New 2ch pocket DSO+SG - Zeeweii DSO2512G

[Aldo22]

Hi tatel!

Please see this animation: https://youtu.be/cUD1gMAl6W4

Thank you! That's a great animation, but do you know this video?



If you like old American science fiction movies, then you will love the music.   
And I think it's still very educational.

Regarding the other information from you, I have to admit that I cannot assess the effect of the antialiasing filter in practice.

All I can say is that I don't see anything unusual from my normal tests (besides the coax test).
For example, if I connect my small FM test transmitter to the Zeeweii, I can use it to set the transmission frequency.
If Zeeweii says that we now have 105 MHz, then the receiver/radio also has a signal at 105MHz. Why should I think this is wrong?
Of course, I will not seek details about the signal in this area.

If I hold the probes to the 16 MHz clock of the Arduino, the Zeeweii tells me that 16.0MHz is oscillating there.
Why should I assume this is wrong?

When I try to pick bad from even worse ne5532 fakes, the Zeeweii shows me the original square signal and the trapezoid-like amplified signal of the ne5532.
Why should I think that instead of a square and a trapezoid, it's actually a butterfly and a balloon? (Image attached).

If you can explain to me that it's not really a square and a trapezoid (more or less), then I'll throw the device away.
Otherwise that's enough for me and it's worth my money.
Above 5Mhz, I don't really care about how the signal looks anymore.

Perhaps what some don't understand is that I never expected more from this device.
It costs as much as a bottle of wine in a restaurant here.

Thank you!

[tatel]

Hi tatel!
Thank you! That's a great animation, but do you know this video?

No, I didn't. Quite interesting video

Regarding the other information from you, I have to admit that I cannot assess the effect of the antialiasing filter in practice.

Perhaps you should read carefully the Nyquist linked stuff should you have any spare time.

All I can say is that I don't see anything unusual from my normal tests (besides the coax test).
(...)
If you can explain to me that it's not really a square and a trapezoid, then I'll throw the device away.
Otherwise that's enough for me and it's worth my money.

I did the very same point to radiolistener on the other thread. I was using DSO154 example  if you remember. radiolistener's point is aliasing could fake any waveform, even lowest frequency ones, which is true. We wouldn't be able to say if we are seeing good results or aliasing. We could be seeing an 10 Hz waveform where it should be, say 10 MHz.

That said, I don't agree with more-papist-than-the-pope approaches, that's why I'm suggesting to carefully test DSO2512. I would do these tests myself should I have one. Perhaps I will do them with DSO154 if you don't with DSO2512. I'm afraid I would need almost as much help as you, but, hey, knowledge doesn't have a price, for anything else, BanksterCard

Obviously Zeeweii is doing something or this matter would be much more apparent. So, there's something to be learnt here. Put aside the device being really good or just a collection of tricks. Anyway it would be interesting to see what these tricks are. It could perfectly turn out it aint't that bad. At the very least, we would know where it can be trusted up to. Without testing, who knows?

Above 5Mhz, I don't really care about how the signal looks anymore.

It's up to you. Sarna con gusto no pica, as they said (scabies with pleasure does not itch). You don't care, that's fine. However I do. Taking it on a €40 device is a thing.  But when I'm getting into the €100 price range, I don't want toys. This is advertised as a measurement instrument. I want my meters to be exactly 1/10,000,000.00 the distance between Equator and the Pole. Even the proverbial Heredia the gipsy's scale, while infamously bad, benefits him. But it harms Heredia's customers.

Perhaps what some don't understand is that I never expected more from this device.
It costs as much as a bottle of wine in a restaurant here.

Please see my previous point. Again, it's up to you. But please note we are Heredia's customers here. Should you have knew it, wouldn't you expend just €25 more and get a good-working, serious device instead? Like Owon HDS242 could be? I would bet yes.

Anyway, If I'm going to recommend an oscilloscope to a newcomer, I want my advice to be rock-solid.

