Author Topic: Very Fast Rise Time 5v Square Wave Generator  (Read 1045 times)

0 Members and 1 Guest are viewing this topic.

Offline iraquois

  • Contributor
  • Posts: 13
  • Country: gb
Very Fast Rise Time 5v Square Wave Generator
« on: September 29, 2020, 04:52:38 pm »
Hi everyone. I actually have two question. I am trying to design a circuit which is capable of generating 5v square wave which has rise and fall time of below 1ns and minumum pulse width is 40ns. I have seen some topics but they were for 1 or 2 volts. Do you have an idea or a chip that can be used for this purpose?


After this project I might try another project idea which is 250 V in 1 - 2 ns rise and fall time. For this purpose can avalanche topology work?
and if yes I might use ztx415 which has a base-emitter saturation voltage of 900mV.

*** If I find a chip that triggers in subnanoseconds with Vpp 1V, is it possible to use it with fast transformer like ztx415 to have 5v Vpp in again subnanoseconds? (maybe in this case it will be inverted)

*** I tried schmitt triggers but too much distortion I got and I got only 1ns

 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 16039
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Very Fast Rise Time 5v Square Wave Generator
« Reply #1 on: September 29, 2020, 05:54:38 pm »
Free running?  Triggered?  What repeat rate?

What logic did you try, 74LVC?  Handles 5V.  Maybe isn't quite under 1ns, or you need too many gates in parallel and then the skew kills it?

Potato Semi comes to mind, but it looks like they don't have a 5V model, only 3.3V.  (AFAIK it's basically "unlocked" CMOS, rated to run at stupid frequencies.)

I suppose you could still make do, using a couple of those but merging their outputs with a pulse transformer (aka 0° power divider).

As for avalanche, yes, that'll do, but you'll need very low loss coax for the delay line, to get nearly as sharp a falling edge.  Possibly some filtering or clamping could be done to deal with that.  You've got plenty of signal to play with, you can afford to burn a few dozen volts here or there with a rather lossy filter.

A lossy and nonlinear filter might be, for example: coupling the pulse through a capacitor, the end of which is charged slightly negative (say -10V).  100V pulse comes along, raising the end to +90V.  Use a series diode to direct the pulse into the output.  Pulse ends, dropping to 10V, 5, 2, etc. over some nanoseconds ("drool" due to line loss); well, the cap end drops to 0V, -5, -8, etc. at the same time -- the "drool" has been chopped off by the combination of diode and reverse bias.  Add some large value resistors to apply that bias, and you're done.


50 ohm coax is about 200m/us so you need at least 4 meters for the delay line.  There's no reasonable way to vary the line length (electrical trombone lines do exist, but, eh..).

ZTX415 are rated for avalanche, and are good when higher power is needed.  Regular types like MMBT3904 and MMBT2369 will also do, if less power is needed (which is certainly the case here).  You may even find it worthwhile shopping for lower voltage transistors; the savings being in less supply voltage required.  Maybe you can get away with a voltage multiplier, rather than an inductive supply.

All transistor types I've tested, will avalanche pulse at some operating condition (Vces, Ices and Rbe).  The main difference is how wide that range is (particularly Rbe).  I don't think I'd be confident enough to design a production device around a 3904 or 2369 abusing this property, but the ZTX415 and relatives are tested for this and can be designed in reliably.  If you can afford an adjustment step, or accept low production yield, then others will do.

Point being, I remember once testing a, something RF PNP, a salvaged Japanese (2SAxxxx) type, which I think was rated around 30V 100mA, and avalanched around 50-60V.  I think it even performed with B-E shorted (i.e., Rbe = 0), which is convenient.  So, if lower supply voltage is an option, consider selecting a few RF parts to test, and see how it goes.


Still other options -- consider a wideband RF amplifier.  This might be a PHEMT or whatever, in the usual bias circuit; or an MMIC; or something more complicated, say using GaN FETs, maybe coupling transformers, etc.  Generate a pulse with say ~100ps edges using something simple and fast, PECL for instance, and amplify it up to 5V.  The amplifier will be class A (more or less), so expect relatively high power dissipation.  (Transistors and MMICs capable of a few watts are readily available, this is no problem.)

The downside is, most of those perform best with a tuned load; they'll do 10GHz happily, but they won't do 20ps edges.  Some are better at pulses than others, and I forget which are best (PHEMTs, GaN FETs, SiGe HBTs..).  The relevant parameter is output capacitance, or its equivalent -- most of these parts have none of the conventional ratings and you have to interpret the s-parameters instead.  In that case, look for the phase and magnitude slope of s22, try to fit an equivalent capacitance to it.

