K, so i made a small battery powered arc welder (to weld 30 gauge nichrome wires to pcb pads) using [URL="http://www.instructables.com/id/Small-Welder-for-joining-Nichrome-and-Nickel-Wires/"]this guide.[/URL] Everything is the same but i used a 2200uF cap instead of a 1000uF one for more oomph (~1.2 joules instead of 0.6).
[t]http://i.imgur.com/e1qHp0A.jpg[/t] [t]http://i.imgur.com/9MqSnoO.jpg[/t]
However, it's not welding the nichrome wire to the pcb. It does create a huge spark and it's literally taking chunks out of the copper layer but it's not welding the wire:
[t]http://i.imgur.com/BfzrgZY.jpg[/t]
You guys know what's going on?
Wash the board with some sort of alcohol solution, I use a cheap solvent/lighter fluid that's 90-100% ethanol for the boards I make at school, it removes all the dirt and oxidization that prevent soldering and such.
That doesn't work, if I understood that schematic correctly, it creates a sustained arc - get a smaller capacitor and go for duration instead of power?
You're using too much power, you don't want the copper being blasted off, as nikomo said you want duration as well to allow the nichrome and copper to properly form a weld.
It would be a better idea to build a spot welder and weld the nichrome to a copper sheet or wire, you can get far better and more consistent results that way.
[URL="http://facepunch.com/showthread.php?t=1466939"]I actually meant to post this thread i made, as a post in here, but maybe anyone who knows about coding and Arduino controllers could pop by, and have a look?[/URL]
[QUOTE=Chryseus;47779207]You're using too much power, you don't want the copper being blasted off, as nikomo said you want duration as well to allow the nichrome and copper to properly form a weld.
It would be a better idea to build a spot welder and weld the nichrome to a copper sheet or wire, you can get far better and more consistent results that way.[/QUOTE]
I sanded down the wire and copper surface gently and then cleaned them with alcohol. I also tried reducing the voltage (35V all the way down to 18V) but i couldn't get it to work. It was either vaporizing the wire or not doing anything at all. So, changed the 2200uf cap with a 1000uf one.
[t]http://i.imgur.com/YRv8zHT.jpg[/t] [t]http://i.imgur.com/c0MKUE9.jpg[/t]
Exact same shit as the 2200uf one. It's still not working fml. I don't know what to do. Looks like i'm going to have to make a spot welder now... Can you even use a spot welder on pcbs? Wouldn't it melt the shit out of the fr4?
Any of you know how i can join resistance wire (nichrome, kanthal, etc) to copper pcb pads?
Does solder not work?
Solder can work but it's rather tricky since nichrome is similar to aluminium in that it rapidly forms an oxide layer, you need to use a strong acid flux such as that used for plumbing.
[QUOTE=aydin690;47784903]Can you even use a spot welder on pcbs? Wouldn't it melt the shit out of the fr4?[/QUOTE]
You'd spot weld it to copper sheet or wire then solder that as normal.
How do I probe this LED with a multimeter to work out what it's voltage and resistance specs are?
We need to know the input voltage. Probing the LED itself will give you nothing.
[QUOTE=RoboChimp;47786576]How do I probe this LED with a multimeter to work out what it's voltage and resistance specs are?[/QUOTE]
Your best bet is making a constant current source (say with a [URL="http://www.robotroom.com/LEDTester.html"]LM317[/URL]) and then measure the voltage drop across it, then you'll be able to get the parameters according to the [URL="http://www.pveducation.org/pvcdrom/pn-junction/diode-equation"]diode equation[/URL].
With I_d being the current flow & V_d is the voltage drop across the LED. The main characteristic you'll need is I_saturation.
Well it's finally time to retire my crusty old Nexxtech (Radioshack/Source brand) multimeter. It's served me well considering it was like $30 and probably 6 years old at this point. After using something so bad for so long, it really makes me want to make a pretty big jump up to something better. I've been looking through that big spreadsheet of multimeters from the EEVBlog forums and I'm leaning pretty hard towards the [URL="http://www.uni-trend.com/en/product/2014_0711_586.html"]UNI-T UT181A[/URL] which can be had for $360. It's a Chinese brand but people seem to be saying good things about them, and it really seems to out perform most the other multimeters in terms of features and capabilities. The actual TFT LCD screen is also a really nice touch. I feel like the multimeter guys are in cahoots with the graphing calculator people, ensuring we're all stuck with 1970s black-on-grey LCD tech until the end of time.
Definitely open to suggestions here though. My price point is around $300 US.
Looking at the insides of that meter it looks quite nice actually ([url]http://www.eevblog.com/forum/testgear/uni-t-ut181a-pictures/[/url]), pretty much a chinese copy of the Fluke 289 by the looks of it. Not sure if I'd want a multimeter with a TFT LCD and overcomplicated functions for daily use however.
Whatever suits your needs I guess.
