Electrical Engineering V3 - Developers

I'm looking into measuring battery levels with an Arduino since multiple groups have already killed off their (7.4v) LiPo batteries by draining them all the way down to 3v, and I know about voltage dividers (I was thinking of using a 10k and a 20k resistor) to get the voltage down below 5v so I can read it with the analog ports but a lot of schematics also include a capacitor between GND and Vout, do I need that (if so why) and how could I calculate the required values for it?

[QUOTE=Goz3rr;47395347]I'm looking into measuring battery levels with an Arduino since multiple groups have already killed off their (7.4v) LiPo batteries by draining them all the way down to 3v, and I know about voltage dividers (I was thinking of using a 10k and a 20k resistor) to get the voltage down below 5v so I can read it with the analog ports but a lot of schematics also include a capacitor between GND and Vout, do I need that (if so why) and how could I calculate the required values for it?[/QUOTE] Not an engineer, just a tinkerer, but the capacitor might be there to prevent sharp voltage drops when demand spikes.

Noise reduction. Also. [media]http://www.youtube.com/watch?v=f4-jbobSll4[/media]

We are not worthy.. now I am really interested in this book.

Fucking hell. The element in my desoldering gun burned out. Edited: my burned fingers indicate it is in fact something else..... [editline]1[/editline] Cold joint on the thermal control board.

There. Controls are now completely wired. [IMG]http://i11.photobucket.com/albums/a166/ballsandy/Computer%20related/IMG_8721.jpg[/IMG] And so we complete the PCB arrangement as well. One main PCB and two risers linked together that can be plugged and unplugged. [IMG]http://i11.photobucket.com/albums/a166/ballsandy/Computer%20related/IMG_8722.jpg[/IMG] [IMG]http://i11.photobucket.com/albums/a166/ballsandy/Computer%20related/IMG_8723.jpg[/IMG] Now the last part which I just really don't want to do. I have to begin building up the PCB.

Sent a board off to OSH Park for manufacturing, mostly because I'm interested in what kind of quality they can produce. [url]https://oshpark.com/profiles/nikomo[/url] If I ever make anything bigger though, I'll probably send it to Dirty PCBs - OSH Park is still fairly expensive compared to the crap quality Chinese alternatives.

Mounted all the pin headers and installed it for one final clearance check. Next time it's in here it will be finished. [IMG]http://i11.photobucket.com/albums/a166/ballsandy/Computer%20related/IMG_8726.jpg[/IMG] So while I was assembling everything I was also streamlining my layout. I've gone from two PCB's with ten relays to one with five. The rest are scattered about elsewhere in their own modular components. Means that my more solid state switching can take place on the other riser. Woo, more modularity. Also, manual wire routing is a bitch. This is what three hours has yielded so far. [IMG]http://i11.photobucket.com/albums/a166/ballsandy/Computer%20related/IMG_8753.jpg[/IMG] [IMG]http://i11.photobucket.com/albums/a166/ballsandy/Computer%20related/IMG_8754.jpg[/IMG]

Alright, so i'm want to pass a lot of current through a small load (less than 5 ohms). I drew up this circuit quickly and i think it should work but the load is only drawing 20mA instead of the expected 2.4A. Do you guys know what's going on? [img]http://i.imgur.com/1NGGyVm.jpg[/img]

Odd, that works in LTSpice. However I'd try swapping out that 1kohm with a lower value resistor as your mosfet has no gate current going to it (Thus the transistor in the 4N25 can fully saturate) and thus allowing your mosfet gate to fully saturate. First test your mosfet to make sure it isn't dead, probably just by removing the 1kohm/4N25 and directly connecting the gate to 12V (with the 10kohm still connected to pull down/remove static). My suspicion may be the 4N25, but why have the 5V square wave source when you could just have that sourcing the LED directly?

[QUOTE=nikomo;47414714]Sent a board off to OSH Park for manufacturing, mostly because I'm interested in what kind of quality they can produce. [url]https://oshpark.com/profiles/nikomo[/url] If I ever make anything bigger though, I'll probably send it to Dirty PCBs - OSH Park is still fairly expensive compared to the crap quality Chinese alternatives.[/QUOTE] I was impressed by the quality when I last used them.