You can of course get ready-made wideband amps, if you don't mind spending a few bucks.  Likely the best option for a one-off.

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 
The following users thanked this post: iraquois

Offline TurboTom

  • Frequent Contributor
  • **
  • Posts: 907
  • Country: de
Re: Very Fast Rise Time 5v Square Wave Generator
« Reply #2 on: September 29, 2020, 06:16:20 pm »
I got surprisingly good results with paralleled LVC TinyLogic inverters (i.e. two gates in each package) since this permits short and equal traces to interconnect the gates and good supply bypassing. I wasn't able to actually measure the rise time due a lack of proper equipment, and moreover, in the application  (Rb timebase) I didn't really bother. Since the scope that I used to measure the slopes has risetimes in the ballpark of 700ps, I'ld expect the LVCs to to produce slopes faster than 500ps. Here's the post I'm referring to.
 
The following users thanked this post: edavid, iraquois

Online chris_leyson

  • Super Contributor
  • ***
  • Posts: 1506
  • Country: wales
Re: Very Fast Rise Time 5v Square Wave Generator
« Reply #3 on: September 29, 2020, 07:44:13 pm »
TurboToms idea of using parallel CMOS buffers makes a lot of sense, you can parallel them into a resistive combiner to get whatever drive current you need and at the same time reduce capacitive loading on each output. There is of course a physical limit on how far you can reduce the capacitive loading and that is down to package geometry and layout. As for wiring up lots of parallel buffers you should use a pair of input and output delay lines to get the temporal matching right. Capacitance is a killer for rise and fall times, you need lots of drive current and therefore the parallel current drivers makes sense. Because of the parallel combined drivers you will also get a phase noise improvement over a single driver but that is going off topic.
EDIT: For 5V go for fast bidirectional CMOS drivers because translating unidirectional avalanche drivers is a real pain, see Tims post. As for 250V maybe thyratrons but I don't think they would be fast enough to get the <1ns rise time. At work I use 1kV/ns EFT pulse generators for EMC compliance testing, I would love to find out  how they do that but I can't take the covers off GRRR.
« Last Edit: September 29, 2020, 08:49:01 pm by chris_leyson »
 
The following users thanked this post: iraquois

Offline iraquois

  • Contributor
  • Posts: 13
  • Country: gb
Re: Very Fast Rise Time 5v Square Wave Generator
« Reply #4 on: September 29, 2020, 07:55:45 pm »
Free running?  Triggered?  What repeat rate?
Actually only thing that is improtant is the rise time and the voltage. Repeat rate is 30 per second. So it is not important for that speed i think.

I suppose you could still make do, using a couple of those but merging their outputs with a pulse transformer (aka 0° power divider).

You mean with potato cmos chips right? Actually I am not experienced enough. So can you describe what you mean exactly. Or I can google it if you tell me what to look for.


A lossy and nonlinear filter might be, for example: coupling the pulse through a capacitor, the end of which is charged slightly negative (say -10V).  100V pulse comes along, raising the end to +90V.  Use a series diode to direct the pulse into the output.  Pulse ends, dropping to 10V, 5, 2, etc. over some nanoseconds ("drool" due to line loss); well, the cap end drops to 0V, -5, -8, etc. at the same time -- the "drool" has been chopped off by the combination of diode and reverse bias.  Add some large value resistors to apply that bias, and you're done.
As I understand it will be very hard for me to do it with avalanche. I didn't even understand half of your words :))

and thanks a lot it was very helpful I will try it with RF amplifiers also.
« Last Edit: September 29, 2020, 07:59:09 pm by iraquois »
 

Offline iraquois

  • Contributor
  • Posts: 13
  • Country: gb
Re: Very Fast Rise Time 5v Square Wave Generator
« Reply #5 on: September 29, 2020, 07:56:40 pm »
I got surprisingly good results with paralleled LVC TinyLogic inverters (i.e. two gates in each package) since this permits short and equal traces to interconnect the gates and good supply bypassing. I wasn't able to actually measure the rise time due a lack of proper equipment, and moreover, in the application  (Rb timebase) I didn't really bother. Since the scope that I used to measure the slopes has risetimes in the ballpark of 700ps, I'ld expect the LVCs to to produce slopes faster than 500ps. Here's the post I'm referring to.

