This is it !

[7imix]

Other important point for that code is he turn it on in sequence.
= pulse 1, pause, pulse 3, pause, pulse 5 pause, .... repeat


maybe faster code is
= pulse 1, pulse 3, pulse 5, .... repeat


Edit, above consider Y refer to channel number.....

Yes... Absolutely correct.

[7imix]

Only the following pins are used as output on the micro.

PORTB: pin1,2,3
PORTD: pin2,3,4,5,6

Hmm, this is very very interesting. And confusing.

The Pulseout function takes the port as the first argument and the pin as the second argument (passed the variable Y in the code)

What does it mean to call Pulseout with a pin that the port is not configured for? Do the two ports have physically separate pins, or are they just abstractions, more flexible and configurable in software?

I suspect, as he even states in the video, that this is not the cleanest code and there are unnecessary bits in there. Figuring out what is important and what is not is just guesswork without the ability to experiment directly.

What I think is important: short sharp pulses and pulsing the coils around in a circle. I used to be convinced that hitting the resonant frequency was essential, but now I just think it is useful, not essential.

[thatsthat]

Agree

What I think is important: short sharp pulses and pulsing the coils around in a circle. I used to be convinced that hitting the resonant frequency was essential, but now I just think it is useful, not essential.

I agree fully! It's the length of the pulse and sequence of pulsing of the primaries on the coil that is important not using this specific micro and code.

But, to clear-up the confusion around the PORTS configuration and uses.

The variables PORTB and PORTD is physical represented by 8 individual pins on the micro respectively. Thus, to pulse all six HEXFETs with the design of the Hex Controller board, it seems like some of those outputs might be extra or not used at the moment.

Something to remember if not familiar with micro controllers. Each PIN can be assigned a specific function. A PIN might be used as normal digital output or input pin as is the case for the above discussion. Alternatively the same PIN might be used as a PWM or ADC or any other function as provided by the specific micro controller being used. The datasheet of the ATMega48 will show exactly what each PIN can possibly do. That is the major advantage of using a micro as it can do so much with high accuracy and full control to the designer.

To note: Only one function can be assigned to a physical PIN at any one time for Atmel and PICs, thus it cannot be a ADC input and also a digital output at the same time for example.

I think we shouldn't focus anymore on what exactly the code does. How to pulse the coil should be the focus.

Cheers.

[7imix]

please find 3g parts list here:
PWM3G Parts List

I never thanked you properly for posting this list. Thank you, thank you, a million times thank you.

I had a hunch and combed over this list more carefully than I had before.

Stop the presses -- check out this gem.

http://focus.ti.com/lit/ds/slus504e/slus504e.pdf

So much to learn, so little time -- luckily so many people have already learned it before me.

Thank you Bob Boyce for your hard work and the trail of breadcrumbs you left for me to follow -- I have learned so much.

[Farmhand]

Pulsing the Coils Sequence

I agree working out how to pulse the coils is the main objective to looking at code. I'm sure nigel won't mind me linking his code.

Yes I think to just pulse the coils like this- pulse 1 - pause - pulse 2 - pause - pulse 3 - pause a bit longer then repeat.

This would cause a pulsating pulse wave kind of thing, like 1-2-3--1-2-3--1-2-3--1-2-3--. A pause between each then a double pause before the next round.

Like that.

Hows that sound ?

off topic now

thatsthat, here's a picture of my board, and a pdf user manual. Is it a mistake to start to learn with old hardware ?

http://9xfdag.bay.livefilestore.com/...629.jpg?psid=1

PIC16F877A Mini development board.pdf - Windows Live

I'm interested in how this could be used for other projects I have in mind at
a later date. Well I know how it could be used, but is it an OK setup to learn with ?

Thanks

[Farmhand]

Hi all internet is down again i'm back on the wireless broadband on a tiny laptop.

Some way to control the length of the pause between the pulses in a cycle as one function and the length of the pause between each cycle or set of three pulses as another function, could be the only variable we need.

I think Greg already has this type of function in mind.

Cheers

Storm here is serious I hope my roof stays on or my stuff will get wet.

http://public.bay.livefilestore.com/...299.gif?psid=1

http://public.bay.livefilestore.com/...191.gif?psid=1

ninemsn - Hotmail, Messenger, News, Sport, Celebrity, Finance, Travel, Cars, Movies, Shopping

The storm is a fair distance from me but the wind is picking up.

[7imix]

Hope everything goes well for you with that storm farmhand.

