[QUOTE=Chryseus;43065976]Yes it is possible to make a transistor using point contact methods although the performance you'll get is absolutely awful compared to any modern transistor, there isn't really any easy method as far as I'm aware to make transistors.
You'd probably have more luck making vacuum tubes, although that of course requires a vacuum pump.[/QUOTE]
I had a pretty crazy idea.
You could 3D print transistors by laying down a multi-material pattern of pre-made powders, sintering them together with a laser, and then doing the same for the next layer.
The powders would be: copper, pure silicon, silicon oxide, uniformly doped silicon (p- or n-) of a given concentration.
Now how do you dope silicon? And uniformly, at that? I'm thinking the basic idea would be, to just get a block of dopant, a block of silicon, weld them together, and then put them into an oven (or kiln) for several hours. Diffusion equations tell us how much time is needed before the concentration of impurity is within 5% of being uniform everywhere. An improvement for this process is to first powderize the silicon and the dopant, then blend them together in a powder mixer, and THEN put them into an oven.
The only problem I see (apart from having to purchase a kiln, a pulverizer, a mixer and a laser sintering printer) is how to buy dopants. For silicon, the acceptors from group-5 are Phosphorus and Arsenic. Phosphorus is a controlled substance and can not be purchased without a license, at least in US. Arsenic is not (but it's supposedly poisonous so there might be some other law for that too).
For the scales I'm aiming at (~1 cm^2) crystalline properties and uniformity shouldn't play too large a role (although I dunno about other silicon impurities). It may be possible to buy broken blank silicon wafers if that becomes a problem.
Though never took a semiconductor physics class so...
[QUOTE=Nikita;43077827]I had a pretty crazy idea.
You could 3D print transistors by laying down a multi-material pattern of pre-made powders, sintering them together with a laser, and then doing the same for the next layer.
The powders would be: copper, pure silicon, silicon oxide, uniformly doped silicon (p- or n-) of a given concentration.
Now how do you dope silicon? And uniformly, at that? I'm thinking the basic idea would be, to just get a block of dopant, a block of silicon, weld them together, and then put them into an oven (or kiln) for several hours. Diffusion equations tell us how much time is needed before the concentration of impurity is within 5% of being uniform everywhere. An improvement for this process is to first powderize the silicon and the dopant, then blend them together in a powder mixer, and THEN put them into an oven.
The only problem I see (apart from having to purchase a kiln, a pulverizer, a mixer and a laser sintering printer) is how to buy dopants. For silicon, the acceptors from group-5 are Phosphorus and Arsenic. Phosphorus is a controlled substance and can not be purchased without a license, at least in US. Arsenic is not (but it's supposedly poisonous so there might be some other law for that too).
For the scales I'm aiming at (~1 cm^2) crystalline properties and uniformity shouldn't play too large a role (although I dunno about other silicon impurities). It may be possible to buy broken blank silicon wafers if that becomes a problem.
Though never took a semiconductor physics class so...[/QUOTE]
Diffusion would only render a singled doped area, you'd want to go the route of [URL="http://ac.els-cdn.com/0022024878904232/1-s2.0-0022024878904232-main.pdf?_tid=f6dce924-5dbd-11e3-b68e-00000aab0f27&acdnat=1386255803_f148d0232f1808fef39811d640e59e99"]vapor-deposition[/URL] which is pretty tough but not impossible (If you have a high enough heat source to vaporize your doping material).
Another route would be with a graphene substrate which is [URL="http://www.youtube.com/watch?v=vQANDuFYvsc"]relatively easy to make[/URL] and sinter silicon powder on it.
I'd also suggest looking into the [URL="http://en.wikipedia.org/wiki/Czochralski_process"]Czochralski process[/URL].
[QUOTE=Nikita;43077827]I had a pretty crazy idea.
You could 3D print transistors by laying down a multi-material pattern of pre-made powders, sintering them together with a laser, and then doing the same for the next layer.
The powders would be: copper, pure silicon, silicon oxide, uniformly doped silicon (p- or n-) of a given concentration.
Now how do you dope silicon? And uniformly, at that? I'm thinking the basic idea would be, to just get a block of dopant, a block of silicon, weld them together, and then put them into an oven (or kiln) for several hours. Diffusion equations tell us how much time is needed before the concentration of impurity is within 5% of being uniform everywhere. An improvement for this process is to first powderize the silicon and the dopant, then blend them together in a powder mixer, and THEN put them into an oven.
