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

[indman]

My interpretation of the 250mA MAX is simply this: If you want to measure more than 250mA, do not use this socket.What is to be marked on the front panel?  If it is OK at 500mA should the marking say 500mA MAX?

This is true if the user knows exactly what current he is going to measure. So a good tip is to always use socket "A" unless you know exactly the magnitude of the current to be measured. And if the user does decide to use the "mA" socket first and the current exceeds the 500mA value for a while. Are you sure that the measuring circuits of this device will withstand such a current if a cheap 600mA fuse is installed?
I, for example, am not at all sure and do not want to check it on my copy. I do not recommend doing such a check to other users.

[xKertx]

I'm trying to save a screeshot, but the "Save failed" message pops up.
Any idea what's causing this?
I dont have any other screenshots saved at the moment, so there should be enough space for that.
It has been working before, but i'm not sure if it has been working with the latest software (v1.39). It might be that i tested it with older software.

[4thDoctorWhoFan]

The fuse ratings are like highway speed signs. 
I take it as a suggestion and not literally. 

[indman]

I'm trying to save a screeshot, but the "Save failed" message pops up.
Any idea what's causing this?
I dont have any other screenshots saved at the moment, so there should be enough space for that.
It has been working before, but i'm not sure if it has been working with the latest software (v1.39). It might be that i tested it with older software.

I have version 1.3.8 installed on my device now and screenshots are saved normally. Check if you can access the memory partition of the device if you connect it as USB Media? Also try formatting that partition again.

[xKertx]

Also try formatting that partition again.

Ah yes. I did the format on the scope and it's working again. Thanks!

[apollo11fan]

Hey. I see you got a response. Yeah I'm on 1.3.9 and can do a save and view.
What's funny though, I tried to press Exit and that did not work.
Tried to go to the Menu and that did not work.
To get back to Scope Mode (from viewing the image of a prior save), I had to press the Mode key.

It seems the firmware is in a state of flux, and while some things are being worked on / improved, other things seem to be getting broken.

I wonder about their "regression testing", either in house, or by external individuals that can influence what is happening with the development of the firmware.
I'll write this again, it sure would be good for someone with the programming chops to enter into an NDA with the manufacturer and do some serious regression testing, and possibly some re-programming of the firmware.

[indman]

What's funny though, I tried to press Exit and that did not work.
Tried to go to the Menu and that did not work.
To get back to Scope Mode (from viewing the image of a prior save), I had to press the Mode key.

Yes, on version 1.3.8 exactly the same condition. I wonder by Exit works on version 1.3.2? I'm too lazy to check it out.

[nikbry]

Just tried it on 1.32 and it seems to have the same behavior, need to press Mode to exit.

[nikbry]

The fuse ratings are like highway speed signs. 
I take it as a suggestion and not literally. 

I try to avoid the hole fuse issue altogether using clamp meters to measure current.
To stay true to your analogy, I just take an airplane instead of the road...no speed signs up in the sky...

[Randy222]

I wonder about their "regression testing",

You are the testor.

[apollo11fan]

except, there is no formal means by which the issues I or someone else identifies makes it back to the people doing the programming and test.  but I know what you mean.

it's a shame. I think there are a number of products built around the same hardware. i wonder if they too have similar problems. I suspect they do not.

this product could be great...  i hope they keep issuing updates. but they really need a team dedicated to the product, and they should make it possible to submit bug reports for their products. 

[Randy222]

this product could be great...  i hope they keep issuing updates. but they really need a team dedicated to the product, and they should make it possible to submit bug reports for their products. 

It may just be limitations of the hardware and the coding just cannot compensate for, so they try and fix via code, but fixing one things breaks something else.
At some point we just classify this item as a no-buy and look for something else, or, just use it for the measurements it can do ok with.

[apollo11fan]

Randy, that is one way to look at it. I find many of the problems are User-Interface related, and that is typical of complex systems, but can be dealt with. It seems the device misses the mark having 50Mhz bandwidth, and there are some issues with trace artifacts (like waveform discontinuities), and annoyances like a partially but inconsistent bargraph display, button labeled exit doesn't exit from the saved image display, long presses on Auto button will do nothing while a short-press does, Auto in 1.3.9 is cough, cough not working quite right... (this list by no means exhaustive)... But these issues (with the possible exception of the bandwidth spec not being met  - not sure about that being impossible... might optimizations in coding help - not sure) are mostly all UI related and very fixable.

[nikbry]

Apollo11fan, I appreciate your perspective and agree with the points being made.

That said, I try to consider the context from the manufacturer's side as well. Being relatively new in the oscilloscope market and offering highly competitive prices means they likely face hardware limitations and resource constraints. They may also outsource some work, which can complicate development and testing, especially when aiming for ambitious goals like the possible release (just guessing) of a 100Mhz scope at a similar price point .

