New oscilloscope from Fnirsi DPOS350P

[Jan Steinman]

I think the major problem with most of the signal generators out there is that they won't generate enough harmonics for displaying a decent square wave.

That's what tunnel diodes are for!

I've modified a Tektronix S-4 sampling head to generate 25 picosecond rise times. It uses six tunnel diodes in a travelling wave gate configuration.

(I used to work at Tektronix.)

[ptluis]

I've read so much negative feedback about this oscilloscope. If I didn't already have one and actively use it, I wouldn't have bought it.
But I definitely like it. Its capabilities far outweigh the price. The only change I had to make was installing good 1 GHz probes, and the picture and accuracy improved significantly. Phase shifts were reduced and became insignificant.
There are still a few drawbacks, for example, that it can freeze or turn off. Or that it doesn't show measurements at low frequencies. But these aren't that significant. I hope they fix this with firmware. The heading measurements still work at all frequencies.
I definitely won't be upgrading to another one in the next year.

So you are saying that by changing the probes, phase shifts become correct? can you post a screenshot?

[ptluis]

I think the major problem with most of the signal generators out there is that they won't generate enough harmonics for displaying a decent square wave.

That's what tunnel diodes are for!

I've modified a Tektronix S-4 sampling head to generate 25 picosecond rise times. It uses six tunnel diodes in a travelling wave gate configuration.

(I used to work at Tektronix.)

I'm referring to out of the box devices, stock units not modified ones.

[ptluis]

Here is an example.  The Fnirsi Bode plot is decent, but the phase is inverted and the log cycles are screwy.

Fnirsi is an enigma.  They seem to put in a good effort and then poop out near the finish line.  Why not spend a little more time and finish what they started

Exactly, and fixing the inverted phase is a simple task and won't take to much programming time. If they do it, they will stop most of the bad comments including my own. Buyers won't realize if they accept the thing has is, fnirsi won't bother to fix it. This device catch the attention for FRA feature, not for the scope part because it is what it is.

[ptluis]

Since you got a dpos350p try triggering an AM sine wave signal carrier 10MHz, mod sine wave 300Hz at 80%, then you see what you bought for 200 bucks.

You keep saying this, and yet, @BillyO managed to post good-looking screenshots of that signal.

To me, that makes it sound like a cockpit problem. Just sayin'.

I think you should go back a few posts and see the answer to my question.

[ptluis]

If you just need a few quick checks, there is a smaller pocket scope for that.

Ah, you must be under 40!

My eyes can't even look at a smart phone, let alone a scope with a screen that is considerably smaller than a smart phone.

I can put up with a lot of issues for a 1024 pixel screen.

If I can't see all the wonderful goodness on a device with a 400x256 screen, what good is it? I want a portable device that I don't have to carry around #3 reading glasses for.

These pocket size devices are not for everyone, those who have bad vision should opt for a 10 inch micsig or one that allows connection to an external monitor. Fortunately, today there's alot of flavours to choose from and fullfil everybody's needs.

[csuhi17]

The funny thing is that with the Micsig I wrote VATO or VTO you can connect any tablet of any size and shape you find.
I just found the 14" miracle of Teclast...

[Atlan]

Here is an example.  The Fnirsi Bode plot is decent, but the phase is inverted and the log cycles are screwy.

Fnirsi is an enigma.  They seem to put in a good effort and then poop out near the finish line.  Why not spend a little more time and finish what they started

Exactly, and fixing the inverted phase is a simple task and won't take to much programming time. If they do it, they will stop most of the bad comments including my own. Buyers won't realize if they accept the thing has is, fnirsi won't bother to fix it. This device catch the attention for FRA feature, not for the scope part because it is what it is.

On the fnirsi 1013, in the 50mV sensitivity mode, a long time base was needed, just multiplying the value by two, and they never fixed it

[ptluis]

Here is an example.  The Fnirsi Bode plot is decent, but the phase is inverted and the log cycles are screwy.

Fnirsi is an enigma.  They seem to put in a good effort and then poop out near the finish line.  Why not spend a little more time and finish what they started

Exactly, and fixing the inverted phase is a simple task and won't take to much programming time. If they do it, they will stop most of the bad comments including my own. Buyers won't realize if they accept the thing has is, fnirsi won't bother to fix it. This device catch the attention for FRA feature, not for the scope part because it is what it is.

On the fnirsi 1013, in the 50mV sensitivity mode, a long time base was needed, just multiplying the value by two, and they never fixed it

As long as they continue selling they won't fix anything. That's why I keep bugging about returning the devices, so they understand the message and have more consideration for the buyers.  Simple fixes are solved really fast, not a huge investment.

