VESC HW Redesign (VESC Mini?)

General topics and discussions about the VESC and its development.
Andreas
Posts: 18
Joined: 04 Feb 2016, 19:02
Location: Switzerland

VESC HW Redesign (VESC Mini?)

Postby Andreas » 04 Feb 2016, 19:44

Hello
I've been working on a redesign of your famous ESC for a little bit. Main goal is to make it smaller, and hopefully even more robust.
My idea for now is to mainly use it in small to medium sized RC projects, and maybe some robotics.
I do plan on using your firmware and config tool without too many modifications. I think the pinout of the controller should work for my layout too, but just in case I've added a pull-up resistor to PD2, so the firmware could detect what kind of board it is running on.

I've been in contact with Oskar (Madcowswe) too and i'm probably trying to help him with his project too if he thinks he can use a helping set of hands/eyes.

Anyway, without further ado, here is the link to a pdf with my current revision of the schematics
(if you see some strange yellow stuff around the components, it's a "smart-pdf", so there are parameters linked behind every component, and it seems to render these links a bit strange sometimes)

As you can see, I work wit CircuitMaker for this project, mainly to get my hands into this tool a bit more. I am an Electronics Engineer working with Altium Designer since the days of Protel 98, so it was the most obvious choice out there while still using a free tool and allow open access to all the files. Maybe i'll even try to use the collaboration features a bit more, if anyone is interested to join in on the project.

Currently I'm still wrapping my head around component placement on the board side. I'm now at a board size of 32x60mm without any tracks, but it looks like it's possible to connect everything up at that size, and maybe even shrink it a few mm more once i've routed everything...

So, if anyone has any feedback for the schematics, other ideas or faults I could/should/must look into, feel free to chime in, i'm always open to question my own sanity ... :roll:

I've attached a little image showing the current progress of the layout work. As you can see, i've changed some things around by having the power inputs at the top edge of the board. I do think this is way better to keep the left side of the board with the controller and digital interfaces a "quiet zone" since the main current does not have to run all the way across the board.

OK, now i'm looking forward for your input, and if it stays quiet around here, i'll just go on anyway and see how it transforms into the real world... I'll try to keep you updated in here with my progress since i don't have any blog or my own website.
Attachments
OPEN_BLDC_ESC.jpg
raw component placement
OPEN_BLDC_ESC.jpg (411.95 KiB) Viewed 3352 times

Jacobbloy
Posts: 7
Joined: 05 Feb 2016, 05:38

Re: VESC HW Redesign (VESC Mini?)

Postby Jacobbloy » 05 Feb 2016, 05:43

Watching this space!

benjamin
Site Admin
Posts: 280
Joined: 15 Dec 2015, 08:38
Location: Sweden
Contact:

Re: VESC HW Redesign (VESC Mini?)

Postby benjamin » 05 Feb 2016, 21:36

Nice that you are getting involved with the project!

The only thing I don't like is that you are using circuit maker, which is closed and only runs on windows. I would very much prefer to support something made with kicad. I enjoy sharing information and working with free software, which is why I spent so much time on this project while sharing everything, so supporting something made with circuit maker is something that I wouldn't like and goes against what I believe in. I hope that this is not discouraging.

Andreas
Posts: 18
Joined: 04 Feb 2016, 19:02
Location: Switzerland

Re: VESC HW Redesign (VESC Mini?)

Postby Andreas » 06 Feb 2016, 10:00

I'm sorry to hear that.
Well, I still intend to keep going with circuit maker. I could have chosen to go all in with Altium Designer, but that would have closed down access to my work even more, so I decided against that and figured CM would be the best option from my perspective. CM can be run on MAC with BootCamp and Parallels Desktop, or in a virtual Windows box on a linux machine, but I know these are just workarounds, and with Altium behind CM, Windows will always be the primary platform. Even if the software is closed source and cloud based, at least the file access is still very much open for everyone with access to some form of Windows. And I intend to share all the files (schematics, layout, BOM, production outputs) publicly too (outside the CM-Universe), I'm just not shure how exactly I want to do that in the end.
I did some trials with other design softwares a few years ago, but you can imagine that someone basically working with Altium on a professional level for a very long time, day in day out, is so much used to the workflow that changing these habits and learn another tool with all its quirks, problems, pitfalls and what not does not come easy. Basically I'm just a little bit too lazy to invest my free time into learning a completely new (from my perspective subpar/inferior) toolset and would rather dedicate that time into the actual project. That's how I decided to set my priorities...

Hopefully there are still some things out of my project that can get merged back into your work on another redesign, if you decide to respin the schematics and board design at some point. I'll happily do design reviews and give my feedback, if you want it.

barney
Posts: 22
Joined: 23 Dec 2015, 08:21

Re: VESC HW Redesign (VESC Mini?)