So been fiddlin' with FreeCAD a while. Hooboy, what a learning curve! I was hoping I could just make some small pieces of stock like 1" square tube or angle or rod or whatever (easy!) and then use them as a sort of parts pallet to assemble whatever I wanted from them by copying them/modifying their lengths. Unfortunately, one would assume the straight-forward method of doing this would be to duplicate the part you want (say, 3/4"x3/4"x1/8" angle), duplicate it, and then modify the length of that part to fit your needs/add holes as necessary.
But no, no such feature yet that I can see. Only way I can see to do that would be to make the "stock" piece, then modify it over and over and save each modification as its own file, then import it later. Gonna have a HUGE clusterfuck of files by the time this is said and done. :v:
This is probably slightly off-topic (and probably an incredibly stupid question because I know [B]nothing[/B] about the topic) but I thought I'd ask anyway since this is the most likely place I know where I could get an answer:
I have an e-bass (cheap) and a sound card ("built-in" but decent), no clue about the technical data of either.
I can connect the bass to my sound card's mic port and use it for Rocksmith (2014), but it's far too quiet by itself (to the point where it doesn't show up on the preview graph).
Windows 7's microphone amplification(?) feature makes it loud enough for the game, but that introduces terrible noise and probably also kills modulation.
Setting it to something in-between and filtering out the noise in Rocksmith [I]kind of[/I] works but it's not pretty.
Is there any way I can amplify the signal [I]for very little money[/I]?
The controls on the guitar are definitely complete garbage so I'd need something I can roughly regulate.
I [I]probably[/I] can't solder without buying stuff for it, but we do have connector clamps and possibly a few random parts around in the unlikely case making one solution from scratch is cheaper... actually I do have a headphone add-on cord with variable resistor somewhere, so regulation is a bit less of an issue. (There's some kind of loose contact / short circuit but it works when you don't move it around too much.)
[QUOTE=Tamschi;47809213]This is probably slightly off-topic (and probably an incredibly stupid question because I know [B]nothing[/B] about the topic) but I thought I'd ask anyway since this is the most likely place I know where I could get an answer:
I have an e-bass (cheap) and a sound card ("built-in" but decent), no clue about the technical data of either.
I can connect the bass to my sound card's mic port and use it for Rocksmith (2014), but it's far too quiet by itself (to the point where it doesn't show up on the preview graph).
Windows 7's microphone amplification(?) feature makes it loud enough for the game, but that introduces terrible noise and probably also kills modulation.
Setting it to something in-between and filtering out the noise in Rocksmith [I]kind of[/I] works but it's not pretty.
Is there any way I can amplify the signal [I]for very little money[/I]?
The controls on the guitar are definitely complete garbage so I'd need something I can roughly regulate.
I [I]probably[/I] can't solder without buying stuff for it, but we do have connector clamps and possibly a few random parts around in the unlikely case making one solution from scratch is cheaper... actually I do have a headphone add-on cord with variable resistor somewhere, so regulation is a bit less of an issue. (There's some kind of loose contact / short circuit but it works when you don't move it around too much.)[/QUOTE]
The only real option without a soldering iron is getting a ready made one, you can get cheap audio amplifier modules off ebay, just search for TDA amplifier module and you should be able to find something that will work for you.
You guys know how to get an arduino to sweep from like 1khz to 12khz and then loop back to 1khz? I was thinking of just writing a for loop with the [I]delayMicroseconds [/I]function but i feel like there should be a more efficient way of doing it.
[QUOTE=nikomo;47785664]Does solder not work?[/QUOTE]
No, you can't solder alloys of chromium.
[QUOTE=Chryseus;47786046]Solder can work but it's rather tricky since nichrome is similar to aluminium in that it rapidly forms an oxide layer, you need to use a strong acid flux such as that used for plumbing. You'd spot weld it to copper sheet or wire then solder that as normal.[/QUOTE]
I tried pre-tinning and strong acid fluxes and it [I]seemed[/I] to work but it dewetted after like half an hour. I ended up spot welding the wire to a sheet of nickel and then soldering that to the pads.
Needed the primary out of this really large old MOT. It was the biggest out of all the other ones I have. I don't know if that's because it's older or just from a higher wattage unit. I'll be unwinding and rewinding it in a different matter later on.
[IMG]http://i.imgur.com/reGfl94l.jpg[/IMG]
[URL="http://imgur.com/a/1jChq"]More pics of the process [/URL]
transformers always look incredibly messy to me, even when not taken apart.
[QUOTE=aydin690;47784903]I sanded down the wire and copper surface gently and then cleaned them with alcohol. I also tried reducing the voltage (35V all the way down to 18V) but i couldn't get it to work. It was either vaporizing the wire or not doing anything at all. So, changed the 2200uf cap with a 1000uf one.
