Magnova oscilloscope

[Martin72]

Could be difficult with the bandwidth.
Oh, I just had to turn it on for a moment...
Booting is fast, it's a bit irritating that there is no boot logo, instead the power on light flashes during this time.
The Scope is dead silent...
The display is very good in terms of clarity and colors, the reaction of the touch panel to finger swipes is almost without delay, very fast.
It seems to be due to the acclimatization, because I have an offset on all channels, let's see how it is tomorrow.

[Geofrey]

USB HS is probably not possible, but FS (12 Mbit/s) and LS (1.5 Mbit/s) should definitely be

[king2]

As always: what would you like to analyse with the scope?
Do you need in depth math functions?

Digital signals, so I need possibility to capture some long sequence, and then analyze it, zoom, find noise/strange edges.
I need long memory for this and interface that can zoom by mouse click (or something more suitable then changing time, scroll by encoder, and aoom in again, like I do now with OWON SDS8202).

Analog signals, for example, response from gamma-photons, ~50-500mV, with possibility to see them with signal persistance. With my current scope such signals not so far from noise floor, so trigger works not so good as I want, persistance works not very well too. Ideally, I want to get hystogram of amplitudes of signals was captured by trigger.

Digital signals, like PWM, with history/trend view, to see, for example, if PWM period was changed over time.

I was very impressed how measurement values displayed on Magnova (on oscillogram itself, showing what I see instead of several columns of values in table separated from signal view).

And, finally, I cannot understand which features I will like, I had no expensive scopes yet

[joeqsmith]

As always: what would you like to analyse with the scope?
Do you need in depth math functions?

Agreed. As far as my experience with Lecroy scopes goes: they are really, utterly great for analysis (like Swiss army knife ^2) but as a general purpose oscilloscope not so much. For starters, one missing feature is peak-detect to prevent aliasing on sub-sampled signals. Lack of peak detect can also lead to missing narrow pulses in a signal in turn leading to mis-interpretation. Think about measuring how many times per second an SPI device is accessed by measuring the CS (select) line over a timespan of several seconds to check uniformity in time while the pulses on the CS lines are in the nanosecond range. Without peak-detect it is possible the pulses won't show or show as irregularities.

Sorry for the disruption and off topic.  See link below

Teledyne LeCroy WaveAce Lab1: Capturing Signals page | 8 of 8
A sampling acquisition samples the waveform at uniform time intervals. A rapid variation in the signal
such as a narrow pulse may therefore be missed. In Peak Detect the highest and lowest values in the
signal are recorded so allowing even narrow pulses to be captured.

https://cdn.teledynelecroy.com/files/tutorials/waveace_lab_capturing_signals.pdf

Lecroy WaveAce series aren't real Lecroy scopes but rebadges from (older) Iwatsu and (newer) Siglent. So it is logical they have features the real Lecroy scopes don't have. Typically the lower end scopes offered by Lecroy are rebadges.

Got it.  For their higher end scopes, the sample rate is fast enough and the triggers have advanced to where, who needs peak detect?  Even with my old scopes at home, I haven't thought about needing that feature in decades. 

***
... and deep memory...

Posts about a vintage Tektronix DSO test board:
https://www.eevblog.com/forum/testgear/show-us-your-square-wave/msg5163681/#msg5163681

Page 6&7 in the manual show their peak detect test waveform:
https://w140.com/tekwiki/images/8/8d/37W-7655.pdf

My 7200 was made in 1989 (35 years old).  This is the same year the Tektronix test board manual was released.   

****
Showing the same Tektronix board using another vintage LeCroy DSO with higher sampling rates and deeper memory.  What a difference 10 years makes.     

Again, sorry for the sidetrack.   

[ralphrmartin]

One minor UI issue, perhaps me rather than the scope. I wanted to find the cursors to measure something. I expected to find them under the "measure" item, but no, they are under the list where you choose channels, triggering, etc.

To me they are a kind of measuring tool, but I suppose to Batronix they are "something to be shown on the screen or not", like the channels...

I should have put this in context.
Other than this minor nitpick, I must say that I find the scope very straightfoward to use, and much more intuitive than my old Keysight scope.
I like having the probe attachments on the side, which gives me more working space overall.

[pmcouto]

I received my Magnova a few days ago, so these are just my first impressions.

