|
The Radiskull posted:Im soon to be building an amp for the sub in my 5.1 somthing around 400w rms. Im not the best with amps so any tested circuits would be cool. You probably won't have a hard time find schematics for this. For example, http://sound.westhost.com/project68.htm
|
#
?
Jan 20, 2008 23:52
|
|
|
|
| # ? Apr 18, 2024 22:32 |
|
|
Squier posted:I am working on my senior year physics project which is a small cyclotron and need some electronics help. I need to create an RF amplifier that is capable of producing around 50-100 watts in the 1-15 MHz range. I haven't had much practical electronics and am hitting brick wall after brick wall trying to design this. So far it seems a push pull tube amplifier would be my best bet but am really lost beyond that. Any help at all would be awesome. Tube amps are very high impedance and voltage devices. Tube circuits are simple and you can get crazy power out of them, but you need to work with very high DC plate voltages that may give you a heart attack if you take it across the chest. Generally you need forced air cooling over 300w or so, but you should be fine. Cooling tubes is way harder than silicon as you can bolt transistors to a heatsink, not so much for a set of 6L6's. Transistor amps deal with lower voltages but high current, and the matching networks are pretty different. In the 100w range you should be able to easily find a single IC that will amplify what you need; if you have to build it discrete, I understand, but the chips are out there to solve it. Have you been given specifications as to how much filtering/etc you need on the output? My only experience is with HF ham amplifiers and we have very stringent requirements on harmonic and spurious outputs.
|
|
#
?
Jan 21, 2008 02:32
|
|
|
Squier posted:I am working on my senior year physics project which is a small cyclotron and need some electronics help. I need to create an RF amplifier that is capable of producing around 50-100 watts in the 1-15 MHz range. I haven't had much practical electronics and am hitting brick wall after brick wall trying to design this. So far it seems a push pull tube amplifier would be my best bet but am really lost beyond that. Any help at all would be awesome. First place I'd look is at APEX's page. They specialize in high voltage amplifier blocks: http://eportal.apexmicrotech.com/mainsite/index.asp From them, I can see the following that might interest you:
For the amps at 14MHz, you get gains of: PA09A - 10.7 PA93 - 0.86 PA98A - 7 Not sure what your target output voltage/driven impedance is, but with those gain values you should be able to figure out how best to place the input/output stages. You'll need some beefy capacitors and I'd probably invest in some high-precision resistors (1% and the like). Beats the hell out of making discrete stuff though. Just make sure you have some good quality HV power supplies. Do be careful with FCC stuff with these, you will be generating some pretty crazy RF interference with it. Try to keep everything you can in a Faraday cage, although since this is for a cyclotron I'm sure you already know that. Also electrical safety, etc. Note also that I've seen a PA97 (1MHz GBP, +/-300V inputs) used, and they were only able to really use it at 300kHz. Granted they didn't use any output stages or whatnot, but just be aware that you may not hit a full 14MHz with that. In that case, look into Class C,D, or E amplifiers for RF applications. Class A and AB work too, but those are gonna be some beastly amps! Tubes can work with that kind of configuration, but I can forsee that being a bit of a pain to create from scratch. Lots of radio systems work with these kind of power levels, there's bound to be plenty of info out there for you. edit: Removed stupid ideas. clredwolf fucked around with this message at 03:09 on Jan 26, 2008 |
|
#
?
