diy - DIY & Electronics

OP here, my last serious project has been this capacitor charger, pretty powerful, can charge them up to ~400V. Ideal for coilguns, close contact EMP devices...

I have videos and detailed instructions about how to make one if someone is interested.

Now I'm pretty busy and I haven't done much recently. I just finished setting up my LED strip lighting, much better than the fluorescent desk lamp I used to have.
Also, getting a new oscilloscope in no time.

I'm also interested in high voltages and flyback transformers, if you have any question please ask.

>>743
>I have videos and detailed instructions about how to make one if someone is interested.
I for one would love to see how you went about making that!
Could you do a write-up for the magazine?

>>743
Did you make the holed board yourself?
Seems easier to make than a PCB, I take it you solder the components with eachother in the back/down side?

>>746
http://www.instructables.com/id/Capacitor-charger-60-400V/
I'm a bit stalled right now, but there are some other projects, like high voltage generators. I was planning to make a hoodie with taser sleeves, maybe this summer.
>Could you do a write-up for the magazine?
Pardon my ignorance, I'm new here, is there a magazine? In that case I wouldn't mind to contribute.

>>747
They are called perf boards, or prototype boards, I bought it at an electronics store nearby. At the other side of each hole there's a copper pad to solder the components and then make the necessary connections.
This is how it looks underneath, this one looks pretty lame because, but you get the idea.

PCB's are not really worth it for me, it takes me quite a lot of time to design them in Eagle, and then the toner transfer method isn't always reliable, specially if the cartridge isn't full. Maybe I'll move on to more advanced circuit printing techniques when I need to make more complex circuits. It is only worth it if you want to make copies of the same circuit.

>>748
Sorry, I finally found it. This is great, definitely worth a read.
https://lainchan.org/zine/latest.pdf
I even feel like watering down it's content by contributing, but I'll try my best if I have to.
I still have a lot to learn, experts in the field would find many flaws and ways to improve my circuits.

>>749
There are several people into electronics on lainchan (including me), that can help you with errors on lainchan.org/irc if you need some help

>>750
I don't usually bother other people unless it's completely necessary, but thanks for your help. I intend to make lots of progress this summer, I'm getting new equipment and tools to do so. I'll be around for a while.
Also, please excuse my English, I'm not a native speaker.

Just finished installing this LED strips, I've even made a brightness regulator. They're much better than the flimsy lamp I've been using all along.
There's nothing else here because I havent moved all my junk yet.

>>752
If anyone is interested, here's the circuit, the 2222 and MOSFET part was taken from a flyback driver, so it might be wise to increase the value of the 330 and 100 Ohm resistances to make the circuit more efficient.
I didn't had a 50k pot, so I ended using a 100k with a 10nF capacitor.
I don't really think that cap is useful, maybe with limited current supplies so energy can be stored during the off state of the dimmer and released afterwards relieving some strain.

>>742
If anyone is interested about the source of that picture, it's from a guy who usually posts in 4chan's diy board. He also posts about his projects, which are quite interesting. An amazing workspace.

>>753
Sorry for making some kind of blog out of this thread since I'm the only one who is posting here, anyway...
I bought a 25W supply for the LEDs for 8€, the one I was using is too old and the voltage varies from 15 to 12.3 depending on the current, which only goes up to 1Amp, I guess it's a transformer-rectifier-capacitor type since it's quite old and it gets very hot. This new one seems much more efficient and will allow me to add a third or even a fourth strip for extra light.

I'll show the completed circuit today if I can finish it in time and I find a nice enclosure to attach to the wall.

>>768
You should write an article for the lainzine about getting into electronics

>>770
I second this inquiry.

>>770
I'm reading the thread at /lit/, English isn't my first language, but I'll try my best. I think it would be nice to have a couple of electronic projects with standard components that people can make and understand to learn about it.
>You should write an article for the lainzine about getting into electronics
I got in in a quite conventional way, although I actively work on my own. I'll try to see what can be done, but I don't think I should give advice about this since I think I don't think have enough experience apart from the hobbyist/student level.

>>770
Seconding this. I bought some electronics things awhile ago but wasn't able to get myself into it correctly. I'd love to hear some tips.

I have several exams before holidays, I'll be quite busy this week and maybe the following one too. I've seen the next number is nearly complete, I hope to be able to finish in time.
I finished the LED lighting, I finally have a decent illumination here.
I didn't wanted the regulator to waste power when at full brightness (even though it just wastes ~0.25W + 0.11W dissipated by the MOSFET in the worst case, it is somewhat insignificant), in a fit of OCD I added a switch to be able to choose between full power and regulated power. I had to modify the regulator and now it doesn't seems to work properly, I can regulate the brightness halfway up to full power, this must mean there isn't enough current so I know I'm losing lots of current somewhere else, but I don't want to fix it now, so I'll use the full power mode until I have time to fix it.

