Keysight 3000T Signal / Tracking generator Frequency Response

[AllTheGearNoIdea]

Hi All

I have a Keysight 3000T with the signal generator option enabled and I must say I am very happy with it but my wife and bank manager are maybe not so pleased. 

 I want to know if I could use this instrument to make some audio measurements on amplifiers. So my questions are, can the signal generator be used as a tracking generator when the scopes in FFT mode and is there a way to measure THD using this instrument.  I note there is a harmonic measurement option under power measurements but this looks to be aimed at switch modes etc.

Many Thanks

[mikeselectricstuff]

I don't think tracking makes sense with FFT as it process it all at once.
You can set up the gen to do a sweep, by frequency modulating with a 0 or 100% duty ramp, and trigger on the modulation source to get a V/F display in normal 'scope' mode. You may be able to combine this with FFT but it may get quite slow.
I don't think there is a way to do a log sweep.

If you have the power supply option, there is a bode plot facility, which is less crippled than the X1000 one, but still pretty slow.

[AllTheGearNoIdea]

Are OK Mike

I thought I could set the FFT to peak hold mode and the somehow do a sweep.  I guess in peak hold mode I could even do the sweep manually. I noticed that I can do this quite easierly with my £99 picoscope so assumed I could do something similar.  Good advice with the modulation setting had not thought of that. 

I think I may have the power option as I see a harmonuc tool under measurements

Thanks for the reply - Big fan of your Videos.

Chris

[mikeselectricstuff]

Daniel just gave a sneak preview of upcoming 3000T firmware that will do more measurements in FFT mode, including THD
https://youtu.be/OgDuL-or12c?t=639

[AllTheGearNoIdea]

Thanks Mike

Loving that THD measurement.

Thanks

[AllTheGearNoIdea]

I found this link for the 1000g and this look like this to exactly what I'm trying to do. I don't think from memory my 3000T has these options. I hope the firmware upgrade will give me the same implementation

Thanks

[JPortici]

question: how good can the THD measurement be if the data comes from an 8 bit digitizer?

[AllTheGearNoIdea]

I have no idea and note that Picoscope recommend there 12/16 bit scope for doing this. If you know the answer please tell. I'm not doing the amplifiers test professionally so i don't need absolute accuracy or to be traceable to a standard.  But be nice to get ballpark figures.

It's frequent response plots I'm most interested in as I have a old leader distortion meter and it only takes a few minutes to hook this up. However manual frequency response sweeps take ages to perform especially if the old anolog sig gen I'm using need to have its output constantly adjusted

Thanks

[AllTheGearNoIdea]

Ps I don't have an axe to grind regarding Picoscope. I just mention them because when I start to look into amplifier characterisation a good audio frequency spectrum analyser is out of my price range and Picoscope appear to have some very good software tools. Although of thread would be interesting in know what anyone thinks about these. Just bought the 8 bit 2000 series it was worth £99 just  to get my unboxing fix for the day - TEA suffer in the final stages!

[Keysight DanielBogdanoff]

We are adding some oscilloscope FFT measurements in the next firmware release for the 3000T, 4000X, and 6000X. I showed some beta firmware in the livestream from Friday: https://youtu.be/OgDuL-or12c?t=10m9s

We're adding channel power, occupied bandwidth, adjacent power ratio, and total harmonic distortion measurements.

[nfmax]

No new firmware for the X2000 series? Seems to be the only model left behind now 

[ruairi]

To the O.P.

I don't think a scope is the right tool to measure audio THD and may have residual values too high for your needs. What you really need is a dedicated audio analyzer - an Audio Precision, Rohde & Schwarz or Prism but of course these are very expensive for a casual user. 

I use a Prism dScope but for field work I use a standard audio interface with the free REW software.  It is very powerful and flexible with excellent FFT too.
 

[MarkL]

No new firmware for the X2000 series? Seems to be the only model left behind now 

AND the X3000A series too.  Neither product line is getting much love anymore.  Last release for both was almost 1.5 years ago.

Dan: Are these product lines dead-ends at this point in terms of new features?

[Keysight DanielBogdanoff]

No new firmware for the X2000 series? Seems to be the only model left behind now 

AND the X3000A series too.  Neither product line is getting much love anymore.  Last release for both was almost 1.5 years ago.

Dan: Are these product lines dead-ends at this point in terms of new features?

I can't say with certainty that they won't see new updates or features, but I'd expect that most things will be for the most current oscilloscopes. The 3000T and 4000X are on the same firmware now, so those two will get the same treatment. For the 3000A vs 3000T, the T is actually cheaper, so for getting a new one there's no reason not to get the T.

[Someone]

I found this link for the 1000g and this look like this to exactly what I'm trying to do. I don't think from memory my 3000T has these options. I hope the firmware upgrade will give me the same implementation

Thanks

The FRA is hidden in the power analysis option on the 3000X, 3000T, 4000, etc:
http://literature.cdn.keysight.com/litweb/pdf/5992-0594EN.pdf

[nfmax]

No new firmware for the X2000 series? Seems to be the only model left behind now 

AND the X3000A series too.  Neither product line is getting much love anymore.  Last release for both was almost 1.5 years ago.

Dan: Are these product lines dead-ends at this point in terms of new features?

I can't say with certainty that they won't see new updates or features, but I'd expect that most things will be for the most current oscilloscopes. The 3000T and 4000X are on the same firmware now, so those two will get the same treatment. For the 3000A vs 3000T, the T is actually cheaper, so for getting a new one there's no reason not to get the T.

