pick-and-place

The fact that I had to drag it out of you.

Actually you only allude to the real issue, that sometimes the internals of a chip are running at a different frequency than the externally supplied timing clock. You kinda get to that with the SERDES, but PLLs are not always used to boost the frequency of the clock inside the chip. There are times when they are used in FPGAs to assure the same phase is used on inputs and outputs from the chip increasing the problem of "synchronicity".

;)

My preliminary write-up of the circuit does cite Peter Baxandall. There's n o reference to Alan Dower Blumlein, and while I do use a transformer in a w ay that Blumlein would probably have liked, I'd be hard pressed to find a w ay to cite one of his 128 patents (and he probably wouldn't have recognised 128 as 2^7).

As hagiography, it probably wouldn't make the grade. The circuit does use o ne of Barry Gilbert's better integrated circuits rather than Jim Thompson's cruder offering, so it might be seen as polytheistic.

I also cite Jim Williams, who failed to cite Peter Baxandall as the invento r of the "current driven Royer inverter" in any one of the application note s AN45, AN49, AN51, AN55, AN61, and AN65 though he does cite his work in AN

-65 - a 1960 paper "Transistor Sine-Wave LC Oscillators" in the British Jou rnal of the IEEE, paper number 2978E, in a discussion of root-mean-square p ower measurements.

A proper acolyte would have written off Jim Williams as a heretic for that.

The Baxandall is overly complicated. No company would ever use that in real life. It's like the papers from IEEE, novel this, novel that. Almost all crap ;-)

Cheers

Klaus

hmm, depends on what scale of "synchronicity" we are talking

when using the pll to make things synchronous with an external clock you'll be delaying the internal clock to compensate for IO and routing delays, so the internal clocking could easily be a nanosecond off anything happing on the outside

-Lasse

n.;)

al life. It's like the papers from IEEE, novel this, novel that. Almost all crap ;-)

As the "current driven Royer inverter" in Linear Technology's application n otes AN45, AN49, AN51, AN55, AN61, and AN65, Baxandall's Class-D oscillator proved to be the most popular circuit that Jim Williams ever publicised, a nd drove the cold-cathode back-lights on a whole generation of portable com puters.

For that specific application, the additional complication was worth the ef fort. And it does inject much smaller spikes into the power rail than class ical Royer inverters.

Pardon the question, no offense inteneded, but WTF is wrong with you?

Yep, that is true for the internal clock, but that is not what I wrote. There is not much reason to do that in FPGAs. They typically use FFs in the IOBs and once you have captured the signals in the IOBs you have lots of slack time internally. However, IO prop times can be significant and IO timing can be very critical. Memory, for example, will often use such timing to keep all the I/Os sync'd to the external clock input.

Different issue.

The formulas are just circuits 101 stuff. Logarithmic symmetry is more rati onally termed ratiometric symmetry.

Don't put words into my mouth, I never stated any such thing about "in and of itself."

The low SRF is indicative of significant ESL. Paralleling C+L+R gets you n C+L/n+R/n. SRF and Q stay the same but effective ESL is reduced. Vias are l ittle lumped capacitors out to microwave frequencies. Whatever the hell is cargo cult engineering...

You can say that again...

The only thing off by an order of magnitude is your brain power. The compar ison is between figures 5 (calculated) and figure 9 ( measured). Were you i n such a rush to discredit the worth of the paper that you failed to notice figure 5 is composed of curves parameterized by h whereas figure 9 is for a 31 mil board and parameterized by probe position??? I really don't see bu t the slightest hint of asymptotic clustering to figure 5, but the 10mil cu rve of figure 5 looks to form an EXCELLENT fit in amplitude and resonance.

You mean mutual impedance? All the hand waving in the world isn't going to make Larkin look like anything more than a hack.

Kind of sacrilegious of you to politicize a strictly analytic argument.

Yeah, what he said!

Jamie

Common Sense is not one of Bill's strong point!

I'll let you know when I figure out what his useful strong point is.

Jamie

Jamie's judgement on this matter is compromised by his lack of any kind of sense - common or otherwise.

useful strong point is.

Waiting for Jamie to figure anything out is rather like watching paint dry - but at least watching paint dry gives you useful information. When paint is dry you can touch it. When Jamie has "figured something out", he's usually got it wrong.

You're so far behind these days and most likely always have been that many look at you like a fart in the wind, gone before being noticed.

Jamie

Or, more often, of a lot of C.

On Thursday, 4 September 2014 11:21:13 UTC+10, Maynard A. Philbrook Jr. wr ote:

of sense - common or otherwise.

dry - but at least watching paint dry gives you useful information. When pa int is dry you can touch it. When Jamie has "figured something out", he's u sually got it wrong.

that many look at you like a fart in the wind, gone before being noticed.

An entertaining conceit, but you don't usually get patents if you aren't at least up with the game (though krw has managed to get a couple of patents

- if while working at IBM, where they patent everything, including their to ilet paper). Similarly, you do have be somewhere close to the state of the art to get papers published in the peer-reviewed literature. One of mine ha s been cited 17 times (only twice by me) which would suggest - to somebody a little more with it than Jamie - that I did manage to get ahead of the ga me at least once.

Quite why Jamie thinks that he can locate the leading edge of the state of the art is is something of mystery. Presumably one of the many things he do esn't quite understand is his own appalling ignorance - what he doesn't kno w could damage his public image, if it wasn't already sea-green incorruptib le.

As Will Rogers said, it isn't what you don't know that's really embarrassin g, but the things you think you know that ain't so.

ein.;)

real life. It's like the papers from IEEE, novel this, novel that. Almost a ll crap ;-)

notes AN45, AN49, AN51, AN55, AN61, and AN65, Baxandall's Class-D oscillat or proved to be the most popular circuit that Jim Williams ever publicised, and drove the cold-cathode back-lights on a whole generation of portable c omputers.

I wasn't actually referring to the Williams driver circuits, but the circui t you posted a while back of a power supply which was overly complicated an d based on the Baxandall

Cheers

Klaus

than Jamie - that I did manage to get ahead of the game at least once.

Do us a favor, pull that plug!

Jamie

mlein.;)

n real life. It's like the papers from IEEE, novel this, novel that. Almost all crap ;-)

on notes AN45, AN49, AN51, AN55, AN61, and AN65, Baxandall's Class-D oscill ator proved to be the most popular circuit that Jim Williams ever publicise d, and drove the cold-cathode back-lights on a whole generation of portable computers.

uit you posted a while back of a power supply which was overly complicated and based on the Baxandall.

Peter Baxandall is actually most famous for his tone control.

Criticising him because I put together what you see as an overly complicate d power supply circuit is a little silly. I don't recall you posting a simp ler version of my circuit. Simple circuits are nice, but real world circuit s tend to become complicated to solve real world problems.