Two Tone Test with Scope and SA

[mawyatt]

I'm not sure I did everything right. I combined the two signals from the generator with a T-piece and terminated them with 50 Ohm on the oscilloscope.

Think you want to resistively combine the two sources for the tones unless you can combine them internally like the Siglent. A resistive "T" should work, think this has 16.7 series ohms on each leg to maintain 50 ohms on all three ports.

You have to be very careful combining the signals because the output amplifiers of the signal generators will produce the same distortion products that you are testing for if they are driven with the other signal.



Ideally a combiner which provides isolation between the input ports is required but it is usually sufficient to pad each output with an attenuator and use minimum loss resistive divider.

That's why you want to look at this with a SA, to see if the AWG is causing any self-generated IMD issues when resistively combined outputs, if so then add some pads. The Siglent AWG has this nice feature where the two tone combining is done internally (digitally I suspect).

Best,

[TheSteve]

Keysight 33522B generating 10 kHz and 10.1 kHz @ 1 VPP, combined via resistive network, 50 ohms everywhere.
FFT on Keysight MSOX3024T and SA on Keysight N9918A(20 dB external atten used here, also at the very bottom of its freq range).





Same test repeated at 10 MHz and 10.1 MHz:

[mawyatt]

I have now combined the signals with 16 ohm resistors on a breadboard, and again terminated with 50 ohm. The spectrum analyzer measures 10 dBm for one tone, the oscilloscope measures 1 Volt peak.
Attached are the measurements for 10 kHz, 100 kHz, 1 MHz and 10 MHz. FFT is normal this time, without Max Hold. The small values jump a little bit up and down around 1 dB.

Generator Rigol DG4162
Oscilloscope R&S RTB2004 (200 MHz BW)
Spectrum Analyzer Rigol DSA815 (with external 20 dB Attenuator)

This shows off the R&S true 10 bit ADC, nice results!!

Best,

[mawyatt]

Keysight 33522B generating 10 kHz and 10.1 kHz @ 1 VPP, combined via resistive network, 50 ohms everywhere.
FFT on Keysight MSOX3024T and SA on Keysight N9918A(20 dB external atten used here, also at the very bottom of its freq range).





Same test repeated at 10 MHz and 10.1 MHz:

The scope results are interesting, they don't show the 2F1-F2 & 2F2-F1 3rd Order IMD products, or they are buried in the noise floor. Seems Keysight is getting more performance out of the 8 bit 5GSPS ADC than I would expect, maybe some sort of manipulation to extend the apparent ADC resolution? Anyway, very nice result for a very nice scope!!

Best,

[mawyatt]

And here the Rigol DS1104Z and MSO2302A. I haven't managed to get better settings. The DS1104Z FFT uses the memory, not the screen as source.

You need a longer time span to separate these two tones.

Best,

[2N3055]

10 kHz and 10.1 kHz 0 dBm from SDG6052X with digital WaveCombine.

Keysight MSOX3104T



Picoscope 3406D MSO (8bit)



Picoscope 4262 (16bit)



SDG6000X WaveCombine works quite well for low frequencies.. And Picos are great for this kind of work...

[tautech]

The $499 SDS1104X-E.......

Signal source as previous, SDG2042X + Wave Combine, 0dB into 50 ohms and max everything...mem depth, 1 Mpts FFT......
Averages engaged this time, just 4.

[2N3055]

The $499 SDS1104X-E.......
...............

That looks good!

[tautech]

The $499 SDS1104X-E.......
...............

That looks good!

They're as good as you drive them. 

Rejected for posting as I didn't take enough time with the settings and particularly not a slow enough timebase:
Compare with last above.



The $379 SDS1202X-E will be next....when I get a chance.

[mawyatt]

Agree, this SDS1104X-E looks really good even slightly better performance than the SDS2104X!! Amazing considering the price!!

Best,

[tautech]

Agree, this SDS1104X-E looks really good even slightly better performance than the SDS2104X!! Amazing considering the price!!

Best,

Siglent's do like a lot of waveforms displayed for FFT so the limits with low frequencies are at a firm 50ms/div when anything slower drops into Auto Roll mode.
FFT Horizontal menu settings (Hz/div, Span) are also limited by timebase settings in SDS1104X-E so some small juggling act ensues to obtain best settings. Practice makes perfect. 

Maybe I'll do a comparison between a few models at just one frequency squeezing the best I can outta each model.

[mawyatt]

The $499 SDS1104X-E.......
...............

That looks good!

They're as good as you drive them. 

Rejected for posting as I didn't take enough time with the settings and particularly not a slow enough timebase:
Compare with last above.



The $379 SDS1202X-E will be next....when I get a chance.

