Magnova oscilloscope

[KungFuJosh]

What firmware version are you running?

I tried the MSO demo board for the dropout trigger, and that seemed to work. What signal generator did you use to setup your dropout test? I'd like to match it if I can.

For the pulse trigger, it worked 100% as expected. Note in the screen captures the trigger says Waiting when the trigger parameters were not met.

Thanks,
Josh

[KungFuJosh]

I quickly made a dumb waveform and loaded it into my SDG2122X. The Magnova triggered fine on falling edge. I can make one that's lower frequency, but I don't think that should matter. If you have a csv file of a different waveform, I'd be happy to try it.

Thanks,
Josh

[rf-messkopf]

Thanks Josh for looking into it. That looks encouraging.

What firmware version are you running?

I suppose it is the current firmware version, but I'll check that as soon as I get back to the scope.

What signal generator did you use to setup your dropout test? I'd like to match it if I can.

That was a Keysight 33521B.

I'll run the tests again tonight and post the results. Curious to see if that's reproducible.

[KungFuJosh]

That was a Keysight 33521B.

I'll run the tests again tonight and post the results. Curious to see if that's reproducible.

Try with the waveform I shared above, or see if you can export the wave and I'll try with yours.

I dunno if they're compatible with each other or not (Siglent/Keysight AWG custom waveforms), but that's interesting to try out too.

Thanks,
Josh

[rf-messkopf]

Okay, here we go. I'm running firmware version 1.5.4, which should be the current.

Strangely, I can still reproduce the problems I observed yesterday, see the attachments. I've indicated the behavior I'm expecting from the scope in the file names (OK/NOK).

Does anyone have an explanation for this? Somehow I feel like I'm making a really stupid mistake and am embarrassing myself.

[rf-messkopf]

Try with the waveform I shared above, or see if you can export the wave and I'll try with yours.

I dunno if they're compatible with each other or not (Siglent/Keysight AWG custom waveforms), but that's interesting to try out too.

I actually used the burst function of the 33521B to produce the dropout waveform. I would have to check whether I can export it as a csv file. But that generator should be able to digest your csv file. I can try that later.

[KungFuJosh]

For the timeout trigger, switch to Falling and it should work. I don't think it will work with Rising because the last point was falling before the timeout. That's my guess as to the reasoning anyway. I used Falling and it worked; rising did not.

For the other ones that don't trigger, there's nothing that appears wrong to me on the screenshot, but it's hard to see what's going on in that case without also seeing the AWG.

Thanks,
Josh

[rf-messkopf]

For the timeout trigger, switch to Falling and it should work. I don't think it will work with Rising because the last point was falling before the timeout. That's my guess as to the reasoning anyway. I used Falling and it worked; rising did not.

I tried rising, falling, and both (see screens 8, 9, 10). The scope only triggers when switched to both, which is odd. I would expect it to trigger on all three, since there is neither a rising nor a falling edge within the 1 ms timeout interval.

For the other ones that don't trigger, there's nothing that appears wrong to me on the screenshot, but it's hard to see what's going on in that case without also seeing the AWG.

The AWG just continuously produces pulses with 100 µs length at a frequency of 1 kHz. And it really does that, I can see the pulses on the scope. 

[rf-messkopf]

I've now pushed all buttons and twisted all knobs within reach, and tried at various voltages, frequencies, pulse widths and other settings, but I can't get that scope to trigger on pulse widths "less than" some value. See the attachment once again.

Can somebody imagine any (hardware) fault that would produce such a result? After all, this scope does not have a hardware trigger and probably does triggering within an FPGA, directly on the ADC data stream.

[KungFuJosh]

Do you have a different signal generator you could try? I had no issues with that trigger at all.

You could try doing a factory reset on the scope, and/or reflashing the firmware. Firmware upgrade is super easy, and you can do it all from the scope's menu over the internet.

[rf-messkopf]

Do you have a different signal generator you could try? I had no issues with that trigger at all.

You could try doing a factory reset on the scope, and/or reflashing the firmware. Firmware upgrade is super easy, and you can do it all from the scope's menu over the internet.

I tried a different signal generator -- same result. I also tried on different channels. I did a factory reset and ran a self-calibration, but no success.

I'll try to re-flash the firmware now. Thanks.

Edit: Re-flashing the firmware also didn't help. This is really odd.

[2N3055]

On pulse and timeout trigger, it should work both ways (positive and negative gong), just trigger on inverted waveform. This can be tested with FM modulation on squarewave. No need for custom waveform, actually a FM modulated squarewave is an all inclusive test vector.
I tried it on a different type of scope and there it works exactly how rf-messkopf expects it.

I would contact Batronix support and ask for clarification. Either documentation is not clear on use and could do with improvement, or there is some anomaly. In both cases, I know they would appreciate feedback. They are proven to be very professional, both support and Magnova team and they are best way to sort you out. If there is an anomaly they will actually appreciate to know about it so they can fix it.

[egonotto]

Hello,

I can hardly imagine that this is a hardware error.

There is a YouTuber called OsziWissen who makes videos about Magnova.
Take a look at his video “Magnova Oscilloscope Pulse / Pulse Width - Trigger Tutorial Finds Outliers” to see if it works for you too.

https://youtu.be/j5r1GXCPrAI?si=ogGDr-SLOJ2ZCnJ2

Best regards
egonotto

[KungFuJosh]

I can't get either trigger to fail on mine. Pulse worked fine for me, and I tested from 1mVpp to 2Vpp. 1mVpp was obviously quite noisy.

