Digital meter-tweezers C-ESR.

[iet]

Finished assembling the digital C-ESR meter. Designed like tweezers. The tips are traditionally made of polished brass.
Measures the capacitance of capacitors from 0.5 uF to 150,000 uF.
Measures the ESR of capacitors from 1 mOhms to 10 Ohms.
Resolution - 1 mOhms.
Accuracy - 1%.
Measurement signal - 90 mV.
https://www.qsl.net/yo4hfu/ESR.html
Many thanks for the information kripton 2035.

[mawyatt]

Nice work!!

Took a brief look at the schematic and noted the test current source has a built in temperature bias of about 800~900ppm/C due to the uncompensated Vbe of VT1 which supplies the test current of ~ 10ma. Seems if you add a Vbe junction in series with the TL431 between Vcc and the base of VT1 this should give some level of compensation for VT1 Vbe, you'll need to change R2 to ~250Ω to keep the ~10ma test current tho.

Best

[iet]

As for the stable current source. I paid special attention to this during the setup of the circuit. I experimented with different TL431 and tested about a dozen of them. As a result, I chose the most suitable in terms of stability and temperature drift. Now the contribution of the temperature drift of the current source to the overall measurement error in milliohms is minimal. In addition, the meter is powered by a highly stable power supply.
So far everything is fine, all tests are running in normal mode. Therefore, if any things are noticed over time, I will definitely use your information.
Thanks.

[iet]

Test board. Measuring SMD resistors.

[mawyatt]

As for the stable current source. I paid special attention to this during the setup of the circuit. I experimented with different TL431 and tested about a dozen of them. As a result, I chose the most suitable in terms of stability and temperature drift. Now the contribution of the temperature drift of the current source to the overall measurement error in milliohms is minimal. In addition, the meter is powered by a highly stable power supply.
So far everything is fine, all tests are running in normal mode. Therefore, if any things are noticed over time, I will definitely use your information.
Thanks.

Maybe this wasn't clear in our post. The TL431 is fine wrt to temperature drift (no need for selection), it's the uncompensated Vbe junction of Current Source PNP Transistor VT1 that causes the drift. Adding a like Vbe junction (diode connected transistor) in series with the TL431 will significantly reduce this drift. You'll need to change the current setting resistor to ~ 250Ω to keep the current ~10ma tho.

No amount of TL431 selection and build care will correct for this built-in temperature dependance, it's basic to the design schematic as shown and will produce significant temperature drift of the current source. Simply adding the mentioned Vbe junction is series with the TL431 will reduce this built-in temperature drift by well over an order of magnitude

Best

[iet]

I am currently making another good meter, but with more advanced capabilities and protection from charged capacitors. The circuit is similar, with two current sources - 10 mA and 2 mA. 10 mA is the same situation. R-180 Ohm.

[iet]

Measuring the capacitance of capacitors. The accuracy of measuring the capacitance can be determined by comparing it with a multimeter. In our case, the accuracy of measuring the capacitance of the multimeter is 5% in the range of 10 mF-100 mF.

[iet]

The maximum capacitance of capacitors that can be measured by tweezers is 150 000 uF. This is all that I have found and connected to the board. Test board. This board can be used to test other devices, decreasing or increasing the total capacitance.

[iet]

The minimum capacitance that can be measured with tweezers is 0.5 uF.Measuring the capacitance of an electrolytic capacitor 0.47 uF 100 v.

[iet]

I tested my digital tweezers in a real situation, where there are 5 capacitors of the Low impedance series in parallel connection. 5 capacitors 2200 uF 35v. In addition, the ESR of these capacitors was checked by two more tweezers at a frequency of 100 kHz. The difference between the three tweezers was minimal and this is good.
https://www.chemi-con.co.jp/products/relatedfiles/capacitor/catalog/KY-LL-e.PDF

[iet]

There are situations in practice when it is difficult or impossible to measure the capacitance of parallel-connected capacitors with different homemade designs. I recently had such a case when inspecting a laptop board at the request of my friend, who had some problems. I tried to measure the total capacitance of SP-Cap 4x330 uF capacitors. The tweezers showed a greatly overestimated capacitance, almost 2.5 times. At the same time, it measured the ESR perfectly. I checked with other devices. The multimeter showed nothing, the capacitance meter too. The capacitance was normally measured only by LCR tweezers at a frequency of 100 Hz. I conducted a small study and found out that the capacitance measurement is shunted by the low resistance of the 6.5 Ohms circuit. That is, with such a low resistance in the circuit where the capacitors are located, it is quite difficult to accurately measure the capacitance with digital tweezers. But these are quite rare cases encountered in practice. Here, naturally, without a meter impedance is not enough.
I checked the ESR value with two tweezers at a frequency of 100 kHz. The difference is minimal and that's good.