Jump to content
43oh

paulpthcom

Members
  • Content Count

    71
  • Joined

  • Last visited

  • Days Won

    1

Reputation Activity

  1. Like
    paulpthcom got a reaction from gsutton in programming a MSP430G2955?   
    The source of mspdebug has been updated with support for the 2955 in case anyone else wants to give that a try.
     
    http://sourceforge.net/p/mspdebug/code/ci/a0499ab6f067e60d387fbd31952c7cc36cbd9027/tree/
  2. Like
    paulpthcom got a reaction from gsutton in programming a MSP430G2955?   
    Another follow up, I was able to program the 2955 as a 2955 using CCS 5.x and the FET430UIF. i sent the USB logs off to the maintainer and hopefully he can extract the needed magic bytes easily enough to get a new build of mspdebug that fully supports this chip.
  3. Like
    paulpthcom got a reaction from spirilis in Switching Power Supply using a TPS61221   
    OK so now with the new 4.7uH inductor I get more weird behavior. From 1.0-1.5 V in I get 3.3 V out, from 1.7-2.7 V in I get around 3.0-2.7 V out.
     
    The new 10uH inductor just doesn't work at all, the output voltage stays in the mOHM range and the current just gets higher and higher as I crank up the input voltage (up to 500 mA).
     
    I'm confused.
  4. Like
    paulpthcom reacted to spirilis in Switching Power Supply using a TPS61221   
    Need 400mA or so. The TPS61221 uses current surges to do its thing... I use a 4.7uH with 650mA rating. Might be able to get away with 250-300mA if you have them though.
     
    Sent from my Galaxy Note II with Tapatalk
     
     
  5. Like
    paulpthcom reacted to spirilis in Switching Power Supply using a TPS61221   
    What's the saturation current on the inductor?  Also what kind of source power supply?
  6. Like
    paulpthcom got a reaction from bluehash in nrf24l01+ increasing distance   
    Tested out the PA+LNA module. Seems to work down to about 1.8-1.9 volts. In power down mode it + the 2452 I'm using consume about 1.6uA vs 0.9uA for the non PA+LNA module. Both of these are actually much better than I had expected.
     
    Not really sure how to go about getting a good measurement of peak transmit current, but setting my multimeter to store 1k readings and scrolling through them the max I saw was around 50mA.
     
    Still waiting to get a couple more sets so I can try both the RX and TX using one and see if it makes enough of a difference in my application to work.
  7. Like
    paulpthcom got a reaction from spirilis in Really Soft UART   
    So I'll start up by admitting this is horrible code, you probably shouldn't use it and it's just generally a bad idea, but someone might find it useful. I'm targeting a 2452 without hardware UART and wanted to be able to printf solely for debugging purposes. I know that there's at least one implementation that uses Timer A and T0.1 to implement a soft UART. That version is much more bullet proof and if you actually need to get every last byte you should use that, or better yet a chip with a hardware UART.
     
    That being said I only need this for debugging and can miss a character once in a while in exchange for being able to use any pin and having simpler interrupt logic. It's just a simple pure bit bang implementation but I wasn't able to find anything like it out there. Seems to work rather well at 57600 with only a junk character once in a great while. Anyways enjoy or laugh at, you won't hurt my feelings either way.
     

    //#define BIT_DELAY __delay_cycles((1000000/9600) - 12) // 9600 BAUD
    #define BIT_DELAY ; // 57600 BAUD

    unsigned char *g_port_out;
    unsigned char g_port_bit;

    void uart_init(unsigned char *port_out, unsigned char *port_dir, unsigned char port_bit)
    {
    g_port_out = port_out;
    g_port_bit = port_bit;

    *port_dir |= port_bit;
    *g_port_out |= port_bit;
    BIT_DELAY;

    // Clear screen
    printf("\033[2J\033[0;0f");
    printf("\033[%d;%df", 0, 0);
    }

    int putchar(int c)
    {
    unsigned int tx_byte = (c | 0x100) << 1; // Add HIGH stop bit and low start bit

    while (tx_byte) {
    if (tx_byte & 0x1) {
    *g_port_out |= g_port_bit;
    } else {
    *g_port_out &= ~g_port_bit;
    }
    tx_byte = tx_byte >> 1;

    BIT_DELAY;
    }

    *g_port_out |= g_port_bit;

    if (c == '\n')
    putchar('\r');

    return 1;
    }

  8. Like
    paulpthcom reacted to jpnorair in ultra low power mcu, is it a narrow segment?   
    EFM32 architecturally almost identical to STM32L, since they both use Cortex-M3.  There are certain universal attributes of Cortex-M, such as the NVIC and APB systems.  In any case, it isn't so bad once you get used to it (it's just turning-on interrupts and clocks in two places instead of one place), but certainly I would estimate about one extra day for driver debugging with Cortex-M than I would with MSP430.  That said, for a system with two or more I/O streams, Cortex-M is actually easier to get right than MSP is, because the NVIC does a lot of work for you.
     
