2 AUG 2010 - jrb
Integration issue: 1-wire sensor did not provide data to the MSP430.
- passively checked all circuit connections between the MSP430, through the connector stack (H1 and H2), to the 1-wire bridge (DS2480) located on the Battery Board. Results were as expected.
- Checked the resistor network that divides down the RX line (from the DS2480 'RX' to the MSP430 UART1) such that the 5V level on the DS2480 is knocked down to 3.3V level at the MSP430 UART1 receive pin. Results were as expected.
- Verified that the EDU Battery board functioned on a Development FM430 board. Results were that data from the DS2480 was received by the MSP430 and placed as expected into a storage array. This demonstrates that the software functions as expected. Note that the Development board was powered with the 5V supply also used by the FM430 EDU.
- A scope was placed on the TX and RX lines on H1. Scope placed at same ground potential as the Development board. Signals were DC coupled at 1Mohm. While software was running, the scope captured the signal levels while the msp430 UART1 communicated with the DS2480 bridge. Results: The transmit from the UART1 to the DS2480 was 3.3V (via scope measurement). The receive from DS2480 to UART1 on msp430 was 4.0V (via scope measurement) with data.
- Results: Sensor data was received by the software as expected.
- Transferred the Battery board from the Development board to the EDU FM430.
- Transferred the 5V power connector from the Dev board to the EDU FM430
- Loaded the same software onto the FM430 board
- Attached scope in the same manner to monitor communication between the msp430 UART1 and the DS2480. (in fact, I left the probes attached to the H1 on the Battery board).
- Moved the scope ground from the Dev board to the Battery board
- Ran the software and measured the voltage levels on the scope (same scope setup as on Development board). Results: The transmit from the UART1 on the msp430 to the DS2480 was 3.3V. The receive from DS2480 to UART1 on msp430 was ~1.8V with data observed transmitted on the scope. HOWEVER: No data received by the msp430 software.
- From the observation, the voltage level from the DS2480 to the msp430 is too low to provide data, even though the DS2480 was transmitting.
It is curious that there is a difference between the Development board and the FM430 board.
One possible solution is to increase the drive current on the receive line. This can be done by lowering the total resistance that makes up the 5V-to-3.3V divider.
Inspection of spec sheets for the DS2480 and msp430f1612 may provide information on the currect value.
It will be also prudent to inspect any differences in jumper settings, or part numbers between the Development board and the FM430 board.
5 AUG 2010
Back in the lab
- verified that the cubesat FM430 board talking to another 3.3V uart (using the Olimex board) also loads the circuit down to about 1.8V
- through experimentation, it was determined that replacing R7 with a 3K ohm resistor, brings the RX line up to about 3V. R7 was 10K.
- using the formulas for a loaded voltage divider, this makes sense for a 5,000 ohm load somewhere on the RX line. Since we do not have schematics for the FM430 board, there may be know way to confirm this observation.
- Then if one assumes the 5K load out of the DS2480 voltage divider, with the original values of R7=10K ohms, R8=39K ohms, the theoretical voltage on the RX line is in the 1.5V range.
- This is matching close to what we observed.
- It is important to note that the load (of about 5K) does not appear on the Development Board. Therefore, until we determine that the MSP430F1612 UART1 RX pin is 5V tolerent, do not use the Battery Board on the Development Board.
- RESULTS: after the resistor out of the DS2480 was reduced to 3K, the software to interface with the 1-wire system performed as expected.