MMIC filter

Mouser also wants $740 a pop. Newark doesn't even carry it and for the ADMV8818 they want $560. That's a lot of bucks. Nearly all my designs have to cost a small fraction of that and it includes everything down to the shipping carton.

When clicking on "Sample & Buy" at the manufacturer nothing comes up:

:-)

Regards, Joerg Designer of "It can't be done" Electronics

Designing and building LC filters is a nuisance, especially if you want to make them out of parts that you have, or can easily get.

I just did a 3-pole lowpass as a DDS filter, 20-40 MHz operating range, 125 ohms in and 10K out. I started with my PowerBasic bessel filter designer and fiddled from there.

That's for my 50-cent 32-bit DDS trigger generator.

Does that cost more?

John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics

At very high frequencies old tricks of the trade come back in. Caps and inductors as traces, little wire stubs that get bent over "just so" to align something, etc.

I usually eyeball it and then massage it on SPICE. It's not a highly scientific method but works.

Which would become a $500 DDS trigger generator if you use the AD filter :-)

To the surprise of my clients it's the contrary. The most client-shocking redesign was an auto-align circuit for ganged ADC-channels. High speed, high phase accuracy and all that. They had an elaborate time domain method with a fat DSP, lots of code and very expensive chips used as programmable delay chips. The NRE alone had been humongous. It never reliably converged so the system hung a lot. I suggested to ditch all that and use time domain. This caused an uproar because I had rocked the boat a lot and usually consultants aren't supposed to do that. "I don't think this can possibly work", "It won't deliver the accuracy", "It won't converge either" and all that.

Yet the boss let me do it. In the end the whole thing dropped from three-digit dollars in HW to under 10 bucks. Instead of expensive discrete-step time delay chips I used inductors, caps and varicap diodes for almost infinite granularity. The DSP became unemployed because the connected PC could easily handle the computations. It converged in less than a second, always. The NRE was low because it took less than two weeks of my time and less than a day for the programmer, and we didn't need an expensive DSP programmer.

I meant they used a frequency domain method.

Embarrassing. Were any of the customers design team later defenstrated?

Joe

No, they were pretty good. It's the usual phenomenon where, in an old German saying, you can't see the forest because of all the trees. Frequency domain, FFT and all that is the classical approach but consultants can help companies see other more unorthodox methods. Most of my design jobs boil down to that. The other jobs are more boring, failed EMC tests and such.

What is the saying in German?

Rendered in English: "can't see the forest for the trees".

Same here. Actually, some of the failures were due to subtle EMI/EMC issues causing failures due to unstable timing, by an unobvious mechanism.

Joe

Is that why you are concerned about tracking?

Oh man, that's the ultimate prepper :-)

It's nice but I like flashlights with Li-Ion better. They can be charged back up even with wimpy power sources. Usually even while staying lit.

Careful ...

Real men don't need fire starters.

"Man kann den Wald vor lauter Baeumen nicht sehen". In Business it is often called "Betriebsblindheit", the thinking that it has always been done this way so why should we consider a change now?

Thanks. For some reason I only remembered the Spanish version "no ver el bosque por los arboles".

Interleaved ADC's ?

LeCroy made one scope where they made a bank of frequency-domain down-converters that covered something like 100 GHz in maybe 5G bands, digitized each one, and then spliced it back together mathematically.

It sounds like a nightmare, and was.

We do picosecond-resolution delays with simple RCs and comparators. Works fine.

There are some programmable silicon-delay-line chips. They are expensive and basically worthless.

John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics

Am 29.11.25 um 04:12 schrieb john larkin:

In a previous life I used a dozen or 2 of AD9500 to control the phase delays in a phased array to sweep the direction of the ultrasonic beam for scanning the inner enclosure of nuclear reactors.

Worked nicely. In the lab, it was set up by a HP-85 desk top computer via HP-IB.

Gerhard

Am 29.11.25 um 04:12 schrieb john larkin:

In a previous life I used a dozen or 2 of AD9500 to control the phase delays in a phased array to sweep the direction of the ultrasonic beam for scanning the inner enclosure of nuclear reactors.

Worked nicely. In the lab, it was set up by a HP-85 desk top computer via HP-IB.

Gerhard

Brainstorming is designed to get around that to some extent, but if you aren't used to thinking outside the box it's difficult to step back far enough to get outside the box.

Frequently poached from other companies that the consultants have helped. It's not about stealing industrial secrets - just starting off from a different place.

There's nothing boring about EMC.

The On-Semi MC100EP195 and MC100E196 are about $30 each in small quantities, about half that in volume.

The delay is temperature dependent and you have to autocalibrate frequently (which wouldn't have been difficult), or stabilise the temperature.

They didn't strike me as worthless, but tastes differ.

There was the MC100EP195 and the Micrel equivalent. I played with those. The MC wasn't monotonic on delay and had a lot of jitter. Both had bad (and unspecified) tempcos. Expensive too.

They were switched-tap silicon delay lines.

They might make sense at really high data rates where ramps wouldn't work and some temperature drift would be OK.

The AD9500 is a ramp/comparator design that's at least monotonic. It's tempco is 7.5 ps/degc typ. It's a Rochester part now.

John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics

Hi, that post of mine never showed up here in the newsgroup. Nor did a retry. Did you get it as personal mail? In the sent-box it is addressed to s.e.d.

Weird. Gerhard

Am 29.11.25 um 16:40 schrieb john larkin:

I saw it in s.e.d. and replied in the usual way. I guess you couldn't see your own post for some reason.

John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics

My experience with brainstorm sessions is not good. The results are often encouraging but then hardly anything of it gets documented and typically none of itv is implemented. All I need is a large whiteboard or a large piece of paper. Plus coffee or mate (having a mate right now).

That I make darn sure won't happen here. All client files are strictly separated and each client's stuff has its own shelf. No crossings allowed.

After a dozen times of telling people not to use anodized aluminum for gear that needs to shield and they did it anyhow it does become boring. Then, explaining for the umpteenth time that split grounds do not work and that pigtail ground connections have close to zero effect at VHF and up.

It showed up twice for me. Not seeing your own or other people's posts or only seeing a header but no text body happened a lot to me over the last year. Not sure if it's the FU Berlin server I am using. However, that will be shut down in 2027 anyhow so no need to diagnose much.

That's exactly what I found. I never saw much value in them versus a homebrew solution. On the contrary, they were single-sourced and that presents a great risk to the business, especially nowadays.

I now design more things with discretes than ever before, not trusting the supply chain anymore. Plus it's usually cheaper and with SC75 and