Postby barney » 06 Feb 2016, 10:58

Hi Andreas,

please take a look to this MOSFETs:
http://www.irf.com/product-info/datasheets/data/irf7749l2pbf.pdf

60V, 1.1mOhm, 140A, 71nC, Coolable.

Barney

Andreas
Posts: 18
Joined: 04 Feb 2016, 19:02
Location: Switzerland

Re: VESC HW Redesign (VESC Mini?)

Postby Andreas » 06 Feb 2016, 11:12

I was considering these direct-FET style packages too, but i'm not quite shure about the hidden pads under the metal can. Very hard to inspect and check for failed soldering. Usually we do run every board with BGA, QFN and similar hidden-pad packages thru x-ray inspection to make shure the soldering process was good. Thats especially important at the prototyping phase where you try to dial in your pad geometry, paste masks and reflow paramaters to minimize failures and reworks for the production runs.
I do plan to make these boards just in a simple reflow oven we have at our local lab (pizza oven with thermocouple feedback loop). Our production facility with all the bells and whistles i know from my job is in an other country, and i can't access theyr tools easily to go and do inspection runs on my private work while they run a 24/7 production line...
So, thats why i have choosen the 5x6mm QFN package, since it is a bit easyer to manage the soldering and it still offers a (limited) way to inspect all the solder joints after reflow.

barney
Posts: 22
Joined: 23 Dec 2015, 08:21

Re: VESC HW Redesign (VESC Mini?)

Postby barney » 06 Feb 2016, 11:52

Ok.
but 40V? But you have alternative up to 60V for 12S usage?

No chance to use KiCad? Nobody can use Altium for free. KiCad is easy to learn and the manual semi manual routing is very helpful.

Barney

Andreas
Posts: 18
Joined: 04 Feb 2016, 19:02
Location: Switzerland

Re: VESC HW Redesign (VESC Mini?)

Postby Andreas » 06 Feb 2016, 13:50

I basically limit my design at 30V to get some advantages in size reduction. Shure there are FETs in the QFN 5x6 package that would allow 60V Vds. If you check the hardware specs tho, a 40V FET allows to get a RdsON way lower than any 60V FET does, because there are always tradeofs in the silicon design. If we just look at IRFs FETs in the PQFN 5x6 package, the best 40V N-Channel FET (IRFH7084) has an RdsON spec of just 1.25mOhms, while the best 60V rated FET (IRFH5006) has an RdsON of 4.1mOhm.
But there are other components to look at too. If you want to go to 12s Lipo (50.4V) every capacitor has to be at least 63V or better 100V rated, and the nature of capacitors is that they get bigger with higher voltage ratings, and less efficient (internal losses).
As i initially said, my main goal is to use the ESC in smaller to medium sized RC projects (mostly 1:10 scale), and i don't plan to use this in a high power skateboard, ebike or anything that has the weigth capacity and motor size to be able to run a 12s pack, i'm more aiming at 6s and lower cell count.
There are still more trade offs if you design for higher max. voltages, so thats why i decided to go with that target i'm aiming at right now.

No chance to use KiCad, well, that would be a strech to nail me down on that, but currently, for this project, i do not plan to switch over to another software, especially at the stage the project is at right now, in the middle of layouting. Maybe for another project, i could try to wrap my rusty mind around another tool... who knows.

I agree with you that Altium Designer is not free, but Circuit Maker (from Altium) is. (well, if you don't count the windows license to run it properly at least).
You can go check it out at https://circuitmaker.com/
It's basically the full Altium Designer core, reduced to limit it's functionality and productivity by slapping on another user interface with "ribbons", removing a few of the features (i miss :roll: ) from the full package and aparently somehow slowing the program down the more complex your design gets (they still want to make money with the full AD licenses of course, CM is for the maker audiences doing smaller projects). Other than that, there are not any hard limits as to what you can do with circuit maker. So, no board size, pin/pad count, layer count or other restrictions that you would not have in the full AD.

tom frisch
Posts: 2
Joined: 08 Feb 2016, 15:40
Location: Portland

Re: VESC HW Redesign (VESC Mini?)

Postby tom frisch » 09 Feb 2016, 12:18

Very interested in your progress. I'm also hoping to use in small robotics.

Keep up the good work!

Andreas
Posts: 18
Joined: 04 Feb 2016, 19:02
Location: Switzerland

Re: VESC HW Redesign (VESC Mini?)

Postby Andreas » 09 Feb 2016, 19:39

VESC_Mini_V2.png
Progress-Update
VESC_Mini_V2.png (329.16 KiB) Viewed 3184 times


Here is a little progress-update. The left side is basically connected up, but not very much optimised yet. The big thing to do now is finishing all the connections between FETs and driver. And then do another optimization. I want to move the driver around a bit, more up and to the left, but now that I know how much space is needed for the controller, all the interfaces and the interconnect to the driver, that should not be too hard to do.


Return to “General”

Who is online

Users browsing this forum: No registered users and 3 guests