Packaging
The scope comes in a single cardboard box, held in place by 2 lateral high density foam pieces. Also held by the foam pieces, there’s a small box containing the probes, a power cord, an hex key to adjust the stand and a “Thank you” letter from Batronix. This box is placed in front of the screen, providing some protection.
The scope itself doesn’t have any protection, so mine arrived with some dust on it, but no scratches. Dust was easy to clean with just a dry cloth.
Cardboard is too thin and can be easily damaged in transit (my box arrived with a hole in a corner, but nothing was damaged inside).
Batronix really needs to improve packaging, otherwise some scopes *will* be damaged in transit. We all know how “gentle” delivery people can be… 

Construction, form factor & screen
As Dave would say: This thing is built like a tank!
No flimsy plastics here – the case is all metal (aluminum) and the only plastic parts are the handle in the back and the knobs and buttons in the front. And the handle is just a plastic part covering a cutout in the back of the case – the weight is supported by the case.

Despite the generous screen size, the scope is compact; Not having the connectors in the front really helps. Unlike others, it doesn’t use a lot of space on the desk.
The stand is small, but more than enough to keep the instrument stable, while allowing a lot of flexibility in adjusting the tilt angle. The stand can be removed, exposing a standard VESA mount.

The screen is bright and crisp and easy to read (a huge plus for my tired eyes!). Did I mention the generous size? 
Full HD resolution allows for lots of information to be displayed simultaneously, without sacrificing readability too much.
It also does a good job reducing the reflections and, very important, is not a fingerprint magnet.

Overall, in my opinion, this is a good example of industrial design.

Boot and Operation
With the current SW version, the instrument takes less than 30s to boot, which is faster than many other modern scopes that run a standard OS, such as Linux or Android.
During boot, the screen is blank – No logo or boot messages.

Although there is a fan, it shouldn’t work in a normal lab environment (<25ºC). This means silent operation! 
There’s an heatsink in the back of the instrument that gets warm to the touch (<50ºC) after 15-20min of operation. Not sure if it gets hotter if the scope is performing intensive processing for an extended period of time. 

Regarding user operation, the instrument is very responsive, with no noticeable delay to user input or action.


UI
The UI is “clean” and sober and I would say it’s intuitive and easy to use.
Functions are organized logically in the menus and are easy to access – No need for “digging” through many levels of menu options. The combination of touch-screen (or mouse) and 4 knobs with functions according to the context make it very easy to operate.
To master the instrument, obviously there is a learning curve. But not too steep.

Although this initial version is good and very usable, there’s certainly room for improvement.
The addition of icons/pictograms/graphics to the menus and wise usage of color would be very welcome. A more flexible and powerful windows manager would also be very welcome. As an example, I can mention R&S UI.

Features
Current SW version offers a comprehensive set of features, probably enough for many users.
There are some useful features not usually found in other scopes in this price range, such as trend charts for measurements and spectrogram with customizable color grading.
Also very useful (at least for me) is the possibility of setting custom fractional attenuation ratios – When using special-purpose custom built probes with oddball attenuation values, being able to set the correct attenuation ratio and have the real value shown is a huge time saver and avoids possible mistakes in manual calculations.

Of course there’s a lot of room for improvements – Batronix has already mentioned some planned new features to be added in future SW updates.

Wish list
In addition to the features and enhancements already mentioned by other EEVblog members, I would welcome the following features:
-Zone triggering (must intersect/must not intersect)
-Video triggering (line/field/line #). Yes, some users still have to deal with analogue video signals… 
-Frequency response analysis (Bode plot). I understand this is already planned, to be launched when the Waveform Generator module becomes available.
-Power analysis

Final thoughts
In my opinion, Magnova is a robust platform with a huge potential. 
As Batronix doesn’t have (yet?) other oscilloscope series, there’s no risk of hurting sales of higher specs models by implementing advanced features in lower specs series. I see this as an advantage over other manufacturers. 

If Magnova specs fit the requirements, I would strongly recommend to consider it when shopping for a new oscilloscope.

[Martin72]

There was still a box in the box...
With the accessories, power cable, an Allen key...
And four very nice probes from Testec.
All in all, it makes a pretty professional impression.
Well, and the side connectors of the scope, my longest BNC cable should just about fit....
The picture also shows how compact a scope with a 15” screen can be.

[shabaz]

Nice seeing that photo showing the size comparison with other instruments on your desk!
I hope Batronix can convert that into a very long-term (i.e. permanent) loan, e.g. by eventually writing it off, since they may need someone in the field to help answer questions as they occur, or try out new features.

Regarding cables, I've had a good experience with superbat. They make up cables to specific custom lengths, and it's low-cost. They are a bit slow, sometimes taking a week before they ship the package. Typically I go for RG316DS (DS = double-shielded) cables from their list.

