O'scope won't sync

[mojoe]

I was trying to adjust a 10 Mhz OCXO against my GPDSO, with the usual configuration - GPSDO on channel 1 and OCXO on channel 2, trigger on 1.

Using my SDS2304X, I could not get the OCXO waveform to stabilize, so that I could adjust it. If I swapped trigger, then the GPSDO waveform was all over the place. Using Auto or Normal made no difference.

When I dug out my old SDS1102CML, everything worked as it should. Triggering on the GPSDO, I had a stable, slowly scrolling OCXO waveform.

Why would the newer, more expensive scope not work for this simple application?

[bob91343]

Digital oscilloscopes are, generally, a solution to a nonexistent problem.  I only use mine for special applications; the analog one is the daily driver.  I have two digitals on the bench but only use them for fun.  I use the XY mode on my Tek to compare the rubidium standard with the counter oscillator and it's a pleasure.  The digital scopes just don't do it.  Well, they sort of do but it's a pain, as you are finding out.

Even the almost basic of old analog scopes can do what you want.

[james_s]

Digital oscilloscopes are, generally, a solution to a nonexistent problem. 

Huh? On what planet? I still keep my trusty Tek 465B around but it's mostly out of nostalgia. My DSO can do so many things the analog scope flat out can't do, and it can do almost everything better. The ONLY thing the analog scope really truly excels at is XY mode. There's a reason DSOs have absolutely dominated the market for >20 years and I haven't seen an analog scope used in a company lab throughout my career.

For what it's worth, my TDS3000 works perfectly to compare frequencies with two oscillators. I've never used a SDS2304X so I can't speak for that specifically but I suspect this is a case of user error, either that or there is something unusual about that particular model scope.

[george.b]

Digital oscilloscopes are, generally, a solution to a nonexistent problem.

I know, right? I mean, for example, why go through all the hassle of digitizing a waveform into a convenient file format for documentation/later use/reference, when you can just take a picture of the scope screen? With a film camera, preferably - because, you know, digital cameras are a solution to a nonexistent problem.

 

[DaJMasta]

Without knowing the full details of how it was setup or seeing how it was reacting, it could be any number of things.  The obvious choice would be some sort of overlooked setting for triggering: level, type, channel, some potential memory or analysis options, persistence, etc. , and the alternative is that your scope's triggering is somehow broken.

As you say, it is a relatively simple task for any scope, so my assumption would be that there was some missed or overlooked setting that has a bearing on the way its triggering and displaying unuseful information as a result.



No idea how it turned to digital scope hate, and so quickly...

[bob91343]

Well guys don't get me wrong.  It does seem that one needs to fuss more and know more to get a proper display on a digital scope.  For easy tasks it's overkill.  Sure there are lots of functions missing on analog units but why drive a Ferrari to the grocery store?  Most of what I do is within the capability of analog and its setup is faster and more intuitive.

Plus I don't have to wait for self test (with the fear it won't pass).  And for many uses the analog is faster to respond and display.

[Anthocyanina]

This sounds like they have significantly different frequencies, and that oscilloscope doesn't have alt trigger, so only one of the wafeworms will look stable. The older siglent you have does have alt trigger so that's probably why you can see both waveforms properly on that one. Without alt trigger, if both waveforms aren't time correlated and are not the exact same frequency, the one you're not triggering on will move on the screen, how fast depends on how much it differs from the one you're triggering on, and to which side it's moving will depend on if it's a higher or a lower frequency.

[james_s]

Well guys don't get me wrong.  It does seem that one needs to fuss more and know more to get a proper display on a digital scope.  For easy tasks it's overkill.  Sure there are lots of functions missing on analog units but why drive a Ferrari to the grocery store?  Most of what I do is within the capability of analog and its setup is faster and more intuitive.

Plus I don't have to wait for self test (with the fear it won't pass).  And for many uses the analog is faster to respond and display.

You have to know how to use the instrument whether analog or digital, I have certainly seen plenty of people incorrectly using an analog scope. I think you are just already skilled with the analog scope and forgetting all the ways to do it wrong. A digital scope isn't much harder, it's just different, it's a different tool that needs different techniques, and it's more powerful which makes it more complex but for basic measurements it isn't any harder. The TDS3000 takes about as long to boot as my analog scope's CRT takes to warm up, and much of the time I can touch the probe to the point I want to test and hit the Auto button and it sets itself up.