For me, as someone who remembers when oscilloscopes were much more expensive and only found in labs, getting a two-channel 40+MHz battery-powered scope with a decent display and a multi-meter for $65 is impressive (probe(s) included). It may have quirks, but it functions as intended. For more serious work, however, this or any other affordable handheld meter might not be suitable. Spending a bit more on brands like Hantek or Owon might offer better options, though they have their own issues too.

Could you provide some examples of the trace artifacts you've encountered, like the waveform discontinuities? That would help in understanding the extent of the issues.

[indman]

it's a shame.
this product could be great...  i hope they keep issuing updates. but they really need a team dedicated to the product, and they should make it possible to submit bug reports for their products.

I completely agree with you. I used to consider the ZOYI company more reputable than the toy FNIRSI since many models of multimeters from this company passed through my hands. This, of course, is not Fluke, but the devices for their price category worked and now work successfully according to the stated characteristics. The ZT-703S leaves me with a very ambivalent impression. On the one hand, as far as the multimeter is concerned, I have almost no complaints about its operation; the characteristics are quite decent, and it is convenient to use. On the other hand, as far as the oscilloscope is concerned, the number of ridiculous and childish errors that can be detected in just a couple of minutes of working with the device is simply amazing. Does it seem like the multimeter and oscilloscope were created by two different teams of engineers?
I believe that a device with so many ridiculous errors in the software simply should not enter the market! Is it really so difficult for an engineer, after releasing the firmware, to check the operation of the main modes, functions and buttons that the user uses often and which must be performed without failure? This is truly a shame for this team!

[apollo11fan]

I sense they likely ported the code from their line of multi-meters then added the o-scope feature, and that is why I think the multimeter part (with the exception of the bar-graph (there is no need for it) and the lack of a beep for diode testing) works fairly well.

If they outsourced the o-scope development, i might better understand the buggy nature of the UI. I wonder if the ANENG version of the o-scope AOS02 and 03 function similarly and have similar bugs. 

[nikbry]

Randy, that is one way to look at it. I find many of the problems are User-Interface related, and that is typical of complex systems, but can be dealt with. It seems the device misses the mark having 50Mhz bandwidth, and there are some issues with trace artifacts (like waveform discontinuities), and annoyances like a partially but inconsistent bargraph display, button labeled exit doesn't exit from the saved image display, long presses on Auto button will do nothing while a short-press does, Auto in 1.3.9 is cough, cough not working quite right... (this list by no means exhaustive)... But these issues (with the possible exception of the bandwidth spec not being met  - not sure about that being impossible... might optimizations in coding help - not sure) are mostly all UI related and very fixable.

Could you provide some examples of the trace artifacts you've encountered, like the waveform discontinuities? That would help in understanding the extent of the issues.

[xKertx]

I wonder if the ANENG version of the o-scope AOS02 and 03 function similarly and have similar bugs. 

Yes. They have same software. Only difference is the badge on the front panel.

[apollo11fan]

Could you provide some examples of the trace artifacts you've encountered, like the waveform discontinuities? That would help in understanding the extent of the issues.

Right, so I went back and used my 60Mhz DDS Signal Generator to supply a number of different waveforms at different frequencies, and did not see the discontinuities I had seen with pre-1.3.9 revisions of the firmware. They took the shape of part of the waveform having a flipped phase that didn't update. I don't see that now. That's good.

However, I do see that the measurement / estimation of frequency is non-existant for some frequencies based on the time scale chosen. Also, in general, unless the signal is of really low frequency, I seem to get better results using the high-sampling rate and leaving that setting alone. And finally, in V1.3.9, as others have mentioned Auto *one must press the button quickly or nothing happens* seems to fail.

Assuming the signal being checked is periodic, I wonder why the firmware can't remove the time-varying estimate of the mean of the signal to remove any DC offset, then simply look at zero crossings per unit time to estimate frequency and duty cycle or am I simplifying the approach too much? I'm personally not expecting these signal estimates to update really fast. Can't remember how quickly my Hantek DSO5202P updates, but it's pretty spot on with these measurements.

[apollo11fan]

I didn't realize the issue with the discontinuous waveforms was documented by someone else in this YouTube video:
https://youtu.be/0r5H0t7_PqA

[xKertx]

I don't know if i'm doing something wrong, but for some reason i get discontinuous waveform every time i freeze the waveform with trigger and then scrolling the display to left or right.
For example, few days ago i was looking at the output voltage rise when starting a powersupply. And when i tried to scroll the display, on all the parts that were out of display originally the line dropped down to 0V.

Edit: i guess that was just a user error.
It think it just runs out of memory when scrolling.

[apollo11fan]

Edit: i guess that was just a user error.
It think it just runs out of memory when scrolling.

To be honest, I'm trying to figure out what you are doing and wonder if there's some feature I missed.
Can you explain your sequence of key-presses, please and volt/div and time/div settings, plus more info on the use of the trigger?

I'd like to try to replicate the behavior. 
Thanks.