I can't say the connectors are top quality, they are just average, but certainly for the price they seem very reasonable.

I sometimes self-assemble cables, too, but that can be impractical on occasion (for instance, if more specialist tools are needed).

[Martin72]

Hi,

Based on a recommendation from user Performa01, I had assembled a cable with Hyperflex 5 myself.
The result was so convincing that I built more cables and disposed of my old ones.
I had bought the plugs and the cable there:

https://kabel-kusch.de/Koaxkabel/HyperFlex-10/HyperFlex-5.htm

https://kabel-kusch.de/BNC-Stecker/BNC-58-155/bnc58-155.htm

[Martin72]

A clip of the Magnova booting up. Normally, you don't see the screen light up after turning it on, which is due to the smartphone.
Only the power button flashes and after about 30 seconds the device is ready.

[nctnico]

Hi,

Based on a recommendation from user Performa01, I had assembled a cable with Hyperflex 5 myself.
The result was so convincing that I built more cables and disposed of my old ones.
I had bought the plugs and the cable there:

https://kabel-kusch.de/Koaxkabel/HyperFlex-10/HyperFlex-5.htm

https://kabel-kusch.de/BNC-Stecker/BNC-58-155/bnc58-155.htm

Personally I never had much luck with those solder mounted BNC connectors. In my case the shielding tears off where the shielding is clamped under the strain relief after a while (resulting in a broken cable). Since then I buy pre-made cables. The hyperflex cable seems to have nice specs so I wonder if it can be bought with BNC and/or SMA connectors already mounted.

[Martin72]

It's also quite annoying, I had added an extra 10 centimeters to my desired length as a precaution – and I ended up with exactly that...

But the shop has instructions for almost everything:

https://kabel-kusch.de/Montagen/BNC-155/mont-bnc-155.htm

But I have to say, I don't think I'll be soldering cables myself anytime soon.

[Martin72]

The hyperflex cable seems to have nice specs

https://www.eevblog.com/forum/testgear/siglent-sds2000x-hd-12bit-(published-for-chinese-domestic-market-only)/msg5211516/#msg5211516

There you can see the difference, the Amazon cable has a more uneven course.

[Martin72]

One more thing, then it's time for a break.
Everyone knows the option of inverting the image when taking a screenshot, or almost every scope offers this option.
What I have not seen yet (please correct me if I'm wrong), however, is that you can set this for normal operation on the Magnova.
I think that's very good.



(UHD 60fps)

[nctnico]

One more thing, then it's time for a break.
Everyone knows the option of inverting the image when taking a screenshot, or almost every scope offers this option.
What I have not seen yet (please correct me if I'm wrong), however, is that you can set this for normal operation on the Magnova.
I think that's very good.

Yeah, they stole the idea from me. I did that to a Tektronix TDS744A (and several other instruments) as part of a TFT screen conversion a long time ago:



A white background is much nicer to work with so IMHO it is fantastic the Magnova has this option.

[king2]

A clip of the Magnova booting up.

Wow. Why only 3 wfms/s at 1ms setting?

[Martin72]

As for the offset I noticed yesterday: it disappears by itself. Unfortunately, I didn't pay attention to when it happened, but I'll find out tomorrow.
I then took my thermocam and took pictures after about 40 minutes.
The temperatures are absolutely harmless, but the scope had nothing to do either.
Maybe I can take some measurements tomorrow, the afternoon is planned and I only have the weekends for something like that, no vacation in sight, too bad.

[Martin72]

Wow. Why only 3 wfms/s at 1ms setting?

No idea, but maybe you like this better...

[king2]

Wow. Why only 3 wfms/s at 1ms setting?

No idea, but maybe you like this better...

With 1ms/div we have 12 ms on one screen, so if we will capture signals screen by screen (in ideal world), we will have 1000/12 = ~83 wfms/s.
Maybe 24Mpts memory takes effect. Can you try to make same test with less memory settings?

[Martin72]

There must be a signal, otherwise nothing useful will be displayed.
Here is a sine at 1ms/div and manually set memory (minimum).

[king2]

Thank you!

Indeed, refresh speed depends on memory depth.
@Andre77, is this normal Magnova's refresh speed?

My old OWON SDS8202 gives refresh speed ~5 times per second (as I can see by my eyes with 1ms/div and 10M memory deep (1 channel active, 500MSa/s) and speeds up when changing depth to 1M.

[FloBX]

Indeed, refresh speed depends on memory depth.
@Andre77, is this normal Magnova's refresh speed?