Jan 21, 2008 03:55
|
|
|
Ok, I was able to clean up these cell phone pics a bit, so onto: Project #2 - BJTs and Kirchoff's Current Law This time, I'm doing this with a breadboard. Feel free to follow along! Today, we're going to use a BJT to light an LED. A BJT is a Bipolar Junction Transistor. It's the first mass-produced type of transistor, created in the late 1940s by Bell Labs. This little device revolutionized the electronics industry, and are only just now being phased out of use (by a similar device, the MOSFET). BJTs are simple devices in principle, but the math and theory behind them is incredibly complex and actually involves quantum-mechanical principles (if you want to take a crack at it, I recommend reading this). Most universities save transistors until well after other devices are taught. I'm bringing it up early because it's a great way to show KCL in action, and to get you all used to using them early (because the more you use them, the more comfortable you feel using them and the more you see you can do with them). When we last left current theory in that weird article of mine, we saw that current is the flow of electrons from one point to another. It's like water moving through a hose, in a way. In order to better understand the transistor, we need to understand Kirchoff's Current Law, or KCL. To understand KCL, we're going to take this water analogy further. Imagine you have one of those Y-splitters (edit: they're called tees apparently) for 3 garden hoses. One hose connects to your house's faucet, and the other two go to lawn sprinklers. (laughably bad artist rendition)  Now for the principle behind KCL. Treat the splitter as if it's a point. Then treat the two levers as switches:  The principle behind KCL is that the sum of the currents flowing through that point is zero. Wait, what? That can't be right. Let's start with the simple case first, which is the one above. There, the two sprinklers are off (as indicated by the switches not completing the circuit). In this case, zero current flows to either sprinkler. Therefore, the current through the hose from the faucet to the splitter must be zero, according to KCL. That's pretty obvious, right?  Ok, now let's turn the top sprinkler on. Let's say that the faucet is a current source that provides 2 amps when 'on'. This means that unless both switches are off (in which case no current can possibly flow) the faucet lets lose 2 'amps' of water:  No current can possibly flow into the bottom sprinkler, so the current through the hose to that sprinkler is 0 amps. The current into the point from the faucet is two amps. Now let's put KCL into an equation. We know that all the current has to sum up. I_faucet + I_sprinkler_Top + I_sprinkler_Bottom = 0 We know I_faucet is 2 amps, and I_sprinkler_Bottom is 0 amps. So, I_sprinkler_Top has to be -2 amps. Negative amperage is the same as saying that 2 amps are going away from the point.  Now if we turn on both sprinklers, we have two devices buying for 2 amps of current. Since both devices are equal, both devices get half of the faucet's current. So:  This preserves KCL, so that: I_faucet + I_sprinkler_top + I_sprinkler_bottom = 2 amps - 1 amp - 1 amp = 0 amps. This should make sense to you. When you use one of those Y-splitters in real life, the pressure behind both sprinklers decreases. If you turn off one sprinkler though, the pressure on the other goes back to normal. So basically, the current through a 'point' has to equal 0. We can pretty losely define a point. A current source can be a point, since the current going into it must be the same as the current leaving it. A resistor or an LED can be a point, for the same reason. An entire circuit can be a point, since no current enters or leaves the circuit. An iPod, not plugged into anything but playing some music, is a point since no current leaves or enters the iPod until you connect it to a computer. And even then, the current entering that iPod MUST be the same as the current leaving it while it's connected. You get the idea. Unlike Ohm's Law or some other approximations of electric circuits, KCL is always true. It may seem fairly obvious to some of you, and a trivial observation to the rest, but it's pretty important when analyzing circuits. Now let's try applying KCL to the real world. Here is what you will need: Part's List
Now to assemble this circuit step-by-step. Apologies for the crappy picture quality. Start with a blank breadboard:  Step 1: Connect alligator clips to the top two rails of your breadboard (top is positive, bottom is ground). Connect those to the 9V battery.  Unplug one terminal of the battery. Some of my pictures have it plugged in due to me testing it. Don't plug that in till I tell you too to be on the safe side.  Step 2: Place the switch somewhere on the breadboard.  Step 3: Place the transistor on the breadboard, with the flat side facing towards the bottom of the breadboard.  Step 4: Place the LED on the breadboard, so that the longer lead (the positive side) is facing the top side of the breadboard. (Oh god this picture is terrible, I couldn't get a better one though)  Step 5: Place the 100k resistor between the middle pin on the transistor and the switch.  Step 6: Time to wire everything up.
