NurseBob

Members
  • Content count

    108
  • Joined

  • Last visited

  • Days Won

    12

NurseBob last won the day on February 9

NurseBob had the most liked content!

About NurseBob

  • Rank
    Level 1
  • Birthday 09/09/1950

Contact Methods

  • Website URL
    http://www.nursebobsblog.org

Profile Information

  • Gender
    Male
  • Location
    St. Helena, CA
  1. > I have used a multimeter and I measure this pin and I get between 1.10 and 1.15V. Is this in reference to the msp432 pin? I thought you were measuring the lm35... So, if you're testing the P5.4 pin, and you have CCS, I'd look in the debugger at the port registers to see if your statements to set P5.4 SEL1 and SEL0 are actually working as intended. Where is the sample code from? Aside from that, I don't have a suggestion.
  2. Do you have another device? I have some lm34s -T092 devices (read F instead of C) and in my home office I'm seeing .70V - or 70 degrees F when attached to the LP 5V tap. My recollection is that it's easy to damage the devices with ESD or if wired wrong... Other than that, I don't have more to offer. Bob
  3. A few questions: What is your power supply? What was the temperature of the environment? The lm35 needs at least a 4V supply. Its output is 10mV/degree C, so at 20 degrees C, the output should be 200 mV.
  4. You might also explore the software for one of TI's older experimenter boards which had the ability to record and play back audio for both specs and firmware. Not energia-friendly, but still instructive.
  5. While I2C is not my strongest area, I'm wondering about the clock in the two situations. Stepping through Energia's wire implementation I have this vague memory that it defaults to a 400k clock??? Could that play a role? And, as often can be the case, could pulldown resistors be playing a role??? Bob
  6. @bluehash Thank you for the hard work. I've had enough experience with running sites to know how time-consuming, and frustratiing, it can be. Bob
  7. One last note - The driverlib example runs at 100kHz and it appears that the Wire.cpp code sets the clock to 400kHz. I believe that the pullup resistors may need to be of lower values with the faster clock, but that's probably not all that's happening.
  8. Danial, I suspect the issues are more with Energia and the Wire library, rather than the board/mcu. Unfortunately, I don't have the time or skills to really delve deeply into the Wire/Energia code to better understand what is, or is not, happening. The fact that the driverlib code works under CCS indicates it's not the hardware. Bob
  9. On further investigation, I've managed to migrate a pair of msp432 driverlib examples from their P1/EUSCI_B0 to a P6/EUSCI_B1 implementation. The code examples selected were: msp432p401x_eusci0_i2c_10 (master) and msp432p401x_eusci0_i2c_11 (slave) Basically, a find/replace of all "EUSCI_B0" instances with "EUSCI_B1", as well as editing the interrupt enable from "EUSCIB0_IRQn" to "EUSCIB1_IRQn", and the interrupt handler from "EUSCIB0_IRQHandler" to "EUSCIB1_IRQHandler". Finally, the port assignments need to be updated from: // Configure GPIO P1->OUT &= ~BIT0; // Clear P1.0 output latch P1->DIR |= BIT0; // For LED P1->SEL0 |= BIT6 | BIT7; // I2C pins To: // Configure GPIO P1->OUT &= ~BIT0; // Clear P1.0 output latch P1->DIR |= BIT0; // For LED P6->SEL0 |= BIT4 | BIT5; // I2C pins As before, 2k2 pullups.
  10. I found I could get the test code to run with a pair of msp430f5529s after importing the energia project to CCS 6.1.3, recompiling and uploading. However, that approach was not successful with the '432 devices. I see the communication attempts with the '432 in the logic analyzer all followed by NAK, rather than an ACK. The successful comms are with the '430s. I note that I could not get the code to work when compiled and loaded with Energia 18. I haven't tried 17... I'm using a pair of 2k2 resistors for the pullups. The setup in both cases is the same, except for the LP boards. SCL is to P6.5 on the '432 and P4.2 on the '430, and SDA is on P6.4 on the 432 and P4.1 on the '430, which is the BP standard.
  11. 1) The "weird" display could be a serial comm mismatch between your PC serial port and the serial output from the boards. So, first check that your serial display is set to the same baud rate as the sketch. 2) I2C requires pullup resistors. While some sketches will use the internal resistors available on some msp devices, those internal resistors are often too "weak" - around 30K to work effectively, if at all. Typically, 1K - 10K work well. Further, I would use the 3.3V supply for your pullups. The msp ports are not 5V tolerant, as they are natively 3.3V devices. The 5V pin is coming off of the USB connection. 3) Posting the actual code you are running is always important and useful to help others sort out issues. Regarding I2C pullups - I wish that TI would offer a LP board with pads where I could add an "optional" pair of resistors. Tying-in external resistors is a pain...
  12. @@curtis63 I think you'll find what you're looking for in this zip file: here or go to this page http://www.ti.com/lsds/ti/microcontrollers_16-bit_32-bit/msp/tools_software.page - scroll down to "Code Examples for MSP Microcontrollers" and click on Show More to see all the examples This is a zip file of plain C or Assembler code, for both CCS and IAR Kickstart as well as a GCC make file. You'll want to look at the set of files named msp430g2xx3uscib0_i2c_01 - _13. I've found similar for the F2013 very, very useful. A couple of days ago I set up a master->Slave multibyte read/write. Since the F series devices share a very similar code base, "porting" the master code to a F5529 is pretty clean. BTW - typically the example use the internal pullup resistors. This approach is risky at best. Use a set of real pullup resistors (1k - 10k range) and comment out the pullup resistor code. HTH Bob Oh, BTW, I usually start the slave first, then the master, seems to be more reliable with the code examples. They are not robust regarding connecting, and on my system the voltage swings when starting the debuggers is often "seen" by the slave, putting the state machine into an incorrect state to connect. Also, I'm not trying any of this in Energia or with the launchpads at this time (I have connected a 5529LP to an F2013 via I2C in the past, just not now) - I use a pair of the FETs and run two instances of IAR to watch the debuggers in both windows. I don't think CCS supports running multiple instances unless you set up separate workspaces.
  13. >As per the hard copy provided by TI with board they say that all the pins are interrupt enabled If you review the user guide for the F5529 page 3 fig. 1-1, you will find that only ports 1 and 2 support interrupts & wakeup (from LPM4), further documented in table 4-1 Terminal Functions Also, posting the code you're running is the only way for others to evaluate what's happening or not happening...
  14. @@amc, I've used the Adafruit 2.8 TFT Cap-Touch LCD with an MSP430F529 - All I remember having to change were the pin assignment from the arduino sketch to match pin assignments on the '430 for SPI, and since the LCD I'm using was a cap-touch device, the I2C pins. HTH Bob
  15. >It's important to say that even if i dont connect the MSP-FET to the MCU at all it still pops the same error. That is expected behavior. Have you put a ammeter on your supply to see what's being drawn? If it exceeds 60-70 mA, you will see that error since the FET is limited to that max. If your system is designed to draw more power than the FET can supply, you need to set up for an external supply. Your comments regarding the board heating do seem to point to a short somewhere in your system... "3.5 Using the Power Supply Feature of the MSP-FET430UIF and MSP-FET All MSP debug probes can supply targets with up to 100 mA through pin 2 of the 14-pin JTAG connector. NOTE: The target should not consume more than 60 mA at peak current, as it may violate the USB specification. Details can found on www.USB.org. Example: If the target board has a capacitor on the VCC line with a capacity of more than 10 ?F, it may cause an inrush current during capacitor charging that may exceed 60 mA. In this case, the current should be limited by the design of the target board, or an external power supply should be used." slau647f.pdf