R&S MXO3/04/05 - Reviews, Bugs & Wanted Features/Improvements Thread

[tooki]

I can't speak for the product line, but if I had to guess, my feeling is that the intent was intended to convey that for the MXO, one only needs to buy the hardware (the logic pods) and that no additional software license (option key you type in) is needed.
…
Personally, I think that not needing a (node-locked) software license in addition to the hardware pods is an improvement, whether one calls them "options," "licenses," etc. 

Of course it is. But then just communicate it clearly and unambiguously.

(And for the record, on one of our scopes, the RTO6, the logic analyzer is actually a separate, field-installable piece of hardware.  I know this because I've personally installed one  )

Yep, I obviously saw that when doing my research for which scopes use which logic probes.

And there, calling it an “option” makes sense, because it really is optional and is sold separately.

[pdenisowski]

And there, calling it an “option” makes sense, because it really is optional and is sold separately.

I know what you're trying to say and agree that there is always room for improvement with regard to language, but the former linguist in me would also like to say that (in common usage) an "option" is not necessarily "optional"   

An "option" is a choice between alternatives (one of which could be "no choice" in some cases), whereas "optional" means not compulsory or not mandatory. 

Attached is an example from Apple (who, admittedly, sometimes play fast and loose with the English language).  When you buy a MacBook Pro, you must pick between two processor "options".  And you also have to choose a payment "option" - paying for it is definitely not "optional" 

[chilternview]

Here's another possible bug on the MXO3, or maybe it's my misunderstanding on how to set it up?

Both HD3 and MXO3 were connected using D0-D3 to an Agilent N2918A demo board set to generate SPI with D0=CS (active low), D1 = SCLK, D2 = MOSI, D3 = MISO.

Then the trigger was set to MISO 03 10 (i.e. 2 hex digits). The HD3 triggers happily on this data.

The MXO3 was setup in a similar fashion with the same trigger setup. But it never triggers. Experimenting shows that if a single digit '03' is used, the MXO3 triggers, but with more than one digit it doesn't.

User error or bug?

[tooki]

And there, calling it an “option” makes sense, because it really is optional and is sold separately.

I know what you're trying to say and agree that there is always room for improvement with regard to language, but the former linguist in me would also like to say that (in common usage) an "option" is not necessarily "optional"   

An "option" is a choice between alternatives (one of which could be "no choice" in some cases), whereas "optional" means not compulsory or not mandatory. 

Absolutely. I literally said that in my original rant:

...why even bother calling it an option (options imply choice!) and cause customer (and internal!) confusion by giving it a part number?

P.S. Always nice to see a fellow linguist — though I didn't major in it, I have my minor in linguistics. (The few extra classes needed for the linguistics major would have delayed my graduation by at least 3 semesters, since some classes were offered so infrequently. As a transfer student, I didn't have the whole 4 years available to me...)

P.P.S. As a former employee of the "fruit stand": I assure you, if there is one thing one can't say about Apple, it's of being "fast and loose with language" — Apple's marketing and communication department ("Marcom") is known jokingly as the Apple stasi. They (and legal, of course) pore over every single word that a customer sees, and of course certain terminology that employees must use (even if the stores are actually surprisingly flexible here; it's not at all like call centers where you have to stick to a script). But every word is a very, very deliberate choice.

[Rydda]

Following up on this, what I found is, every time I start the scope with the logic probes connected, I get an alignment failed error: (Attachment Link) " alt="" class="bbc_img" />
If I run manually align afterwards, alignment passes, but it's rather annoying. Is this the same for other folks or shall I ask for service? It's a brand new unit.

I can confirm that mine behaves exactly the same. I had not noticed it previously, because I find the stiff springy probe cables annoying, so I disconnect when when not in use.

[chilternview]

User error or bug?

Some further testing - with I2C or RS232, a two byte data trigger works. But for SPI, it doesn't. The main difference is the first two were using CH1 and CH2 probes, the SPI uses D0-D3 probes.

[Bernd_2]

Hello chilternview,

Unfortunately, I can't test this myself right now, but possibly it might help to set the Search Mode to "Bit-aligned" and the value in the Position field to "0" in the trigger menu shown in the last image (MXO3_SPI_trig2.png).

If I understand correctly, the word you search starts with "03 10".

A value of 1 in the Position field means that, when set to "Word-aligned," the first word before the value being searched for is ignored, or when set to "Bit-aligned," the first bit before the value being searched for is ignored.

With set Position set to 0, the search should start then with the first Bit in your search string.