 Step 8: Reconnect the Battery. Make sure the positive rail and the negative rail are connected to the respective sides of the battery.  Step 9: Now press the button!  What's happening? If we analyze this circuit, we see that the transistor separates the LED from the switch. We also see that the resistor is far too large a value to turn the LED on by itself, yet the LED lights. The transistor is obviously doing something, but what? Well, the transistor is acting as a current source, that's what. More specifically, it's acting as a current-controlled current source. In less crazy terms, it's acting as a current multiplier. You've probably noticed that the transistor has 3 terminals. Looking at the transistor from it's flat side, the right-hand terminal is called the 'collector'. The middle terminal is called the 'base', and the left-hand terminal is called the 'emitter'. If you measure the voltage between the base and the emitter while the switch is on, you'll see a drop of 0.7V. This is the same as a silicon diode. So there's basically a diode between the base and emitter. Between the collector and the emitter, there's a current source. The current through this current source is directly proportional to the current through the base-emitter diode. For cheap-o transistors like we're using, the current from the collector to the emitter is about 100 times the current from the base to the emitter. This is enough current to turn on the LED, which is connected to the collector. Thanks to KCL, we can also see that the current through the emitter is the same as the current through the base and the current through the collector added. You can verify this with an ammeter if you want. This current is all dumped back onto ground. Then the current is pushed through the battery, back down the LED and switch-sides of the circuit again. Things to try: -Try a much larger resistor, like a 1 Megaohm resistor. Also try a smaller resistor, maybe a 50kohm. -Try placing a capacitor around the switch (between the resistor and the positive rail). Press the switch and see what happens. -Criticize my circuit to the core. C'mon, bring it on. I'll fix the problems with it even. clredwolf fucked around with this message at 14:02 on Jan 21, 2008 |
|
#
?
Jan 21, 2008 06:53
|
|
|
I don't know if you have seen it yet but: http://www.physics.rutgers.edu/cyclotron/ Those are a couple of "homebuild" cyclotrons at rutgers. You might be better off just scrounging for the RF supplies. There are probably still plenty of old tube based amps and signal sources stacked in a corner in some government lab. Who are you getting the rest of parts for the cyclotron from? Huge pieces of iron and magnets are not cheap.
|
|
#
?
Jan 21, 2008 06:53
|
|
|
I think a schematic would be infinitely more useful than a breadboard walkthrough. It's a little hard to follow, and ultimately doesn't leave any lasting understanding of how the circuit is constructed. Other than that, it's useful as a basic example of transistor usage. Also, to add: if you do not have NPN transistors but you DO have PNP transistors, you can construct the same circuit by reversing all devices which have polarity. In this case, the diode and the battery need simply be flipped. This is true of any circuit which utilizes PNP or NPN transistors. For those who are wondering, but don't want to get into difficult reading, NPN and PNP refer to the type of materials used to create the transistor. N-type material has an excess of negative charge carriers (electrons) and P-type material has an excess of positive charge carriers (holes). The choice of NPN vs. PNP is mostly arbitrary for most circuits. Audio circuits might benefit from one type over another, but it usually has more to do with the materials used for the devices rather than their orientation. Edit: do you plan to go over basic amplifier circuits and biasing if you're going to start talking about transistors? I'm not sure it's necessary to go through the more complicated circuit analysis techniques once you get through KVL, KCL, Nodal, and Mesh analysis, and amplifiers and biasing are pretty much essentials. scholzie fucked around with this message at 09:18 on Jan 21, 2008 |
|
#
?
Jan 21, 2008 09:05
|
|
|
clredwolf posted:
Shouldn't that bottom switch be closed?
|
|
#
?
Jan 21, 2008 13:35
|
|
|
Skycks posted:Shouldn't that bottom switch be closed? Fixed now. scholzie posted:I think a schematic would be infinitely more useful than a breadboard walkthrough. It's a little hard to follow, and ultimately doesn't leave any lasting understanding of how the circuit is constructed. Other than that, it's useful as a basic example of transistor usage. I would have made one, but my schematic editor failed to start last night. My intention it to get it working after work today, then slip the schematics in. I agree, it needs those schematics.
|
|
#
?