The only problem I see (apart from having to purchase a kiln, a pulverizer, a mixer and a laser sintering printer) is how to buy dopants. For silicon, the acceptors from group-5 are Phosphorus and Arsenic. Phosphorus is a controlled substance and can not be purchased without a license, at least in US. Arsenic is not (but it's supposedly poisonous so there might be some other law for that too).
For the scales I'm aiming at (~1 cm^2) crystalline properties and uniformity shouldn't play too large a role (although I dunno about other silicon impurities). It may be possible to buy broken blank silicon wafers if that becomes a problem.
Though never took a semiconductor physics class so...[/QUOTE]
Somewhat related, have you watched Jeri Ellsworth's youtube vids on making transistors? She diys it all the way and ends up with useable FETs IIRC.
[QUOTE=nuttyboffin;43076876]Have fun hunting around for the other side of the wire then forgetting it when you go to solder it on and soldering on the wrong wire (i do that a lot)[/QUOTE]
Already ahead of you. The wires are sequenced so it's hard to screw up. Also it's the PLCC sockets that suck so much ass. The DIP TEXTOOL sockets are piss easy.
It will all be packed full of glue once testing is completed.
Got an LED strip from ebay that doesn't quite match the items description.
[url]http://www.ebay.com/itm/121166208243[/url]
Under the specification it mentions that the working current/meter is 1,2 amps while the working voltage is 12. In reality the current is way lesser, as each LED is limited by a 270 ohm resistor and they are connected in series of 3. There are 60 LED's in each meter so 4/270*60 = 0.89 A. I have confirmed that with a multimeter. Am I missing something or I got screwed over 4 watts?
EDIT: On a closer look, it looks like that they have sent me the wrong strip, as the one i have doesn't match the picture slightly.
Welcome to ebay
Circuit analysis final tomorrow morning...mmmm
I don't know jack shit about electronics, but I assume somebody will tell me if I'm wrong.
I have a 5v DC line, that I want to run some LEDs off of. I have 6 5v leds, and wired them in parallel.
The LEDs are really dim, I probably should of use different leds in the long run. Can I use something like a capacitor in front of them all, to increase their brightness?
[t]http://i.imgur.com/lxKPg7Z.jpg[/t]
Just soldered in the headers to my xmega128 adapter board and then soldered the whole thing into my protoboard.
[QUOTE=Drak_Thing;43104062]I don't know jack shit about electronics, but I assume somebody will tell me if I'm wrong.
I have a 5v DC line, that I want to run some LEDs off of. I have 6 5v leds, and wired them in parallel.
The LEDs are really dim, I probably should of use different leds in the long run. Can I use something like a capacitor in front of them all, to increase their brightness?[/QUOTE]
You mean they're designed for 5V, like those ones with the resistor inside ?
All I can suggest is different LEDs, putting your own resistor in also allows much greater control over brightness.
[QUOTE=Angus725;43085445]Circuit analysis final tomorrow morning...mmmm[/QUOTE]
How'd it go?
[QUOTE=DrDevil;43106241]How'd it go?[/QUOTE]
Thought it was good, but with the way they mark it, who knows. Spent an hour trying to derive a 2nd order differential equation, only to figure out that it wanted a first order ODE in the last 15 minutes. Luckily I had the rest of the exam finished already.
In hindsight, doing Laplace transforms on the circuit would have been easier.
[QUOTE=Angus725;43108337]Thought it was good, but with the way they mark it, who knows. Spent an hour trying to derive a 2nd order differential equation, only to figure out that it wanted a first order ODE in the last 15 minutes. Luckily I had the rest of the exam finished already.
In hindsight, doing Laplace transforms on the circuit would have been easier.[/QUOTE]
In my 2 years of electrical engineering, I'm still only able to do circuit analysis. Never did get any shit done practically.
[QUOTE=war_man333;43111175]In my 2 years of electrical engineering, I'm still only able to do circuit analysis. Never did get any shit done practically.[/QUOTE]
That must be really boring, I can't imagine doing a week of circuit analysis let alone 2 years.
I think practical experience and experimentation teaches so much better.
[QUOTE=Chryseus;43111360]That must be really boring, I can't imagine doing a week of circuit analysis let alone 2 years.
I think practical experience and experimentation teaches so much better.[/QUOTE]
I agree; For the first two years of my course, we had labs that (tried to) synchronise with lectures to reinforce what we had learned in lectures. I always found that going through a well explained lab, and being challenged to write down what you think is happening (and why) made me learn a lot more, in a much shorter timescale than otherwise
[QUOTE=war_man333;43111175]In my 2 years of electrical engineering, I'm still only able to do circuit analysis. Never did get any shit done practically.[/QUOTE]
Also 2nd year here.
