Jump to content


  • Content Count

  • Joined

  • Last visited

Everything posted by TomKraut

  1. TomKraut


    You're right, there is something like that for the C55x. Your comment made me google a bit and I found this: http://e2e.ti.com/support/dsp/tms320c50 ... 49849.aspx There seem to be special C55's which have JTAG disabled, so it's possible that they were used. I will know for sure when I get my board, probably sometime next week. Worst case scenario, I will ask one of my colleagues to swap the DSP with one that has JTAG
  2. TomKraut


    I call BS... Of course, TI might be able to disable JTAG functionality on their ICs, but it's not a documented feature. And why would they break out the JTAG header if it's unusable? And as for the MP3 licensing: if speedsthatbeat was able to substitute the BoosterPack .bin on the SD card with the one from the audio connected framework and it worked, there is no reason why it shouldn't work the other way around. So you could just as easily sniff around the MP3 code with an eZdsp5535. I guess it's just to discourage people from buying the ACTBP for DSP development, as it wasn't meant fo
  3. Hearing about these for the first time... Is Olimex still pursuing this? All I could find was a blog post from February stating that there are some prototypes and the boards should be ready for purchase by March... which they aren't, unless you count an eBay listing from the UK. If not, does anyone know about another cheap F5510 breakout board? The Olimex F5510 Starter isn't too expensive, but with LCD etc. I'm sure it should be possible to shave a couple of bucks of the price if there are only power supply, oscillator and USB connector. Cheers TomKraut
  4. Here is what I would do: get a cheap car USB charger from eBay, crack it open and then use something like a LM317 (easily obtainable LDO) to convert the 5V to 3.3V. That way, the nasty irregularities from the car supply have already been taken care of without you having to build a protection circuit yourself. Cheers TomKraut
  5. I see. In that case, forget what I said. I should have opened that datasheet first...
  6. First you say LEDs are power hogs, then you add a 75V to 3.3V linear regulator? Something doesn't add up Have a look at the LM5007 Step-Down regulator. WEBENCH is acting up for me right now, but there may be a reference layout for this. And if not, you might be able to get your hands on the gerbers for the eval kit. Cheers TomKraut
  7. TomKraut


    Sweet! I'm guessing integrated isolated XDS100? ;-)
  8. Talk about a relaunch gone horribly wrong... While the TI.com page was never very Chrome friendly, it is now completely unusable. I cannot get the collapsible sections on the product information pages to open. And besides, these are horrible from a design point anyway, because the page seems cluttered, but there isn't any information on it anymore. Who at TI made the decision to let the intern take a shot at a multi billion dollar company home page? Oh, and right now, I can't even get it to load... Cheers TomKraut
  9. TomKraut


    Have a look here: http://focus.ti.com/paramsearch/docs/pa ... STRY_PGE_T We use a TPS2113 to be able to power an entire Linux system (OMAP3530 based) from a backup source for 30 seconds when the main battery is removed... which to the average user seems like magic The switch decides which power rail to use based an the sensed input voltage. Cheers TomKraut
  10. I have the "Error connecting to the target" and sometimes "no emulator found" (or something) too. Whenever it happens, I remind myself that the Launchpad is only $4.30... And that I have yet to work with an emulator, for MSP430 or otherwise, that is always 100% reliable. These things happen, they may have to do with the fact that precise timing is required for MCU emulation whereas Windows (or Linux or Mac OS X) is anything but a realtime OS, and USB is anything but a realtime bus. As for the UART emulation part of the Launchpad: think of it as a gimmick. It may work, but it is not to
  11. With the example code from TI there is basically nothing more to do for a serial interface but implementing a handler to process the received characters. You have a serial interface running in less than 30 minutes. However, if the only thing you need is a serial-to-usb type of connection, I would go for an FTDI FT232R. There are SSOP packages available, and the PC drivers are easier to handle and especially to distribute. If this is a project you are not only doing for yourself, the Windows drivers with WHQL certificates are invaluable (asking your customer to ignore a driver warning is ju
  12. Sure. They are in the F5x and F6x series. http://www.ti.com/lsds/ti/microcontroll ... earch.page http://www.ti.com/mcu/docs/mcuorphan.ts ... +OT+430usb ;-) Cheers TomKraut
  13. And exposed contacts, the need for a SD card holder, most of the time no reliable information about access times, layout, environment ratings... Of course they are NAND flash, but what Reza was talking about are most likely the 'naked' NAND chips found in smartphones and the likes... usually referred to just as NAND.
  14. TomKraut