Jan 21, 2008 14:06
|
|
|
Would any goons be interested in a very small variable power supply? Might as well make "first" electronics projects useful right? I need to get permission to put the kits together first. However I'm putting feelers out, would any of you buy kits like that? PCB and all parts included? What would you pay?
|
|
#
?
Jan 21, 2008 16:01
|
|
|
If I were actually trying to learn electronics from this thread, I'd be so drat confused.
|
|
#
?
Jan 21, 2008 16:28
|
|
|
Phlegmbot posted:If I were actually trying to learn electronics from this thread, I'd be so drat confused. To be fair, a lot of other hobbyist sites and books and magazines aren't a lot better. I didn't start to truly understand any of the more complicated stuff (Transistors, etc.) until I started studying it in College.
|
|
#
?
Jan 21, 2008 17:12
|
|
|
Phlegmbot posted:If I were actually trying to learn electronics from this thread, I'd be so drat confused. LOL its not what you would call and electronics syllabus is it!. How about If someone has some server space we start an Electronics wiki. Unless there is one and in that case it was all my idea. Honest!!!
|
|
#
?
Jan 21, 2008 17:14
|
|
|
Phlegmbot posted:If I were actually trying to learn electronics from this thread, I'd be so drat confused. Rightfully so. This thread is all over the place. In the OP, there is a link to the free online series, Lessons in Electronic Circuits. Those books start from the very beginning and explain things very well. There is even a lesson on power dissipation that intentionally has you burn up some resistors. I thought that was a pretty neat way of hands-on teaching. Considering that they are free, you can't really do any better. If you would like to print hard copies, be warned that the illustrations are in color and you'll need quite a bit of paper and some way to bind it. I assume you can buy pre-drilled 3-hole color paper, so some of that and some binders may be the way to go. Or maybe Kinko's can print and bind it. I don't know.
|
|
#
?
Jan 21, 2008 19:04
|
|
|
MIT has their first circuits course on their opencourseware website. It has video lectures and such, so if you think you have the necessary mathematical and physics background it might be a way of getting started without actually going to college. http://ocw.mit.edu/OcwWeb/Electrical-Engineering-and-Computer-Science/6-002Spring-2007/CourseHome/index.htm
|
|
#
?
Jan 21, 2008 19:51
|
|
|
Can I ask a stupid question? I'd like to add an extra output to my dac, but I'm unsure if just placing the new one parallel to the existing one is a good idea. Wouldn't that efficitively halve the amplitude of the signal on both outputs? I feel like I have to do something with opamps, but I don't know if that's the correct path to take  .
|
|
#
?
Jan 21, 2008 20:22
|
|
|
pim01 posted:Can I ask a stupid question? I'd like to add an extra output to my dac, but I'm unsure if just placing the new one parallel to the existing one is a good idea. Wouldn't that efficitively halve the amplitude of the signal on both outputs? Nope, remember, paralleled circuits have the same voltage but half the current. This means that if you take your output and connect it to parallel outputs A and B, each one has an impedance twice that of the original output - meaning less current available. However, most stuff these days has pretty high input impedance and you should not appreciably load the circuit. I would recommend placing 1k resistors or something similar in line with each paralleled output to give better isolation. If you want an active solution, just get a single-rail dual opamp and build a pair of voltage followers - + input is signal, - input is connected to the opamp output. If it's stereo, just make a pair of circuits, one for L, one for R.
|
|
#
?
Jan 21, 2008 21:00
|
|
|
I'm taking "advanced lab" for physics and we are using The Art of Electronics by Paul Horowitz. Does anyone have any experience with this book? I have only taken some simple circuit design in a normal university physics class and I am a little worried because the book looks quite complex.
|
|
#
?
Jan 21, 2008 23:27
|
|
|
Phlegmbot posted:If I were actually trying to learn electronics from this thread, I'd be so drat confused. Yeah I have that impression too Oh well, I'll try to keep it going. Would cleaning up the articles help, or should they be rewritten?
clredwolf fucked around with this message at 02:48 on Jan 22, 2008 |
|
#
?