I used a bipolar switch for the connection of the grid to the PSU, then I connected +12V to the center pin of a switch, one way let's pass all the current though the LEDs, the other way gives 12 volts to the regulator which powers the LEDs, after modifying the circuit the MOSFET acts as an output rather than as a sink of current, I did this to use just one switch.
The first thing I did when I got the PSU was to plug it in and check for abnormal readings with the multimeter. In this case the output was 11.2, luckily there was a variable resistor to control the output voltage and I was able to bring it up to 12, it was gunked so I almost destroy it in order to change the resistance.
Another thing I've learnt is the resistivity of the tracks that run parallel down the strip to power the LEDs seems to be a bit high, measuring the voltage at the beginning of the strip gives me around 12 volts, but after the voltage has decreased 0.17V, even though I connected the three stips in parallel to avoid this.

>>775
You mean the LEDs are all powered in parallel off of the power track that's running alongside them, so that you don't need to put like (Vf) * (# of LEDs) = (Hi-Voltage) into the strip to get through all of them? And it's that power track which has so much voltage drop?

>>777
The strip is all in parallel, I mean, there are two tracks running parallel at the sides which power the LEDs. But if I cut the strip into pieces I can connect the pieces in series (as it was before) or in parallel, this means paralleling those tracks.

I know I haven't made myself clear, so I made this. Since the tracks are slim and there's some current flowing through them the voltage keeps falling as it reaches the end, paralleling the strips makes all the strips to use the same power and increase the brightness a bit.

I started with electronics a few weeks ago.
I'm thinking in do this:
https://www.youtube.com/watch?v=OUwGC35vX50

>>779
555 timers are a must-have, they have lots of different applications.
Try to combine it with a counter, like the 4017. Counters basically "count" the number of pulses switching on one of their outputs at a time, you can make an 8bit synth with it for example.

https://www.youtube.com/watch?v=nyTz_mcBOnE

I finally finished my exams, I can start writing now.
>>770
>>771
>>774
Getting into electronics, got it, I'll write about that, but you should take my advice with a grain of salt since I'm still quite new.
I was thinking of doing a series of articles make readers familiar with the basics of electronics so it's easier for them to get into the world, ohm's law, basic definitions, electronic components like transistors, capacitors or inductors, circuits...

>Part 1:
-Short intro (a bit of rambling) about electronics
-Getting into electronics: Indispensable tools and things you might need to know
>Part 2:
Theory: Ohm's law and power dissipation (Too easy?)
Component:(Theory and uses): Resistor ¿Transistor? ¿LED and diodes? ¿Capacitor? ¿Inductors? (often related to theory)
Forgotten technology: Maybe I'll add a subsection about dying or obsolete tech that is no longer in use (e.g: vacuum tubes, magnetic core memory, vacuum fluorescent displays...) I find this quite interesting.
>Part 3:
Circuit: A circuit so the reader can apply the concepts

The main issue I see with this is the first articles will be quite simple and obvious for a mildly versed reader. I was thinking about including the transistor because the resistor is a very simple component and adding it would fill an entire page at least while pleasing more advanced readers, since I'm quite worried about watering down the content and diminishing the overall quality, also if I only explain a component at a time it would take many publications to have a general idea, and at this rate I don't want to do it.
If I include the transistor I may include the LED diode and diodes as well to make a circuit with all those components. I guess all this would occupy several pages, probably more than the average article.

The last thing I was wondering about is how should I handle it, I guess I'll write a word with double column with similar format and looks and handle the pdf.
Excuse my english, it's not my first language, I'll probably advice the editor to take a quick read just in case...
Let me know what do you think please.

im trying to fix an old keyboard, the board got damaged, the copper pads are ripped off, so i will make the schematic in kicad and rebuild the whole thing in a protoboard for testing, if everything works fine i will make the pcb, also the plastic connector from the sheets are ruined, it's impossible to get FCC conector over here.

>>797
Neat, glad to see it being saved from ending in the trash, that's a lot of work but seems to be worth it.
>>753
I wasn't losing current, the LEDs were dimmer because I connected the MOSFET the wrong way, normally the MOSFET should be connected as the picture shows, if you connect the MOSFET before the LED strip the voltage difference won't be enough to activate it correctly, or at least that's what I understood. Now the IRF540 works just fine, with no noticeable change when shorted at full brightness, it doesn't gets too hot even though it's switching around 1.5 Amps.
The 555 was getting too hot, I guessed this was because the resistance at the output was was too low, I changed the 100R resistor by a 1k and now it works just as fine with no noticeable change in brightness, maybe I'll also change the 330R by a bigger value, since there is an average current of 18mA going down the 2222 and maybe it isn't necessary to have that much current flowing to charge the gate of the MOSFET, when I have the oscilloscope I'll measure the frequency to calculate how big the resistor should be to ensure the gate gets charged, that way I'll save some power.
The only problem now is the supply seems to hiss when the brightness is brought to its minimum.
http://www.555-timer-circuits.com/common-mistakes.html

>>780
Holy dicks, would it kill you to boot something up to make a decent schematic?

>>802
I didn't make it, I took it from the internet, it's not my fault.

>>791
Want to give you a heads up on that i've already submitted a zine article about troubleshooting and fixing old electronic equipment. But more in depth about electronic parts, tools, ohms law and stuff would be awesome to add as a seperate article

i will repost if necessary

anyone know if i can fix this soykaf

>>806
I haven't found much, but this made me laugh:
https://www.youtube.com/watch?v=VMbdhu7EeSA
Probably the ribbon cable or something else is loose, try opening it and making an improvised wedge to press the ribbon against the connector or whatever.
>>805
I'll avoid talking about the same things, I don't know if I'll be able to submit it in time, but I'll try.

>>753
Last and final version of the circuit before stuffing the thing in a box. The switch is to avoid power consumption by the regulator, there might be a better way around this.

For those expecting an article about starting in the world of electronics I have some links that might be useful, since I won't be able to write said article, at least not in time for the 2nd number.
http://blog.thelifeofkenneth.com/2011/06/so-you-want-to-build-electronics.html
http://pastebin.com/9UgLjyND

Working on two circuits, an EMP gun powered with a 3.7V lithium battery (when tuned it takes about 10 seconds to charge 200V 2000uF caps) and a heart beeper thing, powered by a 9V.
I'll post full instructions on how to make the EMP gun, maybe the heart thing too if I can get the board right...

I'm currently playing with a python neural net library (pybrain). Was thinking about using the GPIOs on a beaglebone as inputs to some network, but I can't think of any real use for it. Anybody has an idea?

I haven't been able to break out a soldering iron much, but something I think might be pretty nice is a computer system built around a DSP as the main processor, in which main components are connected by a parallel bus that goes around in a simple loop on a double-sided PCP so they can be connected in series without having to worry about turns in the bus causing skew between channels. I dunno how much it makes sense but I figure it could be nice for having a machine that keeps real-time response to a lot of things, and I would the Harvard architecture aspect of it could be useful if you want to isolate things.

At left: artist's depiction

Just posted this to Yotsuba"s /diy/, but thought maybe y'all could help me too.

Cloning an amplifier.
Only thing left to figure out is the power transformer.
Original company wrapped their own, so I'm trying to reverse deduce something compatible that will work.
Mains in, +-40 AC out (that's 80V with a center tap if I'm understanding this correctly).
The whole amp should pull 700W at `rated output'.

What does this mean for the transformer power rating? Does it need to be at least 700W[VA] as well?
Why is the rated current for the primary never specified? Is it always just assumed to be able to handle it based on the rating of the secondary?
If the secondary output is +-40V, is the current figured by 700W/40V because it's all AC or 700W/80V because that's the total potential?
More generally, does this imply a coil ratio of 3:1 (120:40/240:80) or 3:2 (120:80)?

Is Hammond the best transformer brand?
Is this way overkill or the one I should buy? http://www.mouser.com/ProductDetail/Hammond-Manufacturing/182V40/?qs=sGAEpiMZZMvwUzoUXIIvyQPvPmwnNFGyC3HMAk3oIMg%3d

I've never built anything with a transformer like this before, and the real world way they do this is still confusing to me.
Just trying to make sense of it all.
Thank you for any help you can offer.

>>905
Update - I'd forgot to consider that the ratings are RMS and not peak.
I think http://www.mouser.com/ProductDetail/Hammond/1182U30/?qs=%2fha2pyFadui7tzTDjgBcipVOQ4OZkf5x4mtu2fnxels%3d might be the one I want.

60V center tapped will give me (* 30 (sqrt 2)) => 42.5V rails.
1000VA is well above my 700W requirement.
Current through each secondary should then be 11.7A - Well under the 16.7A rating.

Is all this right?
Pic related is what I'm trying for if that helps.

Can anybody help me understand the current limit mode of this LT3750 chip?

I want to achieve constant current (not more than 2mA) at about 150V but not sure I understand how this is possible.

In my understanding, the feedback line will help the chip adjust Duty cycle depending on the voltage, how is that current related?

Or is it that my understanding of a flyback converter is broken?