I find I use single shot capture a lot, that's something an analog scope just can't do unless you use a scope camera.

[Fungus]

Plus I don't have to wait for self test (with the fear it won't pass).

And we don't have to wait for the CRT to warm up with the fear that this might be the final power-on before those old capacitors finally give up.

Plus the wiggly line is never rotated with respect to the grid.

[2N3055]

I was trying to adjust a 10 Mhz OCXO against my GPDSO, with the usual configuration - GPSDO on channel 1 and OCXO on channel 2, trigger on 1.

Using my SDS2304X, I could not get the OCXO waveform to stabilize, so that I could adjust it. If I swapped trigger, then the GPSDO waveform was all over the place. Using Auto or Normal made no difference.

When I dug out my old SDS1102CML, everything worked as it should. Triggering on the GPSDO, I had a stable, slowly scrolling OCXO waveform.

Why would the newer, more expensive scope not work for this simple application?

What was timebase ?
Trigger settings ? Did you use alternate trigger on SDS1102CML ? If you did that was wrong, and new scope shows how it is..

You shouldn't ever write anything like this without detailed explanation of settings and what and how was measured. Your SDS2304X also  has 5 times faster horizontal timebase, and faster rise time.... You will see things you couldn't with SDS1102CML, that were always there, but you couldn't see them.

Please, if you need useful answer, provide more technical details... Otherwise, you will only get random ramblings, speculations and non topic related rants.. Nothing useful.

[2N3055]

Well guys don't get me wrong.  It does seem that one needs to fuss more and know more to get a proper display on a digital scope.  For easy tasks it's overkill.  Sure there are lots of functions missing on analog units but why drive a Ferrari to the grocery store?  Most of what I do is within the capability of analog and its setup is faster and more intuitive.

Plus I don't have to wait for self test (with the fear it won't pass).  And for many uses the analog is faster to respond and display.

It is called progress. Digital scopes are smaller, use less electricity and are many times more powerful. In comparison they have huge color screens (yay, traces are different colour, so you know which one is which!). It needs a bit of learning to be able to use it to full power. If you just want to use basic things, it is as simple and fast as an analog scope. In this story analog scope is a Ferrari (they used to be much more expensive than digital scopes) and only because nobody except few die-hards is using them any more for anything serious, the prices for what today 40 years after it's prime is basically an electronic waste, is still more than some new quite capable small scopes.
Pretty much everything you say is an statement from 20 years ago, when it pretty much was true. Well, not anymore..

While katana was an technological marvel 300 years ago, and is a piece of art to be admired today, a testament to human ingenuity, bringing katana to a modern war would be gross mistake. 

[tautech]

I was trying to adjust a 10 Mhz OCXO against my GPDSO, with the usual configuration - GPSDO on channel 1 and OCXO on channel 2, trigger on 1.

Using my SDS2304X, I could not get the OCXO waveform to stabilize, so that I could adjust it. If I swapped trigger, then the GPSDO waveform was all over the place. Using Auto or Normal made no difference.

When I dug out my old SDS1102CML, everything worked as it should. Triggering on the GPSDO, I had a stable, slowly scrolling OCXO waveform.

Why would the newer, more expensive scope not work for this simple application?

Because your older CML has as 2N3055 points out an Alt trigger.

So how to work around this, I suggest a frequency measurement for each channel and use of XY mode to make the adjustment to get a match.

[mojoe]

I thought I did provide the needed information. I powered on each scope, with the inputs as listed above, and hit Auto. That should give me the default settings, with trigger on channel 1. I didn't change anything else, except to try triggering on channel 2.

[mojoe]

I appreciate the help. It seems that most of the replies are off topic, to say the least. I will try X-Y mode.

[tautech]

I appreciate the help. It seems that most of the replies are off topic, to say the least. I will try X-Y mode.

First you'll need to know how far off your OCXO is from 10 MHz and then adjust it as close as you can and the scope would benefit from being connected to your GPDSO.
Once you have the 2 close then use XY mode.

[nctnico]

I appreciate the help. It seems that most of the replies are off topic, to say the least. I will try X-Y mode.

