Author Topic: Recommendations for a fast analog scope  (Read 2089 times)

0 Members and 1 Guest are viewing this topic.

Offline rhbTopic starter

  • Super Contributor
  • ***
  • Posts: 3476
  • Country: us
Recommendations for a fast analog scope
« on: June 18, 2018, 03:47:38 pm »
After some "interesting" experiences with a DSO, I've decided I'd like to get a fast analog scope.  The Tek 2465B seems to fit the bill at $300 on eBay, but I'm not at all familiar with it.  All my Tek experience was before that came out (e.g. 465, 475 & 485).

I'm looking for something which will be reliable and serviceable should the need arise.
 

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 26755
  • Country: nl
    • NCT Developments
Re: Recommendations for a fast analog scope
« Reply #1 on: June 18, 2018, 03:54:24 pm »
Look for a 470MHz Lecroy LA314H (Iwatsu SS-7847A). This is a reasonably modern analog oscilloscope. I used to own one over a decade ago. The only downside is that it has two attenuation settings and no 50 Ohm mode for channels 3&4.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline Floopy

  • Regular Contributor
  • *
  • Posts: 157
  • Country: us
  • I'm going to need another oscilloscope to fix that
Re: Recommendations for a fast analog scope
« Reply #2 on: June 18, 2018, 04:05:53 pm »
I like the Tektronix 2000 series, but it all depends on what you do.
« Last Edit: June 18, 2018, 06:40:10 pm by Floopy »
 

Online T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 21609
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Recommendations for a fast analog scope
« Reply #3 on: June 18, 2018, 04:08:02 pm »
As not answering the question goes:

Have you tried HP/Agilent/Keysight DSOs?  They're way more responsive than the trash Tek made over the same era.  Classic example, the 54600, though you want one with a bit more bandwidth it seems; I forget which of the series you're looking for.

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

Offline rhbTopic starter

  • Super Contributor
  • ***
  • Posts: 3476
  • Country: us
Re: Recommendations for a fast analog scope
« Reply #4 on: June 18, 2018, 04:56:39 pm »
This was prompted by my experience with an MSOX3104T which I am returning.  I've got 3 DSOs at present, but my analog scopes are socketed transistor units which are way too much work to maintain.  Also they are 60 & 100 MHz dual trace.

I'd get a 7000 series mainframe, but I just don't have the space.  I specifically want an analog scope which I can adjust to have minimal overshoot.  I thought the 3-4% overshoot of my MSO-2204EA  was bad.  Then I saw the 7% overshoot of the MSOX3104T.

I want something that I *know* I can trust.  A soldered transistor 485 would suit me fine, but I don't know how to determine that a unit is soldered.  In fact, now that I think about it, that's probably what I want.  It has no digital parts unlike later models.
 

Offline KrudyZ

  • Frequent Contributor
  • **
  • Posts: 274
  • Country: us
Re: Recommendations for a fast analog scope
« Reply #5 on: June 18, 2018, 05:06:29 pm »
I have a LeCroy LA354, which is a very interesting scope.
It has an internal CRT and then uses a camera to take pictures of the CRT to display on an LCD.
This yields an analog scope with 500 MHz bandwidth with the ability to display single shot events at full brightness in a small portable box.
I also have an 11032 which was the last hurrah of Tektronix analog scopes. Very capable unit with an excellent built-in frequency counter.
It's a bit on the big and heavy side though.
 

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 26755
  • Country: nl
    • NCT Developments
Re: Recommendations for a fast analog scope
« Reply #6 on: June 18, 2018, 05:21:29 pm »
As not answering the question goes:

Have you tried HP/Agilent/Keysight DSOs?  They're way more responsive than the trash Tek made over the same era.  Classic example, the 54600, though you want one with a bit more bandwidth it seems; I forget which of the series you're looking for.
OTOH the Tektronix TDS500/700 series DSOs have more memory and color displays. Somehow these are more appealing to me based on specs compared to HP scopes. I've owned a couple of these as well (510A, 644A -briefly due to no peak detect- and the 744A). Then again I like the Agilent 54845A I currently have for high frequency measurements even though it won't trigger beyond 1GHz (who thought that was a good idea at Agilent?  :palm: ).
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline Hydron

  • Frequent Contributor
  • **
  • Posts: 978
  • Country: gb
Re: Recommendations for a fast analog scope
« Reply #7 on: June 18, 2018, 05:33:03 pm »
You may find that your 7% overshoot issue is due to a maximally-flat rather than the "traditional" gaussian response.

I'm not 100% certain, but I recall from another thread that the 1GHz 3000T series scope uses a maximally-flat filter response to push the bandwidth a bit closer to nyquist without aliasing issues, and that the lower-bandwidth 3000T models have a gaussian response which should give lower overshoot.

A bit of research may be in order (I saw R&S talking about this exact issue in some of their marketing fluff trying to sell their gaussian response scopes).
 

