Zoyi ZT-XT1 Transistor tester / multimeter

[bffargo]

Oh yes they did it.  Starting to look like Fnirsi with the flood of devices.  This one isn't cheap either. Doubt it would be any better than the original  ATMega328 based units with the highly tweaked and supported open source firmware.

https://www.aliexpress.com/item/3256811577169825.html

[Aldo22]

I think it's primarily a rechargeable 25,000 counts DMM with IPS Screen a few extras like the "waveform" graph.
That explains the price, in my opinion.
You get the transistor tester on top of that.

[@]

Looks interesting...

Listing for $42.57: https://aliexpress.com/item/1005011769799035.html Whoops, looks like I got the last one ... that one's now $72.

Specs pros and cons at first glance:
no 250μA or 2.5mA range as on other Zoyi models, despite having two jacks.
frequency measurement tops out at 1MHz
250Mohm range (accuracy is 5%, though)
100mF capacity range

With the firmware being upgradeable, this could be a nice unit for some tinkering. One of the screenshots shows different versions for the "Main" and "Tester" firmware, so I'm wondering if there's actually an ATmega328 (or clone) in there (which would be the "Tester"). The multimeter part looks similar to what they have in the ZT-XR1 battery tester or ZT-R01/R02 infrared cameras.

[indman]

With the firmware being upgradeable, this could be a nice unit for some tinkering. One of the screenshots shows different versions for the "Main" and "Tester" firmware, so I'm wondering if there's actually an ATmega328 (or clone) in there (which would be the "Tester")

Forget about ATmega328 in these combined devices. The basis of the transistor tester will be a Chinese STM clone with firmware without the ability to radically configure or change anything.

[Aldo22]

The multimeter part looks similar to what they have in the ZT-XR1 battery tester or ZT-R01/R02 infrared cameras.

In terms of basic data (counts, basic accuracy), yes, but in detail they are all different.
The ZT-XR1 has no current measurement at all, and although the ZT-R01/R02 have a 25mV range (which the ZT-XT1 does not have), they do not have a 250μA or 2.5mA range either.
In terms of measurement capabilities, none of them completely replace the ZT-225.

[lsqm01]

 :)Hey everyone, I'm an electronics enthusiast from China and I finally came across this review thread for the ZOYI ZT-XT1. I've been debating whether to get it as my first multimeter—how's its overall performance? Really appreciate your thoughts and input!

[Aldo22]

:)Hey everyone, I'm an electronics enthusiast from China and I finally came across this review thread for the ZOYI ZT-XT1. I've been debating whether to get it as my first multimeter—how's its overall performance? Really appreciate your thoughts and input!

I don't know, I think it's too expensive for what it is.
I'd rather buy a ZT-225 and a TC1, unless you really want the IPS screen on the DMM.

For example:
https://www.aliexpress.com/item/1005011825939113.html
https://www.aliexpress.com/item/1005004063983355.html

[lsqm01]

Thanks for your reply! Yeah, even here in China, this meter goes for around $39. What really caught my eye is the transistor tester feature, though it seems like it's based on the open-source LCR-T4 project. But at this price, it just doesn't feel quite right. Appreciate your input!

[@]

I was able to get one for just under USD 40. I would agree that $65 or $70 is too much.

Both meter and component tester work ok, though I don't have any precision references to check against. The operation using rubber buttons is not very ergonomic, I much prefer a rotary switch. On top of that, the missing low current ranges don't make this usable as a primary meter even for a low budget, in my view. That of course depends on whether one tends to to mainly low power electronics. The ZT-303 or ZT-225 are much better as multimeters, but of course without the component tester functionality.

The multimeter functionality is done by an Artery AT32F403A Cortex-M4 talking to a DM1109CEN multimeter ASIC. The MCU's flash memory seems to have read-out protection enabled.

The component tester is a Nation/Nsing N32G031, so a cheap Cortex-M0, as @indman suspected. I was able to read the flash memory and may try to analyse it.