Best Regards
Bernd

[chilternview]

Hello chilternview,

Unfortunately, I can't test this myself right now, but possibly it might help to set the Search Mode to "Bit-aligned" and the value in the Position field to "0" in the trigger menu shown in the last image (MXO3_SPI_trig2.png).

If I understand correctly, the word you search starts with "03 10".

A value of 1 in the Position field means that, when set to "Word-aligned," the first word before the value being searched for is ignored, or when set to "Bit-aligned," the first bit before the value being searched for is ignored.

With set Position set to 0, the search should start then with the first Bit in your search string.

Best Regards
Bernd

Thank you Bernd,
I did wonder at the 'Position' field, but you can't actually set it to 0 - if you do it tells you the minimum is 1.

I also tried bit-aligned with various values of 'Position' from 1 to 16 but nothing would generate a trigger. However, a single (8 bit) data value *does* work with 'Position' of 1.

For I2C or UART decoding, a two byte data value does work.

[Bernd_2]

Too bad—that would have been too good to be true.

Maybe Paul or Daniel from R&S can help.

Best Regards
Bernd

[Rydda]

User error or bug?

Hi.
I tried to duplicate all settings shown in your post, using a 100kBits/s SPI signal from the Batronix demo board as source
It works just as I expected it to.

It is not clear to me, are the 03,10 values even listed in the decode table?

[nctnico]

Here's another possible bug on the MXO3, or maybe it's my misunderstanding on how to set it up?

Both HD3 and MXO3 were connected using D0-D3 to an Agilent N2918A demo board set to generate SPI with D0=CS (active low), D1 = SCLK, D2 = MOSI, D3 = MISO.

Then the trigger was set to MISO 03 10 (i.e. 2 hex digits). The HD3 triggers happily on this data.

The MXO3 was setup in a similar fashion with the same trigger setup. But it never triggers. Experimenting shows that if a single digit '03' is used, the MXO3 triggers, but with more than one digit it doesn't.

User error or bug?

I think user error  . It looks like your word size is 8 bits while the value you entered is 16 bit. What if there is a space between the two byte values?

[minus3db]

Here's another possible bug on the MXO3, or maybe it's my misunderstanding on how to set it up?

Both HD3 and MXO3 were connected using D0-D3 to an Agilent N2918A demo board set to generate SPI with D0=CS (active low), D1 = SCLK, D2 = MOSI, D3 = MISO.

Then the trigger was set to MISO 03 10 (i.e. 2 hex digits). The HD3 triggers happily on this data.

The MXO3 was setup in a similar fashion with the same trigger setup. But it never triggers. Experimenting shows that if a single digit '03' is used, the MXO3 triggers, but with more than one digit it doesn't.

I'm not familiar with hose that N2918A demo board works, so I'll just ask from what I see:

1)  Are you controlling what SPI data the board generates, or is it totally random?  ie are you certain it's generating the 'x03 x10' data pattern?

2) so in that trigger field, are you sure it's looking for hex data there by default, or you think it is treated as decimal numbers unless you designated as hex (ie something like 'x03 x10' or similar?

any mentions of how to use this field in their manuals? 

as others stated, since it's triggering on the 03, I'm inclined to think it's how to use that field properly for > 1 bytes that is the issue....

[Martin72]

Hi,
Tomorrow will be the final tests, and then, unfortunately, I’ll have to send the scope back.
You can take that “unfortunately” literally, because I find some aspects of the device remarkable and would have liked to examine it more closely.
But tomorrow I at least want to briefly test the Power Analyzer Mode.
My demo board isn’t finished yet, so I’ll use my power adapter again to check the power quality measurement.

Speaking of the demo board:
It’s almost finished, but it’s “high voltage” and therefore not without risk, so it’s not suitable for everyone to try to replicate.
So I’ve come up with something else and will continue to pursue this concept.

[CRTbrain]

21 nS is the fastest blind time between updates.  Every 30 mS the display does a refresh with up to to hundreds of thousands overlayed waveforms each from an acquisition.  During display update there's a longer time (> 21 nS).  I haven't measured on MXO 3 , but did measure on my MXO5 and it matches what R&S publishes. 282uS.  This is a far better value than any other scope I've seen when the display refreshes. 
 

See page 10.
https://www.rohde-schwarz.com/us/solutions/electronics-testing/digital-design-testing/landingpages/wp-selecting-your-next-oscilloscope-why-fast-update-rate-matters_257022.html

[KungFuJosh]

I don't wanna read that- I want Paul to read it to me! Where's the video, Paul??