Also make sure the samplerate is high enough (for 10MHz > 30Msps or in time/div: <= 50ns/div) . With a 10MHz sine wave it is easy to get an aliased signal which looks OK but is the result of mixing the oscilloscope's own oscillator with the external signal. IMHO that is the only explaination for what you are experiencing. And try to make it a habit of setting up your oscilloscope manually instead of using the auto button or at least check the vertical V/div and horizontal time/div settings.

[Wallace Gasiewicz]

This is just my limited experience with a digital scope. Agilent 54810A. It is 500 MHz now.
On low freq it is hard to sync, if you use the filters to rid yourself of high freq, it does MUCH better.
I think much of the perceived problem is today's digital scopes, in general, have much higher bandwidth then the analog scopes I am used to.
So there is more opportunity for the scope to trigger (attempt to trigger?) on an unwanted higher freq. Also shows the higher freq crap that I was happy to be ignorant of.
I finally learned to use XY on my digital. Not too bad once you get used to the menus.
The Lissajous figure looks fantastic in magenta.

Wally

[bob91343]

I never got a satisfactory X-Y display on my digital scopes.  Is there a trick?  I spent plenty of time fussing with them and finally got a display but it wasn't good for zero beating two oscillators.  Meanwhile the old analog scope did it with ease.

Yes I may be living in the past but that's not always a bad thing.  I have friends who are far worse than I in this respect.

[nctnico]

I never got a satisfactory X-Y display on my digital scopes.  Is there a trick?  I spent plenty of time fussing with them and finally got a display but it wasn't good for zero beating two oscillators.

Didn't had that problem with an Agilent DSO and that was about 20 years ago. The trick is to catch at least a full period of the slowest clock cycle of the signals you are comparing. This is a samplerate versus memory depth problem. So make sure you have at least a full cycle of both signals on screen and then switch to XY mode. While in XY mode, stay clear of the time/div knob unless you know what you are doing.

[tautech]

I never got a satisfactory X-Y display on my digital scopes.  Is there a trick?  I spent plenty of time fussing with them and finally got a display but it wasn't good for zero beating two oscillators.  Meanwhile the old analog scope did it with ease.

Yes, typically reducing the mem depth to allow the scope to process X-Y faster however unless the DSO has some moderately fast processor other than for simple frequency beats it's gunna struggle.

Some modern DSO's provide a quite useful X-Y mode:

[David Hess]

Using Auto or Normal made no difference.

Switching to normal triggering mode was going to be my first suggestion because the auto triggering mode timeout may be shorter than 1 second.  Are you sure that normal mode triggering works properly?  Maybe it is broken, or has always been broken.  It would not surprise me because most people will never use normal mode triggering.

Also try testing it with single sweep triggering.

[james_s]

I never got a satisfactory X-Y display on my digital scopes.  Is there a trick?  I spent plenty of time fussing with them and finally got a display but it wasn't good for zero beating two oscillators.  Meanwhile the old analog scope did it with ease.

Yes I may be living in the past but that's not always a bad thing.  I have friends who are far worse than I in this respect.

I use XY mode to test vector arcade game boards and the XY function on every DSO I've tried is completely useless for that. That's the one single thing I can still think of where an analog scope is far superior.

[bdunham7]

Because your older CML has as 2N3055 points out an Alt trigger.

Something else has to be going on with his setup, you don't need ALT trigger to sync two oscillators.  In fact it won't work that way.  The only things I can think of is that his update rate is way slow or some setting is in a strange mode.

I don't know how well X-Y is going to work on that scope, but at 10MHz there aren't many analog scopes that will be of any use.  Typically, CRO or DSO, to sync two oscillators you do it just the way he describes--two channels, trigger on the reference channel, the other one drifts.

[Fungus]

I never got a satisfactory X-Y display on my digital scopes.  Is there a trick?

The problem is usually your memory depth. Make it as small as possible.

Even so: Most people don't base their purchase decision on XY mode. There's a saying about not throwing out the baby with the bath water that applies here.

[nctnico]

I never got a satisfactory X-Y display on my digital scopes.  Is there a trick?

The problem is usually your memory depth. Make it as small as possible.

NO!!!!! You need to catch at least 1 cycle of each signal otherwise the XY picture isn't complete. Catch too much and the picture will become blurry.