You can see the 680R and 470k resistors commonly used in the component testers.

[indman]

Both meter and component tester work ok, though I don't have any precision references to check against.

Thank you for the high-quality photographs of the internal world of this device. Show us, if it’s not difficult, photos the capacitors measuring , for example MLCC with a capacity of 1-10 μF?

[@]

Measuring capacitors in multimeter mode works similar in speed and accuracy to other cheap DMMs. Sorry, I don't have pictures at the moment but they wouldn't be very interesting I think.

As for the component tester, I was taking the unlocked Cortex-M0 chip (N32G031) as an opportunity to learn to use Ghidra for reverse-engineering. The ADC measurements are converted to float numbers. Then it's mostly basic arithmetic, but in very inefficient soft-float implementation: multiply by 5 (volts), divide by 4095 (not 4096?) to get a voltage value. This could have been done more efficiently with integer math on the Cortex-M0, or by transferring the raw ADC value to the Artery M4 chip, which has a hardware FPU.

The two MCUs communicate over bog-standard 115200 baud UART, by the way. The N32G031 firmware doesn't implement any low-power/sleep functionality, I believe the master Artery MCU simply powers off the 5V boost converter to it unless there is a component measurement requested by the user (by pressing the TEST MODE button).

The firmware is also built with low optimization level and asserts enabled, which at least helps with reverse-engineering. There also is a bit-banged logging output on PF1 (pin 3) which uses printf, with all that entails...

I think I found a bug with the ESR measurements of capacitors. The measurement is done 10 times, with the first 9 results inexplicably being thrown away. Just the ESR measurement thus takes over 6 seconds instead of under a second. I also very much doubt the accuracy of the ESR measurement: The results are repeatable, but adding discrete resistors <10R in series doesn't change the result anywhere near like it should. I don't have another ESR meter to verify the ZT-XT1's result.

At the moment I don't have a way to attack the main processor, and that's the one that handles the entire user interface. So there isn't a way to extend the functionality of the component tester. I'm hoping for Zotec to release a firmware update, so I could analyse the Artery chip's firmware image.

[indman]

Measuring capacitors in multimeter mode works similar in speed and accuracy to other cheap DMMs. Sorry, I don't have pictures at the moment but they wouldn't be very interesting I think.

I'm not interested in DMM in this device. I asked you to take photos of capacitance and ESR measurements in component tester mode.

I think I found a bug with the ESR measurements of capacitors.

So far, our Chinese friends have not been able to correctly transfer the code from the author’s project to the new architecture. Therefore, there will be errors not only when measuring ESR.

[TheDefpom]

Zoyi sent me one to review, I haven't got that done yet as I have been flat out with other things, but I should get it done in the next week.

[BillyO]

[@]

Thanks for making a video on this device, Scott.

Your unit seems to be missing the plastic frames for the plug-in boards, too. A post on a chinese forum shows these, see below. I think the boards would be less wobbly with these in place.

I poked around the firmware for the component tester and here are some things I found:

• The static measurements such as junction drop voltage, forward and reverse currents are straightforward and I didn't expect or see any surprises. I've only looked at BJT, enhancement-mode MOSFETs, diodes and resistors, though. Darlingtons are also recognized as such but the current gain values don't make sense - hFE is lower than for a single BJT.
• The capacitor measurement for values below 90nF is buggy: it sometimes assigns the wrong probe numbers (see photo below). If that happens, the capacitance value might be slightly off, but not by much.
• As mentioned before, the capacitor ESR measurement is inexplicably done ten times, throwing away the first nine measurements. That increases test time significantly.
• I have some doubts about the ESR results as well, but don't have enough knowledge or the right equipment to make any definitive judgment
• Vloss (dissipation factor) for capacitor is measured differently depending on capacitance values. Below 18uF, the capacitor is charged to 2.5V, above 18uF, to 1.25V. The probe resistor is the disconnected and the voltage measured after 10ms. The voltage difference then gives Vloss. But due to the different charge voltages, capacitors below 18uF show values around 2%...3%, where as higher value parts show around 7%. Again I don't have equipment to check this, but it seems suspicious.

