120 Degree Phase Shift Osc.

rote:

a CMOS

ext, I

to

. 16MHz 5uW/

be very

If I flattered you assiduously enough, you'd suddenly see all sorts of useful strengths, and your favourable opinion would be worth exactly as little as you current negative opinion.

I can live with it - it isn't difficult.

-- Bill Sloman, Nijmegen

800x600 - and it sometimes causes problems on some sites. So, you may have nailed my problem, _again_. :-)

Not failing now, so I can't test. But I'm sticking with 800x600 until/unless forced to go 1024x768 (or whatever).

Thanks, Ed

CMOS

, I

Hz

/

There are pages of PIC's in the Farnell catalogue. Try to be a little more specific.

But you didn't specify the manufacturer, or point me to the data sheet, which does suggest that you are relying on your memory.

-- Bill Sloman, Nijmegen

Hz

/

t's

.

in

h g

we

A half-baked reference to a vaguely similar device isn't a secure platform for pompous pontification. What you've revealed does seem to be more relevant than James Arthur's single-side-band transmitter, but you haven't told us enough about it to make it relevant to Mark Weaver's specific question.

What was step-size in yur approximation to a sine wave, and how much current did your micro draw? These are simple enough questions, so why can't you answer them?

-- Bill Sloman, Nijmegen

er

real

t

=A0I

y -

ot

S z

Since I seem to be the entire defending force, I'm both commander-in- chief and the poor bloody infantry. Most people wouldn't see the point of splitting the functions in this situation.

-- Bill Sloman, Nijmegen

er

real

t

=A0I

y -

not

S

Hz

/

Since you haven't actually thrown in any ideas, and your contributions about what Mark Weaver might actually build have been totally unspecific - without a data sheet or a manufacturer's part number - you need to start taking your own advice.

-- Bill Sloman, Nijmegen

Indeed. :-)

In my upper level classes MOSFETs were drawn exactly like BJTs except that the source and drain legs were horizontal rather than angled. (Like this:

-- but without bothering to add the vertical line indicated the insulated gate.) Definitely a "slang" approach there... I kinda liked it, actually, although unfortunately many other people I've worked with don't seem to have encountered it and hence are justifiably confused. (I have a book somewhere where the guy rants about how awful many of the simplified MOSFET symbols are...)

And I suppose less cumbersome than "variable resistor..."

---Joel

ppenedBillSloman

:

an

n't

So how much is "very little"? With a 20MHz clock?

My PDP-8 was full of interrupts, and I used lots of them. It didn't offer any kind of sleep mode.

For someone who claims to be well-informed, you have an odd unwillingness to specify how much current your microprocessors draw when they are asleep but running a counter at 20MHz.

k to

And - of course - flattering about John Larkin or Inland Electronics ...

-- Bill Sloman, Nijmegen

ed

m not

MOS

I 6MHz

uW/

This started when I threw in the idea that a PIC could easily synthesize the needed sinewaves, and pointed out some of the advantages. The spirit of that suggestion alone was more than enough to guide someone skilled in the art to implement it.

Since then we've spent more time addressing your incorrect criticisms than it would've taken to design the device.

-- Cheers, James Arthur

doesn't insist

without

Wow, do you talk like that in real life? What a pompous fathead.

John

You've obviously not looked at the source.

I agree with that

--
"For a successful technology, reality must take precedence 
over public relations, for nature cannot be fooled."
                                       (Richard Feynman)

Rich would know about that ;-)

-- "For a successful technology, reality must take precedence over public relations, for nature cannot be fooled." (Richard Feynman)

Pretty good, but I quite liked the old one.

Fine here in firefox (ancient 1.0).

Tidy throws up syntax warnings, but nothing unusual about that.

--
"For a successful technology, reality must take precedence 
over public relations, for nature cannot be fooled."
                                       (Richard Feynman)

I miss my 2048*1536 HP monitor.

--
Politicians should only get paid if the budget is balanced, and there is
enough left over to pay them.

ated

I'm not

CMOS

t, I

o 16MHz s 25uW/ a

Since the Farnell catalogue devotes some twenty pages to lisitng the various flavours of PIC that they sell ex-stock, someone skilled in the art would have been busy for a while working out what - if anything - you had in mind.

Since your contributions have yet to be distinguished by the appearance of a specific part number for which one could track down a data sheet, your contribution to any such design would be a big fat zero.

John Larkin has finally got around to pointing out that the magic ingredient in an "appropriate" micro would be a programmable interrupt- generating timing counter which could wake up the micro four times in every - say - 50usec on the appropriate edges of a - say - 20MHz clock to allow it to generate pulse-width modulated approximations to three

10Hz sine waves. He has yet to unambiguously identify which micro he fancies for the job, or tell us how much current its programmable counter will draw while the rest of the micro is sound asleep.

This puts him a little closer to than you to such a design, but not all that close.

He wouldn't have done that much if I hadn't provoked him with my "incorrect" criticism.

John Devereux came up with a specific ARM microprocessor, the ADUC7021, much earlier in the thread - on November 6 - and since it includes 4 12-bit DACs it can implement a DDS solution and thus wouldn't need to be clocked particularly fast or chew up much current, so he's way ahead of the game. His primary motive was merely to heckle John Fields, which does suggest that the creative process benefits from a certain amount of point-scoring.

-- Bill Sloman, Nijmegen

m

doesn't insist

t -

A can

nk you without

you say

ents as

Google

sually

on't.

d

ed

ain

list

and

is

of

e e

Depends on the company.

I said it first - and in more detail.

-- Bill Sloman, Nijmegen

l Sloman

seem

his

t co=3D

.

ween

s,

ple fitler.

hannel.

summing

else none.

le use R2R ladders.

ATTINY84, internal oscillator @ 8MHz

since I didn't bother figuring out the PWM setting for the timer, I just made ~33kHz timer interrupt running three software deltasigma modulators, and a 128byte table for one cycle of sinewave.

no filter on outputs: 2.6mA @ 3V

1K + 2.2u filter on outputs : 3.8mA @ 3V

-Lasse

No slang. That's they way they were drawn. It took me a while to figure out what they were doing with those new-fangled drawings you're used to. ;-)

I used this sort of symbol for a BJT for 20 years:

C E O O | _|_ +---+ \P/ | N | V +---+ +---+ | P |---OB | N |--OB +---+ +---+ (with or without the polarity/sex \N/ | P | notations or horizontal stripes) V +---+ | | O O E C

It took a little while to get used to it but the symbol worked great for diff amps, current switches (AKA ECL), and current mirrors. What other circuits are there? ;-)

Not as obvious, though; more "slang". ;-)

It makes people simpler.