Author Topic: My projects (Read 3117 times)

willib · « **on:** March 22, 2006, 01:22:01 AM »

this spot will be a list of my current projects .

as some have allready read of my goings on i have been dabbling with pwm

to date i've boiled water , cut styrofoam and run a dc motor with pwm.

much more to come...

willib · « **Reply #1 on:** March 21, 2006, 06:55:50 PM »

this is the blue foam i cut, it left a nice coating on the foam

this is how i did it

its really hard to see the wire , but its #30 wire , and when i was cutting it there was 24v and tad over an amp going through the wire..

willib · « **Reply #2 on:** March 21, 2006, 07:03:54 PM »

these puppies are abot 7 feet long x 9" wide x 3 7/8 " deep

i really want to use these pieces to make some some very large foam blades.

i can get more , if they dont empty the dumpster . lol

willib · « **Reply #3 on:** March 21, 2006, 07:15:07 PM »

i actually have two 12v batteries on there now , and two fets in parallel , since there are no heat sinks on the fets , i thought i had better add a second one.so they share the current when it is up around 5A..

http://www.otherpower.com/images/scimages/2965/my_pwm.JPG

willib · « **Reply #4 on:** March 21, 2006, 07:31:02 PM »

this was a little experiment , to see how long it would take to boil water with electricity

it took 1 min 7 seconds to boil , with 24V @ 4A .

the element was #30 wire , several feet of it.. all coiled up in a glass tube ..

the water was cold when i started the experiment , not sure of the exact temp...

not very scientific , but it was cool (err hot) none the less..

willib · « **Reply #5 on:** March 21, 2006, 08:03:34 PM »

http://www.otherpower.com/images/scimages/2965/IM001600.JPG

since i took this picture i have been toying with a 9 coil 12 pole dual rotor setup for those blades, the heart of the mini mill is a 5 1/4 " hard drive motor , which i am using as a bearing..

the blades at this point are not done but measure 8 1/2 inches from the center of rotation to the tip..and i still have 4 more inches to add on..

these are the rotors i want to put on the mini mill 4" dia.

1/2" magnets

willib · « **Reply #6 on:** March 21, 2006, 09:02:03 PM »

check this out ,i found this the other week ..

its 16" diameter!!

last time i had it going it lasted over 8 min. ,till it stoped turning, thats a lot of flywheel action

oztules · « **Reply #7 on:** March 22, 2006, 05:45:00 AM »

Hi Willib

It would be better if the reactive load (motor) had a freewheeling diode across it to supress the back emf spikes caused by the motor. This will improve performance and stability.

As you go for more power, you will need a few more enhancements to protect the fet. Zeners accross the inputs, snubbers for inductive loading, and MOV's for spike protection, etc, etc.

Good learning ground here,..... well done. Power electronics is a whole new world to play with.

..........oztules

dinges · « **Reply #8 on:** March 22, 2006, 05:49:30 AM »

Yes, that was my idea too. Specifically, I was thinking about adding a resistor (10k?) grom gate to ground, to protect the sensitive gate of the FET.

It won't make the circuit work any better than it does now, but it will make sure it will survive for much longer.

A 16F84 as a pulser... Geez. What is the world coming to nowadays. Why do I get this image of a mosquito and a cannon in my mind Never mind, as long as it works and you are familiar with it.

Peter.

oztules · « **Reply #9 on:** March 22, 2006, 06:13:56 AM »

I had that typed out and more but wiped it and went for what was glaringly necessary.

I had hoped the 16F84 had a R to ground on its output internally. I don't know the 16F84 at all.

I used to generate a triangular wave and put it through an op amp against a slope for my pwm, or for double ended a tl494, or 3846 for push pull. However, i now use L4990 (i have 148 of these left over from other projects some time ago....sudenly it looks good, along with 500 x 20A 200v shottky's) for single ended output.

I will use these someday..... thats my story amd I'm sticking to it:-)....I think there are a few hundred e cores out there somewhere too.

