Being very impressed with the specs of the SOC PGA970, an awesome SOC designed to excite, condition then digitize analog LVDT sensors!
Being so enamored with the idea of such a tiny device to utilize LVDT sensors, I invested in a PGA970EVM package from Texas Instruments.
I spent many hours with the device trying to make it operate with no joy.
3 replacements were sent with the comment that new SOCs were soldered to the boards.
Each EVM would produce a sine wave of excellent quality the first time it was powered on.
After the first write cycle to change the parameters of the sine wave (frequency and amplitude), the SOC would not produce a sine wave.
I have carefully verified grounding of my equipment for protection against ESD, power supply voltage quality (very clean DC), AND have provided captures of wave forms during SPI communication during read/write cycles. (also very clean wave forms)
After several replacements and direct support with a Texas Instrument engineer, he believes it is some mistake on my part but can not say what it is.
After reading all of the E2E posts regarding this EVM unit, it is obvious that I am not the only one to suffer this issue.
Using a USB2ANY that was tested functional in the TI lab, mass read/writes to the device produce random data written to the registers.
When performing read/writes to individual registers the expected data is written to the registers.
I have requested an RMA for refund the 3rd EVM module.
at the e2e.ti.com a search for PGA970EVM will show produce the post many have made.
This is a big blow to my confidence..... Is it actually possible that TI has provided an evaluation module with defective firmware for the PGA970?
This is the sine wave output after mass write
This is the SPI communication durring a mass write cycle: channel 1 is MOSI, channel 2 is MISO, channel 3 is SCK
Jaime
Being so enamored with the idea of such a tiny device to utilize LVDT sensors, I invested in a PGA970EVM package from Texas Instruments.
I spent many hours with the device trying to make it operate with no joy.
3 replacements were sent with the comment that new SOCs were soldered to the boards.
Each EVM would produce a sine wave of excellent quality the first time it was powered on.
After the first write cycle to change the parameters of the sine wave (frequency and amplitude), the SOC would not produce a sine wave.
I have carefully verified grounding of my equipment for protection against ESD, power supply voltage quality (very clean DC), AND have provided captures of wave forms during SPI communication during read/write cycles. (also very clean wave forms)
After several replacements and direct support with a Texas Instrument engineer, he believes it is some mistake on my part but can not say what it is.
After reading all of the E2E posts regarding this EVM unit, it is obvious that I am not the only one to suffer this issue.
Using a USB2ANY that was tested functional in the TI lab, mass read/writes to the device produce random data written to the registers.
When performing read/writes to individual registers the expected data is written to the registers.
I have requested an RMA for refund the 3rd EVM module.
at the e2e.ti.com a search for PGA970EVM will show produce the post many have made.
This is a big blow to my confidence..... Is it actually possible that TI has provided an evaluation module with defective firmware for the PGA970?
This is the sine wave output after mass write
This is the SPI communication durring a mass write cycle: channel 1 is MOSI, channel 2 is MISO, channel 3 is SCK
Jaime