Rob,

The 1 and 4 markers should be on the two small tones on each side of the larger tones, they are spaced the delta frequency of the larger tones on each side. Since the two large tones are at 10.00MHz and 10.01MHz, delta F is then 0.01MHz (10KHz), then the 3rd Order IMD products are at 9.99MHz (2F1 -F2) and 10.01MHz (2F2-F1). This looks like these 3rd Order IMD products are over 65dB down (don't know how you squeeze this from an 8 bit converter, but your result show such!!), which is quite good indeed

Nice work

BTW can these results be brought into an app for a better resolution FFT display?

Best,

[tautech]

Rob,

The 1 and 4 markers should be on the two small tones on each side of the larger tones, they are spaced the delta frequency of the larger tones on each side. Since the two large tones are at 10.00MHz and 10.01MHz, delta F is then 0.01MHz (10KHz), then the 3rd Order IMD products are at 9.99MHz (2F1 -F2) and 10.01MHz (2F2-F1). This looks like these 3rd Order IMD products are over 65dB down (don't know how you squeeze this from an 8 bit converter, but your result show such!!), which is quite good indeed

Nice work

FYI nothing special was done with Markers in these shots other than select Peak which is why they are where they are. 
We can select Frequency which will place them on the nearer tones or manually position them and select how many markers we wish to display.
To date I've just let the Markers do their thing with just a Peak setting.
This ^ applies to all Siglent models shown thus far.

BTW can these results be brought into an app for a better resolution FFT display?

Probably but beyond my pay grade and understanding. Most stuff can be exported as a CSV Save.

We can however instead of using Split screen FFT use an Exclusive FFT mode and maybe I need drop a couple of shots in to show how different it looks.
With models shown thus far we can also port to any size display we like using the Webserver feature via a LAN link to a PC.
The LAN speeds on these are pretty quick however there is a wee latency when porting to another device.

[mawyatt]

Rob,

The 1 and 4 markers should be on the two small tones on each side of the larger tones, they are spaced the delta frequency of the larger tones on each side. Since the two large tones are at 10.00MHz and 10.01MHz, delta F is then 0.01MHz (10KHz), then the 3rd Order IMD products are at 9.99MHz (2F1 -F2) and 10.01MHz (2F2-F1). This looks like these 3rd Order IMD products are over 65dB down (don't know how you squeeze this from an 8 bit converter, but your result show such!!), which is quite good indeed

Nice work

FYI nothing special was done with Markers in these shots other than select Peak which is why they are where they are. 
We can select Frequency which will place them on the nearer tones or manually position them and select how many markers we wish to display.
To date I've just let the Markers do their thing with just a Peak setting.
This ^ applies to all Siglent models shown thus far.

On most systems the closer small 3rd Order IMD tones are more dominate than the ones father out as things begin to roll off as you move away from the dominate 2 tones. This appears to be a little different, either case it's really good performance!!

Yes, a single screen image of the FFT would show better. The TD display doesn't shown much as we've seen (all look almost exactly alike) and why this 2 tone test shows more on what's going on within the scope under test regarding DR.

I'm already sold but others may benefit form this.

Best,

[tautech]

Rob,

The 1 and 4 markers should be on the two small tones on each side of the larger tones, they are spaced the delta frequency of the larger tones on each side. Since the two large tones are at 10.00MHz and 10.01MHz, delta F is then 0.01MHz (10KHz), then the 3rd Order IMD products are at 9.99MHz (2F1 -F2) and 10.01MHz (2F2-F1). This looks like these 3rd Order IMD products are over 65dB down (don't know how you squeeze this from an 8 bit converter, but your result show such!!), which is quite good indeed

Nice work

FYI nothing special was done with Markers in these shots other than select Peak which is why they are where they are. 
We can select Frequency which will place them on the nearer tones or manually position them and select how many markers we wish to display.
To date I've just let the Markers do their thing with just a Peak setting.
This ^ applies to all Siglent models shown thus far.

On most systems the closer small 3rd Order IMD tones are more dominate than the ones father out as things begin to roll off as you move away from the dominate 2 tones. This appears to be a little different, either case it's really good performance!!

Yes, a single screen image of the FFT would show better. The TD display doesn't shown much as we've seen (all look almost exactly alike) and why this 2 tone test shows more on what's going on within the scope under test regarding DR.

I'm already sold but others may benefit form this.

Best,

Another point of reference when you asked about better resolution.
Well of course if we know the fundamentals are where they should be we can go chasing more resolution with a different Ref position and increased vertical scaling.
Here I've selected a full screen display that still shows the source waveform yet if that's not required we can select exclusive FFT, neither of which are hamstrung for resolution like the SDS1104X-E 50/50 split display.

[mawyatt]

That's a nice result and display.

I'm sold, are discount coupons available for the SDS1104X-E & SDG1032X?

Thanks,

Best,

[tautech]

Here's another shot from SDS2104X Plus diving deeper towards the noise floor in full screen FFT mode.

[bicycleguy]

Since you seem to be comparing FFT implementations wouldn't it be better to ask for something that can be compared like:
What are the closest 2 tones that can be separated with 3dB of dip between peeks and what's the highest and lowest frequency they can be done at.
Unfortunately the limits are set by the weakest link in the tone generator or the scope.