.........oztules

willib · « **Reply #10 on:** March 22, 2006, 06:38:13 AM »

Gentelmen , the uController is a 16F88 ,it has a built in pwm circuit , as well as a built in A/D ( analog to digital) circuit..

the 16F84 was superceded by the 16F628.

iwas thinking of putting a diode across the IRF3703 ..but what about the body diode?

the sound the motor makes has changed , dont know why..it used to make a high pitch humm ,when starting , but now it makes a low pitch groan or growel. very strange..

spinner · « **Reply #11 on:** March 22, 2006, 06:41:33 AM »

allllllllll right!

neat lil rig

spinner

oztules · « **Reply #12 on:** March 22, 2006, 06:51:38 AM »

At low frequencies, the body diode is fine. The diode across the motor is the one of more urgent consideration.

If you consider your ignition coil in your car, you hold a current on the coil...nothing happens, you open the circuit, we generate a hv spark on the secondary as the magnetic field collapses and "moves with speed through the windings"....

As the field collapses on the armature, the windings generate a voltage of reverse polarity that generated the field in the first place.

This induced voltage will try and spike your fet.

The faster you collapse the field, the higher will be the return voltage (speed of magnets past coil of wire type argument.. higher the speed, higher the voltage. Same with collapsing the magnetic field in the motor windings, or ignition coil).

So we place a big fat diode across the motor input. This shorts out the back emf, while not interfering with the ingoing emf....does that make sense?

if not i'll try again.............oztules

willib · « **Reply #13 on:** March 22, 2006, 06:54:10 AM »

Thank you!

its been real fun since i dont have any room to permantly install a larger windmill outside , this one is installed to fly outside on my windowsill.

of course that isnt going to stop me from testing a larger windmill outside , i could test it at local park i think, when it's done..

willib · « **Reply #14 on:** March 22, 2006, 06:58:00 AM »

yes thanks ! , forgot about that !!

oztules · « **Reply #15 on:** March 22, 2006, 06:59:40 AM »

I forgot. see if your frequency has drifted .

For a high pitch to change to a low pitch, your frequency has changed. The pulse width modulation of the motor will remain the same as the time on/time off ratio will still give the same average current over time as a higher freq.....

Motor efficiency may change as the reluctance of the windings will be different at different frequencies. Flux would be the man to explain this further.....i'm not that flash with magnet theory.

..........oztules

SmoggyTurnip · « **Reply #16 on:** March 22, 2006, 07:25:38 AM »

Great projects. Nice pics too.

Something funny about the boiling water though.

24 volts X 4 amps = 96 watts.

96 wats = 328 btu/hour

in 67 seconds that is only 6 btu's

it takes 1136 btu to bring a gallon of

water from 70 degrees to 212 degrees.

6/1136 = .005

so your set up should only be able

to bring .005 gallons to boil in that time.

.005 gallons is .64 oz

The picture looks like you

have about a cup of water there.

Am I missing something?

willib · « **Reply #17 on:** March 22, 2006, 07:50:43 AM »

the tube is what contains the water , not the bowl. i used the water in the bowl to keep the hotglue from melting.

great calculations!!

the inside dia. of the tube 0.384" and the length is 2.94",

SmoggyTurnip · « **Reply #18 on:** March 22, 2006, 07:56:38 AM »

Now it makes sense - that is .188 ounces.

Should have no problem with that.

willib · « **Reply #19 on:** March 22, 2006, 08:00:50 AM »

That was fast.

willib · « **Reply #20 on:** March 22, 2006, 10:16:08 PM »

i put a new bearing on there , and a new body .

i decogged the bearing also, by taking off the magnets..

i spins very easily now , before it would take a small gust of wind to get it going but now its much better.

the string in there is to keep the blades from hitting the window..

http://www.otherpower.com/images/scimages/2965/IM001782.JPG

in the first photo i had a plastic wind sock on there but it didnt do any good so i took it off..

http://www.otherpower.com/images/scimages/2965/IM001792.JPG

willib · « **Reply #21 on:** March 23, 2006, 10:50:50 PM »

I dont think thats it, because the 16F88 is getting its clock frequency from an oscillator divided down with counters.

willib · « **Reply #22 on:** March 23, 2006, 11:21:33 PM »

i'm trying to think of a source for rotors for my mini mill , since the steel doesnt have to be as thick as large rotors , i'm thinking maybe a steel drum lid would work ..

