Paul & Sabrina's cheap DIY 144v motor controller

[thingstodo]

I know that I found a couple of references, way back in the thread, for adopting mosfets. I can't locate any of them.

I'd appreciate a current link.

Are the price points based on quantity, or is this for the initial order to test out the design?

[MPaulHolmes]

Hello! I'm not taking donations anymore. I'm too busy right now teaching and totally over my head working from home on stuff that I have no business working on for not very much money(as an independent contractor for a company in oregon), and the occasional kit. There haven't been any large group buys to really run prices down I think because the market just isn't very big. Also, it's relatively untested (compared to a commercial product) and not that many people want to risk setting hundreds of dollars on fire.

[sawickm]

Originally Posted by MPaulHolmes

I'm too busy right now teaching and totally over my head working from home on stuff that I have no business working on for not very much money, and the occasional kit. There haven't been any large group buys to really run prices down I think because the market just isn't very big.

Hey Paul,

Don’t throw in the towel yet !!! Hang in there !!! you just have to much going at once!

like you said “the market is small”, but in my opinion the DIY interest is still out there !!!.

I think you have reached the limit for the interest, and need for your “Rev C” controller kit. It was a good first step in proving your ideas, but time to move on ….

When your life slows down my advice would be for you to take the next step, focus on your new projects. Your new Universal AC/SR/DC controller project is an excellent idea, it would be very sad to see that idea die. Your time would be best spent on the firmware development for AC/SR/DC Universal Controller since you have most of the work on hardware already done.

The Wiki now has most of the information on the original and new ReVolt projects for anyone to make their own PCB’s or start a group purchase. Those interested in this forum should also contribute their expertise to ReVolt projects and keep the ReVolt forum alive, Paul can't do it all.

Again, just my opinion.

-Mark

[sawickm]

Originally Posted by MPaulHolmes

I think a Wiki failure page is a great idea!

A Wiki "Open ReVolt Problems/Failures" section was added to the Wiki.

Open ReVolt - EcoModder

-Mark

[MPaulHolmes]

Those sound like really good ideas, Louie. I was going to do another board like you were saying that would be a little easier to solder for people. Actually, everything you were suggesting was also in my head! The multiple driver interfaces is a really good idea.

[flores]

One idea I had was to use a netduino or arduino for the PWM part, because it is far simpler to program and debug (netduino is, arduino a little less) and cheap. We could design a sort of driver shield, that would connect to a netduino and drive an igbt or mosfet power section.. I'm trying this at the moment.

[adamj12b]

Lou,

You have some good ideas.

Originally Posted by LouMan

Everybody get together and put idea's on the table for a new and exciting ver. E one that is newbie friendly, full of new features, maybe a new pic or arm controller with a grafixs output screen with controller data would be neat, the explorer program shows raw data about the controllers workings but how about a advance screen that shows human data form in degrees and graphs that is understandable without mental recalculations about what the raw data means?

There is 3 things about this that concern me. First, changing the main processor would basically make a totally different controller. If it gets moved to a dsPIC chip, its atleast close and the code can be transposed pretty easily. Second, I would consider the contoller to handle graphics display to be a bad idea. It will put unecessary strain on the porcessor and could detract from what its main job is.... controlling the motor. The best idea here would be to build a serial interface display that reads the serial data and prints it on the graphics display of choice. Third, When you say "the explorer program" are you talking about RTD Explorer? If so, It already converts all temps to C and F in the lower right cornor. It is not graphed because a line of 25C on a scale of 0-510 would be so lost. lol If its a big request though, it could easily be added. We have a V.1.0 almost ready for release.

Originally Posted by LouMan

Look at the Commercial controllers and see what exiting options they have that would be neat to incorperate in the new controller.

Actually, The only option that the Revolt controllers dont have that the big guys do is direct battery amperage measurement, battery voltage and motor voltage. Really, all that needs to be added to make up those 3 is battery voltage measurement. The controller does a very good job of calculating battery amps. I tested this by setting battery current limit to 100A and put a load on it. The controller held the battery current to 100.1A measured with my Fluke 337 clamp meter. Other then that, the controller just needs some inputs and outputs for motor speed and tach drive. Also, contactor drives could be added for series/parallel switching, but I dont think it would be used much in the DIY area. Ive only heard of one application so far and thats Jack's bike..... But hes a little crazy.

Originally Posted by LouMan

But foremost build the board were it would be fun to work on, newbie friendly, well labeled, and for god sake no smd parts no need in taking the fun out of electronics.
The revolt projects are open source for anybody to build apon, change, improve apon, so make it were resordering a part due to revisions doesn't make you worrie about sorder bridges or trying to find a sordering iron small enough ( MAKE IT OPEN SOURCE ).

No SMD?! LOL I actually find surface mount more fun then through hole. Designs are more elegant and usually easier to layout and smaller boards. But I guess thats why im going to school to be an EE.

I like the idea of grouping the parts into areas and surrounding them with silk screen box's with lables. A few test pads could be added to each group for checking. Known test values for these pads added to the wiki. It woud greatly increase the ease of testing.

Originally Posted by LouMan

One option I would like to see on a new board is the driver section, make it were you can choose to put parts on to drive mosfet's or igbt's with the powerex drivers chips, paul just came up with a copy of the powerex driver board but how about adding the traces to the new ver.E board and traces for the mosfet driver,
poeple came populate the area of the board for what there going to be driving mosfets or igbt's.
board makers charge for the size of the board not the amount of traces so I think it would be a great feature and exciting.

As for the driver section, I would do it a little different. Board cost can sky rocket with board size, and unless huge qunatities are orders, boards a re expensive. I would design the board with either pin headers for plug on driver modules or connectors for cabeling to go to remote driver boards. This is pretty much the best idea fo IGBT's. You want to mount and connect the driver as close as possible to the modules. Thats the reason for the Powerex board. Its is designed to mount right on the gate drive pins of the module. This is the shortest gate trace possible.

I have a few projects up my sleeve that I think I should start talking about soon. These range from IGBT controllers Charger controllers, and alot of stuff in-between.

-Adam

[chapper]

Give Lou a Gold star.

[Greg Fordyce]

One idea I would like to see is a hardware and software shut down when the throttle is released. Most throttle pedals have a microswitch that operates when the throttle is released. It would be nice to connect this to the controller so that the micro knows when the throttle is released and also have this same switch disable the output drivers as well. Just extra safety, and should be fairly easy to do.

[Tweety]

Soldering SMD's by hand is easy... I do that all the time... But yeah, I know it's not for everyone...

But on the driver part... I think it's an excellent idea to make it a port out to a daughtercard...

LouMan, vibration isn't an issue, just stack the cards on top of each other with spacerbolts and a pinheader... I have done that several times in applications for dragsters/funnycars and similar... Then you can talk about vibrations, and it's completely problem free since the cards go dangling in sync...

I agree with adamj12b, a serial interface to plug in an external "screen driver" would be a much better idea... Why not use the daughterboard idea here too? Make a second port for a small daughther board that reads the serial data and reconfigures into whatever graphs you would like and then sticks it on a display? Much better than loading down the main CPU with that task...