I agree the 3000A is obsolescent, but the 2000A series is still current - unless you know something you haven't told us about yet?
Seriously though, it would be nice to see graticule labelling on this scope, even if the FRA capability which is present now on all the other X-series scopes doesn't make it for whatever reason (though that does make the new R&S look an even stronger competitor to the 2000 series).

[AndyC_772]

I can't say with certainty that they won't see new updates or features, but I'd expect that most things will be for the most current oscilloscopes. The 3000T and 4000X are on the same firmware now, so those two will get the same treatment. For the 3000A vs 3000T, the T is actually cheaper, so for getting a new one there's no reason not to get the T.

That's a shame; I have an MSO-X3104A which I've no intention of replacing any time soon, but it would be really nice to see it still supported with feature updates.

I understand that there's a better business case to be made for upgrading products that are still current, but it's disappointing to miss out just because a slightly improved model has come out. Is the 3000T really so different to the 3000X that it would be a big job to port the new features back?

[nfmax]

I can't say with certainty that they won't see new updates or features, but I'd expect that most things will be for the most current oscilloscopes. The 3000T and 4000X are on the same firmware now, so those two will get the same treatment. For the 3000A vs 3000T, the T is actually cheaper, so for getting a new one there's no reason not to get the T.

That's a shame; I have an MSO-X3104A which I've no intention of replacing any time soon, but it would be really nice to see it still supported with feature updates.

I understand that there's a better business case to be made for upgrading products that are still current, but it's disappointing to miss out just because a slightly improved model has come out. Is the 3000T really so different to the 3000X that it would be a big job to port the new features back?

My understanding is the 3000T is much more like a 4000 than it is like a 3000A. But the 2000A and 3000A are very similar and share firmware. I'm not sure about the 1000 series though - no firmware download seems to be available.

[electrolust]

Ps I don't have an axe to grind regarding Picoscope. I just mention them because when I start to look into amplifier characterisation a good audio frequency spectrum analyser is out of my price range and Picoscope appear to have some very good software tools.

QA401?  ~$400

Without looking up picoscope I seem to remember it being a grand at least, to start.  And up.

[JPortici]

correct, for the 5000 series (which has 8 to 16 bit ADC . hardware, not software interpolated)

[nfmax]

I can't say with certainty that they won't see new updates or features, but I'd expect that most things will be for the most current oscilloscopes. The 3000T and 4000X are on the same firmware now, so those two will get the same treatment. For the 3000A vs 3000T, the T is actually cheaper, so for getting a new one there's no reason not to get the T.

That's a shame; I have an MSO-X3104A which I've no intention of replacing any time soon, but it would be really nice to see it still supported with feature updates.

I understand that there's a better business case to be made for upgrading products that are still current, but it's disappointing to miss out just because a slightly improved model has come out. Is the 3000T really so different to the 3000X that it would be a big job to port the new features back?

My understanding is the 3000T is much more like a 4000 than it is like a 3000A. But the 2000A and 3000A are very similar and share firmware. I'm not sure about the 1000 series though - no firmware download seems to be available.

...and version 2.42 for the 2000A/3000A has just appeared. The DVM facility and the training waveforms are now enabled for everybody, and there are a bunch of bug fixes. Not able to try it out yet, as the 'scope is running an extended test where I'm using segmented memory to capture a short event that should happen every 24 hours. I should do this sort of thing more often if it's going to make software updates happen! 

[Keysight DanielBogdanoff]

Haha, lucky timing I guess

[MarkL]

Haha, lucky timing I guess

No new features, though, in 2.42.

And could they get any less specific with the release notes?  The bug descriptions are all one sentence long and 7 of the 11 say "An issue with <blah blah> has been corrected".  One of them I recognize since I was the one who reported it in October of 2014.

Rather disappointing.

Any thoughts on an A to T trade-in program?

[nfmax]

So - what would I like to see in the next 2000A firmware release, apart from bug fixes?

• In single-shot mode, record and show on screen the date and time of the trigger event (like the time of the first segment in segmented mode)
• (Optionally?) enable the on-screen display of segments as they are measured - saves me having to wait 3 days to see if my fix worked!
• NTP time synchronisation (if the LAN interface is installed)
• Graticule labels! https://xkcd.com/833/
• Spectrum phase (from X1000 series)
• Bode plots (from X1000 series)
• Advanced maths, especially integrate & differentiate (from 3000A series)
• Arbitrary waveform function generation (if WGN option is installed)

I'm not convinced about the usefulness of the additional trigger modes of the 3000A, but then I thought the DVM option would be useless (I bought the bundle when it was on offer) until I tried it, and it's actually very handy indeed at times. The 3000A series will always have the edge of more acquisition memory, faster ADC, wider bandwidth, 50 ohm inputs, and probe interface, quite apart from the 3000T's touch screen & zone trigger, so it ought to be possible to show a little love to the 2000A series as well! Although it's the oldest MZ4 scope, it's still current - mine is still in warranty!

I'd be sort of happy if the last three bullets were a new chargeable upgrade (DSOX2TURBO) but I'd prefer not to have to pay, or at least to have them included in the existing bundle upgrade.

From what I read about the new R&S, it looks like they are exposing the acquired data through sysfs, and making the resulting files available over the network. I have no personal experience with this scope (yet) but it sounds like a neat idea.