Author Topic: EEVblog #306 - Jim Williams Pulse Generator  (Read 294425 times)

0 Members and 1 Guest are viewing this topic.

Offline free_electron

  • Super Contributor
  • ***
  • Posts: 8515
  • Country: us
    • SiliconValleyGarage
Re: EEVblog #306 - Jim Williams Pulse Generator
« Reply #25 on: July 06, 2012, 12:12:57 am »
Here's a quick layout ( still needs work ... )
Small case using a cr2032 coin cell. Basically the same schematic from AN47.
I have a 5mm hole in 6mm pad in the PCb to solder the TO18 body upside down. The collector of this transistor is connected to the case ( i cut one open and doublechecked as well as ohmed it out ) So the case of the transitor gets the 1M /2pF node

Soldermask has been removed in that area to avoid leakage.

The base leg of the transistor is bent down to an SMD pad that goes with 10K to ground. The emitter is bent towards the SMA connector. A 1206 50 resistor is mounted on-end on an SMD pad. The emitter leg lies on top of the other electrode of the 1206. Two dabs of solder and you have a perfect flying lead line there. No leakage anywhere and mechanically robust.

Fits in a little case with a slide on/off switch and LED that tells you its on.

I could not really find Male BNC plugs that solder directly on PCB.  And the high end scopes have SMA inputs... the SMA end launch connector is the right height to pull off the trick with the 1206 standing on end.
So compromise : screw on an SMA to BNC adapter and you can plug this straight into a 'low end' scope input.
for the really speed deamons : use an SMA-SMA male-male connector.
« Last Edit: July 06, 2012, 12:16:27 am by free_electron »
Professional Electron Wrangler.
Any comments, or points of view expressed, are my own and not endorsed , induced or compensated by my employer(s).
 

Offline w2aew

  • Super Contributor
  • ***
  • Posts: 1780
  • Country: us
  • I usTa cuDnt speL enjinere, noW I aR wuN
    • My YouTube Channel
Re: EEVblog #306 - Jim Williams Pulse Generator
« Reply #26 on: July 06, 2012, 03:28:39 am »
Hey tnt - do you have any more of the PCBs for this pulse generator?
YouTube channel: https://www.youtube.com/w2aew
FAE for Tektronix
Technical Coordinator for the ARRL Northern NJ Section
 

Offline vk6zgo

  • Super Contributor
  • ***
  • Posts: 7563
  • Country: au
Re: EEVblog #306 - Jim Williams Pulse Generator
« Reply #27 on: July 06, 2012, 04:19:46 am »
Looking at the shape of the pulse,it looks a lot like the sin^2 pulses we used to test TV system video response.
Back in the day,they used rectangular pulses,but problems with ringing made them less than perfect for the job.
The sin^2 pulse provides a much more repeatable test result.
There was a fair bit written up  about this by J.E.Weaver,& others.
The relationships are different to simple rise time,but that's about all I can remember about the niceties of it all,at this remove.
 

Offline tnt

  • Regular Contributor
  • *
  • Posts: 241
Re: EEVblog #306 - Jim Williams Pulse Generator
« Reply #28 on: July 06, 2012, 05:49:55 am »
@free_electron: Nice assembly :) Looking forward to see it built.

@Lukas: For fun ? There is still much to be learned from how this circuit work IMHO (even just from the stepup for eg, that uses a trick to overcome the limitation of the stepup controller, or from the constructions techniques involved here, ...). Also, doesn't PECL have a pretty limited swing ? Some circuits might behave differently in small signal vs large signal (here the pulse is ~ 10V in 50 ohm).
 

Offline free_electron

  • Super Contributor
  • ***
  • Posts: 8515
  • Country: us
    • SiliconValleyGarage
Re: EEVblog #306 - Jim Williams Pulse Generator
« Reply #29 on: July 06, 2012, 06:17:57 am »
Pecl is about 800mV and is differential. Single ended it swings only 400mV. And does. Ot go to zero. Centerpoint is 1.65 volt.

The JW pulser does 10 volts or so...
I'm going to tweak the layout a bit more over the weekend. I need to move the 2pF cap so one electrode touches the body of the to. I will place additional cap footprints in a circle around the to-18 body so we can play with the cap value.

Will send pcb to itead to run
Professional Electron Wrangler.
Any comments, or points of view expressed, are my own and not endorsed , induced or compensated by my employer(s).
 