Thank you!

My pleasure, man

[Aldo22]

Hi tatel!

I did the very same point to radiolistener on the other thread. I was using DSO154 example  if you remember. radiolistener's point is aliasing could fake any waveform, even lowest frequency ones, which is true. We wouldn't be able to say if we are seeing good results or aliasing. We could be seeing an 10 Hz waveform where it should be, say 10 MHz.

That is probably possible, but when should I expect such a fake waveform?
When using signal generators I pretty much always see what it's supposed to be, sine, square, triangle, sawtooth, noise... as long as I'm within reasonable bandwidth.
Why should there suddenly be a wrong signal?
When is that to be expected? On a rapid rise?

Please see my previous point. Again, it's up to you. But please note we are Heredia's customers here. Should you have knew it, wouldn't you expend just €25 more and get a good-working, serious device instead? Like Owon HDS242 could be? I would bet yes.

Are you sure that this Owon is any better?
Afaik it's an oscilloscope and a multimeter.
I can't imagine that the price jump is enough to deliver a substantially better device.
It doesn't even pretend to have an equal sample rate. (250/125 MSa/s, Rise time ≤ 8 ns)
But I would be interested in this question.

Thank you!

[tatel]

That is probably possible, but when should I expect such a fake waveform?
When using signal generators I pretty much always see what it's supposed to be, sine, square, triangle, sawtooth, noise... as long as I'm within reasonable bandwidth.
Why should there suddenly be a wrong signal?
When is that to be expected? On a rapid rise?

To put it simply, it would depend on the frequency of the harmonics present in your signal and sample rate in your oscilloscope. As I understand it, results could be *wildly* different from reality. To know better, you really should read the linked stuff. There are even exercises about aliasing calculations.

I don't know how zeeweii does it to achieve seemingly good results. Perhaps they have a frequency counter? Otherwise, if they get an aliased waveform, they would probably also show an erroneous frequency. That would be really a red flag, but it doesn't seem to happen so easily.

Again, I don't really know what the juju is but I think it would be interesting -and quite instructive- to know what the trick is. Very probably some of the old grumpy guys in the forum will know better but I'm at a total loss right now.

Are you sure that this Owon is any better?
Afaik it's an oscilloscope and a multimeter.
I can't imagine that the price jump is enough to deliver a substantially better device.
It doesn't even pretend to have an equal sample rate. (250/125 MSa/s)
But I would be interested in this question.

Thank you!

Well, one thing that struck me with DSO2512 is that, even if advertised as a 120 MHz device, it doesn't seem to have a 20 MHz limit button. AFAIK, almost any other oscilloscope able to do more than 20 MHz does have that 20 MHz-limit button, just because it's the industry standard limit to measure PSU ripple/noise. So I guessed it would be a 20 MHz device at most. Antialiasing filter got under the radar. I'm going to take that button's absence as a red flag from now on, at least for devices advertised as able to do more than 20 MHz. I hope that goddam filter does not get under the radar again.

HDS242 does have that button (and much more realistic 40 MHz bandwidth specification). There's little doubt that button activates a low-pass filter with a 20 MHz cut-off frequency. If it really does that, then also little doubt is a better device.

About it being a dual-purpose device, please note DSO2512 is also a dual-purpose device (signal generator). Also, IIRC, DSO2512 sampling rate is 200/100 MS/s, so lower?

But not, I'm not sure it's a better device (yet). I'll have to read more about it. Right now I'm much more interested in aliasing and the probably clever tricks Zeeweii is doing. It could turn out that the knowledge to be gained studying these tricks is worth device's price

[Aldo22]

About it being a dual-purpose device, please note DSO2512 is also a dual-purpose device (signal generator). Also, IIRC, DSO2512 sampling rate is 200/100 MS/s, so lower?