[Farmhand]

Storm was a bit of a fizzer, looks like everyone is ok, so thats good. I am
concerned about Greg, we haven't heard from him for a little while, I hope he
is ok. These disaster events are a reason for our work, to have small
powerfull devices that could be stored in a cupboard, and would tap the
environment to provide excess power for stuff when the grid will be down
and people can't buy fuel. The device could be stored in a water proof
enclosure. And anchored down or put somewhere it won't get blown away.
A source battery and a charge battery to swap around and a load
could be furnished with power for an extended period or indefinately, if managed correctly.

Only the people with generators and fuel have power when the poles are
snapped off. A small solar array to charge the batteries should they get flat
for some reason. All that is really needed for emergency power is enough
to power a refrigerator and some lights. Everything else is luxury. If the sun
don't shine much only a small amount of solar power is availiable, after a
cyclone people would be very lucky if thier wind generator is still there and
fuctioning. A large solar array on a roof would probably dissapear into the
bush with the roof.

.....

What do you think about that pulsing sequence ? Sounds like it might cause
something to happen.

I haven't had any response from loadstone about the tape. I hope I can get
some from town when I go in some time in the next 7 days.

Unfortunately I don't have enough money to buy all the things I want for my
different projects, i'm going to visit the motor winding shop in town
( if it's still trading) find out what i can get from them in scraps and so forth.
I'll also ask them if the can get silver coated pfte insulated copper wire. And tape.

Nice of thatsthat to join the forum to help us . Thanks

Cheers

[sucahyo]

I hope you will be safe farmhand.

I wonder if sequencing stingo is possible. Instead of triggered by own coil, triggered by neighboor coil, a coil trigger the other coil transistor. three 10k pot provide delay for three coil. Pulse depend on pnp type.

[Farmhand]

I hope you will be safe farmhand.

I wonder if sequencing stingo is possible. Instead of triggered by own coil, triggered by neighboor coil, a coil trigger the other coil transistor. three 10k pot provide delay for three coil. Pulse depend on pnp type.

Yes I believe a similar thing could be achieved in a few ways with differing
effectiveness. Even with all the coils in series pulsed by one stingo
with a phase shifter of some kind between the windings or something.

Tesla showa a few ways to acheive a phase delay or shift with coils or
capacitors and such. Anything is possible.

It will take research and experimentation though. When i find something appropriate sucahyo I will show you what i mean, I mentioned it a while back,
it is in the High frequency papers, "Polyphase Currents and High Potential Lighting". Designed for motors but they use coils so.....

Cheers

P.S. Not only was there no lives lost in the storm but three babies were born.
What a coincidence "three storm babies", awesome. + 3 for Queensland.

[7imix]

I hope you will be safe farmhand.

I wonder if sequencing stingo is possible. Instead of triggered by own coil, triggered by neighboor coil, a coil trigger the other coil transistor. three 10k pot provide delay for three coil. Pulse depend on pnp type.

Yes, I'm interested in a sequencing stingo like that. I hope to get back to my low power stingo research soon.

[gmeast]

progress

Storm was a bit of a fizzer, looks like everyone is ok, so thats good. I am
concerned about Greg, we haven't heard from him for a little while, I hope he
is ok. These disaster events are a reason for our work, to have small
powerfull devices that could be stored in a cupboard, and would tap the
environment to provide excess power for stuff when the grid will be down
and people can't buy fuel. The device could be stored in a water proof
enclosure. And anchored down or put somewhere it won't get blown away.
A source battery and a charge battery to swap around and a load
could be furnished with power for an extended period or indefinately, if managed correctly.

Only the people with generators and fuel have power when the poles are
snapped off. A small solar array to charge the batteries should they get flat
for some reason. All that is really needed for emergency power is enough
to power a refrigerator and some lights. Everything else is luxury. If the sun
don't shine much only a small amount of solar power is availiable, after a
cyclone people would be very lucky if thier wind generator is still there and
fuctioning. A large solar array on a roof would probably dissapear into the
bush with the roof.

.....

What do you think about that pulsing sequence ? Sounds like it might cause
something to happen.

I haven't had any response from loadstone about the tape. I hope I can get
some from town when I go in some time in the next 7 days.

Unfortunately I don't have enough money to buy all the things I want for my
different projects, i'm going to visit the motor winding shop in town
( if it's still trading) find out what i can get from them in scraps and so forth.
I'll also ask them if the can get silver coated pfte insulated copper wire. And tape.

Nice of thatsthat to join the forum to help us . Thanks

Cheers

Hi all,

Just had my head buried in some finalizing circuitry (mostly finalizing) and had to design my own digital comparator for the job. So guys, how many phases do we want to potentially be able to do. The Hex Controller gives you 6 of them. I think maybe all we need is potentially 4 ... so that's what we're going to get ... 4 phases (possible).