[t]http://i.imgur.com/YRv8zHT.jpg[/t] [t]http://i.imgur.com/c0MKUE9.jpg[/t]
Exact same shit as the 2200uf one. It's still not working fml. I don't know what to do. Looks like i'm going to have to make a spot welder now... Can you even use a spot welder on pcbs? Wouldn't it melt the shit out of the fr4?
Any of you know how i can join resistance wire (nichrome, kanthal, etc) to copper pcb pads?[/QUOTE]
I don't know much about electronics, but I know a lot about welding, and you're running too hot. This is done by pushing too much amperage too fast through the material, causing the heat affected areas to become brittle.
If you can find a way to regulate the amperage, it would make it better. That design will put out a specified amount of amperage, which may be good for his application, but any change, such as material thickness, will require a change in the current needed.
You also need to understand the difference between arc welding and spot welding. Spot welding slowly heats the materials until they melt together. Arc welding relies on a sustained arc to melt the metals. This process leaves the weld open to external impurities in the air. To combat this, arc welding requires shielding to prevent air from getting into the weld. On stick electrodes, the coating burns creating the shield. In processes like MIG and TIG, the shielding is produced by blanketing the weld with a sheilding gas, such as argon or CO2.
You need spot welding for this application. The discharge of a cap is too quick for you to watch and see when to stop, so you're going to have to find a way to regulate its discharge amperage and voltage.
I found the conformance testing data for the biggest transformer on a PSU I'm still taking apart (in picture, still mounted on the board)
[t]https://i.imgur.com/L88nFr7.jpg[/t]
I reckon I might be able to get a lot of what I need for a power supply, off this one PSU.
[QUOTE=nikomo;47817579]I found the conformance testing data for the biggest transformer on a PSU I'm still taking apart (in picture, still mounted on the board)
[t]https://i.imgur.com/L88nFr7.jpg[/t]
I reckon I might be able to get a lot of what I need for a power supply, off this one PSU.[/QUOTE]
Are you making a switching supply?
I was thinking flyback or SEPIC pre-regulation, into a linear regulation stage.
The ripple after the switching regulation could be a problem, the high frequency component from switching will still be in the final output.
I found a neat application note (Application Note U-165, Lisa Dinwoodie) from a company called Unitrode, it deals with flyback converters, I need to read through it.
I want to know what kind of power output I can get out of what transformer, I don't want to end up building something and then see that it barely pushes 2A out at 5V.
Friend of mine asked me to fix his tablet's USB port for him, said his kid tried shoving the cable in the wrong way and forced the connector inside, which would lead me to believe the kid somehow pushed the connector off the board, peeling the traces and what-not with it. Not fun stuff to repair, but doable.
Cracked it open and found out that the USB connector was perfectly fine, all neatly soldered in place and held there tight with some sort of tough, rubbery red adhesive.
Except, it wasn't perfectly fine, because what the kid actually did was he somehow managed to chip the plastic bit in the middle, leaving one of the connectors just floating in air.
So now I've got him rummaging for a donor phone so I can replace this connector. :v:
I got a small lot of microUSB connectors from eBay (like 10 or 20pcs), just for projects and repairs.
Stuff's useful yo.
Take a guess which cap I replaced two years ago. The others were so badly swollen that they were forcing their rubber bungs out. :v:
[IMG]http://i11.photobucket.com/albums/a166/ballsandy/Computer%20related/IMG_9243.jpg[/IMG]
I didn't have replacements on hand, so I did something dirty and stole replacements from another motherboard.
[IMG]http://i11.photobucket.com/albums/a166/ballsandy/Computer%20related/IMG_9246.jpg[/IMG]
What brand are those?
Old ones were United Nippon Chemicon, KZG series. The replacements were Panasonic.
[QUOTE=nikomo;47822643]I got a small lot of microUSB connectors from eBay (like 10 or 20pcs), just for projects and repairs.
Stuff's useful yo.[/QUOTE]
Got some cheap from Cheapistan, but it says estimated delivery is between June 16th and July 21st. :/
Hi everyone.
I don't know if this will get me hated or anything. But I will jump right to the point.
I am a relatively good programmer, I know a bunch of programming languages and can accomplish a lot with my code. But, I would like to move in to the field of electronics - or something similar, perhaps robotics. I know this is vague, but in the long run I would like to be able to create something like Tile, something like Nest - does this make sense?
Where would be the best place to start with this? Hope this is not TOO off topic.
Thanks in advance.
[QUOTE=mikkeljuhl;47831685]I would like to be able to create something like Tile, something like Nest - does this make sense?
Where would be the best place to start with this? Hope this is not TOO off topic. [/QUOTE]
Picking up a microcontroller development kit like an Arduino would be the best place to start as well as learning the basics of electronics with a breadboard and some components, if you want to make a finished product in the future you'd also need to spend some time learning how to design and layout a PCB.
I also added a bunch of new free books to the op.
Sorry, you need to Log In to post a reply to this thread.
Electrical Engineering V3
