Jump to content
Electronics-Lab.com Community


  • Content Count

  • Joined

  • Last visited

  1. There's a site dedicated to open-source EDA software at: http://opencollector.org/ Not all of it's for Linux/Unix, though, but most of it is.
  2. sec

    "Safe" 5000 V

    Alun has explained the theory pretty well, but on the practical side, you'll want to build something like a Van de Graaf generator, or a Tesla coil. Search around on the Internet; there are plenty of resources out there. A forum dedicated to these (and other) HV topics can be found at http://forum.4hv.org/index.php
  3. Neat! A few things worth noting... Pin 9 is +5VSB. This pin supplies +5v so long as the mains power is connected -- even if the machine is powered down. Well, unless you turn off that switch on the back of the PSU, that is. (You apparently know this; I'm just clarifying for the benefit of others.) Your software could easily be modified to 'press' the power button on startup. With normal ATX machines, if the power goes out, the machine will not come back on when the power comes back, making them unsuitable for things like servers. With this modification, the machine will come right back to life as soon as power is restored. Some motherboards have this feature in their BIOS setup, but not all of them. Indeed, you could easily add a switch which controls whether this feature is active or not. As for your idea for monitoring the voltages, what you'll need is an analogue-to-digital converter (ADC). Some versions of the PIC chip have ADC inputs built in, but the 16F84 isn't one of them. You can still use it if you get an outboard ADC and connect it to the PIC. Check especially the websites of Texas Instruments, Maxim, and Analog Devices, all of which manufacture many chips suitable for this purpose. Keep in mind that for all but the 3.3v rail, you will need to use a voltage divider to scale the voltage to the point where it's within the 0-5v range of most ADCs. That's not very difficult, though. (Yes, it's a good idea to scale the 5v rail, too, since if the output is high, it could quite easily go above the 5v upper limit of the ADC.) As for adding the LCD display, there are plenty of resources available on the web which explain how to do this. With all of the features you're planning on adding, though, you may find yourself running out of I/O pins. There are ways to use a serial shift register like the 74HC164 to reduce the number of I/O pins you'll need to attach the LCD.
  4. Thanks, that pointed me in the right direction. Some googling revealed that pre-cut insulators made out of this material go by the trade name 'Thermo-Pad', and sell for about US$1-1.50 for 5-packs of the most common shapes.
  5. In some commercial devices, I've seen power semiconductors that are insulated from their heat sinks by a grey, rubber-like insulator. Does anyone know what this material is, how its thermal transfer properties compare to the classic mica washer/silicone grease combination, and where it can be obtained from?
  6. In that case, you might want to look into using a FET.
  7. Oscillator coupled with a Greinacher voltage doubler, right? In theory, it should work. I'm not quite sure about your implementation, though. It looks like you've grafted two different schematics together to get that schematic. I'm curious about the left-hand side of the circuit, as it doesn't look like it should work. Do you have a reference for it? I'm also not quite sure how it's supposed to increase the voltage.
  8. Take a look at: http://sound.westhost.com/power.htm for a discussion of the wacky world of power measurements for the purpose of advertizing. :)
  9. I really don't understand what the optoisolators are for. Isolating the fans? Isolating them from what, or isolating what from them? If you were using the output of some sensitive IC to drive the fans, then I could see it, but a simple switch does not need isolation from the fans, nor do the fans need isolation from the switch! Get rid of the optoisolators, and use the switch to switch the intake fan between the 8v source and the 12v source. Don't make things more complicated than they need to be. BTW, all of the computer case fans I have around here have a current draw of 0.1 A or so. Much less than what you estimated. Check the labels of your fans; the current should be printed on it.
  10. Sorry, but without more information about what you're trying to do, we can't really give you a good answer. As for transistors, they will all have a maximum frequency for which they're good. This depends on the model of transistor, and can range from a few kilohertz to a few gigahertz. You'll need to check the transistor's datasheet to find out for sure. Almost any integrated circuit will have some kind of maximum frequency that they will be rated for. Be careful, though, since sometimes this parameter is lumped together with other parameters. For example, op-amps have a parameter called slew rate, which is usually measured in V/uS. This tells you how much change in voltage the op-amp can deal with in a given time frame. What this means is that the greater the voltage swing the op-amp has to deal with, the lower its maximum frequency will be. Another thing you have to worry about at high frequencies is that for simplicity's sake, we usually model the wires connecting the components as perfect conductors. At low frequencies, this is a good approximation, but at high frequencies, if you don't model the capacitances and inductances associated with the conductors, you'll get wildly inaccurate results. Needless to say, this complicates modelling drastically. Read up on transmission line theory for an overview of the issues involved. That said, I have naively (that is, without taking special measures for dealing with high frequencies) built digital logic circuits up into the mid-single-digit megahertz range, without any ill effects. Not so sure about analog circuits, as I've never built anything beyond about 100 KHz. I would assume that high-frequency effects would start being noticeable in analog circuits before they would be in digital circuits, as in the latter, the effects would have to be large enough to start flipping bits before they would become noticeable.
  • Create New...