Jump to content


  • Content Count

  • Joined

  • Last visited

  • Days Won


Everything posted by veryalive

  1. hi - although I'm a bit late to this thread... Regarding +5 volt operation on the LM4F120 I/O pins... Specifically the LM4F on the 'old' Stellaris Launchpad..... I have followed the data sheet explicitly where it states that certain of the I/O pins are 5volt tolerant on input. I have not damaged anything by reading 5volt input signals to the chip. I cannot measure anything to confirm it is true, however; I just trusted the data sheet. The data sheet I'm using is : DS-LM4F120H5QR-13200.2535 SPMS294G, starting in Chapter 10. And, in section 10.1 : All GPIO signals are 5-V tolerant when configured as inputs except for PD4, PD5, PB0 and PB1, which are limited to 3.6 V I'd be very curious if someone else has successfully used this for I2C which uses pull-up resistors (+-3 K ohms) to a +5volt bus. Cheers..
  2. Although you don't mention the resolution, the update rate, your experience level, and your application, you may want to consider this.... For low frequencies such as yours, you can measure the period between rising edges of the signal, then calculate it's reciprocal. Many commercial frequency counters will do this. You'll get better resolution and faster update rate. And averaging is easier, to remove noise. Just mentioning this in the spirit of presenting an alternative. Due to other activities at the moment, I don't have time to code an example, but it would go something like this, in Energia: ... Define a pin interrupt to trigger on the rising edge, using 'microseconds', get the time delay from the previous interrupt. (Or poll the pin for the transition). ... Optionally, average several samples. ... Calculate the inverse of the period to get the frequency. ... And output it on the serial port. Hope this helps, if it's not what you're looking for, then the other methods mentioned on the forum are your best bet. Cheers.
  3. @blankname -- and others interested in the 16KB code limit of the 'free' CS compiler. Yes, I too found I could compile a file with more than 16KB code loaded into an MSP device. (Using the 'free version' on Windows, CCS v 6.1.3) As I have a licence now (yet to be implemented on this machine), I'm not too worried about future changes TI may make in counting / restricting compiled code size. For my part, I sort of assumed it was a TI CCS bug as the compiler version number (v15 etc) looks much different than pervious ones (v4.x). Attached is a text file with my test code and results in a short report format I made for my notes. Cheers. edit .... (note ........ trying to find out how to add a file to this post) (have to select 'use full editor')Code size GT 16KB CCSv6.1.3.txt
  4. @@zeke and others.... cool idea., that USB wifi idea. >>>> did you actually get it going successfully? hmmmm, now I'm wondering if I shouldn't haver ordered TWO licences. Two licences should allow me very flexible coverage of 4 PCs if I interpret things correctly. BTW, I received my CCS licence info by email from TI, now to install it on 2 machines.
  5. @@bluehash Thanks for the offer, but I sorted it out and placed my order. In fact, as someone already mentioned, I just edited away the full-priced CCS and it went through. While I was at the store, I ordered an MSP430 with 4 24bit sigma- delta ADC channels (MSP430i2041pw 28 pin tssop). A big thanks to the forum folks! Cheers
  6. .... and I notice on the TI WIKI for CCS licencing that the node-locked licence allows use on 2 (two) registered machines: >>>>> QUESTION - is this indeed true? Cheers ======>> copied off that page... http://processors.wiki.ti.com/index.php/Licensing_-_CCSv6#Node.C2.A0Locked_Licenses Node Locked Licenses Node locked licenses are tied to a particular machine. You can use all of the features of CCS but you can only use them on the machines noted in the license file. The license file allows you to specify 2 MAC addresses. A MAC address is a unique identifier in network adapter. Most new machines have multiple of these. Node locked licenses must be purchased from TI or one of our distributors.
  7. This is incredibly good news if true. A full CCS licence for 10 bucks. I just tried placing an order on the TI Store, and it failed. So I got onto TI chat, explained the situation and got the following reply: --- excerpt --- Please contact your regional support group via phone so an agent can assist with placing this order for you You will need to contact your regional support group. You can locate your region at the attached link. -------------- And some phone numbers were added.... I'll call the number tomorrow morning and see if I can actually purchase sucha a CCS licence. Cheers