No. The specs are 500MS/s and <3ns rise time for the DSO2512.
Don't know if true, but if, then the Owon is simply less than half as "good" and probably a little more conservative with the bandwidth declaration.
I really don't know much about it, but i expect no "filter-magic" in the Owon from this data.
But really this is just guessing.

[wasedadoc]

Just do some background reading on sampling, Nyquist and aliasing.  It is neither rocket science nor black magic. Once you understand the circumstances under which aliasing occurs and what effect it has you will know when you need to worry about it and when you do not.

[tatel]

No. The specs are 500MS/s and <3ns rise time for the DSO2512.
Don't know if true, but if, then the Owon is simply less than half as "good" and probably a little more conservative with the bandwidth declaration.
I really don't know much about it, but i expect no "filter-magic" in the Owon from this data.
But really this is just guessing.

Please excuse me, but LOL. My GW Instek has about 2.8-3.2 ns rise time, measured with a Leo Bodnard pulser just to make sure there were no errors coming from a DIY pulser. 3.5 ns = 100 MHz bandwith. Rise time <3 ns = Bandwidth >100 MHz. Even with a scope able to do some 109 MHz (3.2 ns rise time) with that pulser, I do need 150 MHz probes to get total system bandwidth on the 85 MHz range. Would need 300 MHz probes to get into the 100 MHz system bandwidth range.

No way DSO2512 really has these specs. Forum member pcprogrammer said it's probably a 200/100 MS/s on the first posts of this thread. He has worked intensively with FNIRSI 1013 which has the same F1C100 brains than DSO2512 and -it seems- similar ADCs (but two of them, not just one). Other forum member agreed with pcprogrammer after some testing. Sorry to say it, but you got harm from Heredia's scale.

I don't really know but I don't believe even a word in DSO2512 published specs.

[tatel]

Just do some background reading on sampling, Nyquist and aliasing.  It is neither rocket science nor black magic. Once you understand the circumstances under which aliasing occurs and what effect it has you will know when you need to worry about it and when you do not.

Yeah, I will do as soon I got some more spare time. However what I don't really understand is how they achieve seemingly good results wrt to show accurate frequencies and waveform. As I understand it wouldn't be so easy after getting aliasing during sampling.

However I throwed it into the slaugtherhouse so hopefully the old grumpy guys will have an answer after all the butchering.

https://www.eevblog.com/forum/testgear/zeeweii-toyscopes-how-do-they-do-it/msg4826186/#msg4826186

[Aldo22]

Hi tatel!

I don't really know but I don't believe even a word in DSO2512 published specs.

I don't know, I don't understand enough about it.
I also think the ZEEWEII specs are a quite optimistic.
Maybe someone experienced can estimate the order of magnitude based on the 40MHz square that I had already posted before, if that is possible.
It was measured with a probe and ground spring (PP-250).
Like I said, I don't know enough about it to say anything.

Buenas noches

[radiolistener]

If you can get me a good, working one, with reasonably easy-to-find spares, dual-channel, one of them invertible, 20 MHz analog scope that can be delivered to my door for €150, no matter if you got it for $80 or for free, I'm ready to pay that to you, just send me a private message and a Paypal link.

Before war it was possible, but now... If you can see where I'm living, you probably should understand that there is more relevant issue here - on how to survive. You even cannot imagine what we had to go through, I'm living in the city which was bombed every day for at least half year. About 700000 peoples were evacuated from my city, many peoples died, many just disappeared. And even now, I can hear blasts in the city, but now they are rare.

[tatel]

Before war it was possible, but now... If you can see where I'm living, you probably should understand that there is more relevant issue here - on how to survive. You even cannot imagine what we had to go through, I'm living in the city which was bombed every day for at least half year. About 700000 peoples were evacuated from my city, many peoples died, many just disappeared. And even now, I can hear blasts in the city, but now they are rare.

Sorry to hear that. Please take care. I wish you good luck.