Offline rhbTopic starter

  • Super Contributor
  • ***
  • Posts: 3476
  • Country: us
Re: Recommendations for a fast analog scope
« Reply #8 on: June 18, 2018, 05:57:41 pm »
I understand *why* the overshoot is there.  I can draw the Fourier transforms on a napkin after several rounds of drinks in a bar.

Other than meeting some marketing weasel's demands, why would you build a scope with 7% (or even 3%) overshoot?  What is the user going to gain? You're not giving the user an accurate representation of the signal. You're giving them a distorted signal so the scope can be sold as something it's really not capable of doing.

For narrowband signals you can easily correct the loss in amplitude of a Gaussian input response.  You've got to sample the waveform and ship it off to a PC to correct for the effects of excessively sharp input frequency attenuation. It is precisely for very broadband signals that amplitude fidelity is important.  A step response is the edge case.  It doesn't get worse.

I want a a fully  analog reference scope which is reliable and can be adjusted to have *no* overshoot on a fast edge.
 

Offline chris_leyson

  • Super Contributor
  • ***
  • Posts: 1541
  • Country: wales
Re: Recommendations for a fast analog scope
« Reply #9 on: June 18, 2018, 07:05:02 pm »
Quote
I'd get a 7000 series mainframe, but I just don't have the space
Was forced to sell all my TEK 7000 gear and in a way I'm glad, they take up far too much space and need constant attention to keep them running. Some of this gear could be 40 years old and so don't expect reliability, and BTW all the transistors are socketed.

Quote
Classic example, the 54600, though you want one with a bit more bandwidth it seems; I forget which of the series you're looking for.
I have a 54610B which has a quoted BW of 500 MHz and measured BW of 750MHz but only 20MSa/s, has some nice features. The later 54616C is 500MHz BW and 2GSa/s. https://www.keysight.com/en/pc-1000002066%3Aepsg%3Apgr/discontinued-54600-series-oscilloscopes?nid=-32416.0.00&cc=GB&lc=eng
 
The following users thanked this post: T3sl4co1l

Online Fungus

  • Super Contributor
  • ***
  • Posts: 16562
  • Country: 00
Re: Recommendations for a fast analog scope
« Reply #10 on: June 18, 2018, 07:51:41 pm »
I want a a fully  analog reference scope which is reliable and can be adjusted to have *no* overshoot on a fast edge.

That's almost impossible; a lot of the overshoot you see comes from the probe cable, the circuit you're probing, etc.

 

Offline rhbTopic starter

  • Super Contributor
  • ***
  • Posts: 3476
  • Country: us
Re: Recommendations for a fast analog scope
« Reply #11 on: June 18, 2018, 08:06:00 pm »
I'm referring to a pulse input at the BNC.  I've got a 36 pS rise time 10 MHz pulser from Leo Bodnar.  What I want is for the ringing to be the cabling and DUT.
 

Online Fungus

  • Super Contributor
  • ***
  • Posts: 16562
  • Country: 00
Re: Recommendations for a fast analog scope
« Reply #12 on: June 18, 2018, 08:17:18 pm »
I'm referring to a pulse input at the BNC.  I've got a 36 pS rise time 10 MHz pulser from Leo Bodnar.  What I want is for the ringing to be the cabling and DUT.

I'm not sure that "analog oscilloscope" is a magic solution to that.

why would you build a scope with 7% (or even 3%) overshoot?

You wouldn't deliberately build one but people's jobs are usually "Make an oscilloscope for $XXX", not "make the best oscilloscope possible with no price limit".

Making an oscilloscope front end with zero capacitance and zero inductance is expensive.

(some might say "impossible").
 

Offline rhbTopic starter

  • Super Contributor
  • ***
  • Posts: 3476
  • Country: us
Re: Recommendations for a fast analog scope
« Reply #13 on: June 18, 2018, 09:04:39 pm »
I've repaired and calibrated a couple of analog scopes, a 60 & a 100 MHz.  In the adjustment procedure, you input a fast rise time pulse (I used a Tek 106) and you adjust one or more trimmers for best rise time and least overshoot.  It's been 20 years so I don't recall how many adjustments each front end has.   But it's not rocket science.  Generally the Tek calibration procedures specify no more than 1-2% overshoot IIRC.  But someone like me is free to choose less if they don't mind a slower rise time.  The adjustment is a compromise between overshoot and rise time.

You *cannot* make an AFE with zero capacitance and inductance.  You *do* make AFEs with inductors and capacitors chosen to give the desired BW.  There's a damn good reason there are not a lot of 1 GHz analog scopes and not many above 350 MHz.

An analog scope is not a magic solution to anything.  But a calibrated analog scope is not subject to FW bugs.  And it's very easy to feed a known signal into a test fixture to verify the response.  That's what so nice about Leo's little gadget.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf