We have used ADE7858A in our product for energy measurement purpose. In few devices on field we have observed that all the electrical parameter value gets stuck and frequency value becomes 409.6 Hz. After some diagnostics we have found that RSTDONE bit in STATUS1 gets set. In this condition if we initialize all configuration register and start DSP again then it works normally for sometime then same behaviour is observed. But we are not executing soft reset or hard reset from microcontroller. ADE7858A reset pin is directly connected to microcontroller GPIO.
Is it possible that some noise on ADE7858A reset pin is causing problem? If yes then why problem is observed only on few devices? And how can I solve this issue?
Please show us your circuit schematic and register configure procedure in your design. Thank you very much.
Thank you very much for your reply. Please find schematic below:-
Register Initialization is as follows:-
1. I2C Lock by writing to CONFIG2 register
2. Write Voltage & Current Gain values in AVGAIN, BVGAIN, CVGAIN, AIGAIN, BIGAIN, CIGAIN
3. Set active, reactive & apparent energy threshold in WTHR1, WTHR0, VATHR1, VATHR0, VARTHR1, VARTHR0. Value VARTHR0 is written thrice.
4. Enable ram protection
5. Start DSP by writing to RUN register
Sorry, I can't see your circuit picture clearly. RSTDONE will be set if the chip receive reset command, transition of PSM mode and hardware reset. Please let the chip reset pin connect the circuit like datasheet test circuit and try again. Thank you very much.
We have already connected reset circuit as mentioned in datasheet but there is no major improvement. We have observed a fact that if current transformers are disconnected then problem is disappears. We are using normal 2 core wire of 40 inch length to extend current transformer secondary connection. Can current transformer cause this issue?
One more observation is that ADE7858A resets only when we communicate with it or another peripheral on I2C. We have 3 chips connected on I2C bus. This issue is only observed in some devices on field. We are not able to reproduce it in lab. Can emi or any other noise lead to such random behaviour?