I'm supposed to be in Computer Engineering (mix of hardware and software), but because it's joint with Electrical Eng 2nd year, I'm also taking a bunch of electrical courses this and next semesters.
It's not so bad when you're with a club that does some electronics though, as you get to work on some more practical stuff.
[t]https://dl.dropboxusercontent.com/u/64514745/2013-12-06%2016.36.51.jpg[/t]
[editline]8th December 2013[/editline]
[QUOTE=Chryseus;43111360]That must be really boring, I can't imagine doing a week of circuit analysis let alone 2 years.
I think practical experience and experimentation teaches so much better.[/QUOTE]
Totally agreed, I have done 2 circuits courses already, and still have another one left to go (transistors course).
Would any of these rectifying circuits work given the components maximum voltage and current limits isn't exceeded?
[T]http://s16.postimg.org/bb9yywwit/N_P_mosfet_rectifier.png[/T]
[T]http://s28.postimg.org/rn6s1ueal/NPN_transistor_rectifier.png[/T]
[QUOTE=Chryseus;43111360]That must be really boring, I can't imagine doing a week of circuit analysis let alone 2 years.
I think practical experience and experimentation teaches so much better.[/QUOTE]
Sure is. Because I'm a very practical person.
It's like when people try to teach you programming but you never get to program.
You're gonna have a bad time.
So i opened a case on ebay about the LED strip and the seller offered me a 1 dollar refund and to keep the strip. Not bad, considering i paid 3$ for the strip and only got 2/3 of the promised wattage.
[QUOTE=O Cheerios O;43121562]Would any of these rectifying circuits work given the components maximum voltage and current limits isn't exceeded?
[T]http://s16.postimg.org/bb9yywwit/N_P_mosfet_rectifier.png[/T]
[T]http://s28.postimg.org/rn6s1ueal/NPN_transistor_rectifier.png[/T][/QUOTE]
I don't even know if they would work, but I suspect on the assumption they did, they'd be horrifically inefficient.
[QUOTE=O Cheerios O;43121562]Would any of these rectifying circuits work given the components maximum voltage and current limits isn't exceeded?
[T]http://s16.postimg.org/bb9yywwit/N_P_mosfet_rectifier.png[/T]
[T]http://s28.postimg.org/rn6s1ueal/NPN_transistor_rectifier.png[/T][/QUOTE]
Keep in mind that MOSFETs act like diodes in reverse polarity due to the nature of their construction.
Also mildly unrelated but whats with the divide sign? Notation I'm not familiar with!
Thyristors are a far better solution for controlled rectifiers.
My dad strikes again, giving me this:
[t]http://o38.img-up.net/CAM00245613d.jpg[/t]
[t]http://v34.img-up.net/CAM002465e12.jpg[/t]
[t]http://n40.img-up.net/CAM002475b08.jpg[/t]
[t]http://a32.img-up.net/CAM00257d328.jpg[/t]
I don't know what he is expecting me to do, I have no way to drive that thing and there is no documentation anywhere on the web.
Take it apart and take a look mang
Can easily take the front off but the back has weird screws...
Hey guys,
Has anyone got a good idea for a project i can do to help me get into University, i need somthing i can write large amounts about and also design a production system
for it as if i was going to sell the product.
It would be a huge help, i cant seem to come up with any ideas saddly.
Also, got 2 new power supplies:
[img]http://i.imgur.com/qIoTvV6.jpg[/img]
If anyone is interested in the UK, i may be able to get more of these (£200 includes P&P each)
So, you want to make something (a PCB), and then have a test-jig for programming and testing the boards?
[QUOTE=DrDevil;43146187]So, you want to make something (a PCB), and then have a test-jig for programming and testing the boards?[/QUOTE]
Not exactly, its more like.... Design a full product including casing (will only make contents however likely)
also design the whole production method for production quanitity of 100 pcs, 10,000 pcs and 1,000,000 pcs
[QUOTE=nuttyboffin;43146194]Not exactly, its more like.... Design a full product including casing (will only make contents however likely)
also design the whole production method for production quanitity of 100 pcs, 10,000 pcs and 1,000,000 pcs[/QUOTE]
Well, regarding the PCB part, I can recommend this video series: [url]http://www.youtube.com/watch?v=VXE_dh38HjU[/url]
Otherwise, good luck!
Sorry, you need to Log In to post a reply to this thread.
