Speed Bomber

chuttney1

100 W
Joined
Apr 3, 2013
Messages
243
Location
Los Angeles, CA
tumblr_n8ddueobMY1tnqm0wo1_1280.jpg

tumblr_n8ddueobMY1tnqm0wo3_1280.jpg

tumblr_n8ddueobMY1tnqm0wo2_1280.jpg


Im making a boat. Its actually a case for my electric skateboard to contain the batteries and speed controller. It is not as flat as I wanted because 6S1P lipo battery are a brick. The plan is to make a fiberglass mold. I had foam blocks sitting around since my brother didn't use it. The design was to be as aerodynamically as possible with inspiration from automotive design for the use of side vents and a rear diffuser. The purpose of the side vents is to provide cooling for the speed controller. Now to figure out how to put a easy-to-open door in the case.
 
Nice side vents. It's a neat idea except everything in there is going to get very dirty and I mean dirty especially if you are going through water. I find the best is to keep specifically only the esc part that needs air flow unless it doesn't..

I've been thinking of making a laser printer enclosure out of wood and waterproof'ing it. Making an easily open lid with a latch or something similar.
 
That's going to look pretty cool. I do agree that things might get dirty. I made a wood enclosure for one of my boards too and haven't been able to figure out a decent door solution...
in deck 3  (2).jpg
 
it seems either a small pineapple or a big enclosure... FYI all my electronics are fitted in a 0.75l box
as Torqueboard mentionned, only the ESC potentially requires cooling
 
Yeah, I've actually found my 120Amp ESC's usually don't need any cooling but if you can add it in it's better to add it in. They should be fine without airflow. I doubt it get's much anyways being on the bottom of the board anyhow.

I've been thinking of a laser printed wood enclosure and gluing things together. I like the idea of an enclosure that's built into the board somewhat and has a flange area where a piece of wood as a cover attachs to the flange on the inside of the board possibly velcro and maybe 1 small finger hole to rip the top off.

Been stumped on a nice cover though for a while not. Can't decide lol.
 
The dirt issue, I plan to cover it with some kind of screen mesh or something. My first ESC case was all boxy and had little side vent for air. I could never tell what was happening and always thought the ESC overheated or something. It does have built in overheat protection.

@made_in_the_alps_legacy: the enclosure is too large for sure. Its 19"L x 5.75"W x 2.75"H
 
psychotiller said:
That's going to look pretty cool. I do agree that things might get dirty. I made a wood enclosure for one of my boards too and haven't been able to figure out a decent door solution...
Nice build! could you please tell me what trucks they are? Look very cool! Thanks
 
Instead of fiberglass, it might be easier to just make a simple vacuum former. There's less finishing needed and all you really need is a piece of MDF and a heater.

Also, nice pineapple.
 
@carbon, totally would do it, but I don't know to build a vacuum table so im stuck using fiberglass. My past experience with fiberglass is it is really tricky to work with for beginners and if you have access to vacuum forming stuff with fiberglass or plastic do it. Makes the final product look much cleaner.
 
chuttney1 said:
@carbon, totally would do it, but I don't know to build a vacuum table so im stuck using fiberglass. My past experience with fiberglass is it is really tricky to work with for beginners and if you have access to vacuum forming stuff with fiberglass or plastic do it. Makes the final product look much cleaner.

You could use a heatgun and an abs sheet and heat every inch and bend around a mold.
 
tumblr_n8km7m2gXH1tnqm0wo1_500.jpg


I pulled the mold apart from the foam and it turned out okay. A slight bit of fixing since some parts did not go well. Overall, not bad. Too bad the foam core had to be cut apart because it was hard to get out. Now to figure out what to do from here and use vacuum forming or make a case using carbon fiber and also how to put a door in this.
 
From this day forward, I rename my eboard build to The Speed Bomber. Since I am now on summer break, its about times to get back.

The next two photos are of a stencil of the VESC for solder paste.

MhFqpLJl.jpg

u1APVfml.jpg


This is a photo of the stencil over the VESC. I don't have a boom microscope so I have to use a digital camera or my ipad mini's camera to do visual inspection. Gotta work with the stuff I have instead of forking $200 for a boom microscope.

b5nSgaXl.jpg
 
As an update, I have finished solder two VESC boards. You know the happiest moment when the LEDs on the VESC light up during the uploading of the firmware works. Next step is to get some bullet and XT60 conectors to finish the wiring and test it with my 270 Kv motor I currently have at the moment before storage and another long hiatus of planning for the batteries.


Some pics of the process. I had no microscope to use other than an iPad and iPhone 6+ for the camera as visual inspection.

I used Kester R276 no clean solder paste with a stencil from OSH Stencils.
uAcBjkjl.jpg


Decided to skip the stencil after a few minutes finding it cumbersome and apply everything by hand.
6ECEDj4l.jpg


Built a toaster oven for this project, but the controller acted on me.
gvLrMqul.jpg


Used a hot air gun. Works well, not without the issue to burning the board. Did encounter some issue such as "tomb stoning" with some resistors due to incorrect amount of solder paste used on the pads.
puyPwERl.jpg


pic of the controller out of the packaging
H4Z70fgl.jpg


Visual inspection of the brushless controller after soldering with hot air gun
1HE5TLml.jpg
 
torqueboards said:
Nice work! Where did you get your stencil made?

