Paul & Sabrina's cheap DIY 144v motor controller

[wjdennis]

I found that 750 Ohms made quite a slow pre-charge time, IMHO. I switched to using a rugged 75-watt incandescent bulb. Incandescent bulbs have a fairly constant current rate. I placed the bulb in the dash. It flashes then dims in about 2 seconds. Then I know the pre-charge is complete.

Bill

[MPaulHolmes]

Numerous people I read about on the EVDL use a 100watt lightbulb, which allows the capacitors to charge in a couple seconds.

[MPaulHolmes]

Hey, dennis, that's just what Lee Hart said from EVTech.

He also said that the resistor value was chosen by curtis not to protect the capacitors, but because 750 Ohms is cheap and small. You can go much smaller Ohms if you have a safe wattage rating.

[GrahamMc]

Hello. I am new to Ecomodder. I am interested in converting a car to electricity, but the cost of the parts is prohibitively expensive. I just do not understand how they can charge $1600.00 for a 144V 500 amp controller. I have been looking around on-line for several days to see if there was a kit I could buy and assemble my own when I stumbled on this thread. I am about 8 pages into it, but I am wondering if there is a summary of the components you used for the car version and the wiring schematic somewhere? Some kind of quick reference here is how I did it and why page? I am very impressed with your undertaking, and would love to build on your successes.

[GrahamMc]

Hey I hope that doesn't sound demanding. If you would like I could help write a How-To manual or something once it is perfected? Just a thought. Nice to meet you.

[MPaulHolmes]

Hello! I've been working on a parts list and directions. Well, I've been thinking about doing it. hehe.

Here's the problem: There are ways to do it cheaper. I only just finished version 0.1. It seems to work really well, but it's not perfect. Some of the components in version 0.1 are more expensive than others that would also work, I think. How about a parts list for version 0.1, even though there will be improvements?

There would also need to be directions on the process. There are some steps that could be improved. I'm not sure the best way for those yet. But, it is what it is, and it works, and if you follow the directions, you should have a working controller. So, where to begin? How about a parts list. The directions are sort of involved. That's quite a bit of writing. Hmm... I guess I better get started. Anyone that wants to help with directions, feel free. It would be nice to have pictures to go along with the directions, but I didn't take pictures of everything last time. Hmm... Also, I'm doing a few things different this time to make obvious improvements to the process. So, how about an improved process for version 0.1, and then more directions later for version 0.2, which will probably use an ATMega16 instead of an ATMega8?

I have a problem. This forum isn't the best format for including all the directions. For example, it won't let me upload my .c file unless I change it to a .txt format. Maybe I could start a website, and have Ecomodder link to it? I don't know how to do that, or if it would be the best option. I don't know! Oh Dear!

[Roger Heuckeroth]

Paul,

Does the program have to significantly changed to go to the ATMega16? What was the reason for going to the 16? Expansion?

I already have a controller for my EV. One of the last Zillas to come out of Otmar's shop. Just think I could have made about 10 of yours for the price of one of his. Not that I'm having buyer's remorse or anything :^)

What I would like to do is adapt your design to run an accessory motor for the A/C. I have a 2 HP DC motor that I will connect to the A/C compressor. When it is called to come on, I want the controller to smoothly bring it up to speed. Its rated at 120V, 14A. My pack is 234V fully charged, so I know I'll have to upgrade the MOSFETs, Diodes and Capacitors to higher voltage ratings. Any thoughts on this as to how easy, or not, it would be to program the controller to do this?

I would like to help you get a PC board made for the "brain" part of the controller. If your agreeable to this, let me know.

Roger

[MPaulHolmes]

Hey Roger! I don't think it would be very difficult at all. Just find some descent higher voltage mosfets and diodes. You might be able to just use a single mosfet/diode pair, and maybe 2 or 4 400v filter capacitors. It's not that hard to get the code to ramp up the throttle slowly if you want it to come on pretty gradually.

It would be much easier to do this than regular code. It would take like an hour of messing with the program. You may not need hardware current limiting in this context. It could be a much simpler control board. Maybe still a thermistor?? I don't know. 14 amps is pretty puny compared to 500. I doubt heating would be an issue, either.

I would love to get the control board made! They are all "through hole" components. I'm not sure how optimal the layout I used is. What I did before was just superglue 2 prototype PCBs from radio shack together, and solder everything with wires, trying to give some thought to layout (like where the digital grounds connect to analog grounds). It was exceedingly tedious.

To have a board, where all you do is push the components through the holes, and then solder away! Man, that would be so dang awesome, I can hardly stand it. I just want to make sure that the layout is done in a pretty good way, and I am NO expert on board layout. What I did works. In fact, my car drives very very smoothly now.

It's true that perfect is the enemy of good, so I don't know! What to just choose a layout that is similar to what I did, and call it good? It would be much easier this way than the radio shack way.

Getting a PCB made would still allow for different values of resistors and capacitors later. You would just plug in whatever you wanted into the holes! So that still would offer some flexibility.

A friend of mine recommended

Futurlec - PCB Manufacturing Service and PCB Production

He said it's the cheapest place he has seen for the hobbyist.

I have ExpressPCB software, which allows you to do the pcb layout and submit it to futurlec.com, but I've never used it before. I only used ExpressSCH, the schematic part of the software. I could try to figure it out. The whole controller would be DRAMATICALLY simplified once the control board becomes just a plug and solder thing! I wonder if they allow labeling on the pcb, so anyone would know where to plug in the components? If not, a master picture could be printed on the computer.

Pricing Example:

for (10) double-sided 4" x 3" PCB's it would be $7.74 per board with $25
setup. That's $10.24 per board.
If you add soldermask, the cost per board would then be $12.64 per board.

(Note: That's the example that my friend gave me. I would probably want to do single sided pcb, because they are through-hole components)

[Roger Heuckeroth]

That Futurlec place sounds good. Did you notice that you don't even have to do the PCB design yourself. They will do the design for you if you just give them a schematic. Unless we have a PCB design expert that is willing to jump in and offer their services, I would just suggest we go ahead and have them do the design.

Maybe once you've got Rev 2 all breadboarded and tested with the ATMega16 you can put together a schematic and we can go have the boards made. I think it would be a good idea to at least bring the extra I/O channels to the side of the board, so that tying in more I/O will be easier.

I'll make you a deal. If you write the code for my little A/C motor controller I'll pay for having the first 10 boards made. I can easily just write off buying 10 boards and the associated design costs to my business. Do we have a deal?

[MPaulHolmes]

Originally Posted by Roger Heuckeroth

That Futurlec place sounds good. Did you notice that you don't even have to do the PCB design yourself. They will do the design for you if you just give them a schematic. Unless we have a PCB design expert that is willing to jump in and offer their services, I would just suggest we go ahead and have them do the design.

That's just too awesome for words!

Originally Posted by Roger Heuckeroth

Maybe once you've got Rev 2 all breadboarded and tested with the ATMega16 you can put together a schematic and we can go have the boards made. I think it would be a good idea to at least bring the extra I/O channels to the side of the board, so that tying in more I/O will be easier.

That sounds awesome! Good idea! I never thought until now that pre-done PCBs could be so versatile!

Originally Posted by Roger Heuckeroth

I'll make you a deal. If you write the code for my little A/C motor controller I'll pay for having the first 10 boards made. I can easily just write off buying 10 boards and the associated design costs to my business. Do we have a deal?

DEAL! Just give me the behavior you want it to have, and I'll take care of the rest. I probably could even use the humble little ATMega8 for this one. hehe. I still have 9 left in the garage.

Ever seen the A-Team? I love it when a plan comes together!