Jan 22, 2008 00:09
|
|
|
I think you shouldn't bother with the articles. Anyone who wants to learn is going to have a wealth of information available to them at any level of experience by using one of the links posted earlier. The other downside to putting all this effort into writing them and they'll be in the archive 6 months from now  I think it's best to just show some projects or links to projects, and post questions and tips.
|
|
#
?
Jan 22, 2008 03:09
|
|
|
Thanks a lot for the info guys, I'm looking through it all now and will surely be back with more questions soon enough.
|
|
#
?
Jan 22, 2008 03:24
|
|
|
Poonpalace posted:I'm taking "advanced lab" for physics and we are using The Art of Electronics by Paul Horowitz. Does anyone have any experience with this book? I have only taken some simple circuit design in a normal university physics class and I am a little worried because the book looks quite complex. This is actually the best textbook there is for electronics. It might look complicated, but reading through it, and with some additional "push" from your Lab TA or Teacher, you shouldn't have a problem at all.
|
|
#
?
Jan 22, 2008 04:09
|
|
|
clredwolf posted:Yeah I have that impression too I would personally start from scratch. This thread was polluted by diagrams of transistors and LEDs and crap on page one. It should have started with volt, amps, Ohm's law, and resistors. Then you learn about DC circuits with multiple resistors and solve for V and I via KCL and KVL. Then you learn about combining resistors in series and parallel. Then you learn about Thevenin and Norton. Then you learn about the capacitor as a circuit element in the time domain - paying no attention to the electromagnetics behind its operation. Then you learn about DC RC circuits. Then you learn about inductors as a circuit element in the time domain - once again paying no attention to how an inductor actually works. Then you learn DC RL circuits. Then DC RLC circuits. You get the idea. Basically follow the curriculum of any first-year circuits course.
|
|
#
?
Jan 22, 2008 05:43
|
|
|
Nerobro posted:Would any goons be interested in a very small variable power supply? Might as well make "first" electronics projects useful right? I need to get permission to put the kits together first. However I'm putting feelers out, would any of you buy kits like that? PCB and all parts included? What would you pay? I am attempting to put a design together right now for a 30W power supply. Features: *Ethernet connection, LXI support and maybe software Class C IEEE 1588 support. *2 output tracking +/- 30v@1A to ~1v@3-5A DC. *.05%+5mV output accuracy. (about 15mV off at 20V) *Adjustable current limit. *10$ or less in one off parts. IE standard resistors and transistors don't count. Also not taking into account PCB costs, people can always build it on perfboard. I think all of this is doable at the price. I'm aiming for something roughly equivalent to an agilent E3611A but a little less accurate, and a couple more bells and whistles. Something like the PIC18F66J60 has it costs about 3.50 in volume. It has a built in 10 Base-T ethernet mac, phy and only needs a magjack(1.50$)for it to work. For power I would probably go with some sort of switcher with a linear regulator for the final output. This avoids an expensive and large transformer. I've actually never designed a switcher for this much power and for low EMI. In any case the PIC18F66J60 would be running the switcher off of its PWM outputs. Not sure what to do about the linear stage yet. Maybe just see if I can find a cheap SPI DAC and have a couple opamps regulate to that. Or maybe get another PWM to set that. Startup is a little problem. Might just use a resistive AC regulator to get the uC and switcher going then have an isolation relay kick in. Not that big of a deal if the uC is isolated from everything else. I originally wanted this thing to be an Altoid form factor device. But at 30W I think I'm asking for a bit much. EMI is going to be a pain as well. I was thinking of trying a psudo spread spectrum switcher to try and limit it somewhat. Anyone have some ideas? 10$ to make the equivalent of a 450$ power supply might be a bit much, but I'd like to get close.
|
|
#
?
Jan 22, 2008 05:44
|
|
|
Phlegmbot posted:You get the idea. Basically follow the curriculum of any first-year circuits course. My first-year circuits course included a moderate amount of physical electronics, and I think I'm a little better off for it. Although we didn't have to know anything terrifically in depth, just having some of the stuff worked out on the board helped a lot in conceptualizing a lot of stuff. One cool thing our teacher did was to bring in a giant inductor, and hook it in to a simple RL circuit with a cutoff in the middle. To illustrate that you couldn't change the current across an Inductor instantly, he threw the cutoff, and you could watch it arc across the several inch gap created. It was cool stuff.
|
|
#
?
Jan 22, 2008 06:00
|
|
|
Jonny 290 posted:Nope, remember, paralleled circuits have the same voltage but half the current. This means that if you take your output and connect it to parallel outputs A and B, each one has an impedance twice that of the original output - meaning less current available. Thanks for the concise answer! I think I'll build some voltage followers, just for the fun of it  .
|
|
#
?
Jan 22, 2008 10:00
|
|
|
I’ve found a video on youtube by some guys that do MAKE magazine. (I've never heard of it) but they managed use a laser printer and an iron to make the tracks for copper clad pcb! They Ironed the Toner onto the clad. http://www.youtube.com/watch?v=urv6jArKp6M&feature=related It looks like it would work really well and the finish was great! I’m going to try this tonight and see how well it works. edit: fixed link The Radiskull fucked around with this message at 10:22 on Jan 22, 2008 |
|
#
?
Jan 22, 2008 10:19
|
|
|
The Radiskull posted:I’ve found a video on youtube by some guys that do MAKE magazine. (I've never heard of it) but they managed use a laser printer and an iron to make the tracks for copper clad pcb! They Ironed the Toner onto the clad. As for the power supply thing, would you pay $7 for one? I was just pricing out the parts on digikey, and I could reasonably sell "kits" for about that.
|
|
#
?
Jan 22, 2008 14:41
|
|
|
Nerobro posted:It works great. Taht's how I've made most of my PCBs. I'd probably buy one, what kind of specs are we talking?
|
|
#
?
Jan 22, 2008 15:12
|
|
|
Nerobro posted:It works great. Taht's how I've made most of my PCBs. In the same vein I was considering putting together a kit for a headphone amp. It would be stereo, and if I can fit it in would also have a fuzz or distortion circuit for guitar players. I made the amp so I could play guitar without bothering the neighbors late at night. I want to make it so I could fit it in an altoids tin, so my next project is turning the circuit into a PCB design. Right now I run the guitar into a marshall blues breakers (overdrive) clone circuit, and then the overdrive into the headphone amp, but I should be able to squeeze them all in and switch the fuzz or overdrive with a switch. It's pretty loud, by the way, and it can run off of a 9V battery or 9V wall wart so it's very, very portable. With the proper input adapter (or if this is a kit, the proper jack chosen at purchase time) it would make an excellent amp for high quality iPod headphones (like the Etymotic ER6s or Grados, which should really use an amp for proper frequency response from the ipod). I was figuring I could offer fuzz, overdrive, or distortion for the first stage (or no effect at all), then your choice of input and output jacks (1/8" stereo or 1/4" stereo) and your choice of LED color for the distortion stage.
|
|
#
?
Jan 22, 2008 17:42
|
|
|
I now have a Tek 453 scope on the way. 
|
|
#
?
Jan 22, 2008 18:17
|
|
|
Skycks posted:I'd probably buy one, what kind of specs are we talking? Amp and a half, linear, up to 35v input. you'll need to maintain something like a 2v over your output for it to work though. Nothing crazy. But a quick eight or ten components to get someone soldering, and soldering something useful.
|
|
#
?
Jan 22, 2008 19:07
|
|
|
I found that toner transfer using paper took too long, and was too much hassle as it took me 3 or 4 tries (at 15 minutes each go) before I got something that I could etch. Using Press-N-Peel blue available here, is much quicker, and nine times out of ten transfers perfectly on the first attempt. mrbill posted:I now have a Tek 453 scope on the way. I received a Tek TDS2012B scope yesterday  with a current probe on the way.
|
|
#
?
