negative supply

I dunno if I'll need synchronous rectification but programmable soft start is a nice feature

Oh and they also bring out the error amp output and inverting input for the compensation network that's nice of them.

I think its little cousin the LM3478 might be sufficient for this but you lose the programmable soft-start.

Do you mean the high current side? Sometimes the channel works as an input. The box can simulate or acquire synchro/resolver/RVDT/LVDT signals.

Au contraire. When a 545 equalled a Chevy, it was the best scope in the world, and the Chevy rusted out in five years.

The best scope in the world today costs quite a few Chevys, and those Chevys last a lot longer.

Of course I can get probably 20 TDS 694Cs (10 Gs/s simultaneous on four channels) for the price of a stripped Malibu.

Cheers

Phil Hobbs

And then there's this: I hope you had tissues ready to wipe up the drool.

Clifford Heath.

Technology is great! I'm guessing both have decreased as a fraction of average income. Better stuff for less money, what could be better?

George H.

What's that cost?

torsdag den 20. september 2018 kl. 03.02.14 UTC+2 skrev John Larkin:

in the

from the video description:

The 110GHz 4-channel variant of the UXR-Series oscilloscope has an MSRP of $1.3 Million US dollars.

Impressive. Even better than the 100 GHz Hypres scope from circa 1990 and the Picosecond Pulse Labs 100 GHz sampler.

Of course preserving that 110 GHz bandwidth requires soldering your circuit to the front of the scope, but never mind. ;)

Cheers

Phil Hobbs

The guy said $1.2M iirc.

That's a lot of Rigols. ;)

Cheers

Phil Hobbs

Clifford Heath wrote

Very impressive indeed.

Wonder how many iterations they had in designing those acquisition boards. And what sort of software for the layouts.

Indium phosphide???

That's close to DC-to-infrared.

I note the comment from user "Keysight Labs" that they offer student discounts.

I was thinking [tm] for Fourier transform (VFFT (very fast FT) ) that when you go that high, then a simple 10$ prism will also give you the FT. Put 30$ [IR capable] camera on it. For a bit below IR was it not gratings? Bunch of gratings with detectors behind it, maybe cheaper. Bit rusty on that, Martin Brown may know?

Then do a reverse FT? Oh well.. LOL

Students with very rich parents I suppose, but maybe a university lab could afford one.

Even at 60GHz it is worth putting an anti-reflection coating on the antennas :)

An old prof of mine, Bill Shuter (may his tribe increase) had a millimetre- wave radiotelescope back in the 70s. He was looking at interstellar giant m olecular clouds, investigating the dynamics of star formation via the emiss ion spectra of carbon-13 and carbon-12 monoxide near 110 and 115 GHz.

The hardware was a cooled paramp followed by an early GaAs FET IF amp at ar ound 20 GHZ, and then a block downconverter feeding a rack full of filter+d etector boards.

I did an undergraduate research assistantship with him, trying to understan d the weird line shapes observed. They were said to be "self-reversed", i.e . there were satellite peaks on either side of a smallish main peak. That w as apparently caused mainly by the outer regions of the cloud being opaque closer to the line centre. (Fun.)

So mm-wave technology has been around for quite awhile.

Cheers

Phil Hobbs

Oh indeed, but there's a little difference between something you can buy from a catalog, and hand-crafted one-off state of the research art!

There seems to be some rule-of-thumb that it takes 50 years to go from fundamental research to widespread availability :)

My only brush with mm-wave was in 1996, for data links in the 60GHz oxygen hole. 20dBi antennas were a thumb-sized ceramic(?) lens, with a vacuum-formed plastic layer acting as an anti-reflection "coating". Definitely not Kansas, for any trad radio engineer.

On Thursday, September 20, 2018 at 4:53:31 AM UTC-4, snipped-for-privacy@gmail.com wrot e:

e-wave radiotelescope back in the 70s. He was looking at interstellar giant molecular clouds, investigating the dynamics of star formation via the emi ssion spectra of carbon-13 and carbon-12 monoxide near 110 and 115 GHz.

around 20 GHZ, and then a block downconverter feeding a rack full of filter

and the weird line shapes observed. They were said to be "self-reversed", i .e. there were satellite peaks on either side of a smallish main peak. That was apparently caused mainly by the outer regions of the cloud being opaqu e closer to the line centre. (Fun.)

Fun, We had a CO2 pumped organic vapor laser... you'd put different molecules into the tube, I think it was ant piss* (formic acid) that had lines near 500 um. (20 cm-1) That's close to 100 GHz. I took some of my favorite data with that laser.

George H.

Transformers do have rather more adjustable parameters than inductors.

Custom design and winding isn't any more of a huge nuisance than getting a printed circuit board made, and in fact if you go for printed windings - which John Larkin doesn't seem to have discovered yet - it's pretty much the same process.

Mostly in places where it's a sub-optimal solution ...

The math is trivial.

Newark and Digikey stock a useful range of cores and matching formers - specialist suppliers stock more, and it's not that hard to find them.

Farnell in the UK stocked a useful range of enamelled copper wire, but I tended to buy single reels from this lot, who had a better range.

If you've got a coil winding machine (and small ones are cheap)it only takes a few minutes to wind a few hundred turns.

Once you've worked out what you want to wind and what you want to wind it around, the parts tend to next day delivery. A week would be generous - a month is total nonsense.

So go for printed windings. They depend on exactly the same gerber files as your printed circuit boards.

A professional winding shop winds to tight specs, and you can test the coils before you wrap cores around them. Resistances tend to come in +/-2% - due to the tolerance on the wire diameter. Air-cored inductance is going to be somewhat more reproducible.

Ferrite cores are well-controlled and at least as reproducible as transistors.

They clearly scare you out of your wits, but there are loads of products that depend on them

Cambridge Instruments had half a dozen of them, and their electron microscopes a sold in the hundred or so per year range.

But you can find coil winding sub-contractors. In the Netherlands I used

to make me a couple of prototype coils. I drove over to Horst with the coil formers, and they posted the wound coils back in a days or so.

It has gotten a lot easier and cheaper in recent decades.

Not all that cheap - too far north for illegal Mexicans.

But transformers aren't trimming potnetiometers. Many more adjustable parameters.