    I don't measure with a timer, I measure with a button.  The first thing in the ISR is a line of code that brings the button pin to ground, and I measure the pulse on a logic analyzer or oscilloscope.  Once I have the number, I can subtract the number of clocks due to the line of code and get the interval.  The interrupt hardware treats a timer event or an external interrupt the same, or at least the difference is negligible.  You can get to that conclusion by looking at the schematics in the users' guide.  On the STM32L in particular, RTC interrupts are handled through the same hardware as the external interrupts.  On MSP430, they are different, but even if there is some synchronous logic in the interrupt hardware (there isn't for external line-in interrupts), I care less about losing half an ACLK period than I do about the amount of power expended during wakeup.
  9. Like
    paulpthcom got a reaction from gsutton in ultra low power mcu, is it a narrow segment?   
    STM32L has an M3 core, is in the same low power range as the MSP430 and has an ADC to DMA scan mode. Also loads of timers, RAM/Flash, cheap dedicated programmer/debugger and (at least on paper) better peripherals than the value series.
     
    That being said, with my limited experience, it feels much more complex to get configured properly, the STM website is brutally slow and finding sample code is non trivial.
  10. Like
    paulpthcom got a reaction from spirilis in nRF24 Temperature Sensor Bug   
    Looks right to me, you'll get an int value that's 100x greater than the actual float value (+/- whatever errors are caused by not using floating point math). Really curious to see what kind of results you get with this, again hoping I didn't screw anything up...
     
     
    Have you found the 46 ms delay between turning on the ADC and reading the value to actually be needed? I've found
    ADC10CTL1 = INCH_10 | ADC10DIV_4; ADC10CTL0 = SREF_1 | ADC10SHT_3 | ADC10ON | ADC10IE | ENC | ADC10SC | REFON; LPM3; ADC10CTL0 = 0; to work well, but can't remember if I've tried a delay to see if it improved things.
     
    And totally unrelated, but did you know you could define interrupts as the following in GCC, seems IMO a bit cleaner as far as wakeups go.
    __attribute__((interrupt(ADC10_VECTOR), wakeup)) void adc10_isr(void) { }
  11. Like
    paulpthcom reacted to cubeberg in Does an MSP430 need any kind of circuitry to drive a "Self Drive" piezo   
    Because the current is so low - you should be fine - I've directly driven small speakers.  You can drive an LED using the MSP430 which takes much more current than 2mA.
  12. Like
    paulpthcom reacted to spirilis in Does an MSP430 need any kind of circuitry to drive a "Self Drive" piezo   
    My understanding is the MSP430 pins should not be driven above 10mA. There are protection diodes in each pin (Vcc+0.7V and GND-0.7V I think) and they shouldn't have more than 2mA going through them.
     
    Side note, the MSP430 F5172's VIO-driven pins can drive 20mA at VIO rail (1.8-5.5V) with two Drive Strength options per pin.
     
    Sent from my Galaxy Note II with Tapatalk
     
  13. Like
    paulpthcom reacted to jpnorair in nrf24l01+ strange behavior when connected to same computer   
    Use components and board parameters spec'ed in the ref design.  El cheapo modules will cut corners wherever possible.  This is the problem.  It's quite possible they are even using counterfeit NRF chips.
  14. Like
    paulpthcom reacted to spirilis in Any good source(s) for launchpad/RF2500/1.27mm connectors?   
    They have a lot of those pin header sets on AliExpress.com, but I still don't see any right-angle female headers.  Weird.  Lots of 1.27mm pitch headers (including right-angle) and for very cheap:
     
    http://www.aliexpress.com/item/10-sets-1-27mm-Single-Row-Straight-Male-Female-Pin-Header-40Pins-X-10pcs/623864215.html
     
    That's a male/female matched set but straight.  I wonder, making your own board that mates to the eZ430 connector, maybe having SMD pads reach near the edge of the board you could solder the straight female headers sideways onto those.
  15. Like
    paulpthcom reacted to RobG in Any good source(s) for launchpad/RF2500/1.27mm connectors?   
    MDFly has a nice selection of 1.27/0.05 headers.
  16. Like
    paulpthcom reacted to spirilis in nrf24l01+ strange behavior when connected to same computer   
    Yeah so my setup is like...

     
    (PS, totally love the new Samsung Galaxy Note II I got on sunday!)
  17. Like
    paulpthcom reacted to spirilis in nrf24l01+ strange behavior when connected to same computer   
    Ohhh, gotcha.  I've seen this before.  The pushbutton should ideally not short to ground directly but short through a resistor, say 1K ohm.  I believe what happens is when you introduce the capacitor to GND it causes a "ground bounce" scenario that screws up the MCU and related circuits.
     
