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

[Edmiri]

What newest version has no or few bugs?

[xKertx]

What newest version has no or few bugs?

They all have bugs.
I think the latest (1.42) works best, but some may disagree.

[apollo11fan]

It's strange. The company's website indicates 1.39 is the current revision. 
Are they just not bothering to post updated download links?
I hope they are working on bug fixes. 

[xKertx]

It's strange. The company's website indicates 1.39 is the current revision. 
Are they just not bothering to post updated download links?
I hope they are working on bug fixes. 

https://zotektools.com/products/triple-in-one-instrument-combining-oscilloscope-multimeter-and-signal-source/

The "firmware download" link is directing to V1.42

[metehoca]

Is it possible to compile the issues we encountered in version 1.4.2? Perhaps it could be possible to forward them collectively to ZOYI. From what I understand, they are interested in addressing the issues, but they might not be aware of their existence.

[tol]

Is it possible to compile the issues we encountered in version 1.4.2? Perhaps it could be possible to forward them collectively to ZOYI. From what I understand, they are interested in addressing the issues, but they might not be aware of their existence.

They did not even answer to the Chinese seller who I bought this device from.  But if you can contact them, this is reasonable to send them a list of bugs.
Here's one from me. In V142, when you have some calculations selected, make a screenshot:

Now turn off the calculations and make another screenshot:

You see a black rectangle where the calculations used to be. This will happen until you turn off and on the device.

[LucSVK]

Storage Depth  1Kpts

I think this in too few memory.

About the Zeeweii:
http://www.zeeweii.com/productinfo/dso3d12.html

It has bigger screen and more storage memory.

What storage depth does Zoyi ZT703S actually have?
And Zeeweii DSO3d12?


Edit: https://zotektools.com/products/triple-in-one-instrument-combining-oscilloscope-multimeter-and-signal-source/   says record length: 128KByte..  so if it's 8bit ADC then it's 128kpts?  ( 1 point is 8 bits, 128KByte = 128 000Byte =  128 000 * 8 bits = 128 000 * 8 /8  points ? )
Fnirsi 2c53t says 1kpts
Zeeweii DSO3d12 - 128k .. means again 128kByte? Do you know how many bit ADC does Zeeweei have?
I am confused, maybe somebody can explain this.

Thanks L.

[BOPOH]

Zeeweii DSO3d12 - 128k .. means again 128kByte? Do you know how many bit ADC does Zeeweei have?
I am confused, maybe somebody can explain this.

...
My assumption about the ADC being an MXT2088 (AD9288 clone) seems to be correct.

AD9288 is 8bit ADC
Considering how much captured waveform can be zoomed in/out (timescale), the "128k point" assumption seems reasonable.

[csuhi17]

Based on similar discussions, it's somewhat less than that.

The question is whether the amount of usable memory is related to the time base.

If I remember correctly, for the Fnirsi 1013/14, it was written somewhere here that the number of samples saved depends on the time base, so the number in the specification is the best case.

Zoyi also writes: "record length: Max 128Kbyte"
From this, I can conclude that the 128Kbyte is not constant here either.

Also, it is questionable whether there is a bug in it that prevents you from fully utilizing it.

Like for example in the Fnirsi 2C53P, where if you stop recording you cannot zoom in or out of the wave, you cannot move it sideways, etc... but according to its description it says "Storage Depth Up to 64kByte ".

[mightyohm]

Mine has the BNC issue as well (BNC cables won't go on far enough to "click" into place, very frustrating).
How hard is it to remove the plastic behind the BNC connectors? I may go this route so I don't have to make a custom "connector saver" that fits the BNCs.
 

[ajar171]

With the Zoyi, the only way to do a circular capture buffer (while the LCD is on), is to use the AHB SRAM1 and 2, which are 64K each, so 128kBytes combined.

With the 8bit ADC, maximum is 64k points when in dual channel, and 128k points in single channel. What the stock firmware does, is unknown, these are the maximum values.

Some update with my firmware is attached

[kwass]

Some update with my firmware is attached

From this picture, are we to understand that you've made enough progress on your custom firmware for the 703s to master the DMM functionality?

Awesome!

[ajar171]

From this picture, are we to understand that you've made enough progress on your custom firmware for the 703s to master the DMM functionality?

Awesome!

Yes, all the original DMM functions work again, also found the frequency for AC voltage . The DMM chip only sends its actual measured values, everything else is up to the main MCU, so I took the liberty to reorganize the soft button functions to my liking

Sorry for my potato phone camera pictures...

[Ilie321]

Hello 142 is higher size that 139 and for me it says not enough space when i try to copy it on device

[eljjc]

With the Zoyi, the only way to do a circular capture buffer (while the LCD is on), is to use the AHB SRAM1 and 2, which are 64K each, so 128kBytes combined.

With the 8bit ADC, maximum is 64k points when in dual channel, and 128k points in single channel. What the stock firmware does, is unknown, these are the maximum values.

Some update with my firmware is attached

That's amazing! —thanks for sharing!

[ajar171]

Hello 142 is higher size that 139 and for me it says not enough space when i try to copy it on device

First format the internal storage of the meter, the command should be in the oscilloscope menu (do not use your computer to format). When finished, swith to usb mode, that should also be in the scope menu

[LucSVK]

With the Zoyi, the only way to do a circular capture buffer (while the LCD is on), is to use the AHB SRAM1 and 2, which are 64K each, so 128kBytes combined.

