New toy(?) scope Fnirsi DPOX180H, claimed 180MHz/500MSps (May 2023)

[radiolistener]

Mine seems do not have this problem.

Can you please take WAV snapshots taken for open BNC input and for sine on the input? Both for 5 or 10 ns/div.

If you don't have sine source, then use square wave, but use max horizontal resolution (5 or 10 ns/div)

Attach these WAV files here, so we can analyze it. Thanks

[RAPo]

That is not possible.
I've modified my scope in order to have the signal gen output on the bottom using bananajacks. Planning to add some circuitry in order to change offset and amplitude.

Has anyone found a way to change the signal generator amplitude and offset?
Or is it supposed to be centered around 0V as in the above post and users manual?

[RAPo]

As indicated before, it is not a standard wav file. It has not a RIFF Hheader (see image).

The wav file contains measurement and scope settings data.

On GitHub there are repositories that decode the .wav output from a FNIRSI 1013D.
FNIRSI is not willing to share the .wav specs for the DPOX.

Some people complain that you don't see vertical lines with square waves.
What do you think?
Bug or feature?

it depends on horizontal resolution setting. If you're using slow horizontal resolution, the rise edge is too fast, so it will not be noticed. But it will appears when you switch to a fast horizontal resolution, for example 5 ns/div.

I asked for a description of the .wav file

WAV file is a standard container to store waveform data as a data chunks, it also allows to store additional information in a custom defined data chunks. Here is WAV format description: http://soundfile.sapp.org/doc/WaveFormat/

If it supports save to a WAV file it's nice feature, because standard WAV file can be opened and analyzed with variety of software tools.

Just wonder, how much samples it can capture to a WAV file at once?

[RAPo]

@radiolistener In this file  you'll find the requested wav files. It is a sine with 1V or 3V amplitude, generated by my MSO5104. Please report your analysis here.

[radiolistener]

@radiolistener In this file  you'll find the requested wav files. It is a sine with 1V or 3V amplitude, generated by my MSO5104. Please report your analysis here.

Thanks, but it seems that it has unknown format, so I even don't see a sine wave inside....
The format described from github link is incompatible with these files.

Could you please give also just one snapshot with a full display sine in both - WAV and BMP/PNG format (and maybe there is CSV or something like that, it also will help to decode it), because there is needs to understand how it should looks like in order to try decode it...

this is how it looks if try to read it as BE int16 from 0x2c0A offset. does it look like what was shown on the display?

[RAPo]

The scope does not generate .csv files.

I think fnirsi changed the header/scope settings block to accommodate (in respect to the fnirsi 1013D) the new features (e.g. FFT, auto bandwidth amplitude).

The GitHub description can be used as a first start but needs adaptation. E.g. the conversion from byte value to triggertype:

fn triggertype_val(s8 value){
  str rval="";
  if (value==0) {rval="  Auto";}
  if (value==1) {rval="Single";}
  if (value==2) {rval="Normal";}
  return rval;
};

is still valid.

How did you generate the image in your post?
Why is the value of 0x2c0A offset? It seems to me the offset should be greater to disregard the first part of the image.
Also, do mind there are two sets of measurements data in the .wav file:
  raw measurement data and displayed measurement data, which one did you choose to make the graphic?

In this file, you'll find the requested .bmp (and associated .wav files) for open,5MHz-1V/5MHz-3V and 10MHz-1V/10MHz-3V.

@radiolistener In this file  you'll find the requested wav files. It is a sine with 1V or 3V amplitude, generated by my MSO5104. Please report your analysis here.

Thanks, but it seems that it has unknown format, so I even don't see a sine wave inside....
The format described from github link is incompatible with these files.

Could you please give also just one snapshot with a full display sine in both - WAV and BMP/PNG format (and maybe there is CSV or something like that, it also will help to decode it), because there is needs to understand how it should looks like in order to try decode it...

this is how it looks if try to read it as BE int16 from 0x2c0A offset. does it look like what was shown on the display?

[mebel]

[radiolistener]

How did you generate the image in your post?
Why is the value of 0x2c0A offset? It seems to me the offset should be greater to disregard the first part of the image.

I just view the file and found a data which looks like raw waveform stream. It's definitely has 2 bytes per sample, but it's not clear what is the format of each sample.

Also, do mind there are two sets of measurements data in the .wav file:
  raw measurement data and displayed measurement data, which one did you choose to make the graphic?

yes, I understand that. But unfortunately the format of FNIRSI DPOX180H is completely different from FNIRSI-1013D. So, it cannot be decoded with description provided on github page for FNIRSI-1013D. I will try to decode it with your files which include BMP with actual picture on display. But it seems that it may be not so easy, because format looks a little bit mixed and complicated.

Can you please share also wav+bmp for a square wave in that way to keep several periods on the display?