I believe I said I received my driver IC's but haven't tested them yet. They are the UCC27321P chips. The controller is now a very simple device. I'll release details as I test ... and pictures too.

Hope you're safe there Farmhand ... 'nother storm?

I thought we were staying away from the micros here ??????????????????

Using a micro is easier for this, but it would leave many replicators out in the cold. Anyway as I said, I had to design my own comparator to simplify this thing. Until I can find the comparator to do what I want and not a bunch of other crazy s*h*i*t to add to the overhead, I'll have to use a gate array or gate IC's for the logic. After trimming away the fat, this thing is very generic.

After reading earlier posts, the sequencing 123***123***123***, etc with the pauses, I can do ... we just use all 4 phases ... 3 actives to fire coils, and the 4th phase is null .... it just takes up space and times out. I can add more null spaces ... it's just more phases ... how many do you want? ... it just adds a comparator. The only thing is, the pulse width for all 'firings' will be the same ... not independently adjustable ... just globally. But since the windings will all be nearly the same, that won't be a problem.

And once again, please do the math. For higher frequencies, it may not be possible to have a .3% pulse width AND a .125usec pulse. 42KHz is 2.38 x 10^-5sec pulse period and .3% of that is 7.143 x 10^-8sec. .125usec = 1.25 x10^-7sec. On 42KHz, .125usec is close, but more like .5% duty cycle, but I'm not sure the UCC27321P chips can actually give us that. Even if the circuitry pulses a .125usec gate pulse, the MOSFET's 'Turn on Delay Time', 'Rise Time', Turn off Delay Time' and 'Fall Time' gobble up some of the time as well as similar delays in the driver chip. I'm shooting for .125usecs just in case the driver chip can deliver. My pulse width right now is just at 0.3usec, but I'm still using my Op-Amp PWM for the clock. I need a real clock .. please suggest one.

I told everyone I had to do the $$$ thing for a couple of days, so I haven't posted.

Later

[Farmhand]

Hi Greg, Excelent, that all sounds very good, no no one is going to use any
micro's haha. We were just looking at some code to work out what he was
doing, for a sequence.

What do you think about that sequence, excelent idea with the fourth phase
you have. If we do it like that it will create another frequency.

I think a good way to try it would be at 21Khz or 10.5Khz three in sequence like
Pulse pulse pulse delay pulse pulse pulse delay. Then every cycle
of three would be a frequency aswell, don't ask me what it would be though.
Maybe 1/3 of the frequency used for each coil. It would be like sets of three pulses. Awesome.

I wound the third coil, now i just need to space and wax. Looks good. Full.
http://public.bay.livefilestore.com/...003.JPG?psid=1

Cheers

[marxist]

duty cycle

Hi all

... they would inject "you need to be around 0.3% Duty Cycle and around 2.5usec pulse width" ... and off the cuff advice like that.

I am not competent to offer any advice, but I feel that he,who first posted about "0.3% Duty Cycle" may have made a mistake of the kind of a typing error.
He may have meant to say:
"approx. 30% ON-time per cycle"
which is the same as 0.3
which would again be the same as saying "approx. two thirds OFF-time per cycle".

Again: I think it could have been a simple communications error (but that just is some gut feeling of mine and I may be wrong).

EDIT
In the meantime I have realized that the comment of '0.3% duty cycle' was made by watkykjy in his video #3
YouTube - Battery charging with the hex controller (3 of 4)
and after looking at the scope shot of the voltage spikes in this video
it became clear to me that my gut-feeling was a lot of BS and that the duty cycle should really be as short as possible. So a short rise time & a short fall time & a (ultra) short duty cycle seem to all be important to achieve the self charging effect.

[Farmhand]

Hi marxist, I think you're right, I think anything up to 30% on time could be
used myself, per coil, depending on the frequency, and required output.

For self charging purposes I think we should just do as short as is practical
and usefull. Thats all we can do.

OK I have set up with a board to pulse two coils with a fet each in unison
and a single fet board which will be free running from the other one. I've got
two scope's and several meters. So i'll see what I can see.

http://9xfdag.bay.livefilestore.com/...001.JPG?psid=1

Any frequency requests ?

I can't promise I can play you're tune but i'll try.
I won't be doing any cop testing or self charging either just looking at
waveforms and charging a second battery or three.

Cheers


Edit. Scope shot Video coming

http://9xfdag.bay.livefilestore.com/...001.JPG?psid=1

Here's the short vid,

YouTube - Three.wmv

Seems like they try to sync with each other.