[chilternview]

1)  Are you controlling what SPI data the board generates, or is it totally random?  ie are you certain it's generating the 'x03 x10' data pattern?

The board generates fixed data, it's not random - and yes, the 0310 pattern is there, and the HD3 can trigger on that pattern.

2) so in that trigger field, are you sure it's looking for hex data there by default, or you think it is treated as decimal numbers unless you designated as hex (ie something like 'x03 x10' or similar?[/b]

There is the choice of binary, hex, ascii data. I've tried both binary and hex.

as others stated, since it's triggering on the 03, I'm inclined to think it's how to use that field properly for > 1 bytes that is the issue....

It *does* work for I2C and UART data or if just a single '03' is used, and as mentioned it works for the HD3 under similar setup. There really are not many ways the trigger on data dialog can be used.

[core]

21 nS is the fastest blind time between updates.  Every 30 mS the display does a refresh with up to to hundreds of thousands overlayed waveforms each from an acquisition.  During display update there's a longer time (> 21 nS).  I haven't measured on MXO 3 , but did measure on my MXO5 and it matches what R&S publishes. 282uS.  This is a far better value than any other scope I've seen when the display refreshes. 
 

See page 10.
https://www.rohde-schwarz.com/us/solutions/electronics-testing/digital-design-testing/landingpages/wp-selecting-your-next-oscilloscope-why-fast-update-rate-matters_257022.html

Thanks for this information, I hadn't found this document.

That's exactly what I was asking here :
https://www.eevblog.com/forum/testgear/rs-mxo3-a-short-review/msg6223223/#msg6223223

An MXO3 is about to arrive, so I'll be able to test this soon.

[pdenisowski]

See page 10.
https://www.rohde-schwarz.com/us/solutions/electronics-testing/digital-design-testing/landingpages/wp-selecting-your-next-oscilloscope-why-fast-update-rate-matters_257022.html

I don't wanna read that- I want Paul to read it to me! Where's the video, Paul??

   You mean this one?   



I didn't write that whitepaper, but I was one of the proofreaders / reviewers.  The author of that whitepaper also reviewed my video, so both pieces of content should mostly agree with each other

[pdenisowski]

Maybe Paul or Daniel from R&S can help.

Daniel's got an MXO3 - I "only" have an MXO5 - sorry! 

(Would be happy to test this with the 5 if you like)

Edit: completely gratuitous picture of a cart full of MXO5s

[core]

Maybe Paul or Daniel from R&S can help.

Daniel's got an MXO3 - I "only" have an MXO5 - sorry! 

(Would be happy to test this with the 5 if you like)

Edit: completely gratuitous picture of a cart full of MXO5s

Wow !!! beautiful photo !

[Bernd_2]

Daniel's got an MXO3 - I "only" have an MXO5 - sorry! 

(Would be happy to test this with the 5 if you like)

I assume that the menu options and behavior of the SPI decoder are the same for all three MXOx devices.

Best Regards
Bernd

[KungFuJosh]

I don't wanna read that- I want Paul to read it to me! Where's the video, Paul??

   You mean this one?   

Perfect! Thanks.

[KungFuJosh]

completely gratuitous picture of a cart full of MXO5s

You know, that's too many for that cart. You should probably send one of those to me. 🤔

[le_yum]

completely gratuitous picture of a cart full of MXO5s

You know, that's too many for that cart. You should probably send one of those to me. 🤔

you are too late, Dave already found them in the dumpster 

[Martin72]

Power Analyzer Mode

Saving the best for last...
As I mentioned, I was only able to test the aspects covered by my homemade adapter, namely power quality and harmonics.
I used an inexpensive Chinese switching power supply as the test unit and my Siglent electrical load as the load.
The main menu is well-organized; I really liked that you can add functions, which are then displayed as individual “channels”—in other PA modes, you can only switch between functions and view just one at a time.
I also like that you can choose which individual measurement values should be displayed.
In general, there are more than enough options available.
There are so many R&S models stored in the probe selection that I was able to find a suitable preset for both the Testec differential probe and the Siglent current clamp.
The results of my small measurement seem plausible; I’ll double-check them this weekend with a Siglent scope for comparison.

A small request:
Power factor: I'd like to see this displayed without units, with a 0 before the decimal point.
For example, from the measurement screen:
0.573 is more common than 573 m (milli-what?)

I’ll now post a series of pictures from my last session with the MXO38; it’s already packed up again and will be shipped out tomorrow.
I’ll then share my final thoughts on this scope over the weekend.


Martin