As I wrote some posts before, not all fnirsi stuff is bad and we can live with that because their cheap, but when you buy something above 200€....

[onefabis]

The funny thing is that with the Micsig I wrote VATO or VTO you can connect any tablet of any size and shape you find.
I just found the 14" miracle of Teclast...

Wow. That's really nice fresh idea: connect the tablet to the scope with a stick tape, one usb wire and two somehow connected devices for the portability will make it look like DIY project, so when both of them will be discharged, you have to charge them in parallel
Once again, Micsig VATO2004 - 270€ without vat, plus tablet at least 100€ and still it will look like diy connected together for the portability. So 400€ for this frankenstein is what users are looking fore 

[csuhi17]

You don't have to use double-sided tape, you can also use magnetic stuff.
You can detach the tablet at any time, the operating time is twice as long because of the double battery.
wireless connection would be better...

The User manual doesn't say whether USB-c recharges the tablet...

Since the scope is not used in all cases, you can use the detachable tablet separately.

This is not a 2 in 1 device, but a 1 + 1 device...

You don't need to buy a separate tablet that you can use to watch videos and other things.
My Micsig tablet sometimes lags on youtube videos

Be creative or be rich

There was a time when I dreamed of being able to take the display out of my desktop scope separately in wired or wireless mode, but that's just a dream.
It's like a dream like having a physical button controll panel that can be attached to my tablet oscilloscope.

[onefabis]

You don't have to use double-sided tape, you can also use magnetic stuff.

Hmm, I have to find really strong neodym magnets that will hold 640g of the scope and all this should not fall apart in the middle of the work accidentally.

This is not a 2 in 1 device, but a 1 + 1 device...

I don't need 1+1, I need one device that will do the work at any time without needing to construct it once I want to use it: catch -> turn on -> work.

You don't need to buy a separate tablet that you can use to watch videos and other things.

Don't have free tablet for this, what next?

Be creative or be rich

In this case I have to be creative and rich.

I don't know why noone mentioned something similar Hantek TO1254D, TO1154D, etc. But they're expensive (+50€ at least) than Fnirsi and have smaller bandwidth. But anyway it way less expensive than micsig tablet scopes.

[ptluis]

The funny thing is that with the Micsig I wrote VATO or VTO you can connect any tablet of any size and shape you find.
I just found the 14" miracle of Teclast...

Wow. That's really nice fresh idea: connect the tablet to the scope with a stick tape, one usb wire and two somehow connected devices for the portability will make it look like DIY project, so when both of them will be discharged, you have to charge them in parallel
Once again, Micsig VATO2004 - 270€ without vat, plus tablet at least 100€ and still it will look like diy connected together for the portability. So 400€ for this frankenstein is what users are looking fore 

VTO and VATO2004 although they operate just fine, both are poor in features, because those two models were developed for auto industry. I had both and they operate flawless.

[Serhii74]

Passive probe, from an Tektronix oscilloscope, P6158

[yangyao]

Here is an example.  The Fnirsi Bode plot is decent, but the phase is inverted and the log cycles are screwy.

Fnirsi is an enigma.  They seem to put in a good effort and then poop out near the finish line.  Why not spend a little more time and finish what they started

Phase reversal? That's simple. I've been too busy these past few months to check the forums. The new product's ADC has encountered many technical challenges.

[yangyao]

A big request to the programmers at fnirsi who worked on this device: completely redesign the spectrum analyzer block—it doesn't perform its functions at all! This is completely unacceptable! I don't understand why bloggers partially approved this feature. At frequencies of hundreds of kilohertz, the peak detection error is tens of kilohertz!
At least create an FFT function in this mode as a replacement for your spectrum analyzer—after all, you managed to display this function correctly in oscilloscope mode. Move the program code to spectrum analyzer mode.
What does it even show in this mode? We see hills.

Developer is here, his nickname is yangyao. Simply send him PM with your setup, use case and results. Nothing more, just: setup -> steps to reproduce measurements -> spectrum analyzer errors.

Because it uses the FFT method with a maximum of 32768 points, the highest resolution at a 1GS sampling rate is only 30kHz. Measuring hundreds of kHz will definitely result in poor accuracy. Regarding your suggestion to reduce the sampling rate for FFT, this requires many high-roll-off low-pass filters to remove more than half of the spectrum. Otherwise, the spectrum you see will be superimposed with frequencies exceeding the Nyquist frequency, leading to spectral errors. Such filters are very expensive and require many of them.

