Tonghui TH2827C LCR meter

[chilternview]

Anyone got one of these? From what I can see they are the same as a BK 895, which sells for about 3 times the price.

Is the IEEE488 option available? I can't see it on AliExpress.

thanks,
cv.

[KungFuJosh]

TongHui makes excellent hardware, and they license it to other brands who relabel it. TongHui's support is absolutely garbage though, so you're better off getting it from somebody like SourceTronic who rebrands the models with ST in place of the TH. SourceTronic's support is excellent.

[mawyatt]

We don't have this meter, but have the Tonghui TH2830 LCR meter which is very good.

Might caution on buying thru AliExpress as Tonghui indicted to us that they Do Not Authorize any AliExpress vendors to sell their products, so best to find a legit source like Josh mentioned.

Also, at this level of performance and cost, consider the Hioki IM3536 which supports 4Hz to 8MHz ranges and a selectable 10 ohm source impedance. We also have this LCR meter and it's also quite good.

Either of these don't support built-in graphics, so we developed our own plotting routines if interested.

Best,

[KungFuJosh]

Mike makes a good point. His Hioki goes to 8MHz, and the TongHui you're looking at only goes to 1MHz, but both are close enough in price. I don't know if it's hackable or not, but it is a possibility.

[nctnico]

You could try ask Batterfly to see if they can get the Tonghui for you for a reasonable price.

[chilternview]

Thanks, some good things to think about here. Time to check a few more suppliers and prices.

[ch_scr]

If you're not dead-set on a bench instrument, the Analog Devices ADXM2001 0.2Hz-10MHz LCR meter "eval kit" for ~1000€ is good value for money IMHO. If you can live with the "only UI is a serial console" part. Some might see the latter as an upside.

[chilternview]

That ADI board looks impressive. It would be quite a nice project to hook that up with a microcontroller and touch screen... getting tempted!

[zrq]

I have been tempted by the ADMX2001 board for a while, but it doesn't come with any calibration worries me a bit. I don't know what kind of reference one would need to calibrate it. Can electrically small SMD resistors calibrated by a DMM used as a transfer standard? The only manual (the wiki) implys that future deliveries of the board may come with a reference calibration, don't know when will that happen.

[mawyatt]

The ADMX2001 seems like an ideal candidate for a high performance DIY LCR Meter for someone with good software skills to create a nice GUI.

Best,

[nctnico]

Then again, making the test fixture will be very problematic at several MHz.

IMHO a better approach would be to use a network analyser to examine frequency dependant behaviour.

[mawyatt]

From the views at ADI it appears the PCBs have direct 4 BNCs for the respective LCR Kelvin connections at the proper spacing and orientation, altho wouldn't put much faith in the Kelvin leads supplied*. The external fixtures could be the same as we use with the Hioki IM3536 (Tonghui TH26048A and TH26009B) which are good to at least 13MHz according to Tonghui specs, so one can place some confidence in those specs, and why we use them with the Hioki IM3536 to 8MHz.

The cheap SMD LCR meter fixtures we have appear to be direct copies of the Tonghui TH26008A which is speced to 120MHz, so these might be OK to use.

If you check some R&S LCR meters/fixtures they "appear" to be rebrands of Tonghui, so speaks highly of Tonghui equipment. The stuff we have is excellent and not surprised if R&S chose to rebrand some items.

*Wouldn't consider Kelvin Clips leads for anything above ~100KHz for serious measurements, the quality Tonghui TH26011CS we have are speced only to 100KHz.

If we had any serious software skills we might consider the ADI ADMX2001, as it seems to be a fun project to just "see" how good this could be.

We'll be happy to make LCR device measurements with our TH2830 and IM3536 plus fixtures for device comparisons for anyone that wants.

Best,

[Dagobert1]

I have been working with the ADMX2001B since January. The ADMX2001 is operated via the serial interface using a Raspberry Pi 5 with 8 GB and an LCD touchscreen. The Raspberry Pi is running the latest Ubuntu 23.10. I develop the software under Qt6 in C++. In the project wiki on github I have added a few screenshots of the ADMX2001 Software and an installation guide for the tools required for development: https://github.com/Nordlicht123/ADMX2001_LCR_Meter

The sources of the software are not yet available. As soon as the software has reached a certain level of maturity, I will make the software available.