I tried the dropout waveform at different frequencies from 500Hz to at least 1MHz, and it always worked as expected when the dropout time was set appropriately for the frequency.

Playing with an FM modulated square wave sounds fun too. 🤔

[KungFuJosh]

I can hardly imagine that this is a hardware error.

I agree, that's one reason I think reflashing the firmware and factory resetting might be helpful.

His screenshots look correct to me (for whatever that's worth 😉), but we can't see what the AWG is doing.

[Martin72]

Hi,

I decided to play around with the pulse trigger briefly before the weekend.
Generator:
Pulse signal, 1 kHz, pulse width 50 µs.
Trigger level 0 V.
Less than, greater than, inside, outside...
In addition, I deliberately selected settings where the scope should not trigger.
Everything worked fine.

[rf-messkopf]

His screenshots look correct to me (for whatever that's worth 😉), but we can't see what the AWG is doing.

See the attachment. I'm 100% sure that there is nothing wrong with the AWG. For whatever that's worth. 

[rf-messkopf]

I would contact Batronix support and ask for clarification.

I shot them a mail and will keep you posted about any progress.

Thank you all.

[KungFuJosh]

His screenshots look correct to me (for whatever that's worth 😉), but we can't see what the AWG is doing.

See the attachment. I'm 100% sure that there is nothing wrong with the AWG. For whatever that's worth. 

I believe you.

My interest was to compare the AWG settings to the scope's trigger settings. For example, if the scope channel scale is not set correctly for the AWG output level, that could cause issues triggering.

[KungFuJosh]

See the attachment. I'm 100% sure that there is nothing wrong with the AWG. For whatever that's worth. 

From that attachment I got that your AWG was set to 1kHz, 500mVpp, 100.5us pw. I set up each pulse trigger with similar settings to your scope in the screenshot. Each trigger option responded as expected.

Ignore the -2uV trigger, that was leftover from screwing around at 1mV earlier.

Thanks,
Josh

[KungFuJosh]

I tried a ton of different ways to get the pulse trigger to fail. I went through every sampling rate, HiRes mode, etc. The one and only way I could get the trigger to fail was if the vertical scale on the scope was set wrong for the AWG's output level.

With a 500mVpp waveform and 500mV/div for the vertical resolution: I was able to trigger down to 200mVpp, but once I dropped to 100mVpp, the trigger failed. Increasing the vertical resolution to 200mV/div quickly resolved that, and I stepped through to fill the screen at 15mV/div.

That all seems normal to me, but at least it was entertaining.

[tooki]

Hi,
Most of the casing is painted black, so it's fine.

Color has no impact on emissivity or reflectivity when measuring midrange IR  (= inhouse temperatures).
Emissivity is only changing (getting higher) when radiated by higher temperatures like sun IR waves ( shorter waves).

For example: sun IR radiation gets through glass because of its short wave, but once inside can’t get out anymore because of re-radiation with long wave IR I lower temperatures.
It’s known as the greenhouse effect.

So no, you still have to determine the (very) low emissivity of the metal housing and compensate your measurement with it.
Otherwise it is way off.

In other words:
Highly reflective surfaces like polished metals ( like the Magnova) have low emissivity and are poor emitters of thermal radiation.

Ummm… what?

You can literally create emissivity targets (for accurate measurement) by applying a coating of known emissivity to an object. The classic example is ordinary black electrical tape. Thermal cameras have preset emissivity numbers for that!

I do that all the time at work, where we do a lot of stuff in stainless steel, which has grotesquely bad emissivity and just acts like a mirror to IR. Putting bits of electrical tape on it gives nice, accurate measurement targets.

When you’re dealing with a painted object, what then matters is the emissivity of the paint, not the substrate beneath it.

[mawyatt]

The classic example is ordinary black electrical tape. Thermal cameras have preset emissivity numbers for that!

I do that all the time at work, where we do a lot of stuff in stainless steel, which has grotesquely bad emissivity and just acts like a mirror to IR. Putting bits of electrical tape on it gives nice, accurate measurement targets.

When you’re dealing with a painted object, what then matters is the emissivity of the paint, not the substrate beneath it.

The electrical tape is a good idea, thanks

Best

[Sorama]

Putting tape of a know emissivity is the classic way they will teach you when going through ITC thermography certification.

But only to measure the actually emissivity of the object by changing the emissivity until you read the same temperature as when focusing on the tape.

That is how you determine the emissivity of an unknown material.

Btw, the emissivity is not only determined buy its surface, there are 6 factors that influence the emissivity of an object:

Geometry
Surface structure
Material 
Viewing angle 
Wavelength of the camera 
Target temperature

[rf-messkopf]

I shot them a mail and will keep you posted about any progress.

Regarding the trigger problem that we discussed above: Batronix took a look at it and suspected that this is due to timing tolerances in the trigger engine within the FPGA. Today I received a new optimized firmware version, presumably with a design synthesized with tighter timing constraints. With that, my scope behaves just like the others.

I'm glad the problem was solved so easily for me. I could already see myself lugging parcels around. 

Thank you Batronix for the excellent service and the truly great scope.