Update: still not sure that I decoded it properly, but at a glance it looks that there is 300 samples per screen. Since screen is 5*12=60 ns, it means that there is . Which means 0.2 ns per sample or 5 GS/s. Probably they using equivalent time sampling to average sample points using different phase shift for several captures.

The raw data is started at 0x2F38, each sample 2 bytes, 300 samples per screen. The screen is repeated many times, probably for digital phosphor accumulation.

But 5 GS/s looks crazy even if they used equivalent time sampling, so I'm not sure if I decoded it properly... 

[RAPo]

Here is the square wave set, with an extra bonus: ch2_open;-)

Can you please share also wav+bmp for a square wave in that way to keep several periods on the display?

[RAPo]

I discovered a tiny bug. If you go to the param measure menu and then channel 2, remove all measure items, and go out of the menu. The screen shows 1 items instead of 0.

[Aldo22]

I discovered a tiny bug. If you go to the param measure menu and then channel 2, remove all measure items, and go out of the menu. The screen shows 1 items instead of 0.

Hugo Ferreira also found a bug in this area:
https://youtu.be/XoOh0gypVII?t=1605

[RAPo]

But  I can disable all measurements from channel 2 ...

[radiolistener]

Here is the square wave set, with an extra bonus: ch2_open;-)

Thanks, yes, I'm right. Samples are started at offset 0x2f38.
Each sample has 2 bytes in a Little Endian integer format.
Sample count per screen is at offset 0x011f in a Big Endian integer format.

100 kHz square wave file has 24000 samples per screen. Here is how ADC waveform actually looks like in comparison on how it's displayed on DPOX180H display:

[radiolistener]

We have 5 uS * 12 = 60 uS per screen / 24000 samples = 2.5 nS/sample = 400 MS/sec sample rate.
Not bad

But as I understand, this 400 MS/sec for 5 uS/div is probably equivalent sample rate, the same as 5 GS/sec for 5 nS/div.
If ratio for 5 uS/div is 10, then ADC works at 40 MS/sec for 5 uS/div.

Since it works as stroboscopic osciloscope at equivalent sample rate up to 5 GS/sec, this is definitely interesting device.

Here is more detailed ADC noise:

[radiolistener]

I still not sure how they achieved so high sample rate, classic equivalent sampling should not produce so high noise...

[RAPo]

Thanks. For reference, here is the input from the MSO50104 signal generator.

[svetlov]

forum users who already have FNIRSI DPOX180H in their hands
could you apply to the inputs of this oscilloscope
    square wave with steep leading edges? it’s just that no one has done this so far - this would remove a lot of questions about the device

[RAPo]

forum users who already have FNIRSI DPOX180H in their hands
could you apply to the inputs of this oscilloscope
    square wave with steep leading edges? it’s just that no one has done this so far - this would remove a lot of questions about the device

Please specify what is this .

[RAPo]

In this file , you'll find screen dumps and wave files for a range of TinyUltra generator (clean mode) from 10MHz-170MHz in two versions:
  tiny Ultra->Dpox->auto->save bitmap+save wav  and
  tiny Ultra->Dpox->manual best vertical resolution->save bitmap+save wav.
Because the auto function does not work well for freq>100MHz.

Dear RAPo ! you wrote that you will have a TinySA Ultra - could you use it to check the FNIRSI DPOX180H by applying a rectangular signal at high frequencies from 10 megahertz and above to its inputs, which is available on TinySA Ultra?

[Aldo22]

RAPo-  is a square wave with a fast rise time less than 1 nanosecond
TinySA Ultra seems to have MAX 2870 installed, it gives rise to less than 1 nanosecond
set to square wave TinySA Ultra

Are you sure, TinySA Ultra can output square waves below 800Mhz?

[RAPo]

you want to input a 1.2GHz freq @-20.0dBm in a 180MHz scope? Below 0.8G there is only a sine.

[RAPo]

Are you sure, TinySA Ultra can output square waves below 800Mhz?

The tiny SA can do an awful lot, but not that.

[Aldo22]

A Si5351A module can do that up to 160MHz with a rise time of ~1ns.
It's almost for free but you need an Arduino, ESP etc. to control it.
https://www.mikroshop.ch/Home.html?gruppe=10&artikel=2058

[svetlov]

did not immediately see that  tinySA Ultra it can synthesize a square wave only from 240 megahertz
 
Si5351 judging by the data of this seller, at a frequency
   100 MHz rise time -920 ps at 149.7 MHz rise time 660 ps
https://aliexpress.ru/item/1005002591845151.html?sku_id=12000021298933740&spm=a2g2w.productlist.search_results.0.20f34aa6VcnQ8C

[Aldo22]

There are many different implementations.
Everything is "on board" with this one (The second "color" for ~$30: SI5351A STM32 W TFT).
Neither an external microcontroller nor PC/software is required.
There is also a Chinese video there.