With Siglent SDG2042x and Keysight DSOX2014A, both enhanced.

Example: 119.99MHz and 120MHz high freq. limited by SDG and resolution by DSO FFT


Example 10 Hz and 10.05 Hz  and low, resolution limited by taking too much time to scan 

[tautech]

A new attempt with the Rigol DS1104Z, this time it worked better with 1 MHz and 1.01 MHz tone.
But no comparison with tautech's SDS1104X-E.

Try a faster timebase as AFAIK Rigols FFT works better with lesser waveforms on the display.

[tautech]

A new attempt with the Rigol DS1104Z, this time it worked better with 1 MHz and 1.01 MHz tone.
But no comparison with tautech's SDS1104X-E.

Try a faster timebase as AFAIK Rigols FFT works better with lesser waveforms on the display.

At 5 ms/DIV there is only 125 kHz/DIV resolution for the FFT.
I tried it for half an hour, I can't do better.

Yes similar problems exist with other entry level DSO's.
Lower frequencies allow for better results. 

Have a try with 10 or 100KHz and see if that makes any difference.
You won't find the limits unless you go looking for them. 

[mawyatt]

Since you seem to be comparing FFT implementations wouldn't it be better to ask for something that can be compared like:
What are the closest 2 tones that can be separated with 3dB of dip between peeks and what's the highest and lowest frequency they can be done at.
Unfortunately the limits are set by the weakest link in the tone generator or the scope.

With Siglent SDG2042x and Keysight DSOX2014A, both enhanced.

Example: 119.99MHz and 120MHz high freq. limited by SDG and resolution by DSO FFT


Example 10 Hz and 10.05 Hz  and low, resolution limited by taking too much time to scan 

In my case I was interested in the linearity performance of the scope front end and ADC. Also looking at the AWG output with a SA provides a view into how good the AWG signal is. The 2 tone tests was a simple means to show such and why I asked. If you are interested in the limits of the scopes FFT capability, then other tests like you've suggested might be considered.

Honestly I'm impressed by all the scopes behavior, even the low cost "entry' level devices seem to be performing better than I expected.

Now that I've retired (chip design) my plan is to purchase a scope and AWG, then a modest SA for starting a home lab. Have a good initial start with a couple repaired Tek 2465 analog scopes, a couple repaired 34401A DVMs and a new Keysight 34465A DVM (superb DVM).

I realize a scope can't compete with a SA regarding linearity & dynamic range, thus including the SA in the desired equipment list.

Anyway, this has created lots of responses and I wish to thank everyone for jumping in on the fun

BTW the two images show exactly the same screen capture??

Best,

[bicycleguy]

BTW the two images show exactly the same screen capture??

Best,

I was modifying the post.  Maybe try reloading the page.

[gf]

And here the Rigol DS1104Z and MSO2302A. I haven't managed to get better settings. The DS1104Z FFT uses the memory, not the screen as source.

If you can't increase the number of FFT points, then you can alternatively reduce the sampling rate in order to improve the resolution (i.e. choose a timebase which happens to lead to a lower sampling rate). For 10MHz sine waves you don't need a sampling rate of 1GSa/s, but (say) 25 MSa/s should basically suffice. Reducing the sampling rate from 1 GSa/s to 25 MSa/s is supposed to improve the FFT frequency resolution by a factor of 40 then (if the number of FFT points remains the same).

I was modifying the post. Maybe try reloading the page.

Unfortunately still the same. The two preview icons look indeed different, but the large images are still the same. Maybe a bug in the forum software I have noticed this effect a couple of times at other posts, too.

EDIT: The image in this posting https://www.eevblog.com/forum/testgear/two-tone-test-with-scope-and-sa/msg3168886/#msg3168886 seems to be broken as well.

[_Wim_]

Here's a SDG6022X into a RTO1024 oscilloscope and FSP spectrum analyzer.
(Attachment Link)
(Attachment Link)

This is superb performance, what you would expect from a 16 bit ADC based scope!! However it's well out of the price range of these other scopes, and I see it's been replaced with the RTO2000 starting at only $14,500!!

Best,

I think the RTO1024 is an 8-bit scope, but increases its ENOB by using massive oversampling at these low frequencies.

[nctnico]

In my case I was interested in the linearity performance of the scope front end and ADC. Also looking at the AWG output with a SA provides a view into how good the AWG signal is. The 2 tone tests was a simple means to show such and why I asked. If you are interested in the limits of the scopes FFT capability, then other tests like you've suggested might be considered.

Honestly I'm impressed by all the scopes behavior, even the low cost "entry' level devices seem to be performing better than I expected.

AFAIK the two-tone test is primarily a way to see how good the mixer is in a spectrum analyser. A DSO doesn't have a mixer so it is not a surprise this effect doesn't occur.