With the 8bit ADC, maximum is 64k points when in dual channel, and 128k points in single channel. What the stock firmware does, is unknown, these are the maximum values.

Ok but 64k or 128k points sounds so much superior to 1k  the FNIRSI 2c53t I wanted instead of zt703 claims .. It's very suspicious.

[ajar171]

Ok but 64k or 128k points sounds so much superior to 1k  the FNIRSI 2c53t I wanted instead of zt703 claims .. It's very suspicious.

If the Fnirsi really has an FPGA, then it has still better hardware, despite of the low memory size. I also have my doubts about the 250MSPS, thats probably just an overclocked MXT2088, like in the Zoyi... Zoyi and Fnirsi have a strange way of "calculating" the MSPS of their scopes anyway   
I just don't like Fnirsi firmware and the absolute lack of updates.

[IC_Toaster]

Ok but 64k or 128k points sounds so much superior to 1k  the FNIRSI 2c53t I wanted instead of zt703 claims .. It's very suspicious.

If the Fnirsi really has an FPGA, then it has still better hardware, despite of the low memory size. I also have my doubts about the 250MSPS, thats probably just an overclocked MXT2088, like in the Zoyi... Zoyi and Fnirsi have a strange way of "calculating" the MSPS of their scopes anyway   
I just don't like Fnirsi firmware and the absolute lack of updates.

I think the most of the issues with this device are related to the scope. Bad voltage measurement, bad triggering and complex menus.

How did you get the firmware source code?

[ajar171]

I think the most of the issues with this device are related to the scope. Bad voltage measurement, bad triggering and complex menus.

How did you get the firmware source code?

I started a new firmware from scratch, I don't want their sourcecode. I have reversed almost all of the hardware needed to make a new firmware from scratch.
This scopemeter is now a devboard for me. First I want to program the DMM part properly. After that comes the scope, where I have very specific needs (for work) and the hardware is totally capable of fulfilling these needs

Bad voltage measurement: this can be done properly relatively easily, although calibration will be required.
Bad triggering: can be made better, but the hardware is still very limited
Complex menus: yes, this has to be done better

[IC_Toaster]

I think the most of the issues with this device are related to the scope. Bad voltage measurement, bad triggering and complex menus.

How did you get the firmware source code?

I started a new firmware from scratch, I don't want their sourcecode. I have reversed almost all of the hardware needed to make a new firmware from scratch.
This scopemeter is now a devboard for me. First I want to program the DMM part properly. After that comes the scope, where I have very specific needs (for work) and the hardware is totally capable of fulfilling these needs

Bad voltage measurement: this can be done properly relatively easily, although calibration will be required.
Bad triggering: can be made better, but the hardware is still very limited
Complex menus: yes, this has to be done better

I feel I am not able to do all this. I have not idea, your skill surprises me.

[LucSVK]

Ok but 64k or 128k points sounds so much superior to 1k  the FNIRSI 2c53t I wanted instead of zt703 claims .. It's very suspicious.

If the Fnirsi really has an FPGA, then it has still better hardware, despite of the low memory size. I also have my doubts about the 250MSPS, thats probably just an overclocked MXT2088, like in the Zoyi... Zoyi and Fnirsi have a strange way of "calculating" the MSPS of their scopes anyway   
I just don't like Fnirsi firmware and the absolute lack of updates.

In the Fnirsi 2c53p thread people say 250MSPS is valid for OCed ADC. It has MXT2088 which is according to their datasheet 100MSPS.  But they don't know if 250MSPS is for a combined value (125MSPS per channel) of the OCed ADC or what. It would make sense for 50Mhz oscilloscope.

Btw what is the FPGA in oscilloscopes used for? 

[ajar171]

In the Fnirsi 2c53p thread people say 250MSPS is valid for OCed ADC. It has MXT2088 which is according to their datasheet 100MSPS.  But they don't know if 250MSPS is for a combined value (125MSPS per channel) of the OCed ADC or what. It would make sense for 50Mhz oscilloscope.

Btw what is the FPGA in oscilloscopes used for? 

That is just not the way MSPS is advertised on a scope, this is just marketing BS. If everyone would do it like this, than my bench scope would have 8GSPS 
Both are 100MSPS/channel, and are overclocked to 140MSPS (Zoyi) or 125MSPS (Fnirsi).
If you feed in a 50MHz signal, you have 2.5 (Fnirsi) and 2.8 (Zoyi) samples per waveform, the rest is just interpolation. Dave made a video about that:



The FPGA does all the realtime calculations, memory management, generates precise clocks, trigger detection, etc.

Edit: just to be fair to Zoyi (and I am sure Fnirsi did it the same way), it is possible, to route channel 1 to the 2nd input of the adc through an analog switch. This way it is possible to double the sample rate, but only, when using 1 channel.

[EU1]

Mine has the BNC issue as well (BNC cables won't go on far enough to "click" into place, very frustrating).
How hard is it to remove the plastic behind the BNC connectors? I may go this route so I don't have to make a custom "connector saver" that fits the BNCs.

I simply trimmed some plastic off the sleeve on the probe's connector with a sharp knife, and now it fits perfectly.

[kamil004]

Any chance to download this custom firmware and test?