MMIC filter

That's not EMC - that's just human stupidity. You need to hire a primary school teacher (albeit one with electronics skills) and let them handle those kinds of jobs.

Brainstorming is great, done right.

We sign and date our whiteboard scribbles and photograph them and stash the pics in the project notes folder.

Our ideas are certainly implemented.

We had a great one last week. I designed a relay/circuit breaker module and we came up with a nice way to let users gang breakers.

John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics

If it wasn't monotonic on delay you were doing something wrong - the signal is propagating through a sequential structure

The temperature dependence is specified in Figure 4 on page 6.

It shows a non-monotonic progression, which is nuts - but application engineers can get stuff wrong too.

You can autocalibrate them often enough to keep the temperature drift under control;

I used that in 1988, in the Cambridge Instruments multi-sampling electron beam tester. I'd used a Motorola ECL monostable in an earlier version of the machine, with a separate DAC to provide fine control of the delay and that had worked rather better. Having the DAC inside the AD9500 injected extra rail ripple where we couldn't filter it out.

It didn't matter - we were just using the AD9500 to control when the secondary electron sampling was turned on. The half nanosecond wide (or wider) primary electron sampling pulse was derived from an 800MHz clock edge via a nominally 500nsec ramp and comparator delay - the ramp was actually 1nsec long and we only looked at at the middle 500nsec.

I saw it on the newsgroup last night, and tried to respond to it, but the news group went off-line at that point, and I went to bed.

I've now responded to the content further down the thread.

I compared the AD9500 unfavourably with the MC10198 ECL monostable with an external DAC which I had used earlier. Motorola had dumped the MC10198 by 1987, much to my disgust.

It's a desperation measure to break up some kind of intellectual log-jam.

It isn't going to work all that often. Documenting it is a chore, but it does need to be done

Great for people who like visualisation.

By which you mean what? Tripping one circuit breaker trips associated circuit breakers? And the user can set up groups, all of which trip at the same time?

Perhaps not. They speak other languages in Siberia. It's not that hard to learn - I passed science Russian when I was a graduate student. I should have got it while I was still an undergraduate, but I got over-loaded in my final year as an undergraduate, and didn't need it to graduate (since I had already passed science German, and high school French was accepted as adequate for reading scientific papers, and the Nouveau Traite de Chimie Minerale), and repeated it the following year with better success.

Trump really can't be consistent enough to brainwash anybody. The problem with being as "transactional" as he is that it exposes the fact that what he really wants to do is to make a deal rather than achieve some specific target.

He may have recently adopted the goal of winning the Nobel Peace Prize, but that's been discredited since Henry Kissinger won it in 1973.

Or are literate.

Yes. Just like pinning a group of real breaker levers together.

The brainstorm session came up with a nice way to do it in the FPGA without elaborate state machines, and more important, a way to design a user interface that's easy to understand and won't provoke a flood of phone calls for support.

Here's the product:

I guess it could do primitive power metering too.

John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics

Look at fig 3. If the 512-delay block is 511, it won't be monotonic.

Look at fig 4.

If it had 1024 delay elements and 1024 switches, it would be monotonic. But otherwise just as bad.

My customers wouldn't appreciate that.

John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics

Joerg is real, and has been around here for many years. You should be aware of that.

He wants people who will flatter him. Competence is an optiona extra.

It would be much cheaper to have him assassinated, but that would have unfortunate side effects - you should be able to remember the side effects of Pim Fortuyn getting shot.

A nuclear fire-storm would do a great deal of damage. Trump and Putin are both quite demented enough to contemplate it but their associates aren't quite that silly. I'm pretty sure that the White House has a couple of psychiatrists on speed-dial so that they can certify Trump rapidly if he eventually goes completely insane. Putin's handlers would just shoot him.

If you did it right your customers would never notice. I figured that I could get it done in a couple of milliseconds, which surprised me when I ran the numbers.

When my customers trigger my gear, they expect it to work. Every time.

And if it's not monotonic, which it ain't, it can't be calibrated to 1 LSB.

There are much better ways to make delays.