Reducing the memory depth can increase the refresh speed. Nonetheless, you have missed some aspects when it comes to the actual performance at maximum and reduced sample rates. I will try to explain based on the screenshots/captures already posted by Martin:

A clip of the Magnova booting up.

Wow. Why only 3 wfms/s at 1ms setting?

In this case there were no trigger events. What you saw was the (quite slow) refresh rate of the Auto trigger.

As for the offset I noticed yesterday: it disappears by itself. Unfortunately, I didn't pay attention to when it happened, but I'll find out tomorrow.

The screenshot attached in Martins following post already shows a different performance (11 wfms/s when displaying all four channels) at the same settings, but here with an actual triggering based on signal noise.
Looking at the memory depth selected (24 MSa) I would assume that Martin has "Extended Capture" activated in the Acquire menu (compare to 12 ms * 1 GSa/s --> 12 MSa). This results in additional time required for capturing further data left and right to the screen leading to a total capture length of at least twice the "screen size" (thus 2 *12 ms * 1 GSa/s --> 24 MSa).

Rendering less channels will further increase the wfms/s (due to reduced render effort).

There must be a signal, otherwise nothing useful will be displayed.
Here is a sine at 1ms/div and manually set memory (minimum).

Reducing the sample rate will also further increase the wfms/s due to reduced required render effort. But here again we need to consider the length of the actual capture frame (looking at 20 kSa/s at 1 MSa/s) when comparing expected and actual capture rates.

So altogether:
3 wfms/s is not the general refresh speed at 1ms/div (no actual/Auto trigger).
Nonetheless, reducing the memory depth will very likely increase the capture and update rate due to a reduction in render effort (which I consider to be the case for most digital oscilloscopes available).
We could provide some screenshots showcasing the actual performance values in the next days, unless Martin attempts to challenge the above explanation first. 

[Andre77]

Is there any plan to add USB (1 or 2) to the list of decoded and triggered protocols ? Not a deal breaker, but nice to have.

USB low and full speed are possible. We have it on our "wish list" but not with high priority so far.

Booting is fast, it's a bit irritating that there is no boot logo, instead the power on light flashes during this time.

Yes, booting without a boot screen looks a bit empty. We will add a boot screen in one of the next updates.

Final thoughts
In my opinion, Magnova is a robust platform with a huge potential. 
As Batronix doesn’t have (yet?) other oscilloscope series, there’s no risk of hurting sales of higher specs models by implementing advanced features in lower specs series. I see this as an advantage over other manufacturers. 

If Magnova specs fit the requirements, I would strongly recommend to consider it when shopping for a new oscilloscope.

Thank you very much for your detailed and very nice feedback! Some points such as the windows docking system, zone trigger and power analysis are already on our wish list. We didn't have the video triggering on it yet - maybe we misjudged it and it is still needed. Frequency response analysis (Bode plots) are already fully supported, but are not yet visible as long as the internal generator is not installed.

[Martin72]

So altogether:
3 wfms/s is not the general refresh speed at 1ms/div (no actual/Auto trigger).
Nonetheless, reducing the memory depth will very likely increase the capture and update rate due to a reduction in render effort (which I consider to be the case for most digital oscilloscopes available).
We could provide some screenshots showcasing the actual performance values in the next days, unless Martin attempts to challenge the above explanation first. 

It is clear that if there is nothing to trigger, the rate is small.
It gets “more fun” in the opposite case, so I probably want to verify the information from the Magnova with a second scope at the trigger output first.
If you take measurements and also the trend display, it doesn't slow down, quite the opposite.
This is new to me, normally you have to hide as much as possible for the naked update rate, here it is the other way around (see pic below, updaterate rises when meausring and trend is on)- hard to believe...
OK, then I accept the challenge, the update rates will be the first measurements.
I can start later today, I don't know yet if I can finish them.
The Magnova offers many possibilities in this regard, with four different display modes and at least three different memory modes.
That's quite a few...
Report to follow.

(Display modes: None, Linear, Sample, Sinc.....Memory options: Auto(fast), Auto(max), Manual)

[Martin72]

Today I hardly had time, until just now.
So I did a quick and dirty measurement with SDG2122X and the BMO350 in default condition.
I connected the other scope to the aux out for control; you have to activate the output accordingly in the Magnova's trigger menu.
A few checks were enough to determine that the waveforms/sec display can be trusted.
The “real” measurements will be done with SDG2122X and SSG3032X and then also with different configurations regarding display and memory.
This table has to suffice for today, but it already shows that the scope is not a slow one.

Martin