Jump to content
43oh

olivluca

Members
  • Content Count

    51
  • Joined

  • Last visited

  • Days Won

    1

Everything posted by olivluca

  1. Yes, the chip is a pl2303hxd, in minicom I can set the speed exactly to 500kbps, but it doesn't work. I guess that the driver accepts the speeds but internally converts it to another one (Edit checking the source that's exactly the case). Either that, or simply the elm327 isn't working at 500kbps. Interesting that the MSP430 can work at 1Mbps, is that possible with the launchpad or does it need custom circuitry?
  2. I did a stupid thing: I set an elm327 clone to 500kbps, then I discovered that my usb to serial doesn't work at that speed, so I can't communicate with it (or maybe it's the elm327 that doesn't work at that speed, but I don't have an oscilloscope to check). Will a stellaris launchpad be able to send at 500kbps using energia? Nothing fancy, just blindly send the commands to set the elm327 to a lower speed. TIA.
  3. cd /home/ad/Desktop/energia-0101E0009/lib/ mv librxtxSerial.so librxtxSerial.so.orig mv librxtxSerial64.so librxtxSerial.so
  4. I asked a colleague with a steadier hand than me and a proper soldering/desoldering station to remove R1 and I can say it works perfectly with both monitors in parallel now. In fact the modified monitor works perfectly even without the other one connected (i.e. with no 75 Ohm termination in sight).
  5. I'm not familiar enough with C, but I think the "(byte) cast", just casts "cast" before shifting, hence only the last bytes shows., i.e. your code is equivalent to aa = 0xDD >> 24; bb = 0xDD >> 16; cc = 0xDD >> 8; dd = 0xDD cast; Maybe you could try aa = (byte) (cast >> 24); bb = (byte) (cast >> 16); cc = (byte) (cast >> 8); dd = (byte) cast; note the extra parenthesis. Edit: oPossum beat me, but I don't like the last solution since it's not portable. Aren't there standard functions/macros that do this taking care of the endianness of the
  6. OTOH, since I won't be cascading the next monitor from this one, it won't probably do me too much good even if it could work, after all impedance matching is to avoid reflections in the cable, it's not only about giving the correct voltage levels.
  7. Are you sure it's not just R1 that determines the input impedance? After all, R2 and R3 are in series with the (supposedly) high impedance input of the ASIC, so, in parallel with R1, their contribution should be irrelevant. OK, C1 will put R2 in parallel to R1, but only for frequencies above the video bandwidth (otherwise the input impedance would be ~37 Ohm). And the other monitor will still have 75 Ohm, so it would be equivalent to R1. It's been many years since I studied these things though, so I'm probably wrong.
  8. Hello, I bough a couple of cheap lcd monitors (one 4.3" and one 7") to put in my car. Both are to be fed with the same video signal, but, as I expected, I cannot simply put them in parallel since the signal degrades too much. The obvious solution would be to use a video distribution amplifier (maybe I could adapt this simple circuit), but I thought maybe I could modify one of the monitors for high impedance input. I traced the input path of the 4.3" monitor and it is like this (I don't know the values of what I suppose are capacitors, there's no marking on them, they're light brown smd c
  9. Maybe is this issue: https://github.com/energia/Energia/wiki/Linux-Serial-Communication
  10. You could move the existing module somewhere else (outside the /lib/modules/xxxx directory), then put the cdcacm.ko in its place then run "sudo depmod -a". Remember to repeat the whole procedure (clean/compile/install) if you change the kernel (it can be automated with dkms, in fact the supplied "install.sh" script should do that, but I doubt it works and I'm not going to try it).
  11. Download the zip file from the page linked by Rickta59 Unzip it somewhere cd where_you_unzipped_it cd linux-3.0/cdcacm-0.1/ make clean make This should produce a cdcacm.ko file sudo modprobe -r cdc_acm (just in case it's been already loaded) sudo insmod ./cdcacm.ko
  12. Thank you for the release, is there an ETA for the Linux version?
  13. Wait, does energia allow debugging? I though that the only debugging tool available was to add a bunch of Serial.print and/or toggling some debugging outputs.
  14. It's not the serial that's broken but the String class that has some problems with reallocating memory: viewtopic.php?f=38&t=3296
  15. Maybe you need this patched version of cdc-acm (I do) https://github.com/energia/Energia/wiki ... munication
  16. And that's because I didn't notice there was a patch against Energia.h.
  17. I just tried both libraries: CapTouch doesn't seem to be working (it causes Serial to send garbage), Touch compiles after commenting out a couple of lines in the constructor, I'm not 100% sure it works: when I touch the pin I can see isPressed() returning 1, but sometimes it stays at 1 even when I stop touching. Just using a wire though, maybe with a copper plate it works better.
  18. I ordered September 22 (better late than never) and the estimated delivery (to Spain) is for December 5.
  19. I tested and with 80mV voltage across R11, there's no difference in the analog value read (using energia) with or without a 10K resistor in series, so I think that the input resistance of the msp430 is high enough and a 10K resistor should be sufficient protection.
  20. I thought that I could use a 10K resistor on the analog input, to limit the current to the clamp diodes in case the voltage is too high, but then if the input resistance of the analog input is low that wouldn't really help. Any other suggestion?
  21. Hello, I have to control a small DC motor with no limit switches on the end positions, so I have to stop it when I detect a current surge. It's a 12V motor, when running it uses about 15-20mA, and when stalled it goes to 300mA. Since I don't have an h-bridge available but I have lots of transistors, for such a small motor I though I could use a tilden h-bridge, adding a small resistor on the low side and connect it to an analog input on the msp430. I simulated the circuit on circuitlab (OK, a resistor is not a motor but I thought that in a pinch it would do) and I see that with a 5 Ohm
  22. Since I'm running an old distribution and the bundled binaries require a newer glibc, I used that script and (after installing all required dependencies) it worked. I had to use the mspdebug in the binary distribution (the script didn't build it).
  23. While the example works, I had some problems with it: when you push the button it goes to GND, so the comparator should give a 0, but it actually gives an 1 with the button pressed. I thought maybe the reference is sent to V+ and the button to V-, so I studied the diagram to reverse the assignment and I don't understand why your example works: according to the block diagram, you are sending both the reference value and the input pin to V- while V+ is left floating. In the schematic, with CARSEL=1 and CAEX=0 the reference is sent to V-, and P2CA1, P2CA2, P2CA3 also select the input to feed to
  24. I realized that the msp430 has only 512 bytes of ram, maybe that's too little for the String class.
×
×
  • Create New...