[MrCreosote]

I'd consider the Zoyi, but all I really need is a "T7" - and it looks like any variant will do what I want.  So that leaves Cheapest one on Aliexpress vs Cheapest one on Amazon.

However, if the Zoyi added oscilloscope functionality, it would have everything I need in my hand and I'd probably buy it. 
(Wouldn't be surprised if such a tester shows up within 6 months.

[bffargo]

As the one that first spotted this and started this thread calling out Zoyi as turning into another Fnirsi with the flood of similar and odd franken-tester things, I didn't think I'd end up buying one of these, especially before the hardware hacking experts got to it. But a random login to the AliExpress app last night presented me with not only a special offer price discount but also a huge 40% off with Coins deal + $7 off 49 all stores coupon, bringing it down well below the ~$42 price level I had seen it hit during the anniversary sale last month.

So for $24.50 + tax, I ended up succumbing to the curiosity. Though still not expecting much out of it other than to finally get a graphing DMM I guess. Will eagerly watch if anyone figures out its secrets on the xtor side.

As far as the plastic shells around the modules; are you sure those weren't just 3D prints by the user vs. something official from ZOYI?

[oskimac]

I just came across Defpom's review video. I think Zoyi really missed the point here. To me, it’s just a component tester without anything new, plus a poor DMM attached. They had a lot of opportunities to improve the concept, but they failed.

[indman]

I just came across Defpom's review video. I think Zoyi really missed the point here. To me, it’s just a component tester without anything new, plus a poor DMM attached.

In my opinion, the multimeter in this device is of good quality and can be used. But the component tester raises more questions than answers about the need to use it in the future practice of an amateur radio operator?
I did not see in the video review above a detailed analysis of all the capabilities of the component tester, testing only a few copies of transistors is not a serious test.

[bffargo]

Mine arrived today (the ~$24+tax special deal). Main 1.0.8 MM / Tester 1.1.3 firmware like everyone else's.  The 2 adaptor boards came with a black plastic shell around them, not just the bare PCB. Mine came with an extra ZOYI logo'd semi hard shell case (not just the nylon bag inside the box) which smelled like awful burned plastic but was in normal shape.

The limited button toggling modes is annoying as are the orders of up and down/left and right accordingly. Haven't put it through its paces much. The continuity mode at least is very sensitive and latched.  In all modes it seems there is a delay between the first meter readout vs. when the graph starts updating.

[coromonadalix]

its a mix of many things, while each section is not the very best,   it is not a full fledged multimeter   etc ...

Do not buy this thinking you're all set, same for Fnirsi  etc ...   all compromise, not fully safe and CATxx  compliant

[SammysHP]

New firmware "110 116" available:

http://www.szzotek.com/h-col-159.html

The new XT1 update includes the following:
1. Improved measurement error after 100mH inductance measurement.
2. Improved inductance and resistance identification and display.
3. Added JFET identification, such as 2N5457.
4. Added detection for some IGBTs, such as 25N120.
5. Improved multimeter hold display.

[uncle_sem]

with this new firmware transistortester says "resistor" without value when nothing is connected, and capacitor with some pF when socket is inserted.... 
where can I download older firmwares? 

[Aldo22]

with this new firmware transistortester says "resistor" without value when nothing is connected, and capacitor with some pF when socket is inserted.... 
where can I download older firmwares? 

Have you performed a calibration after the FW update?

[SammysHP]

with this new firmware transistortester says "resistor" without value when nothing is connected, and capacitor with some pF when socket is inserted.... 
where can I download older firmwares? 

Sorry to hear that. Mine came with 1.0.8-1.1.3 and it shows the same behavior.

Other firmware is unfortunately not available for download.