New Zoyi multimeter+oscilloscopes - ZT-70xS, up to 50MHz/250MSps (nov 2022)

[tol]

That's why even earlier I wrote something like this
https://www.eevblog.com/forum/testgear/new-zoyi-multimeteroscilloscopes-zt-70xs-up-to-50mhz250msps-(nov-2022)/msg5722843/#msg5722843

Yes, it's typical for such companies to "solve" firmware issues by releasing new hardware version, but this doesn't gain much respect from the customers. Basically, there is limited number of people who are interested in such equipment, and once they are left with a buggy device and abandoned support, most of them will remember this bad experience and never buy the products of this company again. Once I had exactly the same situation with a TrendNet router, they knew there was a bug in their firmware but instead of fixing it they just suggested to buy a new model of their router! I've never bought any TrendNet product since that time and I never will! Zoyi seemed more promising in this respect as they had at least several firmware updated for 703, maybe they still work on it, who knows. At least when I bought it in November it had a fresh production date and the latest firmware pre-installed, so they still continue to produce these things and probably think that this model will still be interesting to customers for some time even when it's full of bugs Why they never answer to anyone - this is not smart for them to treat customers like this, but this is the way they work... I tried to write to their youtube channel, but I wouldn't expect miracles there

[LucSVK]

I got mine from the official Zoyi shop on Aliexpress and with the recent firmware 1.4.2 as well. But mine looks like already used. Somebody returned it before. So maybe they just refurbish returned units slightly, update bootloader / firmware and ship them to customers again, who knows.

But this isn't the problem, bugs are.

It's pity that they can't release firmwares for older devices as an OSS. This would be great, because somebody who wants to give it the effort could mine a hidden reserves from it and improve it.

[lchao]

I've just found this in the comment section of a youtube video by the guy tim1tim2 who used to be the marketing manager (if I remember correctly):
Q: Temperature measurement can be added?
"We originally planned to include it, but for some reason, the code for temperature and the code for waveforms conflicted with each other. We tried to resolve the issue for three months, but to no avail. In the end, we had no choice but to remove the temperature feature."
It clearly shows they are struggling with programming the device...

[Atlan]

 

[indman]

It clearly shows they are struggling with programming the device...

[Edmiri]

Can i use a ch341a to flash your firmware?

[ajar171]

Can i use a ch341a to flash your firmware?

As far as I know, the CH341A does not have JTAG, it can only program EEPROM and FLASH chips through I2C or SPI. The CH347 should be able to do it, but I never tried that myself.

[Grandchuck]

I have been following this thread.  I own one and fall into the camp that feels it has value as is, and also feel that it is reasonably priced.  I have a lab oscilloscope but it resides in a different part of the house from where I solder and assemble things ... the Zoyi lives there.  (a link to my lab scope: https://www.eevblog.com/forum/beginners/help-with-a-simple-three-leds-lamp/msg5781355/#msg5781355)

Scopes are now like DMMs in that lots of folks own more than one.

[chebo]

May be someone knows, what is this - 'Signal output, time base limitation'. What does it mean?

https://youtube.com/shorts/iYAqnu6dGlY?si=5HcJ_Na9OHH24KzA
(how to insert here youtube video correctly?)

[Edmiri]

 Is it posible to use zoyi zt-703s with sigrok using your firmware?

[LucSVK]

May be someone knows, what is this - 'Signal output, time base limitation'. What does it mean?

https://youtube.com/shorts/iYAqnu6dGlY?si=5HcJ_Na9OHH24KzA
(how to insert here youtube video correctly?)

Hi, the problem is that you set output signal of your signal generator to sine 1Khz  instead of square 1Khz.
Go to menu and find output. Set square 1Khz and it should be ok.

[chebo]

May be someone knows, what is this - 'Signal output, time base limitation'. What does it mean?

https://youtube.com/shorts/iYAqnu6dGlY?si=5HcJ_Na9OHH24KzA
(how to insert here youtube video correctly?)

Hi, the problem is that you set output signal of your signal generator to sine 1Khz  instead of square 1Khz.
Go to menu and find output. Set square 1Khz and it should be ok.

But what does this have to do with the internal generator if I'm looking at a completely different signal?

[csuhi17]

But what does this have to do with the internal generator if I'm looking at a completely different signal?
[/quote]

There are two signal generators in this device.
One that you can set in the oscilloscope app.
The other is a separate application.
If you use the one in the oscilloscope, it will degrade the oscilloscope parameters.

The device is probably not able to use both applications at the same time.

reset the signal generator in the scope to a square wave of 1 kHz.

[chebo]

Go to menu and find output. Set square 1Khz and it should be ok.

reset the signal generator in the scope to a square wave of 1 kHz.