    Alright, alright, I will order one when my next paycheck arrives And I will try to build that baseboard I was talking about, with an integrated XDS100 emulator... which can pull double duty as a USB to UART converter, btw... Cheers TomKraut
  15. What are your requirements (capacity, write cycles, environment...)? What is your annual volume? Since SD cards are a consumer product which almost everyone needs in small quantities, it is the easiest and cheapest way to get a lot of capacity if your not making tens of thousands of your product, even if you don't need to remove your storage. If you don't need a lot of capacity and the ruggedness of solid mounted ICs is a necessity, you could go for NOR, FRAM, EEPROM, depending on the number of write cycles. I would be surprised if there are any dedicated wear-leveling ICs out there, b
  16. This is so cool! I want one! I'm with bluehash on the footprint issue, both options would be best. If there isn't enough space, you could solder a socket into a DIP footprint and just swap MSPs. Oh, and just put as many optos on there as possible. One can always leave them unpopulated... Where do you get the W5200s? I only found them at Mouser, and they are out of stock with a lead time of 17 weeks. Cheers TomKraut
  17. TomKraut


    Unfortunately, no. The only way to get I2S would be to lift the TLV320AIC from the board and solder wires to the pads. But since it is a QFN package, this wouldn't be practical... But then, the ezDSP C5535 is not that expensive... ;-) Cheers TomKraut
  18. TomKraut


    Schematics are online now. The JTAG port is completely exposed on a non populated header (easily fixed ;-) ), so this has suddenly become veeery interesting One could either connect a LP + XDS100v2 emulator board, or one could build a board which supplies power to the booster pack and has an integrated XDS100 (but no MSP430). Cheers TomKraut
  19. I debug the G2553 in CCS4 all the time. Maybe you need an update? Cheers TomKraut
  20. I received my code yesterday and immediately ordered my CCS ... which is supposed to ship on April 2nd ??? Why do I have to wait almost a week for an Email with my license to 'ship'?
  21. TomKraut


    I find this booster pack veeeery strange. You have a powerful DSP which does nothing but MP3 en-/decoding and thus should have a lot of spare processing power. Then you connect a MSP430 for the UI, which tells the DSP what to do, going so far that the OLED is connected to the DSP, but you control it's contents via the MSP... Overall, this is a very complicated design which has no relevancy in real life. I guess this might be a fun thing to play with, but a multi processor design where none is needed is always a bad idea. Still: if it had at least an integrated XDS100 emulator, this wou
  22. My stick is in transit, but no code so far...
  23. TomKraut


    Thanks, I quickly read through that thread, but couldn't find anything about synchronous transfers... UART mode is no problem (I did that myself with a TI code example to control a RGB LED from a C# GUI ;-) ), but there doesn't seem much information about USART. I guess I will just have to experiment, which isn't so easy, because I have no counterpart to test with...
  24. TomKraut


    I need to implement a USART on a G2553, has anyone done something like this? In the family manual there is a block diagram (Figure 15-1), where 'UC0CLK' appears to be a possible source for the baud rate generator. However, I wasn't able to find anything about this clock. The only synchronous forms of communication described are I2C and SPI, but USART is not mentioned. The G2x53 datasheet refers to a 'UCLK', but again, only for SPI mode. Is there a better way to do this than a pure software implementation? Cheers TomKraut
  • Create New...