Thank you. I will look at that as soon as possible :)
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 12058
  • Country: us
  • DavidH
Re: Very Fast Rise Time 5v Square Wave Generator
« Reply #6 on: September 29, 2020, 08:31:00 pm »
Hi everyone. I actually have two question. I am trying to design a circuit which is capable of generating 5v square wave which has rise and fall time of below 1ns and minumum pulse width is 40ns. I have seen some topics but they were for 1 or 2 volts. Do you have an idea or a chip that can be used for this purpose?

AC and LVC logic can achieve that, or come close, but driving a significant load, like a 50 ohm transmission line, will mean using many gates in parallel which can be a layout challenge.  The layout will be a challenge no matter what technology is used though.

A higher performance option is to use faster 3.3 volt logic to drive the emitter of an RF bipolar transistor which does the level shift to a higher voltage, but this will require multiple supply voltages.  Reference level pulse generators often work this way and the output level can be variable.

Quote
After this project I might try another project idea which is 250 V in 1 - 2 ns rise and fall time. For this purpose can avalanche topology work?
and if yes I might use ztx415 which has a base-emitter saturation voltage of 900mV.

Absolutely that is possible.  Zetex even wrote an application note showing how which you might want to search for.
 
The following users thanked this post: iraquois

Online chris_leyson

  • Super Contributor
  • ***
  • Posts: 1506
  • Country: wales
Re: Very Fast Rise Time 5v Square Wave Generator
« Reply #7 on: September 29, 2020, 08:56:05 pm »
Quote
The layout will be a challenge no matter what technology is used though.
Use a pair of transmission lines, Tektronix did it over 50 years ago
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 12058
  • Country: us
  • DavidH
Re: Very Fast Rise Time 5v Square Wave Generator
« Reply #8 on: September 29, 2020, 09:23:10 pm »
Quote
The layout will be a challenge no matter what technology is used though.

Use a pair of transmission lines, Tektronix did it over 50 years ago

Where it mattered, Tektronix had access to hybrid construction and parts with dual base and collector connections so that the inputs and outputs could be placed inline to a transmission line.  But you can get below 1 nanosecond even using through-hole parts so modern surface mount parts should make it easy.
 

Offline 0culus

  • Super Contributor
  • ***
  • Posts: 2647
  • Country: us
  • Electronics, RF, and TEA Hobbyist
Re: Very Fast Rise Time 5v Square Wave Generator
« Reply #9 on: September 30, 2020, 01:18:20 am »

snip

50 ohm coax is about 200m/us so you need at least 4 meters for the delay line.  There's no reasonable way to vary the line length (electrical trombone lines do exist, but, eh..).

snip


Slightly off topic, but your comment here made me wonder...I've never seen one in person, but I wonder how Tek achieved variable delay line in the 7A29 with the variable delay line option?
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 16039
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Very Fast Rise Time 5v Square Wave Generator
« Reply #10 on: September 30, 2020, 03:55:44 am »
Can't find a picture of its insides, but it appears to be a fairly large module, with a ten turn dial, and only a 1ns range -- 20 cm or so.  Seems likely it's a trombone (or several) driven by a coarse thread. :-+

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 
The following users thanked this post: 0culus

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 12058
  • Country: us
  • DavidH
Re: Very Fast Rise Time 5v Square Wave Generator
« Reply #11 on: October 01, 2020, 06:19:51 pm »
Tekwiki has a picture of the module inside of the plug-in:

http://w140.com/tekwiki/wiki/File:Tek_7a29_left.jpg

And another photograph showing scale better here:

https://www.amplifier.cd/Test_Equipment/Tektronix/Tektronix_7000_series_amplifier/amplifier_7A29.htm
 
The following users thanked this post: 0culus

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 16039
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Very Fast Rise Time 5v Square Wave Generator
« Reply #12 on: October 01, 2020, 06:21:17 pm »
Yeah, I saw those but not what's under the cover. :-//

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline 0culus

  • Super Contributor
  • ***
  • Posts: 2647
  • Country: us
  • Electronics, RF, and TEA Hobbyist
Re: Very Fast Rise Time 5v Square Wave Generator
« Reply #13 on: October 02, 2020, 12:19:04 am »
Thanks...I REALLY want a pair of those with the delay line for my 7104. My 7A29s do not have the option...but they seem to never come up for sale and I'll bet they are expensive.
 