  8. Hi, Is there a special reason why it has to be a (N-channel) mosfet, @@Fred? How about a plain old bipolar (2N2904 npn) with base current limiting resistor; you'll need to use a flyback diode anyway. So there's just one extra passive component (diode + resistor). Cheers.
  9. Here's a BLDC project that I learned quite a bit from..... see the cut / paste at the bottom of this message ..... (author = @lgbeno) The project uses CCS, a '2553 at 16MHz, and the code is very modular, the documentation (including use of the async port to annotate the scope traces) is cool. Regarding your questions, though : My own view is that CCS is a great IDE, has a solid compiler and the memory / HW register / variables mods one can do while the test processor is halted can really save a lot of time during debugging. CCS, along with Energia is a formidable tool for me for doing small projects and actually finishing them. Best of luck, ====== search the forum on BLDC ======== Posted 10 July 2013 - 05:18 AM Cool project, I did something very similar as well! https://github.com/lgbeno/BLDC-Booster ======
  10. @@RobG Super big thanks. I downloaded your main.c (5.49KB, your post just above) and replaced the old main.c with the new one. But not PCD8544.h (so as to try one step at a time). >>>>> It worked. It compiled and dowloaded into the 2553LP. No 5110 attached at the moment. No errors in the original PCD8544.h file. I'm on two machines here: Win10 and Vista. Vista worked, next I'll try Win10. I'm also on CCS = on the Vista, 6.1.3 on Win10. (but I'm quite sure I got the same prob on the Win10 machine with CCS 6.1.3) I have no clue at all why this happened --- and FWIW, the same thing happened with your 4 channel voltmeter onto Nokia 5110. Maybe the forum download put in non-printing chars? But I'm not going to debug that one... Anyway, thanks again. Next step is to wire up the Nokia 5110 and try out your excellent code. Cheers.
  11. About the 'off topic' 16 / 25 MHz thing for MSP's: ======================================= Its likely more about marketing and silicon economics than physics or architectures, imho. And likely no EE degree is needed on this one. A possible scenario could be: - MSPs are used in a defined market area of small machines, low cost, low performance demands, and likely high volume. (and there are a few other key drivers for the product certainly) - 16MHz would cover a very large majority of the applications that an MSP would likely face in its market - and 16 MHz is a pretty low speed for a modern CMOS chip process..... This speed means, among other things, that the actual silicon chip size for an MSP is pretty small. A higher speed would mean, again among other things, a bigger chip size in square millimeters. - and that the process rules (line widths / parasitics) across the MSP line are reasonably fixed - --- more variants = more differences from design to design === a more difficult to control development environment for the TI MSP chip designers. (thus, a 'bad' thing). etcetera. etcetera. etcetera. And finally - the point of it all - is that TI, or its fab partner, can make a huge number of MSP's per silicon wafer - a bigger number than for instance a faster / larger / higher-featured product. A small chip, like the MSP, can have thousands per wafer, each wafer raw cost being 100's or so USD. So, they strive for: minimum chip die area / maximum yield thru conservative design / maximum working chips off a wafer. So much of the chip industry is economics and it can become complex with lots of variables rather quickly. Does that get close to the open / off-topic query? Cheers.
  12. Hi @@RobG ... If you have some additional advice on how to get your code working - I too have a huge number of errors - exactly like the poster on 3 aug 2015. My CCS6 build looks the same as his screenshot with the first error starting at line 64 ( int c = 0x20;) then onwards. I compiled the code today. I followed your advice (see the note I cut / paste immediately below) but with no positive result. It is very strange. And, in fact, when I tried your ' 4 channel - voltmeter - on - a 5110' - elsewhere in this forum - but I got the same huge number of errors. I had no time to investigate then. But thought to bring it up now - I have 5110 displays on hand and would love to use them ! Thank you. Cheers. ================= RobG's advice..... Posted 04 August 2015 - 04:14 AM This is weird, I got the same error when I created new project. Comment out line right above first error, //setAddr(0, 0); Build, then un-comment that line and build again, error should disappear.
  13. Shipping in The Netherlands is (only) $7 USD. But I think TI's Euro disti hub is in this country.
  14. veryalive