There are 2 things to consider, frequency and duty cycle, is one or the other proportional to current and the other voltage?

http://cds.linear.com/docs/en/datasheet/3750fa.pdf

What? Needed more GPIOs...

>>928
Doesnt work :<

>>806
What is soykaf?

>>967
Newlain detected.

It's a wordfilter on the word s.hit it comes from a cyb game called shadowrun

>>969
whidgle detected.

>>873
Well a neural net acts a control system, so anything you would like controlled by automation. What mechanical stuff are you interested in?

I'd like to try implementing neural nets in hardware, as the neurons themselves seem pretty simple: The weighting part would just be variable resistance, and a relay or Zener diode would perform the function of only activating when signal (either current or voltage) is high enough. Less obvious to me is the method of implementing an update function without having it controlled by microchip, which goes against the whole idea of using discreet components that can be made without clean rooms ( as I describe here >>1008 ).

>>912
>>905
Well, I bought all the pieces guys.
Wish me luck.

>>1031
Hope you have better luck than me!

>>1031
How did it go?

>>1036
They haven't moved.
I actually prolly won't be able to really get to this for another few weeks.
I will of course keep you posted though.

>>1028

Apparently Minsky made a thing pretty similar back in the 60's with vacuum tubes and motor-controlled resistors. I doubt he used any microprocessors/microcontrollers in his system. It was his thesis project, so it's probably not too hard to find the schematics.

Hello fellow Lainons.
This isn't really a project I'm working on but more of a help request.

You see I was using my blender when it suddenly stopped working. And when it didn't work even after cooling down I tried opening it up and saw that yellow goo on some of the parts. After reading up on it I guess it might be some sort of silicon glue, maybe? I just assumed it might have melted because of the heat and short-circuited. It appears to be in contact only with the wire of the "D5" element (there were some other spills not visible in the pic).

So my questions are mainly:
Can I fix this?
What kind of element is this D5?

I should also say I have only a very basic electronics knowledge.
Will of course provide info/pics if somebody would like to help.

>>1210
the big red thing looks like some sort of capacitor that's blown it's guts everywhere (as they do), if you can find it's rating then it should be easy to find a replacement. D5 is just an ordinary resistor, very unlikely that there's any issues with it. new parts should be very cheap.

>>928
I don't know anything about that particular chip but in general you can make a constant current source by just sticking a resistor between the "source" and "sense" inputs of a voltage regulator. Then it's just ohm's law to find voltage in terms of current or current in terms of voltage.
>>1212
It could just be some glue. Electrolytics are the ones that get "interesting" when they get overloaded but that isn't one. My guess is whatever's under that heatsink (or something giving it input) is buggered. Also R5 is a resistor but D5 is clearly a diode.

>>1212
Correction, D5 is a diode. It still looks fine.

It looks to me also as though that cap is not blown; it appears to be covered in glue. That appears to be a thin film capacitor, also called dielectric film or polymer film. They don't usually go as dramatically as electrolytic capacitors.

>>1210
It may not be easy to fix, but also it may be. Do you have any tools at all? (multimeter? anything?)

>>1214
Yes I do have a multimeter (Mastech 1015B, analog), which I played with for the first time. Quite fun.

R5 and R7 (left from the heatsink) show me 3 and 10 Ohm.
R1 makes the needle move very slightly between 10k and 1k, it basically stays on the left while it should go from the right, correct?. Had to change the setting from X10 to X1k to get it to move even that far.

The Diode doesn't seem to give me anything. Only if I lower the resistance it bumps very slightly for a second, but I guess that might be caused by some leftover currents or so. The meter reacted though when I accidently touched the resistor, but only when I connected the right leg of the Diode and the left leg of the resistor. I tried with both ends of the multimeter, same reaction.

Couldn't get much from the capacitors since their legs aren't easily accessible with the exception of the right one (C1, which is also covered in that stuff).

I tested all parts without detaching them.

I also have a soldering iron and the basic screwdriver, plier set.

>>1217
Measuring individual components while they're in the circuit isn't going to tell you what you want. You'd have to draw out the circuit diagram and calculate what the resistances ought to be. Also it's pretty rare for resistors to fail without visible burn marks. More likely it's the semiconductor (triac?) under the heatsink or bad contacts in one of the relays. You did change the fuse right?

>>865
Seems cool but I wouldn't fuarrrk with it out of fear of accidentally hurting someone with electronic medical devices.

Does anyone know the legality of electronic weapons?

>>1217
The diode should conduct in only one direction. Check conductivity in both (by switching which side you have the power and ground on).

I notice in the rightmost picture that some of the glue looks pretty discolored, possibly due to heat. Under the heat sink you have a high power part, probably a FET which turns on the motor. If it burned out, that could be the cause of the heat.