Offline 0culus

  • Super Contributor
  • ***
  • Posts: 2647
  • Country: us
  • Electronics, RF, and TEA Hobbyist
Re: Very Fast Rise Time 5v Square Wave Generator
« Reply #14 on: October 02, 2020, 12:20:06 am »
Yeah, I saw those but not what's under the cover. :-//

Tim

We need to find someone with one for a teardown. Anyway, not to derail the OP but the discussion of a variable delay line made me think of this. :)
 

Online Weston

  • Regular Contributor
  • *
  • Posts: 102
  • Country: us
Re: Very Fast Rise Time 5v Square Wave Generator
« Reply #15 on: October 02, 2020, 01:17:08 am »
Try using a GaN gate drive chip, they are 5V nominal, take logic level inputs, and can drive quite low impedance loads.

The TI LMG1020 is rated for a 400ps rise and fall time into a 100pF load https://www.ti.com/product/LMG1020 . It is in a WLCSP package that can be difficult time solder though.

There is a DFN variant, LMG1025 , that is only mildly slower 650ps rise and 850ns fall time into a 220pF load. I think the only difference between the two parts is the packaging actually, so the  LMG1025 should have a faster rise time into a higher impedance load.

« Last Edit: October 02, 2020, 03:00:11 am by Weston »
 
The following users thanked this post: iraquois

Offline iraquois

  • Contributor
  • Posts: 13
  • Country: gb
Re: Very Fast Rise Time 5v Square Wave Generator
« Reply #16 on: October 02, 2020, 06:21:08 am »
Try using a GaN gate drive chip, they are 5V nominal, take logic level inputs, and can drive quite low impedance loads.

The TI LMG1020 is rated for a 400ps rise and fall time into a 100pF load https://www.ti.com/product/LMG1020 . It is in a WLCSP package that can be difficult time solder though.

There is a DFN variant, LMG1025 , that is only mildly slower 650ps rise and 850ns fall time into a 220pF load. I think the only difference between the two parts is the packaging actually, so the  LMG1025 should have a faster rise time into a higher impedance load.



I saw those. But they need an external pulse. So if my trigger pulse is slow so will the output. Afaik I still need fast pulser even if I but this. Am I right?
 

Offline 2N3055

  • Super Contributor
  • ***
  • Posts: 3034
  • Country: hr
Re: Very Fast Rise Time 5v Square Wave Generator
« Reply #17 on: October 02, 2020, 08:47:14 am »
Try using a GaN gate drive chip, they are 5V nominal, take logic level inputs, and can drive quite low impedance loads.

The TI LMG1020 is rated for a 400ps rise and fall time into a 100pF load https://www.ti.com/product/LMG1020 . It is in a WLCSP package that can be difficult time solder though.

There is a DFN variant, LMG1025 , that is only mildly slower 650ps rise and 850ns fall time into a 220pF load. I think the only difference between the two parts is the packaging actually, so the  LMG1025 should have a faster rise time into a higher impedance load.



I saw those. But they need an external pulse. So if my trigger pulse is slow so will the output. Afaik I still need fast pulser even if I but this. Am I right?

It has Schmitt-Trigger input, so it will switch faster than input, but you would want to have input reasonably fast to minimize jitter.
But even 74CHT  series will be fast enough to drive input...
 

Offline JoeyG

  • Contributor
  • Posts: 44
  • Country: au
Re: Very Fast Rise Time 5v Square Wave Generator
« Reply #18 on: October 02, 2020, 08:57:52 am »
So fundamentally  a very low impedance (resistance) overcomes system/component/pcb capacitance.  as  tr~ RC.
 

Online tggzzz

  • Super Contributor
  • ***
  • Posts: 12796
  • Country: gb
    • Having fun doing more, with less
Re: Very Fast Rise Time 5v Square Wave Generator
« Reply #19 on: October 02, 2020, 09:02:52 am »
I got surprisingly good results with paralleled LVC TinyLogic inverters (i.e. two gates in each package) since this permits short and equal traces to interconnect the gates and good supply bypassing. I wasn't able to actually measure the rise time due a lack of proper equipment, and moreover, in the application  (Rb timebase) I didn't really bother. Since the scope that I used to measure the slopes has risetimes in the ballpark of 700ps, I'ld expect the LVCs to to produce slopes faster than 500ps. Here's the post I'm referring to.

See https://www.eevblog.com/forum/testgear/show-us-your-square-wave/msg1902941/#msg1902941 for an example.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline joeqsmith

  • Super Contributor
  • ***
  • Posts: 6875
  • Country: us
Re: Very Fast Rise Time 5v Square Wave Generator
« Reply #20 on: October 02, 2020, 11:08:50 am »
Is it for a 50ohm system?  Could you provide details on the load?   Is the 5Vp-p measured across the load?  Can you use a differential system?   

How electrically robust is your meter?? https://www.youtube.com/channel/UCsK99WXk9VhcghnAauTBsbg
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf