![]() Having keys Up, Down, Activate would leave you three conventional "fast" keys. Six buttons are enough in the spirit of my foreword. ![]() connecting the free pin to each button through diodes and resistors to the tristatable display lines might do the trick (I hope.). Oh, you need buttons for that too, huh? A tip: when the enable line of the LCD is deasserted, the display controller, to my best knowledge, swithes the remaining lines (DB4-7, RS, RW) to tristate and applies weak pull-ups on them. After checked back, I've noticed that K8LH has given a similar answer, here it goes anyway: (the first part gave you ideas of using multi-level menus on the On Screen Display, but was quite trivial anyway.) Good thing I could wrote down the last page of my text ( ) witth paper and pencil. I've been writing down my thoughts since some time, until I had a crash because of a hardware failure (hard drive.). as usual the posts here are all top notch. and this for me is the joy of design work. the tradeoffs in hardware/firm/software must be negotiated in every system. all the same issues apply whether analog or digital. with ten switches there are 1024 variations. ) there are 8 A/D levels with the no switch level undefined. and note that for this circuit (I know everyone knows this stuff. if so how will that impact the resistor ladder?. will some form of pull up or pull down be required?. if an A/D is utilized what is the no key state?. what is the voltage on the input with no switches closed?. in your 3 switch example what is the difference in voltage between the middle switch being closed and all 3 switches being closed?. This circuit is an interesting idea but may I ask. (the following are observations not intended to critique anyone here but to point out issues that I recognize and would have to deal with if a hardware engineer gave me this circuit to code) My best regards to you all gentlemen (gentle lady?) So a debounce routine after the initial detection is needed. This C= 1uF is more than anything to ensure that the newer switches with fast bounces do not return briefly lose the +5V (divider goes go to zero) and give false readings. The longest 5CR time constant is with R37 being 50ms, the shortest with R52 being about 0.25ms. 1% resistors will not have this problem and the theoretical divider value can be assumed for the look up table plus a little bit of interpolation To avoid ambiguity with 5% resistors a specific look up table should be held in memory or written into the read routine. This 0.5V margin is to ease the rail to rail swing limits allowing the use of CHEAPER opamps. They have been chosen value spaced so that the 16 input voltages are differentiated by steps of at least 0.2V, and that range is bracketed betweenĪbout half a volt above the ground, and about half a volt below the +5V, lines respectively. The resistors themselves are not high tolerance critical. ![]() The unity gain opamp voltage follower is to meet the ADC requirement of a low impedance input. The top resistors/switches are all standard and form a voltage divider with the bottom 1K. īut getting back to the one pin 16 inputs from switches circuit, (the appended circuit) I want it to meet various conditions. AGC ON/OFF, Direct keyboard frequency entry ( called from another button function), lights ON, lights OFF, bandwidth filter. ON/OFF +Vol, -Vol, tone, BFO, SSB selector, tune up, tune down, station memories, band select, Lock detect. Just off the top of my head, it needs a pin for many things, which although very very simple, do need pins. My present project is a multiband radio which to accomodate all the bells and whistles keeps asking for more and more pins. Tip #7 requires an ADC input pin which I have, AN0, ( I wish I had a complete port free!!!) but thanks anyway. wow!Įven so, perhaps a little extra care is needed with the resistor tolerances. Perhaps this #7, Tips 'n tricks, is the most elegant 1 pin solution, a resistor matrix needing only nine resistors. The Assembler code given will be useful to me. Yes indeed AN529 is another possibility the matrix approach uses less resistors but at the same time seems to need more port pins. Combining suggestions gives rise to the attached circuit and which I comment at the end, so if you see anything wrong please come back. Your "just a thought" has hit the jackpot. The second sugestion was more fruitful and I found another pin that wasn't doing too much. I would need to have a little more depth to grasp the first suggestion, as being a beginner it has got to be spelled out for me. Specifically I would reply to the following gentlemen's suggestions/ideas.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |