Jump to content
43oh

simpleavr

Members
  • Content Count

    725
  • Joined

  • Last visited

  • Days Won

    32

Everything posted by simpleavr

  1. @@PhilM The way I am direct driving the LED is not the "correct" way to drive LEDs and could reduce the lifespan of LEDs. The proper way would be to drive them via transistors and current limiting resistors. Which in this project (or most of my projects) I decided to skip them to keep the component count down and to save PCB space. To me it is more important to get this on a easier to build thru-hole layout. LEDs can allow excessive current on a very brief period and since we are doing multiplexing already, we are not constantly turning on each individual LED at the same time. I think
  2. I would recommend this for those who are not regular builders. I had designed this to be all thru-hole to accommodate most builders. But if you haven't touch a soldering iron for years it could still be challenging. My first build takes took two hours (consider that I design the thing and had breadboarded it already) and one full day of debugging. I am sure it goes the same for blue#. Thanks blue# for sending a board over for verification. It's a good move as the board is a little bit "newer" than mine. /EDIT correct engrish
  3. I also prefer green for the TMS080x emulation. The white PCB is actually for another emulator (WIP).
  4. @@abecedarian Ultimately you will want to go direct 'c' (ccs / gcc) to get accurate timing. You cannot rely on your loop w/ Energia. It has this built-in timer WDT interrupt that would throw your own counting off. The current code is not working as (1) you need to also multiply 100 to your t2 value (orginal 440), but then in the loop it will never reach 44000 as your angle value / degree is incrementing by 1125. I changed it to the closest value of 43875. I make use of the Energia function "micros()" to see how long it takes for a 720 degree loop. The code changes are as follows;
  5. All the ones I made were w/ white buttons. Just matches well w/ the white silkscreen on the green PCB. As long as they are two pin 6.5mm spaced they are good. I pick the green PCB as when I started the project Elecrow has this 5x10cm green pcb for $11.90. Now they are offering any color for $12.90, so I cannot resist and have to make this.....
  6. @@rhb It's a Linux issue. You need to install ia32-libs. I am on Mint 11 which should be similar to Ubuntu. ie. sudo apt-get install ia32-libs Newer distros may require separate different packages, you can google and find out.
  7. Both (DataMatch and Sinclair) are contained in the same github project. I found that it is listed on the sinclair page so you would find it from the TI page. The core is similar enough they are using the same code, with a few switches inside based on the 03/05 subtype, and the register masks are a little bit difference. There is this source_code.js which is the ROM for the TI calculator and source_code_sinclair.js for the Sinclair. It looks like (I only check the 1st block of bytecodes) for the TI ROM, it's from a TI patent, and for the Sinclair, it was mentioned it's read from an
  8. @@PhilM @@distefanom @@bluehash Thanks for your interest in the calculator. The group buy was initiated by bluehash. I believe he is procuring all the necessary parts and will turn them into kits, and will have them shipped from US when ready. Let's give him a couple of days to update all on the progress. He's part-time managing this site. The software and hardware are open sourced for anyone who's interested to build, see the build thread for more detail. If anyone needs the Gerber files for fabrication, I will be happy to provide via PM. The circuit is also simple enough to rol
  9. Should compile w/ CCS though, I haven't really take the time to try it. But the template / skeleton used to be compiled on both. And there are tons of left-over space, so should be fine. Also be careful when prototyping, there are no limiting resistors for the LEDs. If they looks "very bright", it's time to disconnect power. I know it's wrong, but I don't want led drivers and extra parts as I want this to be easy to solder and minimal PCB estate.
  10. @@PhilM Hello Phis, Thanks for great resource. Looks like we are adding a 3rd rom (Sinclair Cambridge) to the build thanks to Ken Sherriff. I am really surprise to see the actual Sinclair did not compense the brightness of less segmented digits (say '1'). I guess that was the limitation on the H/W. May be the msp430 implementation should be changed to observe that. I don't have 3D printing resources. The PCB is 50mm x 85mm, i.e. same width, but shorter. I guess one can lengthen the PCB if desired. I know The Sinclair was offered as a kit, so one can design a new PCB that fi
  11. @@enl I am doing what you had in mind. I had serial uart and I am capturing 100ms bursts of the digital I/Os from the 1st interrupt. Still, if problem is on the op-amp stage, a scope could help. May be I should do FFT and get the mic via ADC, that way I will have more control (S/W filters, etc). @@abecedarian It's not the gain. It's more like matching both left and right to have the same gain and behave the same.
  12. @@abecedarian I had the same thinking. Trying to find a way or ways to filter out false alarms. Add redundancy and let them vote (say both set of mics agrees, etc) could be a solution. But I am finding that may be it's the quality of the front-end that's affecting the results. So I would 1st try to improve on them. The resistors are typical 1% types and there is nothing I can do on them. I may swap the electret mics to see if I can get a better match. It would be nice though to have a scope to see what actually happens. Now I am just doing trial and errors. I tried to use interrupt tri
  13. @@Mac Yes you are right P1.3 is high and should be anode, I always confuse about anode and cathode, they are better described IMO in Chinese as being yin and yang (like everything else), where yang goes to V+. If it's Excel that you had used for the schematic, I will skip. But they do look nice. In the github source code, there is also the original tms0800.c which uses a different layout saved 4 pins. The extra pins were used for serial UART and that version of code can connect to a terminal and the emulation runs as a single stepping disassembler where you can read the instruction
  14. @@enl Thanks for all the helpful pointers, I am certainly going to dig into them. I don't understand everything yet but I will educate myself w/ your pointers. I am trying to build a turret shooting where I want to direct a turret towards whoever claps in a room. May be to point a video camera or throw candies at kids. So I am not trying to be really really accurate. Still, it is currently way off (inconsistent, 1/4 times I can get reasonable readings). I am not doing ADC w/ the mic inputs. I am trying something simpler. I have the LM324 split for both channels. Each one has a op-a
  15. @@cubeberg @@zeke It's a sound locator thing that I am having trouble with. I can catch which mic got signal 1st, but the time lapsed between them does not match what I expect. I am trying to calculate by triangulation and get the fine direction of the sounding object. But I think I am getting not every clean signals from the front-end. I.e. every knock / clap generates ripples on both mics and I am not capturing the correct / definite pair. I think I can only see them w/ a scope or logic analyzer. May be time to get / build one.
  16. @@Mac Thanks for trying to decode my badly organized schematic. Comparing yours w/ what I had put out review wrongs on my side. I had mistaken the "negative" led be P2.2 and P2.3, It should be as what you noted (I assume you had observed it from the Fritzing generated schematic, but couldn't figure out the polarity as I just used a two pin header to represent the LED). I will correct all the ASCII schematics as soon as I can. Checking w/ the code reviewed that for my PCB they should be P1.2 anode, P1.3 cathode, to match how I am driving it in the code. if (g_state&
  17. Can't wait to see your prototype. Try squeeze in some real functions to it to make it more interesting. Ex. Couple of them sync / pulse when in close proximity, etc.
  18. @@Druzyek Don't you want to make it standalone and independent of a terminal? With 7 segment LEDs and keypad. Those trainers are very common in the '70s. Like the KIM-1. They usually have a few KBytes of "Monitor" code allow user interfacing. May be even a bootloader that load programs from audio tape drive. What I meant is I think it would be interesting to see the buttons, LEDs and screens be real / physical instead of virtual. I.e more electronic h/w As for my mysterious "device", I will show a short video tomorrow.
  19. I am working on and off w/ this. Guess what it is and what do / will I use it for. No prize, guess right and I'll do a 1 minute video on it.
  20. @@Druzyek Very interesting. Do you have a monitor program / OS? Did you just roll your own bootloading in ASM? The trainer using functions in your ROM? You should move it a peripheral board w/ LEDs and buttons.
  21. /EDIT Sep-24-2014 include feedback for builders. Adding some assembly tips here. General Make sure you got the orientation of the msp430 mcu and led modules right. Square pads on the PCB means pin 1, other pins have round pads. Dots printed on ICs and display modules means pin 1. Match them up. The top-most two tactile button positions are not to be populated, the firmware does not use it. I just put it there because it may be useful to have the extra buttons if you "roll-your-own" firmware and need them. You should use a temperature controlled iron w/ clean (as clean as u can clea
  22. Update. It turns out I can use still use the smd buttons and there is a bonus. I use pliers to clamp down the horizontal contacts and turn them into very short pins. When placing them on the PCB they are kind of wooby wooby as "pins" did not penetrate to the underside. And I solder on the top. And the bonus is that the backside is no longer poke-y. @@swampdonkeykami if you can accept the smd buttons and want your kits quicker, please response and PM your address to me. The differences will be a slight difference on the PCB (bypass cap position and some back-side silkscreen text).
  23. @@akdes You can checkout the github package and start from there. https://github.com/simpleavr/boot430 @@elpaso had contribute a hid430.c as an example build HID usage. I believe it's a mouse. All you need to do is to find and similar V-USB project (may it be joy-stick, keyboard, etc) and mesh them up. HIDs are similar enough and should be able to port easily. If you still encounter problem. Please start a new thread and ask more specific questions (providing enough info for us to help you).
  24. @@bluehash You may want to avoid DX on this one. After a 6 weeks' wait. My tactile buttons comes to me wrong. They (or whoever supplies them / drop shipment?) just filled the order w/ SMD parts instead of thru-hole. May be if you can match them from newark or other US suppliers even better to guarantee quality. I could as well order (which I did last week) from ebay and get them in 2 weeks. The 1st batch (100s) I got from DX are good quality, I guess they ran out and just try to make me put up with whatever they had. I had filed a ticket w/ them, I've been getting stuff from th
  25. @@2lostkiwis @@zeke That reminds me of sick-of-beige. The acrylic won't fit the retro theme for this project though. The original DataMath is actually a beige soapbox design. I do prefer the slim look of the Sinclair. I don't think we need to protect the front as this is really a show piece that asks to be touched. A slide-in shell can protect the back (i.e. protect your hand from the back) and also serve for storage when reversed.
×
×
  • Create New...