Author Topic: Fastest DIY pulse generator? 20ps?  (Read 9794 times)

0 Members and 1 Guest are viewing this topic.

Offline dcarrTopic starter

  • Regular Contributor
  • *
  • Posts: 117
Fastest DIY pulse generator? 20ps?
« on: August 23, 2017, 03:41:12 am »
Simple question: What's the fastest pulse generator that can be (home)built without custom silicon?

I've got a Tektronix SD-24 that can do ~20ps.  How close can one get to that?  Can it be matched?

The closest I've seen so far is ~50ps using a NBSG11.  It seems like snap recovery diodes might have some promise but I'm not sure about DIY availability.  Perhaps some of the very fast bipolar transistors (eg: BFR740) could be useful in a standard or avalanche configuration?  Non-linear transmission lines (shockwave lines) seem very interesting, but I don't think those can be assembled from discrete components with the wavelengths we're talking about.

David
 

Offline karkoon

  • Contributor
  • Posts: 35
  • Country: in
Re: Fastest DIY pulse generator? 20ps?
« Reply #1 on: August 23, 2017, 10:02:55 am »
This video by w2aew shows how to generate a fast pulse with rising edge of 2ns.

https://youtu.be/9cP6w2odGUc

Doesn't know any better way. Very curiously watching this thread. :)


Sent from my iPhone using Tapatalk
 

Offline peter.mitchell

  • Super Contributor
  • ***
  • Posts: 1567
  • Country: au
 

Offline Tom45

  • Frequent Contributor
  • **
  • Posts: 555
  • Country: us
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 16544
  • Country: us
  • DavidH
Re: Fastest DIY pulse generator? 20ps?
« Reply #4 on: August 23, 2017, 01:49:38 pm »
There are some CML (common mode logic), aka ECL (not actually ECL, my mistake), comparators with internal terminations which can generate 10s of picosecond edges but their outputs are not particularly clean.  I was thinking of mounting a tunnel diode into a coaxial cavity like the Tektronix 284 pulse generator does.

Both of these designs point to the immediate problem; the unterminated stub length between the output device and termination limits the high frequency performance.  I estimate that except for this, you should be able to get below 100 picoseconds using NPN RF transistors.  In old Tektronix oscilloscope designs above 200 MHz, they got around this by running two bond wires to the base and collector of the transistor so it could be inserted between the input and output signals and their terminations but I do not know of any modern parts packaged like this.

Unless you want to go the CML comparator route mentioned above and for a printed circuit board design, I think the way to go is a coplanar waveguide implementation of the NBS (National Bureau of Standards which is now NIST) reference level pulse generator design which is very similar to the Tektronix PG506 but I have not tried it yet.  These designs use a series diode in the output to disconnect the transistor from the 50 ohm termination so the most critical part of the circuit also has the simplest layout.
« Last Edit: August 25, 2017, 01:32:48 pm by David Hess »
 
The following users thanked this post: evb149

Offline Leo Bodnar

  • Frequent Contributor
  • **
  • Posts: 801
  • Country: gb
Re: Fastest DIY pulse generator? 20ps?
« Reply #5 on: August 23, 2017, 05:48:30 pm »
What is the pulse amplitude and width you are aiming for?
 The devil is usually in the details.  You can make a fast pulse but struggle to get it out cleanly.
I have achieved about 30-40ps for 1Vpp 10-90% thresholds out of BNC plug. The fall time is better than the rise time.
https://www.eevblog.com/forum/projects/yet-another-fast-edge-pulse-generator/
And if you bend the rules and move the goalposts to 20-80%, yes, you can have 20ps :)

I'd say it's reasonably easy to have two out of three:
1) risetime
2) amplitude
3) pulse width

Non-linear transmission lines (shockwave lines) seem very interesting, but I don't think those can be assembled from discrete components with the wavelengths we're talking about.
They can and have been, but their performance is not that spectacular and only of academic interest (literally.)  And you still need reasonably clean and fast edge to send down them.
http://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0149645&type=printable

Looks like NLTLs are coming out of vogue in favour of photonics (lasers + APDs) recently.  I remember reading somewhere that PSPL who pretty much pioneered ASIC NLTLs are not that keen on making them.  This is third hand unreliable information so don't hold me to it.

Going below 20-30ps is like approaching absolute zero K - it gets exponentially more difficult and you need a solid reason and motivation to keep going.

Leo
« Last Edit: August 23, 2017, 06:07:31 pm by Leo Bodnar »
 

Offline dcarrTopic starter

  • Regular Contributor
  • *
  • Posts: 117
Re: Fastest DIY pulse generator? 20ps?
« Reply #6 on: August 25, 2017, 05:29:54 am »
Wow, it looks like we have some real pulse generator gurus on this forum.

I checked on the SD-24 and the specified TDR amplitude is 250mv (and 28ps 10%-90%).  It sounds like the easiest solution to get close is to try a high speed comparator, but I'm intrigued by some of the other approaches mentioned here too.

Based on Leo's results, I wonder if an ADCMP572 with an optimized transition to an edge launch SMA could get down <28ps 10/90%.  It's definitely in the ballpark.  Are there other similar comparators that are worth a look?

Update: I'm note sure the '572 is what Leo was using in the screenshot that shows ~30ps fall time.  Perhaps it's a MAX3949?

David

« Last Edit: August 25, 2017, 06:44:23 am by dcarr »
 

Offline Leo Bodnar

  • Frequent Contributor
  • **
  • Posts: 801
  • Country: gb
Re: Fastest DIY pulse generator? 20ps?
« Reply #7 on: August 25, 2017, 08:21:23 am »
I checked on the SD-24 and the specified TDR amplitude is 250mv (and 28ps 10%-90%).  It sounds like the easiest solution to get close is to try a high speed comparator, but I'm intrigued by some of the other approaches mentioned here too.

Update: I'm note sure the '572 is what Leo was using in the screenshot that shows ~30ps fall time.  Perhaps it's a MAX3949?
Yes, these results are for MAX3949, I have switched the ICs before finalising the design.

I have tried SD-24 yesterday and it shows 27-28ps risetime (10-90%) for its own incident pulse and 24ps risetime when it reaches SD-30.
This sort of suggests SD-24 TDR pulse being around about 22ps risetime for this particular unit give or take connecting hardware.

Leo
« Last Edit: August 25, 2017, 08:33:46 am by Leo Bodnar »
 

Online Marco

  • Super Contributor
  • ***
  • Posts: 6692
  • Country: nl
Re: Fastest DIY pulse generator? 20ps?
« Reply #8 on: August 25, 2017, 08:43:19 am »
This is the fastest edge on the forum I know off, at least the trailing edge. I assume the less impressive leading one is due to flaws in the differentiator.

https://www.eevblog.com/forum/projects/pulse-generator-150ps-1vpp/

Paper with a little more info.

http://www.adv-radio-sci.net/2/7/2004/ars-2-7-2004.pdf
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf