RIGOL MHO98 and MHO900 Oscilloscope Series

[kmoonwalker]

same here - tried to order one in the morning but they removed it from website while I was on call with them and told me it is not available... :/

[0xdeadbeef]

It's still available at Rigol, either through Amazon or from their own website.
But in might take 2-3 weeks until the next shipment arrives in Europe:
See https://www.eevblog.com/forum/testgear/rigol-mho98-and-mho900-oscilloscope-series/msg6075785/#msg6075785

[Martin72]

It was quicker than I thought, and a huge thank you to Batronix for making it possible—the service is simply sensational.

Now it's here, an MHO984—I had forgotten how small these things are...
This series also comes in good packaging with foam inserts.
The power supply is a Liteon 100W, as with the MHO98 (anything else would have been surprising), the included probes are 500MHz types and look rather cheaply made, and the accessories are also rather sparse.
The most obvious difference is the BNC connectors, which are not gold-plated.
The housing is transparent, which gives it an interesting look.
Not much has changed in terms of connections compared to the DHO800, except for the two generator outputs.
Still only one USB-A port, which is very meager.
I didn't notice a protective film for the display; it actually seems to be that shiny.
Here are a few more pictures; initial measurements and further impressions will likely follow over the weekend.

[Martin72]

Remaining images...

[smk]

I have just begun playing around with the MHO98.
I think the anti-aliasing filter can be seen in FFT-mode with no signal attached to the inputs.
The first picture show the settings I used for the single channel picture.
With only one channel active the noise level starts dropping somewhere between 1.1GHz and 1.2GHz.
Activating the second channel automatically reduces the frequency range to 1GHz and adjusts the filter to 800MHz.
With 3 or 4 Channels the frequency range is limited to 500MHz and again the fiter is adjusted.

Pretty nice results!

If something seems too good to be true, it probably is not.
The filtering shown in the FFT noise plots seems to be done in software, at least a part of it.
I could reproduce the aliasing problems mentioned by faveri97 in reply #67 https://www.eevblog.com/forum/testgear/rigol-mho98-and-mho900-oscilloscope-series/msg6046907/#msg6046907
I should have read all posts in this thread before posting.

[Martin72]

I think I've discovered a bug.
I turned it on briefly and played around with it a bit (The fan is running at full power and is therefore quite noticeable.), then I wanted to turn on the AWG—but it didn't work.
“Option not available”...
All options are listed in the system info and marked with “Limit”—and none of them can be selected.
That doesn't make sense. They (Rigol) should either rename it or make it available as limited as usual, rather than displaying it as limited and then not allowing it to be used.
This is a setback in terms of testing.

[0xdeadbeef]

This "Limit" thing was already mentioned for the MHO98 before
https://www.eevblog.com/forum/testgear/rigol-mho98-and-mho900-oscilloscope-series/msg6070189/#msg6070189
I guess they might have meant "locked" since "limit" doesn't make any sense.

[Fungus]

I think I've discovered a bug.
I turned it on briefly and played around with it a bit (The fan is running at full power and is therefore quite noticeable.), then I wanted to turn on the AWG—but it didn't work.
“Option not available”...
All options are listed in the system info and marked with “Limit”—and none of them can be selected.
That doesn't make sense. They (Rigol) should either rename it or make it available as limited as usual, rather than displaying it as limited and then not allowing it to be used.
This is a setback in terms of testing.

That means you didn't get the licenses yet.

[0xdeadbeef]

I think that's clear. Still, it's a wrong naming and uncommon behavior.
Usually the unlicensed options are available for testing and only get locked after the trial period.
To lock them by default and call this state "limit" is just weird and confusing and hence could be called a bug.

Anyway, I fear that an MHO984 demo unit won't get any licenses - unless Rigol is available to provide limited test licenses.
While Batronix currently has a promotion for a free bag and AWG 50MHz option, I would think that neither was included with Martin's demo unit.

[Fungus]

While Batronix currently has a promotion for a free bag and AWG 50MHz option, I would think that neither was included with Martin's demo unit.

Martin72 can probably get them from Batronix.

One question for owners: Are the settings cleared if you had down the "default" button on powerup?

That was a feature I went on and on about to Rigol after firmware 1.03 bricked my DHO804. I wonder if they ever implemented it.

(I know Rigol are probably reading this... Hi, Rigol!)

[0xdeadbeef]

(I know Rigol are probably reading this... Hi, Rigol!)

If so, I hope that also look into the reported aliasing issue:
https://www.eevblog.com/forum/testgear/rigol-mho98-and-mho900-oscilloscope-series/msg6046907/#msg6046907

[axantas]

While Batronix currently has a promotion for a free bag and AWG 50MHz option, I would think that neither was included with Martin's demo unit.

Martin72 can probably get them from Batronix.

One question for owners: Are the settings cleared if you had down the "default" button on powerup?

That was a feature I went on and on about to Rigol after firmware 1.03 bricked my DHO804. I wonder if they ever implemented it.

(I know Rigol are probably reading this... Hi, Rigol!)

You can press the Default button once every time and are welcomed with the message "Restore default settings?" No need to hold it on powerup.

[smk]

(I know Rigol are probably reading this... Hi, Rigol!)

If so, I hope that also look into the reported aliasing issue:
https://www.eevblog.com/forum/testgear/rigol-mho98-and-mho900-oscilloscope-series/msg6046907/#msg6046907

I hope so too.

I'd like to show just one exampe of the aliasing. From the frequency response graphs I picked a frequency where I expected serious aliasing:1.11GHz. I don't have have a sine wave source for frequencies as high as that so I had to use my LBE-1420 square wave source.

The first picture shows the 1.11GHz with only one channel active at 4GHz sampling rate.

With a second channel active and a sampling rate of 2GHz the scope only shows a flat line but the counter shows 890MHz = 2GHz-1.11GHz.

With three channels active and a sampling rate of 1GHz the scope shows a 110MHz sine wave and counter and measured frequency also read 110MHz.

The last picture is for reference only and shows a 110MHz square wave from the LBE-1420.

[Fungus]

If so, I hope that also look into the reported aliasing issue:

I don't really see that as an "issue", it's just math combined with incorrect usage.

eg. I can make my DHO800 alias on the 1kHz probe compensation signal by setting the wrong timebase. Should Rigol fix this?

[Martin72]

@smk:
When I was using Rigolaktiv yesterday, I was surprised by the “fat” waveform display, as you can see in your images.
The solution was simple: the intensity grading seems to be set to 100% at the beginning.

[0xdeadbeef]

I don't really see that as an "issue", it's just math combined with incorrect usage.

Maybe I misunderstood the aliasing issue but I figured that the MHO98 doesn't implement low-pass filters fitting the current sampling rate (even though the datasheet implies that) and with an actual frontend bandwidth of 1.2GHz or so this creates all kind of weird aliasing issues at 1GSa/s.
There is also the video by this Japanese guy who observed strange behavior with all channels activated:

But honestly it was a bit exhausting to watch with the instable trigger and only so-so subtitles.

[Martin72]

Anyway, I fear that an MHO984 demo unit won't get any licenses - unless Rigol is available to provide limited test licenses.
While Batronix currently has a promotion for a free bag and AWG 50MHz option, I would think that neither was included with Martin's demo unit.

Yes, both AWGs are inactive, as are Bodeplot and the additional decoder types.
I can't imagine getting licenses for this (unless Batronix wants to make it the “real” demo device and should include everything), and can only hope that Rigol will release an update.

[Fungus]

Maybe I misunderstood the aliasing issue but I figured that the MHO98 doesn't implement low-pass filters fitting the current sampling rate (even though the datasheet implies that) and with an actual frontend bandwidth of 1.2GHz or so this creates all kind of weird aliasing issues at 1GSa/s.

That's correct, but what can be done?

As noted: Every 'scope I've owned has been able to alias on the 1kHz probe compensation signal if you set them up wrong.

Even the mighty Siglents have warnings about this in their manuals:

[Martin72]

As noted: Every 'scope I've owned has been able to alias on the 1kHz probe compensation signal if you set them up wrong.

I can't manipulate the sample rate so badly that the 1 kHz signal can no longer be displayed correctly.
Image DS4 shows the lowest sample rate, image DS6 shows the sample rate when the memory is set to “Auto.”
I only took image DS5 because I really like the full-screen function.

[0xdeadbeef]

Anyway, it's one thing if you get aliasing because you reduced the sampling rate willingly or if you get aliasing even at the maximum sampling frequency.
I understand that Rigol uses digital filters here instead of a configurable low pass filter in the frontend and that the attenuation is somewhat inconsistent.

When the sampling rate is 2Gsps, the amplitude response is greatly attenuated at frequencies around 1GHz and rises again at ~1.2GHz. This is more apparent for the 1Gsps sampling rate, where the frequencies around 500MHz and 1500MHz are attenuated.

This is also what I understood from the Japanese youtube video.

[Fungus]

I can't manipulate the sample rate so badly that the 1 kHz signal can no longer be displayed correctly.

This is my DHO800 showing 500Hz for the probe compensation signal:



Note: The hardware counter is correct. What does the counter say for the 1GHz signal on an MHO98?

(does it go that high?)

[ptluis]

This video on YouTube shows a DHO804 using FFT up to 625MHz, I ask those who own MHO98 or MHO900 series what is the max FFT frequency on your model?

The video:
https://youtu.be/w2o6ZWDMqHA?si=JIoCRy3ckgEDAp-U

[thm_w]

This video on YouTube shows a DHO804 using FFT up to 625MHz, I ask those who own MHO98 or MHO900 series what is the max FFT frequency on your model?

The video:
https://youtu.be/w2o6ZWDMqHA?si=JIoCRy3ckgEDAp-U

You already saw and responded to the post above? https://www.eevblog.com/forum/testgear/rigol-mho98-and-mho900-oscilloscope-series/msg6074381/#msg6074381 1.1-1.2GHz

[ptluis]

This video on YouTube shows a DHO804 using FFT up to 625MHz, I ask those who own MHO98 or MHO900 series what is the max FFT frequency on your model?

The video:
https://youtu.be/w2o6ZWDMqHA?si=JIoCRy3ckgEDAp-U

You already saw and responded to the post above? https://www.eevblog.com/forum/testgear/rigol-mho98-and-mho900-oscilloscope-series/msg6074381/#msg6074381 1.1-1.2GHz

Yes you are correct, I forgot I already saw it  To many things to think and read about. Thanks

[EEVblog]

I have the bling version on the way