120 Degree Phase Shift Osc.

oman

xample:

I'm sure you can. What I'm not sure about is why Mark Weaver would want to expend battery life on producing three PWM approximations to a sine wave, when he could cut the current dissipated in producing the sinewaves by several orders of magnitude by going to a DDS approach driving three conventional DACs. The three coils are going to draw about 12mA in total, so "a few" mA might be acceptable, but a few extra hours of operation would also be attractive.

-- Bill Sloman, Nijmegen

separation.

am

to

apart from

voltage,

much

through

four

waves

Hill

it

logic

wired to

degree

three

each,

smaller,

part of

function of

PWM

relay-driving

Engineers have to eat, and most engineers like to cook.

You probably don't cook. You don't seem to have any skills, other than trusting peer-reviewed off-topic stuff that's outside your personal expertise. The continuous droll insult stream isn't much of a skill, because you're really bad at that, too.

I made some really good pasta, in 20 minutes, during the breaks in the World Series game. Mo was impressed.

John

insist

Have you looked at the ADI isolators with integrated power? They're not cheap (~$5) and not terribly efficient, but they look to be perfect for this sort of product. I would have used them if they were around a couple of years ago (too expensive to change anything now).

I like it, but perhaps a little busy on some pages. At least there isn't any flash.

It's only by self-restraint and a lot of scenery peeping that I keep the ratio down to about 2:1. If I didn't crash so often and so hard, it would be even worse.

Up for another try?

John

Sloman

example:

t.

I'd like to see with components you'd use to get 10 times better than a few mA for 3 dacs, a looktable of some sort, a cpld or something to drive it and an oscillator

-Lasse

That's what The Brat thought. So she took a couple of courses on search engine optimization and went for it. I just did what I was told, mostly cleaning up content.

John

t insist

without

ted

Not if it is keeping track of a 20MHz clock. Jan Pantelje has rubbed my nose in the fact that I haven't looked at microcontrollers for a few years, so the transistors that you have to keep flipping to keep track of a 20MHz clock aren't as big as they used to be, but you can't both sleep and control the width of successive pulses to 50nsec.

Since when were VME boards sold as "low power"?

The last time I did that- back in 1968 - was on code running on an IBM

That you've only just discovered the idea, and think it novel enough to boast about, does suggest that you skipped a lot of your undergraduate lectures

-- Bill Sloman, Nijmegen

an

t i=3D

wit=3D

ted

Jan, you may know a shiny new microcontroller that will only use "a few" millimaps in the application, but James Arthur was talking about an 8051 ....

-- Bill Sloman, Nijmegen

insist

without

Yes, they look great for some apps. But this one needs about 30 volts at up to 24 mA, too much for the ADI things. We are using their

That was one rule I got to make, no flash. I wanted no registration or forms too, but I lost that battle to marketing, who want to harvest contact info. At least I got them to keep it down to a minimum of info. I hate filling out forms.

John

oman

n't i=3D

u wit=3D

ay

ghted

't

theres many 8051s, the architecture is old, the mght implementation not be. I'd think that the ones you can buy now could very well be a few mA at

-Lasse

i=

wit=

Sure, though you hardly need 20MHz for a 10Hz PWM. 8051s are even available as part of an FPGA or PSOC, with all of the analogs, except the power stage. Slowman knows little about electronics and insists on demonstrating just how little every day. Well, on his better days he limits his incredible ignorance to American politics or AGW. You know, things that are not provable, even though he claims they have been.

insist

without

Some modern micros have on-chip peripherals that can handle three phases of PWM autonomously. You preset the PWM values, then tell the micro to go to sleep until it's time to revise the settings. Most of the chip can sleep most of the time.

Best regards, Spehro Pefhany

insist

without

More like a few decades

so the transistors that you have to keep flipping to keep

The 12 isolated channels are.

What idea and boast do you refer to?

My undergrad education didn't include digital logic or microprocessors, the latter of which hadn't been invented. No software courses, either. I taught myself all that stuff. I was never very good at learning stuff in class; it's too rote and too slow. I prefer a real problem.

John

The others have already clarified that a uC draws far less power than your proposal to use a DDS.

My project--20-odd years ago--used an 83c751 and operated, among other things a 100mW 920MHz SS transmitter I built from scratch and first principles. The uC generated the SS modulation and complement in real time, and performed the data modulation too, all at 1.2MHz output rate.

There were no hardware development tools, especially none compatible with my power constraints, so I wrote a single-stepping debugger that ran resident in the uC's 2k bytes of EPROM, while still leaving room for my application software, which I'd hand-coded in 8051 assembly. I communicated with my debug terminal over an RS-232 line using two device pins and a software UART I wrote for the purpose. I made my own device programmer from scratch too, because I had to.

Total average power consumption was 13uA from a 9v battery.

In short Bill, you don't understand what you're talking about or to whom you're talking, if you'll pardon my speaking frankly.

I don't know what to say to someone who worries that switching a CMOS uC output too quickly will draw significant power in this context, I really don't.

If you don't think you can get a uC to generate a 10Hz sinewave, okay, I guess I have to believe you, I just don't think it's that big a deal. I've not laid out every detail, I've explained to you in the language I'd use in speaking to a colleague, with the assumption that they'd not get stuck on trivial details and would make any necessary obvious adjustments as the need for such arose.

I do like John Devereux's ARM idea too--that's slick--or one could use any one of a number of today's uC's with 3-phase PWM motor control hardware built-in too.

-- Cheers, James Arthur

Ed

ist

ut

On the contrary, John's a great human being, smart, kind, and usually right.

I don't understand why you think that needs attacking, I just don't.

-- Cheers, James Arthur

st

Sure. You'll have to promise not to go Sonny Bono on us though...

-- Cheers, James Arthur

Works okay for me ....Firefox 3.5.6, 1,280 x 1,024 display.

-- Cheers, James Arthur