[yangyao]

A big request to the programmers at fnirsi who worked on this device: completely redesign the spectrum analyzer block—it doesn't perform its functions at all! This is completely unacceptable! I don't understand why bloggers partially approved this feature. At frequencies of hundreds of kilohertz, the peak detection error is tens of kilohertz!
At least create an FFT function in this mode as a replacement for your spectrum analyzer—after all, you managed to display this function correctly in oscilloscope mode. Move the program code to spectrum analyzer mode.
What does it even show in this mode? We see hills.

Developer is here, his nickname is yangyao. Simply send him PM with your setup, use case and results. Nothing more, just: setup -> steps to reproduce measurements -> spectrum analyzer errors.

The oscilloscope you mentioned with its FFT function enabled actually doesn't actually have a usable FFT function. It's only accurate at the highest sampling rate; otherwise, don't use it, as it will mislead you, and the spectrum you see may not actually represent that frequency.

[Grandchuck]

Here is an example.  The Fnirsi Bode plot is decent, but the phase is inverted and the log cycles are screwy.

Fnirsi is an enigma.  They seem to put in a good effort and then poop out near the finish line.  Why not spend a little more time and finish what they started

Phase reversal? That's simple. I've been too busy these past few months to check the forums. The new product's ADC has encountered many technical challenges.

I assumed it would be easy enough to fix the inverted phase (multiply by -1?).  Can we expect a firmware update?

[SiliconWizard]

The funny thing is that with the Micsig I wrote VATO or VTO you can connect any tablet of any size and shape you find.
I just found the 14" miracle of Teclast...

Wow. That's really nice fresh idea: connect the tablet to the scope with a stick tape, one usb wire and two somehow connected devices for the portability will make it look like DIY project, so when both of them will be discharged, you have to charge them in parallel
Once again, Micsig VATO2004 - 270€ without vat, plus tablet at least 100€ and still it will look like diy connected together for the portability. So 400€ for this frankenstein is what users are looking fore 

The unfortunate part of those Micsig devices is that an Android tablet is apparently the only way to get anything, they haven't released any Windows/macOS/Linux software and they don't support SCPI, so the benefit of having a USB scope is basically not even there, that's pretty much a failed product IMO.

[ptluis]

The oscilloscope you mentioned with its FFT function enabled actually doesn't actually have a usable FFT function. It's only accurate at the highest sampling rate; otherwise, don't use it, as it will mislead you, and the spectrum you see may not actually represent that frequency.

This is serious stuff! it should be mentioned in datasheet and user manual!

[mishail71]

Here is an example.  The Fnirsi Bode plot is decent, but the phase is inverted and the log cycles are screwy.

Fnirsi is an enigma.  They seem to put in a good effort and then poop out near the finish line.  Why not spend a little more time and finish what they started

Phase reversal? That's simple. I've been too busy these past few months to check the forums. The new product's ADC has encountered many technical challenges.

I assumed it would be easy enough to fix the inverted phase (multiply by -1?).  Can we expect a firmware update?

Perhaps this flaw with the phase display will be fixed by the firmware author... Although it's hard to believe!!!

[ststefanov13]

The developer has appeared......

Let's make an updated list of bugs and additions:

Bugs:
1.Vertical markers in zoom mode show .........
2.Inactivity shutdown timers do not work.

Requests:
1.Trigger holdoff functionality

Anyone can add so we can see if the developer is well-intentioned or is here just to advertise.

It would be nice for the developer to take the time and read the entire topic.

[CNCmann]

Hi

Just got my DPOS350P. Basically device is OK for its price, cooling fan is quite noisy.
Need to 3D print desktop stand for it.

1) I found a problem with waveform display at low signal level, input straight from scope's built-in generator.
It can't display waveform if input is lower then 14mV RMS - please look at photos attached.
DPOS350P CAN display 10mV RMS waveform correctly if input is signal level gradually reduced for example from 50 to 10 mV RMS.
3rd photo it is unable to properly trigger something and input Vrms is not correct at all.

Zoom images to see RMS input voltage.

2) Firmware enhancement suggestion - last used setting must be remembered by default between power on/off cycle. Its really inconvenient to load last from menu each time scope is powered on.

[Ivont]

I believe they will not expose themselves.  They will fix the analyzer, at least from 0 to 1MHz.

[Atlan]

Hi
1) I found a problem with waveform display at low signal level, input straight from scope's built-in generator.
It can't display waveform if input is lower then 14mV RMS - please look at photos attached.
DPOS350P CAN display 10mV RMS waveform correctly if input is signal level gradually reduced for example from 50 to 10 mV RMS.

I guess nothing has changed since the 1013D, I see that this feature has been brought to perfection but you have a nice signal waveform without any noise