R&D Projects

[trash]

I have been having a peak around other sites today. Work is slow,
and I have nothing but an internet connection for entertainment.

I noticed that ph2t was playing with more than flashing led’s,
some pretty interesting stuff going on around his place. What projects and ideas do you have on the drawing board ?

micro_amps got my attention with his Dyno. Do you have any more
pics of what is inside the box ? I’ve been working on a dyno
for about a year now, I just don’t get enough time at home to develop it properly. It’s designed for motorcycle engines, but I could be
easily scaled down to micro size. I originally used a PC with a
parrallel port, but I found that I couldn’t get accurate timing because of PC interupts and other things. So I have had to teach myself about pic’s and move the project onto a 16F84.
This is a much better platform for this project. It makes it much smaller and portable. The dyno is 5 truck flywheels, so portable is
rather vague. I’ve have continued working on it while I’ve been
working the past week or so. Not much to do here at night.
So far I’ve been writing the basic code. Every new step uncovers new problems. I was going to write the flywheel data to eeprom, but
it’s too slow and too small. I’m going to move this out to external
ram, more so, I hope to be able to use smart cards to store the data.
The end result is you’ll be able to use a gold card for each test.
just plug it in, run the dyno and the test run is saved on the card.
Take the card out of the dyno, plug it into you computer and the
data will be read out and displayed on the screen with some nice graphs. power, accelleration, torque, speed etc.
I hope to add an LCD to the dyno, this will help make running it easier and should be able to give you live results like RPM etc.
My biggest concern is the time it takes the pic to do things, in
particular writing data to memory. On a smaller scale the timing between flywheel pulses may be so short that accuracy suffers.
clock pulses of 256uS would not be good enough, and reducing this
to 64uS causes more interupts and less time to do other things.
The way around this is to gear down or divide the flywheel pulses
external to the CPU freeing up some tasks. The result is throwing away
resolution. I’d like to keep the resolution as high as possible on
a piston engine so that I can see things like fuel burning etc.
On an eletric motor this wouldn’t be so important, but on tiny motors
you can go from zero to full speed in a second or so.
Any other ideas ?

What other projects are people working on out there ?

[KingRX7]

im working on reading that post its gonna take me hours:D

[trash]

hehe. told ya I was bored, work is slow today 🙂

[trash]

Was also thinking about an electric piston engine today. little V6 using tiny rare earth magnets as pistons and coils to drive them.
Has anybody ever seen anything like that before ? I think it would look pretty cool if the parts were made from perspex.

[micro_Amps]

Yes Trash, I have a working model of your idea. I thought it up about 3 years ago. Unfortunately its not that efficient. Magnetic field strength is based on flux density and that means huge currents pulsed into the ‘head’ coil. It creates massive EMR interference and uses heaps of current. I was switching 150 amps and not getting a great deal of power output, it ran at about 4800rpm on a single cylinder.
Fun to build and watch, but ultimately useless.
:)uA

[trash]

oh I figured EMR would be a downside. How big was you head and coil ? From the sounds of it what you built was large. 150A is about starter motor size. Large motors for this kind of thing
would be ineffecient. I was thinking small, piston size about 3mm dia.
Was your motor a 1 stroke or 2 stroke ? power
on the up stroke as well as the down ?
I was thinking of a flywheel circuit to tune each cyclinder, varying this as the running speed
changes. A real flywheel to smooth out the stroke
and help transfer engery into the next stroke.

[micro_Amps]

My version was about equivilent to a garden trimmer motor (whipper snipper)
The ‘piston’ was about 25mm accross (1 inch), and the head coil was hand wound enamelled wire about 1mm in diameter, power pulsed with an SCR relay, timing was achieved using a hall effect transistors and rare earth mag on the flywheel. Speed and torque were variable and controlled byvoltage and timing ‘angle’, max revs produced at 8deg BTDC. It was torquey but pulsing 100 amps at 80Hz (4800 rpm) is almost like running fully switched on, and there’s a whole lot more you can do with 100 amps than just spin a plastic piston motor.
It was 2 stroke, firing only once just before TDC. The next version would have had the piston running inside a coil ‘tube’ and fired both directions to produce the 1 stroke cycle.
:)uA

[trash]

sounds interesting.
I was thinking about lap counter/timers for mini/micro cars. A bar code scanner seemed simple but expensive. I was writing some software
for I2C last night when it dawned on me to put
a pic in the car with an LED and use it to transmit a code to the lap counter. It then looks
for codes IR as cars pass under it. There is more
than enough processor time to scan the track width for multiple cars.
It’s then simple to display the results on an LCD screen. The code for this is drastically simple… I just to get home, buy some pics and
write the code in my head onto them.
Add this one to my list of uncompleted projects. 👿

[Impreza]

What about a rotary engine?

http://www.mazdarx8.co.uk/rotary/rotaryhow.asp?documentid=1331

[Author]

Posts