires
-power on wire (coming directly from the battery positive terminal or from your key contactor, I forgot how you wired it on the vespa)
-Main power from the battery, positive and negative

That's all. Didn't connect anything else at this point to limit the unknowns, so that's the strict necessary.

Then, lift the rear wheel before you power on the controller, you never know what can happen, you don't want the bike to suddently go full speed (almost never happens, but better safe than sorry)

So, next step is to download the software, for that you go to the kelly website, then support section and you choose the app corresponding to your controller. I suggest you to use the desktop app at first instead of the bluetooth, you can always use the bluetooth android app later.

Once you downloaded the software, extract it somewhere on your hard disk drive.

Then, connect the RS232 cable to your controller, and the usb end to an usb port on your computer.
Power on the controller (so to answer clearly your question, yes, it needs to be connected to the battery at that point and everything should be connected as I instructed before. Very important: never touch the throttle when your controller is connected to the PC or you'll destroy the controller. The Kelly manual repeats this thing over and over again so I guess this is really sensitive. The motor cannot spin under its own power during the programming.

Once you plug the controller into your PC, wait a little bit for the drivers to get installed. In my case, they refused to install themselves the first time, so what I did was simply to unplug the USB cable and then plug it on a different usb port on the same computer. This time it worked. Don't ask me why, it's magic.

Once the driver is installed, then you can open the app.

The first thing you'll have to do is to make the self learning test, where the controller will determine the hall sensors orientation. The manual is quite confusing and the app is also quite weird, but basically you just do the followings:
-Switch off your brain, just follow the steps
-Click on "vehicle" at the very bottom of the window
-Locate the line "Identification angle"
-There should be a number in it, probably 85
-Change this number to 170. Don't ask me why 170, it doesn't make any sense but apparently that's the magic number to get this whole process starting.
-Click on write and then do not touch anything
-I forgot if I had to restart the controller or not at this point, so if nothing happens after ten seconds, then restart the controller
-wait for another ten seconds, if I remember correctly the motor will make some weird click noises, move slightly a few times, then the controller will beep once finished.
-Restart the controller again
-Now it should display 85 again at the line "identification angle"

After that, I just restarted the controller, closed the app and disconnected the usb cable.
I twisted the throttle and the motor was spinning.

Let me know if you have any problem, I'll try to help. But basiclly, read the manual with patience and without rushing and you should be good to go.
Took me maybe 20 minutes to figure out so that's intimidating for sure but that's really no rocket science

 

[JimVonBaden]

Awesome! Thanks for taking the time to outline the steps. The manual is definitely not perfectly clear.

I will let you know if I have issues, and how it works out!

Thanks again!

 

[Dui ni shuo de dui]

My pleasure, hope it works well for you!

Not done much this weekend, basically I just made a few rides to check that everything is fine on the powertrain.
Tested some settings on the controller on the road.
Since I still don't have mechanical brakes, I enabled the slide regen to at least be able to slow down. I tested the controller up to 50% of its maximum power and it felt almost as powerful as my other motorbike already, so I think that is going to be really impressive once finished, if the motor can take it!

Took some pics outside:



The first saddle I made using the silicone clearly wasn't comfortable enough, so I'll make another one, a bit thicker with a slightly different shape.
Aside from that so far everything's fine, the frame and the powertrain are done!

 

[SlowCo]

Nice :thumb:

 

[Dui ni shuo de dui]

Thanks SlowCo!

A little side quest yesterday: I designed and printed a little enclosure for the BMS screen.
Nothing really fancy, I only plan to use it while tuning the controller and testing the bike, the thing can be removed or put back on the bike in just a few seconds.






I had a few short accelerations just to see if the thing was working fine and briefly saw a 200A on the display... that's already what my other bike was pulling at it maximum but this controller is only set up to 50% right now... Can't wait to test it further!

 

[mannydantyla]

Kick ass!

And thank you so much for the Kelly setup advice! I'll have to order the usb connector. They gave me two Bluetooth connectors (one from QS and one from Kelly) but not USB connector. I'll also have to find a Windows laptop unless I want to drag my desktop PC to the garage, or install windows on my macbook. My wife has a Windows 9 or 10 laptop but that won't work I guess. And of course I could use the Android app with one of the bluetooth connectors but from what I've seen, it's even more confusing. I saw a youtube video of someone tying to use it and connect to the controller and it looked like a big headache.