Does anyone know how thick a steel drum lid is?

i'm looking for some steel that is anywhere from 1/8" to 0.175" thick and 4" to 5" in dia.. ?

i'm sure the steel drum isnt even 1/8" thick, but it mightwork

they closed my favorate junkyard !! sucks ! man i could spend hours roaming around there..

elvin1949 · « **Reply #23 on:** March 24, 2006, 01:32:31 AM »

Willib

How thick is the steel in cast off washing machine bodies.Might work

later

elvin

oztules · « **Reply #24 on:** March 24, 2006, 02:01:42 AM »

Sorry Willib

But I am not familiar with the chip that you are using. When I have a problem like this I tend to try and disprove the idea, and then come up with another hairbrained reason and test that.

In this case, just vary the oscillator and see if you can emulate the change in sound. If varying the osc doesn't mimic the fault, you only have a few things left....and none that make sense. To get a change in pitch, by definition that is a change in the vibration frequency, and at this stage I can't see what else would change it.....unless the fet is loading down the pwm output, and that is changing the available voltage inside the chip..(you dont have a high current driver, but you didn't before either I guess)...but that doesn't seem to make much real sense either, but you can check that by checking vref (if that chip has it available, or put the scope on the fet gate, and see if the wave is diminishing in amplitude as well as width when you vary the pulse, or if strong harmonics are being generated, maybe you are hearing them??

Keep us imformed on this one

........oztules

ghurd · « **Reply #25 on:** March 24, 2006, 06:39:05 AM »

Speaker plates!

G-

kitno455 · « **Reply #26 on:** March 24, 2006, 08:18:31 AM »

saw blades.

allan

willib · « **Reply #27 on:** March 24, 2006, 08:53:18 AM »

thanks to all !!

Elvin that might work but i've found another source

Ghurd that would work , but i dont have any of the right size

Kinto that would work also

i've found that i have several bicycle drive sprockets that are only a tad larger than my 5 1/4 " maximum dia and one is 0.095" the other is 0.110" thick ..YAY.., no flux exits the other side.

i have been useing electric outlet box lids but they are only 1/16" and some flux was exiting the backside ,meaning essentially they were not sticking as well as they could have been.

even when i doubled them up, the magnets still flew off at 500RPM. still missing one btw..

since that incident i've put a ring of #17 wire around the magnets to keep them from flying off..

willib · « **Reply #28 on:** March 27, 2006, 12:31:34 AM »

willib · « **Reply #29 on:** March 27, 2006, 08:55:12 AM »

the test bed is composed of 48 -N40 neos 1" x 7/8" dia. $120 worth of magnets.

the rotors are motorcycle front disk brake rotors 10.8" (27.5cm) i'm trying to get a handle on metric system, converting is tough though.

the bearing is from the front of a lumina front wheel drive .

i've gotten a phenominal amount of current from a #10 gauge coil,

above is the coil with #20 wire soldered across its terminals.

still i got .6V(Peak) across its terminals the load is a calculated 0.005833 ohms

that equals 55A peak and 39A rms from one coil @ 166 rpm.

load is 6.6" (16.9cm)of #20 wire .

with what i found out recently about partially filling the hole with wire i should be able to get more turns/voltage .

coil dimentions are 1.75" od (4.5cm)x .76" ID (1.95cm) x .510" thick (1.3cm) all dimentions are measured with a ruler except the coil thickness ,'imperial units', which was measured with a dial caliper.

an interesting side note , when i was turning the test bed with the #10 coil , it would get easier to turn once the load heated up and its resistance increased.

The enamel coating on the load is also breaking down from the heat.

News:

Author Topic: My projects (Read 3117 times)