[Aldo22]

Before war it was possible, but now... If you can see where I'm living, you probably should understand that there is more relevant issue here - on how to survive. You even cannot imagine what we had to go through, I'm living in the city which was bombed every day for at least half year. About 700000 peoples were evacuated from my city, many peoples died, many just disappeared. And even now, I can hear blasts in the city, but now they are rare.

I am sorry to hear that.
i wish you a lot of strength and confidence
Hopefully this stupid war will finally stop!

[Aldo22]

Hi 4thDoctorWhoFan

Like I said in my previous post, I get 35Mhz bandwidth in 1 channel/standard mode & about 40Mhz in 1 channel/OC mode.
I was using a 1Vrms input signal and adjusted the frequency until the amplitude dropped 3dbm to about 707mvRMS.  Connected using a 50ohm coax with a 50ohm load.

I received the TinySA today and wanted to check your measurement.
I connected it via a coax (RG-58/U 50 Ohm RoHS) and a 50 ohm load resistor. (See first image).
In the frequency range you mentioned, I can't see any tendency towards a drop in amplitude (3db) at all.
It dances a bit, but...
The second picture is a reference at 501KHz.
What am I doing wrong or different?
The measurements are all made in normal mode. No overclocking.

Thanks!

[Evi]

Till April 18 everything was OK.
But after absolutely idiotic discussion began.
It seems they are talking about a $10000 brand-name device,
whose seller deceived the buyer with special cynicism.
No time or pages were spared.

[4thDoctorWhoFan]

Hi 4thDoctorWhoFan

Like I said in my previous post, I get 35Mhz bandwidth in 1 channel/standard mode & about 40Mhz in 1 channel/OC mode.
I was using a 1Vrms input signal and adjusted the frequency until the amplitude dropped 3dbm to about 707mvRMS.  Connected using a 50ohm coax with a 50ohm load.

I received the TinySA today and wanted to check your measurement.
I connected it via a coax (RG-58/U 50 Ohm RoHS) and a 50 ohm load resistor. (See first image).
In the frequency range you mentioned, I can't see any tendency towards a drop in amplitude (3db) at all.
It dances a bit, but...
The second picture is a reference at 501KHz.
What am I doing wrong or different?
The measurements are all made in normal mode. No overclocking.

Thanks!

It's hard to read the settings of your TinySA in the picture.
What is the level output set to on the TinySA?  Did you adjust the level output as you increased the frequency or did it remain the same?

[wasedadoc]

Hi 4thDoctorWhoFan

Like I said in my previous post, I get 35Mhz bandwidth in 1 channel/standard mode & about 40Mhz in 1 channel/OC mode.
I was using a 1Vrms input signal and adjusted the frequency until the amplitude dropped 3dbm to about 707mvRMS.  Connected using a 50ohm coax with a 50ohm load.

I received the TinySA today and wanted to check your measurement.
I connected it via a coax (RG-58/U 50 Ohm RoHS) and a 50 ohm load resistor. (See first image).
In the frequency range you mentioned, I can't see any tendency towards a drop in amplitude (3db) at all.
It dances a bit, but...
The second picture is a reference at 501KHz.
What am I doing wrong or different?
The measurements are all made in normal mode. No overclocking.

Thanks!

It's hard to read the settings of your TinySA in the picture.
What is the level output set to on the TinySA?  Did you adjust the level output as you increased the frequency or did it remain the same?

That is old firmware on the TinySA.

The scope is wrongly set to x10.  Actual amplitude is a tenth of what the scope says.

The maximum output amplitude of the Low output of TinySA in Signal Generator mode is -7.0dBm.  That appears to be what it is set to in the first photo. -7.0dBm is 0.283 Volts peak to peak into 50 ohms.

[Aldo22]

It's hard to read the settings of your TinySA in the picture.
What is the level output set to on the TinySA?  Did you adjust the level output as you increased the frequency or did it remain the same?

Everything default: -7.0dBm
An no, I did not increase the level output of course.
I didn't change anything except the frequency.