Here's the advide I copy+pasted into my notes about using the bluetooth app:

1. Wire battery, ensure 11V across the red/black [i.e. +V/GND].
2. Plug in bluetooth adapter and pair like you would any Bluetooth device [PIN is 1234].
3. Open Kelly AC Aduser App on Android phone.
4. Tap Bluetooth COM [tap Yes in pop-up].
5. Tap OP at bottom [tap device in pop-up].
6. Tap R....On motorcycle [wired with throttle as per Kelly pinout], try moving the throttle and view the TPS Pedel value on AC Monitor tab [i.e. it should change values from 44 to 200ish]...."I now have to figure out why the motor isn't moving when operating the throttle."

 

[Elektrosherpa]

A little side quest yesterday: I designed and printed a little enclosure for the BMS screen.
Nothing really fancy, ...

If this is "nothing really fancy", I am curious to see what you call "really fancy"
I find it quite impressive.

 

[Elektrosherpa]

Took some pics outside:
....
The first saddle I made using the silicone clearly wasn't comfortable enough, so I'll make another one, a bit thicker with a slightly different shape.
Aside from that so far everything's fine, the frame and the powertrain are done!

Looks really awesome.
I wish I were already so far with my project

As for the saddle:
I was sure about that when I saw the pics.
In German we call such saddles "Hämorrhoidenspalter"
I leave it to your imagination what that means

 

[Dui ni shuo de dui]

Haha, yeah, hopefully I'll find a way to make the saddle both small and comfortable, pretty sure it's doable, my new saddle design is already much better even though it's hard plastic right now.

Anyway, I printed the left side battery cover yesterday and installed it on the bike to see how it fits:



I think it looks really nice, so I'm printing the other side right now. It took 20 hours to print and 1Kg of PLA filament. Keep in mind these are temporary, the final version will be made out of cabon, so I don't really care if these are not UV resistant, it doesn't really matter, they are just for checking the dimensions and actual looks before creating the molds (but I'll probably use them for a few months, during the time I'm making the carbon covers).
The 3D printed covers are surprizingly tough, these things aren't easy to break.

So the next step is to wait for the second cover to finish printing (hopefully this evening), then create some brackets to attach the covers to the frame.

Once this will be done, I'll start working on the molds, but in the meantime I'll sand the 3D printed covers smooth and give them a nice coat of paint.
Also, once the brackets will be done I'll be able to start painting the frame. Problem is, I can't find the paint I want so far... I'm searching for a nice Ferrari metallic red. The paint I've purchased was supposed to be red pearl, but it is way too orange for my taste so I have to keep searching, I made a few sample tests on some scrap tubes and it wasn't looking like I was expecting.

 

[amberwolf]

For some metallic colors and looks, you may get a better result using the "candy" method, of using a metallic undercoat, or even white, and then using a transparent "lacquer" color (with or without metallic flakes in it, depending on result desired). For some deep colors it may require multiple layers of the transparent color.

Also, this page may be useful:
https://rossoautomobili.com/blogs/magazine/15-shades-of-ferrari-red

And this forum has a few discussions (haven't read any of them)
https://www.google.com/search?q=Ferrari+metallic+red+site:www.ferrarichat.com

 

[Dui ni shuo de dui]

For some metallic colors and looks, you may get a better result using the "candy" method, of using a metallic undercoat, or even white, and then using a transparent "lacquer" color (with or without metallic flakes in it, depending on result desired). For some deep colors it may require multiple layers of the transparent color.

Yeah but I'd really like to avoid doing that, I'm quite bad at painting usually, so the easier the better.
But if I find no other alternative, then I'll try.
Thanks for the suggestions though :wink:
I do know the color code I want to use, but the difficulty is to find a company that agrees to make it. Especially since I won't need a lot, half a liter will already be enough to paint 3 bikes I guess :lol:

Anyway, while I keep looking for paint, I printed the second cover. So for the first time the battery is fully enclosed!
Now I'll have to come up with a bracket system.

Just finished printing:



And after removing the supports, installed on the bike:

 

[Elektrosherpa]

Looks awesome.

Too bad you are in Shanghai....

I am imagining some kind of battery cover for my Bultaco, too -
in my mind it would be something like what you made there,
just continuing the form of the former gas tank downwards.



(rough sketch)

 

[Dui ni shuo de dui]

I am imagining some kind of battery cover for my Bultaco, too -
in my mind it would be something like what you made there,
just continuing the form of the former gas tank downwards.