Offline BravoV

  • Super Contributor
  • ***
  • Posts: 7547
  • Country: 00
  • +++ ATH1
Re: EEVblog #306 - Jim Williams Pulse Generator
« Reply #30 on: July 06, 2012, 06:44:17 am »
Just a noob question regarding PECL vs Avalanche type's slew rate, I thought avalanche type range is in GigaVolt/second which is much-much higher than PECL ?

CMIIW

Offline blackdog

  • Frequent Contributor
  • **
  • Posts: 736
  • Country: nl
  • Please stop pushing bullshit...
Re: EEVblog #306 - Jim Williams Pulse Generator
« Reply #31 on: July 06, 2012, 07:45:19 am »
Hi Free_electron

Cant you make the emittor shorter en use 2x 100 Ohm parr on the output, maybe less ringing?

Just 2 cents..

Bram
Necessity is not an established fact, but an interpretation.
 

Offline EEVblogTopic starter

  • Administrator
  • *****
  • Posts: 37664
  • Country: au
    • EEVblog
Re: EEVblog #306 - Jim Williams Pulse Generator
« Reply #32 on: July 06, 2012, 10:16:26 am »
Where is this app note that shows the coax mod? anyone know the number?

Dave.
 

Offline jahonen

  • Super Contributor
  • ***
  • Posts: 1052
  • Country: fi
Re: EEVblog #306 - Jim Williams Pulse Generator
« Reply #33 on: July 06, 2012, 10:19:19 am »
Where is this app note that shows the coax mod? anyone know the number?

Dave.

AN94 seems to have one, at least.

Regards,
Janne
 

Offline eV1Te

  • Regular Contributor
  • *
  • Posts: 186
  • Country: se
  • Your trusted friend in science!
    • richardandersson.net
Re: EEVblog #306 - Jim Williams Pulse Generator
« Reply #34 on: July 06, 2012, 11:36:35 am »
Does anyone know good tunneling diodes for achieving <100 ps rise times? The only tunnel diode I found at my local electronics store was the 1N3716, although this particular diode does not seem to be optimized for fast rise times (but I could not find typical values for the tunneling speed in the datasheet).

I also found the shortcut way to making ~40 ps rise/fall-time generator with the Silicon Germanium ADCMP580 comparator (built in 50 ohm resistor and drive capability)  :P
 

Offline Neganur

  • Supporter
  • ****
  • Posts: 1138
  • Country: fi
Re: EEVblog #306 - Jim Williams Pulse Generator
« Reply #35 on: July 06, 2012, 12:38:02 pm »
You could try these from ebay:

1I308J tunnel diodes 2pcs

They're supposedly russian Germanium tunnel diodes with 20mA peak point current and 1...4 pF valley point terminal capacitance. Less than 100 ps could be possible with these. Fast tunnel diodes are very rare, and very expensive.
 

Offline vk6zgo

  • Super Contributor
  • ***
  • Posts: 7563
  • Country: au
Re: EEVblog #306 - Jim Williams Pulse Generator
« Reply #36 on: July 06, 2012, 01:09:39 pm »
The old Tektronix TU-5 Pulser used a tunnel diode.
We used these to check our 545B 'scopes back in the day.
They had to be fed with a high voltage square wave--The 545B Cal signal was switchable up to 100volts.
When the tunnel diode operated,it produced a very fast rise time leading edge on a very attenuated version of the input sq wave.
I don't remember how fast it was--after all we were testing 545Bs with it,so it didn't have to be that spectacular!
 

Offline eV1Te

  • Regular Contributor
  • *
  • Posts: 186
  • Country: se
  • Your trusted friend in science!
    • richardandersson.net
Re: EEVblog #306 - Jim Williams Pulse Generator
« Reply #37 on: July 06, 2012, 01:18:19 pm »
You could try these from ebay:

1I308J tunnel diodes 2pcs

They're supposedly russian Germanium tunnel diodes with 20mA peak point current and 1...4 pF valley point terminal capacitance. Less than 100 ps could be possible with these. Fast tunnel diodes are very rare, and very expensive.
Interesting, it seems like tunnel diodes are quite common on ebay, you can even find ones with <1 pF terminal capacitance for just a few dollars each. Unfortunately I don't have access to equipment capable of measuring the performance if I would try and build a step generator with them.