John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics

Am 01.12.25 um 06:13 schrieb john larkin:

I have made one from six 1-to-6 SMA relays and enough semi-rigid SMA cables that I can bring everything in phase quadrature for phase noise measurements.

When it was all done, I got hold of a Miles Design Timepod that does all the magic digitally by cross correlation. They share a 19" box now.

Cheers, Gerhard

The illiterate can still look at pictures. Being able to write doesn't guarantee that you can organise your ideas into a chunk of text that other people will understand in the way that you want them to. Whiteboard scribbles aren't structured text.

A user interface which you hope will be easy to understand. The only way to be confident about that is find a bunch of naive users and test it on them. Engineers aren't naive users.

Then you need two swappable delay units, so that you can calibrate one while the user is using the other.

Real life customers can be more flexible than you seem imagine. Marketing hates talking to them about that sort of compromise - it makes their life more difficult - and tends to try to stop the engineers talking to customers, if higher management lets them get away with it. Tom Peters thought that letting engineers talk to customers was one of the features of an excellent business, and Analog Devices must have though the same. I once got to talk to Barry Gilbert at Cambridge Instruments.

It should be monotonic. It's possible that there's enough voltage noise inside the chip to wreck that, but any sensible development process should have been able to get rid of that.

Clearly the application engineers who wrote the data sheet didn't get monotonic delays, but they might have been able to set up a measuring set-up that was clean enough to deliver monotonic delays if they had been given the time to do the work more carefully.

And if it's not monotonic each step can still be calibrated to to better than 1 LSB, but then you have to re-order the calibration table to make it monotonic.

The data sheet claims that the jitter is around a picosecond, and the circuit is designed to offer a roughly 10psec LSB

Sure. But the MC100EP195 and MC100EP196 are pretty cute parts and a lot more compact and cheaper than most of the better ways I can think of.

The better way that I did get to work (back in 1990) involved using an

I had to use Gigabit Logic's GaAs counter chips back then to get 800MHz coarse clock. ECLinPS was being promised back then but you couldn't buy it.

A photo of an FPGA register map with notes is a serious document.

Given a group of smart people, a good brainstorm session is a way to add (or multiply) their intelligence.

But the personalities and the attitude have to be right. Some people can't brainstorm.

You attitude that most people are dumber than you are would (no doubt has) poison a brainstorm session. We've had junior people, even interns, see something or say something that inspired cool ideas. The audience must be grateful for all ideas, to encourage ideas.

"That violates conservation of energy, but there might be something there..."

A brainstorm session is a way for a group of people to explore an infinite solution space by using parallel processing.

John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics

Did you ever actually read the EP195 data sheet?

GBL logic didn't last long. Their chips were weird expensive power hogs. Essentially GaAs RTL.

John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics

They do not know this. They see a metal enclosure and assume it's almost like a Faraday cage for RF. Yet it isn't, because of a tiny detail in the surface processing. I have met scores of seasoned engineers who did not know this. Except in defense-related comanies, there everyone knew.

It didn't. I don't make a habit of telling people that they are dumb. If they tell me stuff that is demonstrably wrong I do try to get them to recognise that they have to change their opinion, but it's a lot easier to do that in a collaborative fashion in face-to-face encounters. Some people do dig their heels in, but that doesn't happen in a well-run brainstorming session.

That's a well-known feature of brain-storming sessions

No finite brain, nor any loosely linked collection of finite brains, is going to explore an infinite solution space.

Brainstorming is much less ambitious - it's just a mechanism for getting people to think of less obvious potential solutions to the problem that needs to be solved. There's a well known problem in multi-parameter curve fitting where the search algorithm latches onto a local minimum in the search space, and you can need to get the search process to check out a bunch of more or less arbitrary points to get a feel for how deep the minima may get.

Of course it can. It just can't find every possible solution.

Even one brain has a lot parallel processing capability. The advantage of brainstorming is that it randomizes searches to break out of conventional wisdom, namely local minima. That's why an intern can inspire a genius.

What's fun is to brainstorm a project that's well along, and maybe reboot the design. Sunk cost thinking is the enemy there.

Parts obsolescence isn't all bad.

John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics