This is a Fairchild FSA3157 analog mux.
It's impressive, with switch rise/fall times below 500 ps, if you ignore the fine points. But it's too ugly for the application that I had in mind. And around 5 ns pulse width on S, the output pulse gets weird.
It would be fine for slow switching, 8 ns or so.
-- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
What are you hanging on it? That 5 ns sag looks roughly right for some nasty reflection arriving from the other end of a cable.
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
looks like a little reflection (transmission line match problems)
Jamie
Puh-leeze! I don't do nasty reflections.
I used a 470 ohm resistor from the mux output, then a coax to my 50 ohm scope, to make a roughly 10:1 passive probe. It looks like there's about 2 ns of break-before-make in the mux, and it goes hi-z there. So the 520r load discharges the capacitances during that 2 ns.
If I use a higher impedance load, the output pulse is nearly flat (well, there are some modest charge injection bumps) and the rise and fall get radical, below 300 ps.
I wanted to use this to make adjustable-amplitude fast pulses, which it will apparently do as long as I switch it for > 5 ns. That's too slow for my current app, so it won't work. But I thought it was interesting... the data sheets don't characterize switch rise/fall times. It's amazing the kinds of speeds you can get from 10 cent CMOS parts these days.
I guess I'll have to do something else.
-- John Larkin Highland Technology, Inc picosecond timing laser drivers and controllers jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
Well, you didn't post a schematic, and I thought all that bling might have gone to your head. ;)
I use a similar TI part (TS5A23157DGSR) in lock-ins, and they work great.
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
The scope shot has the schematic, on the little yellow Post-It. I can never keep track of which scope shot is which, so I comment them with stickies.
I redid the test with an SD14 (hiZ 3 GHz sampling probe) and got this:
The USB switches are actually analog mux's. TS3USB30EDGSR for instance, 900 MHz analog bandwidth.
I got a bunch of different "3157" type parts to test, but since they all have break-before-make behavior, there's no point in testing them all just now.
So, to make a fast, clean, pulse with variable width and height, I need something else. Maybe an analog multiplier. Maybe a PHEMT pulling a voltage to ground. Maybe some diode steering thingie.
TEK and HP tended to do this, in their faster pulse generators, with custom ICs.
ADG604 is a nice signal-type mux. 280 MHz analog BW, +-5 supplies, and
1 pC. Only 20x the price.-- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
That's the kind of thing you get with too much inductance in the leads/pads or the 2V source line, it's a less than critically damped overshoot.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.