It would also be interesting to compare the performance of the pulse generator to a mechanical switch (preferably a mercury switch) which apparently also can go as low as 70 ps.

What is the rise-time if one mechanically touches the legs of a charged small ceramic cap to the input of the scope?

The old Tektronix TU-5 Pulser used a tunnel diode.
We used these to check our 545B 'scopes back in the day.
They had to be fed with a high voltage square wave--The 545B Cal signal was switchable up to 100volts.
When the tunnel diode operated,it produced a very fast rise time leading edge on a very attenuated version of the input sq wave.
I don't remember how fast it was--after all we were testing 545Bs with it,so it didn't have to be that spectacular!

I've seen tunnel diode step generators that have rise-times below 10 ps, but more common is around 30 ps. Unfortunately the amplitude you get is only as large as the tunneling zone (~300 mV).
 

Offline vk6zgo

  • Super Contributor
  • ***
  • Posts: 7563
  • Country: au
Re: EEVblog #306 - Jim Williams Pulse Generator
« Reply #38 on: July 06, 2012, 01:43:46 pm »
I found the details of the TU-5 here-:www.diybanter.com/.../4131d1223498570-tek-tu-5-td-pulser-tu-5-p..
It is quite a lot slower at 1.5nS.
It's still good for about 233mHz,unless I got my sums wrong!
OK for the 545Bs,which If I remember correctly,were supposed to be 35Mhz at the -3dB point,although they could see about twice that.
I remember looking at CH2 Vision & sound carriers with one,at different times.
 

Offline tecman

  • Frequent Contributor
  • **
  • Posts: 444
  • Country: us
Re: EEVblog #306 - Jim Williams Pulse Generator
« Reply #39 on: July 06, 2012, 03:09:40 pm »
Here's a quick layout ( still needs work ... )
Small case using a cr2032 coin cell. Basically the same schematic from AN47.

Fits in a little case with a slide on/off switch and LED that tells you its on.

Free_electron:

If you finish it up are you thinking about selling a few of them ?  I would buy one since I do not want the hassle of working from scratch.


Dave:  what about the design you have ?  Are they available for purchase from someone ?


I would be interested in one for the bench.


paul
« Last Edit: July 06, 2012, 03:12:42 pm by tecman »
 

Offline free_electron

  • Super Contributor
  • ***
  • Posts: 8515
  • Country: us
    • SiliconValleyGarage
Re: EEVblog #306 - Jim Williams Pulse Generator
« Reply #40 on: July 06, 2012, 03:28:13 pm »

Free_electron
If you finish it up are you thinking about selling a few of them ?  I would buy one since I do not want the hassle of working
[/quote]
I'll have spare boards yes. I can sell blank boards.
Professional Electron Wrangler.
Any comments, or points of view expressed, are my own and not endorsed , induced or compensated by my employer(s).
 

Offline samarkh

  • Contributor
  • Posts: 13
Re: EEVblog #306 - Jim Williams Pulse Generator
« Reply #41 on: July 06, 2012, 03:50:25 pm »
Could you show this being used as a tdr (Time Domain Reflexometer )? To show cable impedance mismatch / breaks / shorts and the like?

Yours Simon M.
 

Offline free_electron

  • Super Contributor
  • ***
  • Posts: 8515
  • Country: us
    • SiliconValleyGarage
Re: EEVblog #306 - Jim Williams Pulse Generator
« Reply #42 on: July 06, 2012, 04:02:21 pm »
For tdr you really need a step as opposed to a pulse.... But i guess a pulse could work too as long as the reflected energy does not come back before the pulse has decayed
Professional Electron Wrangler.
Any comments, or points of view expressed, are my own and not endorsed , induced or compensated by my employer(s).
 

Offline robrenz

  • Super Contributor
  • ***
  • Posts: 3035
  • Country: us
  • Real Machinist, Wannabe EE
Re: EEVblog #306 - Jim Williams Pulse Generator
« Reply #43 on: July 06, 2012, 04:39:05 pm »
Could you show this being used as a tdr (Time Domain Reflexometer )? To show cable impedance mismatch / breaks / shorts and the like?

Yours Simon M.

w2aew's video on that

Offline Pat Pending

  • Regular Contributor
  • *
  • Posts: 161
  • Country: us
Re: EEVblog #306 - Jim Williams Pulse Generator
« Reply #44 on: July 06, 2012, 04:45:46 pm »
I didn't hear it mentioned, but was Daves Agilent scope set for 50R input?
 