Thank you!

[Aldo22]

That is old firmware on the TinySA.

It's from January 2023, but yes, the are upgrades available.

The scope is wrongly set to x10.  Actual amplitude is a tenth of what the scope says.

OK, but does it matter in this case?
The only finding is that the amplitude does not change significantly with increasing frequency.

Thanks!

[wasedadoc]

That is old firmware on the TinySA.

It's from January 2023, but yes, the are upgrades available.

The scope is wrongly set to x10.  Actual amplitude is a tenth of what the scope says.

OK, but does it matter in this case?
The only finding is that the amplitude does not change significantly with increasing frequency.

Thanks!

No, the x1/x10 does not really matter if only doing relative measurements as in this case.

[4thDoctorWhoFan]

It's hard to read the settings of your TinySA in the picture.
What is the level output set to on the TinySA?  Did you adjust the level output as you increased the frequency or did it remain the same?

Everything default: -7.0dBm
An no, I did not increase the level output of course.
I didn't change anything except the frequency.

Thank you!

Okay, I figured out the problem.
The scope cannot handle amplitudes higher than about 300mVrms and maintain bandwidth specs.

I retested the scope using your -7dbm (100mvRMS) settings and I got the same results you did.  The bandwidth specs was much better.  However, once I got above 300mVrms input level, the bandwidth dropped like a rock like in my initial test with 1Vrms.

Not sure why this is the case.  Perhaps someone here with more knowledge can shed some light.

[Aldo22]

Okay, I figured out the problem.
The scope cannot handle amplitudes higher than about 300mVrms and maintain bandwidth specs.

I retested the scope using your -7dbm (100mvRMS) settings and I got the same results you did.  The bandwidth specs was much better.  However, once I got above 300mVrms input level, the bandwidth dropped like a rock like in my initial test with 1Vrms.

Not sure why this is the case.  Perhaps someone here with more knowledge can shed some light.

Thank you.
Sounds weird.
Unfortunately I can't test this with the TinySA.

Btw: Is it possible to connect the Si5351 directly to the TinySA, or does it get damaged without attenuator? (+10dBm)
I only got the TinySA today, hence the stupid question. 

[4thDoctorWhoFan]

I am not familiar with the Si5351 but the specs for the TinySa states:

Low input mode:
1)  Absolute maximum input level of +10dBm with 0dB internal attenuation.
2)  Absolute maximum short term peak input power of +20dBm with 30dB internal attenuation.
3)  Suggested maximum input power of +5dBm with internal attenuation in automatic mode.

High input mode:
1)  Absolute maximum input level without attenuation of +10dBm.

To be safe, I would use some attenuation externally.

[JeremyC]

Everything default: -7.0dBm
An no, I did not increase the level output of course.
I didn't change anything except the frequency.

It's a good sign. Try to increase the frequency in 5MHz increments to point when your Vpp reading will drop to 0.2Vpp, it's -3dB and the frequency it's the scope bandwidth.

[Aldo22]

To be safe, I would use some attenuation externally.

Thanks.
Unfortunately I don't have any SMA-SMA attenuators and I can imagine that some tinkering interferes with the signal.

dBm means power (like watts) right?
If I have no idea what the signal generator is doing, how do I measure it?
Do I measure the voltage and then calculate something against 50 ohms impedance?
Sorry for noob questions.

[Aldo22]

It's a good sign. Try to increase the frequency in 5MHz increments to point when your Vpp reading will drop to 0.2Vpp, it's -3dB and the frequency it's the scope bandwidth.

Thank you!
The funny thing is that the amplitude almost doesn't drop at all (certainly not -3db) to the limit where it can no longer keep up.
See the attachment 1MHz vs 90MHz.
This is still measured in standard mode, not overclocking.
But as 4thDoctorWhoFan said, there seems to be a problem at higher amplitudes. I can not test this.