Yes! You were right! As soon as I switched the internal generator to square, everything worked as it should! A clean signal appeared, and the sweep parameters began to switch correctly. Thank you very much. But for me it was a big surprise - I could not even imagine that there was some kind of correlation! On the contrary, I set the sine to remove the harmonics of the meander, which, as I thought, could cause some kind of interference. And look how it turned out...

[OLderDan]

I just got mine, love it so far. I hate the measure display placement, but if you only have 2 choices selected in ch1 at least the measurement boxes stack to the left. I wish measurements were displayed in a single line across the bottom.
I wonder which is convention for the directions to manipulate time base and v/div, I got used to my hantek, but now it is 'backwards' or maybe this is true correct way?!!
I also discovered cursors won't display if persistence is set anywhere above min (off) after fiddling for ages!

Has anyone actually tried to use the signal generator in the circuit being measured? Did you just use alligator test leads because the frequency is so low, or is there a lead I can buy\make?

[OLderDan]

I don't know, probably it's a bug. The trigger place of the signal is shown correctly in the center up to 500 ns, but when we reduce the time scale to 250 ns or less, the trigger place is lost from the screen. It can be still found somewhere if we move the signal picture to the sides, but as we reduce the time scale it can be hardly made manually as it takes unreasonable amount of time (several minutes). Besides, the position of the marker on the scroll bar becomes lost (out of range) and it's even impossible to understand what part of the memory is shown at the moment. Moreover, getting back the trigger pointer to the center of the screen may takes minutes to roll it back with the arrows buttons - I couldn't find a button that could get it back at once (except reseting the hole device to defaults). I think it's the most annoying issue that I found in this device by the moment.

long press the menu button will recentre the triangle.

[tol]

long press the menu button will recentre the triangle.

Thanks, where did you find this information? Maybe there's something else I miss in there

[OLderDan]

I too spent far too long pressing the direction buttons to recentre the waveform, so i experimented!

[BOPOH]

Thanks, where did you find this information? Maybe there's something else I miss in there

https://www.dropbox.com/scl/fi/qm8vlwby6lqrolhy36m8r/ZT-703S-EN.pdf?rlkey=iaqfg4i1jcn2gumccvhv6msx7&e=1&dl=0
Zt 703s user manual page 12

[tol]

I seem to be stupid and don't understand something here, maybe someone can explain.
I have a signal like in the picture.

I used Cursor feature to measure period T, which is approximately 50 ns. This is also proved by the grid: 2 cells 25 ns each is 50 ns.
Now, frequency F = 1/T, which is 1/0.000000025=40Mhz.
So, why the oscilloscope shows 20Mhz, twice less? What is the real frequency in this example?

[tautech]

I seem to be stupid and don't understand something here, maybe someone can explain.
I have a signal like in the picture.

I used Cursor feature to measure period T, which is approximately 50 ns. This is also proved by the grid: 2 cells 25 ns each is 50 ns.
Now, frequency F = 1/T, which is 1/0.000000025=40Mhz.
So, why the oscilloscope shows 20Mhz, twice less? What is the real frequency in this example?

One cannot know unless the trigger level is set within the amplitude of the signal. 

[Atlan]

1/20Mhz is 0,05us -- 50ns means 2 divisions with a 25ns base, or 5 divisions with a 10ns base

[csuhi17]

I seem to be stupid and don't understand something here, maybe someone can explain.

There are two options for measuring frequency.
One is "Freq", the other is "Fcnt".

As Tautech suggested, set the trigger on the signal, in the middle.

If I try your method, then for me Freq alternates between 20MHz and 40MHz.
If the signal is not clean enough, there are intermediate values.

[tol]

There are two options for measuring frequency.
One is "Freq", the other is "Fcnt".

As Tautech suggested, set the trigger on the signal, in the middle.

If I try your method, then for me Freq alternates between 20MHz and 40MHz.
If the signal is not clean enough, there are intermediate values.

Seems to be my mistake in dividing values. In my case putting the trigger within the wave doesn't change anything, FREQ is always around 20MHz and  FCNT is 424 kHz (not sure what this value means). I tried to find explanations and there was answer from kloetpatra in this forum that:

FCNT: hardware counter based on the trigger. It will display 0.000Hz when trigger level is set outside of waveform.
Freq: software analysis of the waveform in the display memory. It will display --.-- if it can not determine the frequency.

So it looks like the trigger position is not important for Freq but important for FCNT, however, as I understand, they are expected to show the same value, and in my case they don't.

[chebo]

I seem to be stupid and don't understand something here, maybe someone can explain.
I have a signal like in the picture.

I used Cursor feature to measure period T, which is approximately 50 ns. This is also proved by the grid: 2 cells 25 ns each is 50 ns.
Now, frequency F = 1/T, which is 1/0.000000025=40Mhz.
So, why the oscilloscope shows 20Mhz, twice less? What is the real frequency in this example?

The signal period is 50 ns, why do you divide by 25?