Jan 23, 2008 00:56
|
|
|
for a project I need some good antistatic protection for the pins of my microcontroller. It has built in protection diodes, but I don't really trust them since their characteristics aren't specced in the datasheet. The easiest way would be to just put a resistance in front of the input, but the thing is I will be switching it to an output also, and I don't want it to have too high of an output impedance. So it seems my best option is to put on some redundant protection diodes. However, I'm not sure what kind to get. I'd assume Schottkys with good peak current ratings would be good, but I'm not sure. Any advice?
|
|
#
?
Jan 23, 2008 02:48
|
|
|
mtwieg posted:for a project I need some good antistatic protection for the pins of my microcontroller. It has built in protection diodes, but I don't really trust them since their characteristics aren't specced in the datasheet. The easiest way would be to just put a resistance in front of the input, but the thing is I will be switching it to an output also, and I don't want it to have too high of an output impedance. So it seems my best option is to put on some redundant protection diodes. However, I'm not sure what kind to get. I'd assume Schottkys with good peak current ratings would be good, but I'm not sure. If you can get reasonably decent resolution on your PCB, you can try using spark gaps. They are pretty effective at ESD protection. Just have small spots where there is an unobstructed .5 mm gap to ground.
|
|
#
?
Jan 23, 2008 03:09
|
|
|
mrbill posted:I now have a Tek 453 scope on the way. Did you get it on ebay? I was looking at one of those auctions and it was ending soonish.
|
|
#
?
Jan 23, 2008 03:19
|
|
|
ValhallaSmith posted:If you can get reasonably decent resolution on your PCB, you can try using spark gaps. They are pretty effective at ESD protection. Just have small spots where there is an unobstructed .5 mm gap to ground.
|
|
#
?
Jan 23, 2008 03:52
|
|
|
mtwieg posted:well I'm not sure if I'll be making custom pcbs for this particular project, since it's reasonable simple. I'll keep that in mind though. The minimum gap allowed is 8mil, so that should be small enough. I assume for each pin I will want gaps going to both ground and the positive supple rails correct? and there shouldn't be any soldermask over the gap? I'd only do that between signal traces and ground. For input power protection I'd use a MOV. But yea, no solder mask at the contact point. If you google for ESD spark gap PCB you should get to see some decent foil patterns for doing it. When you get boards that have this type of ESD protection be sure to do a continuity check to make sure there isn't a short. Edit: This gives a good overview of available ESD devices: http://www.eetkorea.com/ARTICLES/2007APR/PDF/EEKOL_2007APR19_EMD_AN_02.PDF?SOURCES=DOWNLOAD ValhallaSmith fucked around with this message at 04:41 on Jan 23, 2008 |
|
#
?
Jan 23, 2008 04:29
|
|
|
scholzie posted:Did you get it on ebay? I was looking at one of those auctions and it was ending soonish. Nah. One of the websites I run trades advertising space for equipment from vendors of the market it supports; I found out that one of the vendors also has a test equipment division so they're sending me a 453 instead of another computer this year.
|
|
#
?
Jan 23, 2008 05:15
|
|
|
When you buy a scope on Ebay, how can you tell if the scope will function or not? It seems like most ads are as vague as possible about the condition of the scope. At the cost of paying a little more extra, I would like a guarantee that what I'm buying will work. What are some other options for me to get an oscilloscope?
|
|
#
?
Jan 24, 2008 03:56
|
|
|
|
| # ? Apr 18, 2024 22:32 |
|
|
Almasy posted:When you buy a scope on Ebay, how can you tell if the scope will function or not? It seems like most ads are as vague as possible about the condition of the scope. Depends on how much you are spending and what specs you need. If you are spending 800-1000$ buy new. Less than that buy from someone in your home country. Scopes are ridiculously expensive to ship if you are trying to get an old one. I've got an Agilent 54100A and its a loving anchor. Anything made pre-90's is going to heavy as hell. As far as whether it will work, make sure they have a decent rating. I bought mine from this guy: http://stores.ebay.com/Test-Equipment-Connection he's fast at shipping and everything worked fine. You didn't say what type of circuits you wanted to measure. Someone can probably give you a better idea of what to look for if you say that.
|
|
#
?
Jan 24, 2008 04:38
|
|