Nice! Don't hesitate to get creative on the battery box, it's difficult to make it look nice but it's important since that's the most visible part.

On my side, I made a few progress:
-I designed a very simple system to keep the covers in place, it seems to work well. Only three screws, barely visible.

-Started smoothing the panel with bondo and primer, mostly for fun until I start working on the carbon parts.
Applying the bondo:



Ater a few coats of primer:



So now at least I know that the battery covers work and won't need a redesign. I can work on making the molds.
But first thing I'll do will be to finish the frame, so some more bondo, more sanding, coats of primer and then some nice coats of red.

I also have to start working on the lights, especially the rear one.

 

[mannydantyla]

That looks really good!!

 

[amberwolf]

Once painted, red lights recessed away from but pointing out of the "vents" in the side covers would make the bike more visible at night, and more interesting to look at. These could be lit all the time, or only when braking or turn-signalling, for example, as extra side-viewable indications (not really rear visible due to your legs). If used for indications, they could be left on all the time but brightened for braking and blinked *off* for turning.


Also, I've seen some "stylish" signal-cluster lights with large surface areas on aliexpress and the like lately, though I have no idea how bright they actually are. I have been considering these designs (or something similar I'll have to find the links for) for the back of SB Cruiser's successor, and/or CrazyBike2's resurrection:
https://www.aliexpress.com/item/4000790034144.html?spm=a2g0s.8937460.0.0.227a2e0efX4fNa

They could be moutned vertically at the back of your casing under the seat, and their light would not only be rear-visible, it would also light up the whole rear wheel/frame, making the bike more visible from the side as well as other non-directly-rear angles. They can also be mounted vertically on the forks below the top of the tire, to similarly light the front wheel up, etc.

 

[Timvmax]

Loving this build! I'm a complete newbie, been a "died in the wool" petrolhead for 40+ years but after riding an electric off roader I really want to build something.
Thanks for explaining everything so well, there doesn't seem to be much good information out there, your thread is refreshing.
I don't 3D print but I can draw in CAD and have a CNC mill , CNC Router and manuallathe, so I can make a lot of what I need.
For your info, the fork parts are called the top yoke and bottom yoke
Good luck with the build, I'll be following

 

[Dui ni shuo de dui]

Once painted, red lights recessed away from but pointing out of the "vents" in the side covers would make the bike more visible at night, and more interesting to look at. These could be lit all the time, or only when braking or turn-signalling, for example, as extra side-viewable indications (not really rear visible due to your legs). If used for indications, they could be left on all the time but brightened for braking and blinked *off* for turning.

Thanks for the suggestions, I quite like the idea of intalling some leds inside the battery cover to make the vents more visible, I'll give it a try once the build will be finished, haven't thought of that.

For the tail light, I'll install it right inside the square tube that supports the saddle. I'll custom build something from scratch, using LEDs, 3D printing and colored epoxy. I think it'll look good, but I'm trying to find an easy way to actually build the thing and make it look clean.

I haven't started thinking about the design of the front light, I'll try my best to keep it very discreet with the smallest possible footprint.
I won't install turn signals because I never use them anyway, and I'll try not to install anything that isn't absolutely necessary for driving, not only for weight, but also to reduce the amoung of ugly wiring going everywhere, and also avoid having millions of switches on the handle bars. In the end, after the bike is properly tuned, even the dashboard display screen will be removed.

 

[JimVonBaden]

Looking better and better!

 

[Elektrosherpa]

For the tail light, I'll install it right inside the square tube that supports the saddle. I'll custom build something from scratch, using LEDs, 3D printing and colored epoxy. I think it'll look good, but I'm trying to find an easy way to actually build the thing and make it look clean.

I was expecting you to do this

I won't install turn signals because I never use them anyway

This sounds kind of weird to me:

Me too, I'm not going to install turn signals-
but I am driving 90% in the French countryside on agricultural gravel roads with absolutely no traffic.

If I remember right, you said you are using the bike for riding to your work in a city.
Don't you think it's quite dangerous driving in city traffic without turn signals?

 

[Chalo]

Don't you think it's quite dangerous driving in city traffic without turn signals?

 

[Elektrosherpa]

Don't you think it's quite dangerous driving in city traffic without turn signals?

LOL
If you would use theses signals around here where I live, you wouldn't survive more than 10 minutes in city traffic
And my wife just said, in Russia you wouldn't even survive ONE minute