Offline Lukas

  • Frequent Contributor
  • **
  • Posts: 412
  • Country: de
    • carrotIndustries.net
Re: EEVblog #306 - Jim Williams Pulse Generator
« Reply #45 on: July 06, 2012, 07:16:48 pm »
I don't think that 10Vpp output is desirable. The maximum vertical scale of my fastest 'scope (7A19 in 7834) is 1V/div. Some very fast 'scopes such as the Agilent 90000A series have an input voltage range of only ±5V. All in all, around 400mVpp is more appropriate today than 10V. Of course, you may use attenuators, but that's another thing in the signal path that may decrease signal integrity.
 

Offline Alexei.Polkhanov

  • Frequent Contributor
  • **
  • Posts: 684
  • Country: ca
Re: EEVblog #306 - Jim Williams Pulse Generator
« Reply #46 on: July 06, 2012, 09:58:29 pm »
You could try these from ebay:

1I308J tunnel diodes 2pcs

They're supposedly russian Germanium tunnel diodes with 20mA peak point current and 1...4 pF valley point terminal capacitance. Less than 100 ps could be possible with these. Fast tunnel diodes are very rare, and very expensive.
I got 6 of those from same seller on ebay 2 months ago for laser distance measuring project that I was planning. I checked one diode for basic diode action when I got the package - it is fine, but I have a problem however - I need a holder for these tiny devices. Germanium ... most likely not solderable at all. No clue where I can get nice holders. I soldered a needle to a small pad and tried to use it - looks like I need something better than that and preferably gold plated. Also I am scared to touch them even with wrist band on - they must be uber sensitive to ESD.
 

Offline G7PSK

  • Super Contributor
  • ***
  • Posts: 3859
  • Country: gb
  • It is hot until proved not.
Re: EEVblog #306 - Jim Williams Pulse Generator
« Reply #47 on: July 07, 2012, 08:56:50 am »
You could try these from ebay:

1I308J tunnel diodes 2pcs

They're supposedly russian Germanium tunnel diodes with 20mA peak point current and 1...4 pF valley point terminal capacitance. Less than 100 ps could be possible with these. Fast tunnel diodes are very rare, and very expensive.
I got 6 of those from same seller on ebay 2 months ago for laser distance measuring project that I was planning. I checked one diode for basic diode action when I got the package - it is fine, but I have a problem however - I need a holder for these tiny devices. Germanium ... most likely not solderable at all. No clue where I can get nice holders. I soldered a needle to a small pad and tried to use it - looks like I need something better than that and preferably gold plated. Also I am scared to touch them even with wrist band on - they must be uber sensitive to ESD.

Woods metal. It melts at little over blood heat  70 deg C and was often used for mounting germanium in the past. There are alloy metal that melts even lower.   http://en.wikipedia.org/wiki/Wood's_metal
« Last Edit: July 07, 2012, 09:00:40 am by G7PSK »
 

Offline PChi

  • Frequent Contributor
  • **
  • Posts: 264
  • Country: gb
Re: EEVblog #306 - Jim Williams Pulse Generator
« Reply #48 on: July 07, 2012, 10:46:59 am »
I have just tried using an old SGS BC549C VCBO and VCEO = 30 V maximum which I happened to have in a copy of the avalanche part of the circuit. It didn't avalanche at 120 V which is where I ran out of courage. So just confirming that the transistor type and particular sample are important.

Another factor to consider is that the formula relating rise time and bandwidth assumes that there is no large signal / slew rate limitation.

I'm not sure how much I trust digital oscilloscopes near their bandwidth limitation. Older models had the ability to turn off the curve smoothing and just show the raw samples on the screen. In the single sweep mode it was informative to show the sample points and then to show the smoothed curve.
 

Offline max-bit

  • Frequent Contributor
  • **
  • Posts: 668
  • Country: pl
Re: EEVblog #306 - Jim Williams Pulse Generator
« Reply #49 on: July 07, 2012, 10:48:00 am »
Hi This is my first post :) Gentlemen, if you want I can arrange a lot of transistors 2n2369A Production Motorola and Philips (years of production 80 - 90) Are desoldering Motorola, Philips is not soldered Price 1.5 USD plus shipping and handling.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf