Paul & Sabrina's cheap DIY 144v motor controller

[MPaulHolmes]

dcb, I incorporated your ISR code in, just changing a little. It was really easy to understand. Thank you for such clean code! I'll test it out later.

[dcb]

Thank you Paul.

re: adc bits, worth considering, 255 steps might be noticable on a throttle, dunno. Also don't know what frequency you need to measure throttle at.

Have you experimented with the stimuli files in the avrstudio? That is what I was going to look at next, and maybe figure out how to graph the log files or something for easier validation.

[MPaulHolmes]

I changed a few things in your code including to reflect the heat spreader temperature sampling every 2 seconds, and throttle 1000 times a second. I'll have to test stuff later. I can't wait! ya!

I'll have to check out the stimuli files. I haven't yet.

As for Futurlec, a friend of mine just got his PCB shipment back in a reasonable amount of time from them. It was good quality, and they even sent an extra board for free. If Futurlec doesn't come through with the 3 pcbs, I'll just order from another place, my treat! It's all good! Fear not! Worst case, we get delayed by a couple weeks.

[68 TT]

Originally Posted by MPaulHolmes

I think a metal enclosure would be much better thermally speaking than a Lexan box with a heat sink on top.

Lexan is a very efficient thermal break so it will not transfer heat well at all. An aluminum panel mounted over the Lexan would not work even 25% as well as if it was the actual enclosure material. A very small Lexan window would be the most I would use it for in this application where you want maximum heat loss through the enclosure.

If you want heat dissipation then aluminum is the way to go. The more surface area the better. A finned extrusion would give you the best bang for the buck.

Direct contact with a thermal grease will offer the best heat dissipation.

[68 TT]

I found this place with an appropriate extruded heat sink material. It is about the right width for the controller base and has a 5/16" thick base so it will be plenty strong enough to mount in any manner you choose and strong enough to anchor an enclosure to.

Extrusion Profiles

They sell it by the foot so it would be perfect for the 11" long controller.

You can even get it black anodized.

A simple aluminum box "lid" could be mounted to it on a hinge to allow you to access the internal components easily.

Filling the end panels of the top box with 1/4" to 1/2" of potting material or silicone would make them safer to work with.

[steiner]

Adam,

Nice job on the drawings. I had completed mine before I saw that you had already posted yours. I imported yours and layed mine on top to compare. They are identical so it looks like we both interpreted Paul's notes the same. I have included a pdf version for anyone that can not read dwg or dxf files. Do you need me to generate the g-code or do you already have software for that?

Paul,

The two files that you sent me were the same. That is they were both etching descriptions. Neither of them had drill coordinates. Either Adam or I can add them when you get a chance.

[dcb]

Originally Posted by MPaulHolmes

I'll have to check out the stimuli files. I haven't yet.

I wouldn't bother, I finally convinced myself that they are truly digital only in avrstudio simulator. You can log or simulate a 1 or 0 on portB,C,D only. Cannot log pwm or simulate voltage levels. Back to the hardware discussion on the testing thread I recon.

[adamj12b]

Originally Posted by steiner

Do you need me to generate the g-code or do you already have software for that?

You can create code also, I will compare them and use the quicker one. LOL

-Adam

[JayC]

Adam, are you planning to cut some of these out? If so, what are you using? I see you have the X2 and a Jet baby Bridgeport you converted.

I have the Widgitmaster Wide CNC router so not too worried about the board being too big for it.

Jay

[68 TT]

I came up with a simple enclosure that will work very well and can be fabricated by anybody with a drill, a jig saw and a simple metal bending tool.

I am working on plans for a simple bench mounted sheet metal bending tool that anybody could build for a couple bucks to do the job on their own that i will share with everybody here when I have it perfected.

For a small production run of a dozen the material cost would be $20 for the box and lid plus $15 to $20 for the heat sink base.

A simple box housing secured to the heat sink with a separate lid seemed to make the most sense and was the least expensive and quickest way to put it together.

The housing could be fit with a cooling fan easily and access to the internal components would be as easy as pulling a couple screws to remove the lid. None of the wiring would have to be touched to check it out inside.

The lids could be very simple if you were going to tuck the unit out of sight or they could be machined and/or engraved to make them look nice if you want to place it somewhere visible. A nice ball milled lid with the ecomodder logo or something of your choosing could easily be done with the number of people on here with CNC milling machines.

With one person doing a dozen units it could be produced in 20 minutes or less for the basic enclosure with simple lid. Add in 10 to 20 minutes on a CNC milling machine for a cool lid. Five minutes of finishing and it could have a nice brushed aluminum finish and a heat loss friendly black paint job.

I will finish up the plans and put together a materials list as soon as I can get a few spare minutes to work on it.