The ADMX2001b recognizes 7 impedance measuring ranges that are used for the autorange. Each of these ranges must be calibrated. I am currently in the process of writing the software for calibration over the entire frequency range. At the same time, I have found someone who can measure long-term stable reference capacitances over the frequency range of 0-10 MHz. This data is read into my software and then serves as calibration values, which are used for calibration with the real reference capacitance.

[ch_scr]

I have been working with the ADMX2001B since January. The ADMX2001 is operated via the serial interface using a Raspberry Pi 5 with 8 GB and an LCD touchscreen. The Raspberry Pi is running the latest Ubuntu 23.10. I develop the software under Qt6 in C++. [...]

Pretty nice project (and nice to see someone tackling it at all), but a RasPi 5 and full-blown Ubuntu seem "a bit" over the top for "just" a user interface.
"Complexity is cheap" I guess (I'm aware this allows for speedy development and interoperability).
I guess on the upside, the application could run on a "normal" PC with a standard Ubuntu install (or in a VM), with no need for any extra HW aside the LCR meter itself?

While the high-vis readout and preset buttons seems like they would do nice on a small-ish touch LCD already,
the rest of the UI still seems more appropriate to a "mouse & keyboard" interface and "PC-size" screen?
I'd suspect making one UI fit both the "touch only small screen" and "PC app for general use" without unwieldy compromises is quite impossible.
(e.g. the touch UI might need an on-screen Numpad for number entry, everything bigger, ...)
Which poses the question if there could be 2 "UI presets"?
IMHO if you go build a box with RasPi and touch LCD, it's much more effort to also have physical buttons on there and it makes more sense to "just" use bigger touch LCD
in the first place and implement all UI there?
I'd suspect most would just use the application for convenient communication with the device from "a computer already there" and fewer would build the "dedicated bench box"?

[Dagobert1]

Of course, you could also operate the ADMX2001B with a PC using an adapter from USB to RS232. The software can certainly be adapted so that communication can also take place via a virtual COM port. However, I would like to have a stand-alone device in the lab and connect to the device via Ethernet if necessary. As far as operation is concerned, I'm taking the pragmatic route and I've already tried out the possibilities offered by Qt for the main measurement page. I first had to make the calibration page mouse and keyboard operable, because the calibration is rarely needed during normal operation and I absolutely need the calibration data to calculate the measured values. It is also possible that functions such as calibration are only implemented via the web interface. For me, the measurement currently has absolute priority. What the Gui will ultimately look like will emerge over time. I would also like to implement the "Onscreens Numpad" window, I already had the idea. I have tried this in the pictures with the different outputs of measurement types. In the implementation, this is nothing more than an input field realized by buttons. What is actually written on the buttons is irrelevant.

The main task of a measuring device is high-performance and accurate measurement. All these things in the direction of the GUI are implemented quickly and are therefore currently less important. The high-performance communication to the ADMX2001 and especially the calibration is very complex and has already taken me many weeks of work.

[mawyatt]

Totally agree with your approach. Getting the most out of this ADI setup will require a stable mechanical and electrical environment, especially considering the 10MHz upper limit.

Would think a custom high quality, low noise PS based around a torroidal transformer within sturdy enclosure allowing the PCB 4 BNCs to directly interface outside, or short quality cables to chassis mount BNCs should be in order. Actually having the chassis mount BNCs is probably better since this removes the risk of PCB damage due to external BNC mishandling.

Personally I like the idea of including a RPi with built in touch display so the final instrument can act stand alone and not "tethered" to a PC, never cared for this and one reason our superb performance Pico 4262 rarely sees service. Awhile back we had considered using a dedicated RPi with touch display to allow direct visual displays of Impedance Plots with our 2 Lab Grade LCR Meters. Sort of a Stand With LCR Meter Box rather than having to create the plots on the controlling tethered PC Laptop (still a project on the back burner).

The GUI can evolve over time into something that might please folks, but the focus on the electrical performance at the beginning, believe is the correct approach.

You can put Lipstick on a Pig, but in the end it's still a Pig

Anyway, seems this ADI device has the ability to produce extremely good results over a wide frequency range in the hands of capable folks like yourself, so we are eagerly awaiting your progress.

As a suggestion maybe start a dedicated thread to this subject?

Best,