This is my first post on this forum. I have a problem with an isolation amplifier and i hope there's someone here who can help me out.

Let me explain my situation. I use an ACPL-C79B from Avago to isolate measurement signals. At the input there's a AD8226 In-AMP, which buffers the measurement signal (0-60mV or 0-200mV) to the ACPL.

enter image description here

The power supply is 15V, but because i want to keep the input offset from the AD8226 as low as possible, i create a virtual ground. The virtual ground is created at 10V. So i have 15V to 10V = 5V positive voltage, and 0 to 10 = -10V negative voltage. It's shown below.

single to dual power supply

The ACPL is powered with the 5V positive voltage from this virtual ground (at the primary side). At the secondary side it is also powered with 5V, but from a floating power source.

The datasheet of the ACPL-C79B shows: 60mV input -> 0.5V output. When i use a new ACPL, the output voltage is exactly 0.5000. Perfect! But after a while, the output becomes 0,4860 (still 60mV input). I tried multiple chips (there not very cheap), they all give the same result after a while.

I can't find the problem. Please ask me, if there's something not clear.

(Because the ACPL creates a differential signal, the output voltage (0.5V) is Vout+ - Vout-. This is measured on the chip directly.) (The filter between the AD8226 and the ACPL is not used!)

Has anyone an idea, what causes the problem?

Could it be the virtual ground? Could it be the input signal? Because the - of the input is connected with the virtual ground.

  • \$\begingroup\$ To debug this, first try with a real +5, GND and -10 volt power supply. I haven't read the datasheet for your circuits but in your circuit, your virtual ground will randomly vary due to the thermal coefficients of R1 and R4, and the input voltage offset of IC1 which can also vary with temperature. Since it drifts after a while, this may be the problem. \$\endgroup\$ – pipe Apr 21 '16 at 16:55
  • \$\begingroup\$ Such capacitance on opamp is not a good idea. Remove it. \$\endgroup\$ – Gregory Kornblum Apr 21 '16 at 17:57

Please insert 7805 regulator that supplies only the ACPL for the stability of the IC. Refer to the datasheet.

  • \$\begingroup\$ Please add more information to your answer. It is too short for being helpful. Some more lines should be spent to improve the readability and to explain, why you came to the conclusion. \$\endgroup\$ – Ariser Dec 20 '18 at 21:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.