    In my original example where I encountered this problem, I just removed the capacitor and relied on software debouncing of the button.  In future designs I chose a 2.2K ohm resistor between the button and GND.
  18. Like
    paulpthcom reacted to SugarAddict in nrf24l01+ strange behavior when connected to same computer   
    only thing I can think of is shared ground among same hub usb ports perhaps. try both on 2 separate computers and see what happens?
     
    Tapatalk Mobile Device
  19. Like
    paulpthcom got a reaction from bluehash in MSP430G2955 wireless temperature sensor   
    In case anyone is interested (though arguably off topic) this is the STM32L version of the "same" board that I'm going to use to compare to the MSP430 version. Crappy soldering job is somewhat due to having a short somewhere and I had to desolder everything to find it (ended up being on the STM32L, ugh).
     
    This one is a bit more complex than the MSP430 version since I added a UEXT like port and used the 64 pin version instead of 48 so I could screw around with the same design and use some of their other chips.
     
    Lots of stupid little mistakes on this one but only two major issues. 
    The RTC crystal requires caps on the STM32 series and this one doesn't have any (doh!). I didn't put in any jumpers to allow switching between boot modes (flash vs SRAM boot), you wouldn't think this is a big issue if you aren't familiar with these chips but without this option its very easy to get the chip in a state where it's "bricked". Because of issue 2 I won't be able to play around with external clock, which rules out USB for this version of the board. But hey at least it's good enough to blink and give me some more experience with this family while I wait for yet another couple weeks till a new version gets delivered.

  20. Like
    paulpthcom got a reaction from spirilis in MSP430G2955 wireless temperature sensor   
    Nothing so far seems to work pretty well, I picked 10nF because that's listed on the MSP-TS430DA38 board (http://www.ti.com/lit/ug/slau278n/slau278n.pdf
     page 57), but I see now that it has a DNP on it, so no idea why they even put it on there. Will try < 1nF to see what happens.
     
    BTW thanks for the help starting out with DIPTrace!
  21. Like
    paulpthcom reacted to spirilis in MSP430G2955 wireless temperature sensor   
    Reminds me, I need to calibrate my G2452-based wireless temp bugs and deploy them.
     
    I would not use anything larger than 2.2nF on the reset pin if you ever expect to program it; prefer much less than that, I often use 470pF since I have 'em in my bin.  I'd switch to 1nF if I had them.  Something higher than nothing just to help stabilize it, but in reality many have no problems even without a cap there.
  22. Like
    paulpthcom got a reaction from spirilis in MSP430G2955 wireless temperature sensor   
    Yeah I know everyone has done one of these but it's a good learning project. Basically a 1x1" board with a 2955, button, LED, thermistor and a header for an NRF2401 board. I also did a similar board for the STM32L series (but not back yet from OSHPark) and plan to compare both to see which performs better overall.
     
    End goal is to end up having a number of these sensors throughout my house, periodically logging temperature to a central USB connected device. Preferably with 1+ year battery life on 1-2 AAAs.
     
    Overall the board came out pretty nicely, only changes I'd make would be.
    A bit more spacing between the 10uF and 0.1uF capacitors. Reverse the SBW connector (makes it easier to connect up to the FET) Switch the serial header to either have only T and R, or T, R, + and - Maybe put all the components on top, though moving the caps to the bottom saved a ton of space Smaller or no RTC crystal Nothing really bad and for a noob I think it came out pretty well.
     
    Only issue I have is that putting a 2.2nF capacitor between GND and RESET caused it to not always properly restart. Removing the capacitor fixed the problem and I'll probably try replacing it with a 10nF to see if it still causes a problem. Not really sure what the proper value is, there's some TI docs that state no more than 2.2nF, there's at least one schematic for this chip that shows a 10nF.
     
    Feel free to comment, critique, ask questions, etc...
     

  23. Like
    paulpthcom reacted to caa028 in Noritake VFD promo   
    Got an e-mail saying they are running a promo on CU-Y series - 6 models (3 configurations with 2 different interface options) all for $18 a piece. Limit - 5 per order.
     
    http://noritake-vfd.com/cu-y.aspx
  24. Like
    paulpthcom got a reaction from ike in programming a MSP430G2955?   
    Another follow up, I was able to program the 2955 as a 2955 using CCS 5.x and the FET430UIF. i sent the USB logs off to the maintainer and hopefully he can extract the needed magic bytes easily enough to get a new build of mspdebug that fully supports this chip.
  25. Like
    paulpthcom reacted to ILAMtitan in Recommended AVcc supply for G2955?   
    I typically just decouple DVcc and AVcc with a low impedance resistor.  10ohms is typically good enough to reduce the noise carryover to an acceptable level for high performance analog applications.  Depending on the accuracy you need though, you may not have to even do that.
     
    Any reason for the extra 22uF cap?  The standard bulk and bypass caps should be sufficient for this part.
×
×
  • Create New...