    OV7670 and MSP430

    Yes, 10K resistors at 400KHz I2C rates would not be recommended.
  15. Clearly, I'm diverging from the point of your question, @@yyrkoon (your question was about assigning specific MSP functions to selected physical pins using the internal MSP registers)) Perhaps you already know .... But if its speed and transfer efficiency you're looking for on a '2553 ADC channel(s), why not look at the DTC mode? It is effectively a dedicated DMA channel which transfers ADC10 data to memory. Cheers, CHAPTER --- within MSP430x2xx Family User's Guide. Literature Number: SLAU144J 22.2.7 ADC10 Data Transfer Controller
  16. Well, I'll try this.... I've heard of folks (hobbyists) who've gotten CCS for a very low price. Question, then, to TI, is if they can provide the stock CCS to us hobbyists which goes far beyond the 16K compiled code limit. And Energia is fantastic without doubt. Sometimes though, I could use more documentation on the functions all in one place. Bluhash, thanks for your efforts. Cheers
  17. veryalive

    OV7670 and MSP430

    Is your i2c address correct? Do you have i2c pullup resistors installed? About 3k ohm. You are loading 0x21 into the MSP I2C address register. I thought I read in the camera data sheet: ,, address 43 for reads ,, 42 for writes The data sheet does not say if those are hex or decimal addresses. Assuming they could be decimal, then i2c address reg in the MSP should be loaded : Write - 42dec - 0x2a Read. - 43dec - 0x2b I have no experience with this camera. But i2c addressing can occasionally be ambiguous. Just use them all ,,,, all 4 ,,,, until you get a reply on the bus. Hope this works, and I hope you've got a logic analyser. Cheers
  18. @@spirilis I checked the large amount of MSP430 code you wrote for the nRF. And the BBQ monitor - quite good stuff. QUESTION --- do you think there would be much problem to port it to the MSP432 (432) Launchpad? Please, no need to get too specific, but I'd be interested to hear your opinion. Thinking aloud --- I'd like to use an MSP432 LP plus a nRF with the PA + LNA + antenna as a master station. Slave stations based on MSP430G2553 plus nRF on their low-cost PCBs with the track antenna. (alternatively, I have a couple of old Stellaris LPs which I could use as a base station -- did I see that someone ported to Stellaris?). cheers...
  19. @@spirilis Super thanks for this excellent work. It is brilliant! I've used your Energia library with success for testing nRF modules. Now I'm keen to try the CCS code on my systems. I'll check your github. cheers.
  20. ...... and from the ccs wiki ..... http://processors.wiki.ti.com/index.php/Compiler_Installation_and_Selection MSP GCC 5.x not seen by CCSMSP GCC compilers have an installation issue with CCS v6.1.2 and earlier where it will not properly register the 5.x msp gcc release. The 4.9 release will properly register and show as an available compiler version. In order to fix this issue, save below xml file, MSP430_GNU_5.0.xml, to this location: <ccs>\ccsv6\eclipse\plugins\com.ti.ccstudio.buildDefinitions.GNU_6.1.1.201512151800\resources\buildDefinitions\msp430\ File:MSP430 GNU 5.0.zip Restart CCS. From your CCS project, update the compiler on the General tab to point to the new 5.x msp gcc release. Click ok. You should now see both msp gcc compiler versions as available choices. The issue will be fixed in CCS 6.1.3 and CCS 6.2
  21. a question .... WHAT IS YOUR 5529 LP BOARD REV ??? Rev 1.4 had weird problems with a kind of power switch driven by the emulator side. I had such problems with the Rev 1.4 board which I resolved by reflashing the EZ FET. At TI'er on the E2E Forum helped me; this was about 2 years ago. In Rev 1.5 5529 LP, those power switches were removed.... o
  22. Not to forget that TI has the excellent CCS IDE which, in the free version with 16KB compiled code limit, allows professional development and debug of most any 2553 project you can imagine.
  • Create New...