High speed pulse generator to test oscilloscope

Greetings I have a Tek TDS694C I will be selling, and I'd like to demo it's high speed performance by showing it detecting a rise time - I'm trying to remember exactly what that is - somewhere in picosecnd range?

Is there a chip which will show a fast rise time? I believe they use some kind of high speed fpga?

thanks in advance jb

Reply to
haiticare2011
Loading thread data ...

It's something like, 1/3 or 1/pi times 1/bandwidth, so 1GHz is 300ps of something.

There's a nice J. Williams article about fast edges/ pulses AN47? (maybe?)

George H.

Reply to
George Herold

An old Philips 74LVC04 chip can do about 250ps risetime:

formatting link

(It is Philips not TI as mentioned in the video.)

Unfortunately that particular board didn't have sufficient low-frequency decoupling (it was intended for use at much higher frequencies where the decoupling was adequate), so the power supply cable inductance rings badly with the small decoupling capacitors on the board. If he had looked at the fall-time instead of rise time, and left my ac-coupling and back termination in, it might have been a bit cleaner. The best option would have been to improve the local low frequency decoupling, with some big caps.

Chris

Reply to
Chris Jones

John Larkin is the expert here. When I had to do something similar - a long time ago - I used a well-specified step-recovery diode from HP.

If you could find one in a low-inductance - which is to say, a small circuit mount - package, and mount it in a transmission line environment, above a good ground plane, you might have got down to a 100psec risetime.

John seems to do better. Some ECLinPS parts are supposed to be in the same ball-park, and will drive 50R terminated transmission lines, as will a step recovery diode.

--
Bill Sloman, Sydney
Reply to
Bill Sloman

Well, the real answer of course is to hang onto the TDS694C, and get an

11801C and SD-24 TDR head to test it with. ;)

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

hmmmm What is the tds64c god for, practically? jb

Reply to
haiticare2011

Yeah, the TDR step (and the calibrator output) are metrology-class rising edges, in the 30 ps range.

A lot of cmos chips will make pretty clean edges in the ballpark of 500 ps.

NL37WZ16US is great, 11 cents, but the US8 package is nasty.

DS90LV012ATMF is basically a r-r comparator with a 500 ps output. SOT23, reasonable to solder down on some hacked copperclad.

The Onsemi Gigacomm parts can make really sweet edges down around 50 ps, but need to be handled very carefully, definitely not x-acto knife stuff.

You can reasonably breadboard with EclipsLite and EclipsPlus parts

formatting link

225 ps typ rise/fall for that EL07.
--
John Larkin                  Highland Technology Inc 
www.highlandtechnology.com   jlarkin at highlandtechnology dot com    
 Click to see the full signature
Reply to
John Larkin

I guess a fall time would also count. How high is "high speed"? A simple BFR92 driven hard can give you fall times in the 100psec range. That would be a part that could already be in a parts bin somewhere. The BFS17 is good as well but it depends which kind you have.

--
Regards, Joerg 

http://www.analogconsultants.com/
Reply to
Joerg

It's my favourite general-purpose high speed scope. I have a couple of

500-MHz ones, which are about the fastest that still have the 1 Mohm input option, but for anything I can use 50-ohms for, or can hang a FET probe on, the 694C is the proverbial bee's knees. 125 ps rise time, no overshoot, whee!

Mine has the 2Msample memory as well, which is a win, and a nice bright display.

Sampling scopes require many triggers per waveform (512 in the case of the 11801 series), and can't look backwards in time before their triggers. (That's why the 11802 comes with two coiled hardline delay lines in the mainframe.)

Sampling scope triggering is also external-only, which takes a bit of futzing around if you're not doing stimulus-response type measurements.

The 11801 will give you a factor of 8 faster response than that with the ordinary plugins, and a factor of nearly 20 with the fastest ones.

That's fast enough that you have to put the sampling head right at the circuit, which is where the extender cables come in.

Cheers

Phil "Scope junkie" Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

formatting link
formatting link

-Lasse

-Lasse

Reply to
Lasse Langwadt Christensen

FPGA's are slugs. You'd do better with a discrete solution, or even a mercury-wetted pulser (and with a length of good coax cable, that makes a nice Blumlein pulse generator).

Reply to
whit3rd

2M was not available in the TDS600 series. The maximum memory was 120K per channel for the TDS694C (option 1M) The TDS694X was the only one that had any memory options.

What's interesting is that the TDS684C acquisition board is laid out for larger memory chips - I've always wondered what would happen if I were to populate them...

Reply to
JW

Thanks Lasse, I was thinking of this one where he (J.W.) does a pulser from a mercury reed relay near the end... maybe the diode turn on/ off time app note. here,

formatting link
AN94 (taming of the slew) is nice too.

George h.

Reply to
George Herold

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.