How long did that take you? Looks fairly quick with that hot air gun?

I got the stencils from OSH Stencils. They have the choice for either 3 or 5 mil thickness in laser cut Kapton tape. From start to finish, 14 hours for two boards. Only because I did it in stages to get a feel for the process. I have no pick and place machine so it added time. Hot air gun is quick once solder melts with a few seconds. I estimate it would probably be closer to 2-3 hours per board assuming no problems with the application of paste solder and everything went well. I did forgot to mention I used the leaded solder paste because I fear I'm going to burn my stuff. The lower melting point of leaded solder makes it slightly easier.
 
The past few weeks has been demoralizing for me in my quest to redo my eboard build before going back to school. The past few times the VESCs I have soldered work, but encountered the DRV8302 error. Every time I try to replace the DRV8302, one solder trace lifts off from the board forcing me to solder a new one. Most recently, I have encountered the openocd error trying to upload the firmware to the VESC and have asked Vedder for help. My first stm32f4discovery board had the MCU flashed with the VESC firmware when I did not remove the CN3 jumpers. I thought this was a problem, but was not the root cause of the openocd error. I have gone on STM's website forum for help in looking for an answer. What I found is the openocd error can be cause by a hardware, USB drivers issue, or relabeling the SWD ports as GPIO on the Discovery board. I decided to get another stm32f4discovery board in case the first one I have is bricked but highly unlikely. I have eliminate probable sources of the error relating to the USB driver as I am able to connect to the Discovery MCU when writing the openocd script. This leaves me to believe the real root of the cause of the openocd is hardware related to the something on the VESC I have soldered. The voltage regulators do work as I have set up a systematic process to ensure one component is not mixed match with another during the placing of SMD parts on the VESC before hot air soldering. At this point after having 3 VESC I have soldered not work while reusing some of the old components from previous sudo-failed boards and from a financial perspective, Im just going to cut my loss and get a fully working/tested one.
 
chuttney1 said:
The past few weeks has been demoralizing for me in my quest to redo my eboard build before going back to school. The past few times the VESCs I have soldered work, but encountered the DRV8302 error. Every time I try to replace the DRV8302, one solder trace lifts off from the board forcing me to solder a new one. Most recently, I have encountered the openocd error trying to upload the firmware to the VESC and have asked Vedder for help. My first stm32f4discovery board had the MCU flashed with the VESC firmware when I did not remove the CN3 jumpers. I thought this was a problem, but was not the root cause of the openocd error. I have gone on STM's website forum for help in looking for an answer. What I found is the openocd error can be cause by a hardware, USB drivers issue, or relabeling the SWD ports as GPIO on the Discovery board. I decided to get another stm32f4discovery board in case the first one I have is bricked but highly unlikely. I have eliminate probable sources of the error relating to the USB driver as I am able to connect to the Discovery MCU when writing the openocd script. This leaves me to believe the real root of the cause of the openocd is hardware related to the something on the VESC I have soldered. The voltage regulators do work as I have set up a systematic process to ensure one component is not mixed match with another during the placing of SMD parts on the VESC before hot air soldering. At this point after having 3 VESC I have soldered not work while reusing some of the old components from previous sudo-failed boards and from a financial perspective, Im just going to cut my loss and get a fully working/tested one.


OpenOCD error is possibly due to you need to install what is called "synaptic package manager" on ubuntu in order to install openocd dependency prior to loading the firmware.... I noticed that after installing synaptic, then installing OpenOCD from there it worked. This is something I have tried to bring to Vedders attention as I believe the guide may need to be modified for fresh installs of ubuntu.

you can download synaptic package manager by finding this in the ubuntu application manager.


You can see the install instructions for openocd here: https://help.ubuntu.com/community/OpenOCD

it clearly states int the guide "Installation Instructions
Look for the package named openocd in the Synaptic Package Manager and install it."
Good luck.
 
jamesonotc said:
OpenOCD error is possibly due to you need to install what is called "synaptic package manager" on ubuntu in order to install openocd dependency prior to loading the firmware.... I noticed that after installing synaptic, then installing OpenOCD from there it worked. This is something I have tried to bring to Vedders attention as I believe the guide may need to be modified for fresh installs of ubuntu.

you can download synaptic package manager by finding this in the ubuntu application manager.


You can see the install instructions for openocd here: https://help.ubuntu.com/community/OpenOCD

it clearly states int the guide "Installation Instructions
Look for the package named openocd in the Synaptic Package Manager and install it."
Good luck.

I don't think it would be necessary to do this method as Vedder does address the stm32f4discovery udev rules with the file he has. I did check the code and it seems fine. My openocd error is now a problem with transport, but Im not sure what is really wrong if everything else is alright. I am using VMware on windows 7 64-bit
 
Back
Top