Transistors are a buttfuck for me to work with compared to relays. :v:

[QUOTE=pentium;47419085]Transistors are a buttfuck for me to work with compared to relays. :v:[/QUOTE] mosfets ftw! :D

I'm getting my ass kicked by a 555 timer, for some reason it just won't work. :suicide: Will try again tomorrow.

[QUOTE=Leestons;47419321]I'm getting my ass kicked by a 555 timer, for some reason it just won't work. :suicide: Will try again tomorrow.[/QUOTE] Schematic?

[QUOTE=LoneWolf_Recon;47419391]Schematic?[/QUOTE] [img]http://cdn.instructables.com/F0U/NQNH/H742SU5Y/F0UNQNHH742SU5Y.LARGE.jpg[/img] 47uF instead of 0.47uF The only output I'm getting is when I move the wires, like an inteference.

Even at the lowest setting on the pot, your 555 is only oscillating at ~2Hz. Non-audible, also I'd suggest driving the speaker with a transistor/resistor combo instead of directly from the output pin as the 555 typically can only source ~10-20mA.

That's possibly where I'm going wrong. I'll take another look at it tomorrow when I'm not tired. Thanks.

I'm trying to use [URL="https://github.com/tmk/tmk_keyboard/tree/master/converter/adb_usb"]tmk_keyboard[/URL] with an old apple keyboard. How do you compile using a makefile on windows 8.1? Winavr crashes on the make command, and I can't find any other tools that will get me a hex file for my teensy. [editline]29th March 2015[/editline] Okay then now it's just working. :v:

So, someone I know has a Wiltron / Anritsu 7N50 RF detector for his scalar network analyzer. Apparently it doesn't work, I wonder why. [url=http://i.imgur.com/iSVsQPf.jpg][img]http://i.imgur.com/iSVsQPfl.jpg[/img][/url] By the way, these RF detector diode assemblies are tiny. [url=http://i.imgur.com/GIINkRw.jpg][img]http://i.imgur.com/GIINkRwl.jpg[/img][/url]

I repeat. Transistors suck. Consider the following: [IMG]http://i11.photobucket.com/albums/a166/ballsandy/Computer%20related/blink.png[/IMG] OH is +5v that comes in from a previously buffered signal. It passes through this circuit which makes the line blink. On its own it will work with an LED but I'm using a tiny Grain of Rice bulb which it refuses to play nice with. So say we buffer the BLINK output with a 2N3904, collector powered from the same OH signal as we are trying to avoid using external voltage rails. The result is it works, but the emitter output form the 3904 is about 2,5v and the bulb simply too dim to be useable in anything but the dark. We can up the 3904's collector voltage to 12V (by adding a connection to an external rail, which isn't what I really want...) but this has no effect on the voltage being sent to the bulb. I simply cannot make the buffer switch the bulb on and off in accordance to the BLINK input without a massive voltage drop. We tried to sub in a 3906 and just use +12 on the collector and a 220 ohm resistor on the emitter so we don't kill the 6v bulb but a new problem arose. If the blink OH input was 0v the light still comes on and goes into a weird dim/bright blink even though the blinker is not being powered at all. It does blink fine and bright though when OH does have +5 but that's irrellevant if we can't keep the lamp off. Me, Ddrl66 and Chryseus are all frustrated at this point. All we want is to make a fucking lamp blink as bright as it can.

Keep in mind that 3904/3906s are BJTs thus current controlled transistors. Currently the massive voltage drop is due to the BJT buffering the BLINK output not receiving enough current on it's base pin so that it can amplify the current and thus drive the grain of wheat lamp bright enough. To be honest, the 3904/3906s are really shitty BJTs ([URL="http://www.robotroom.com/BipolarHBridge4.html"]This is a good evaluation why[/URL]). If you tried lowering R5, you'd be drawing too much current on the buffer and thus the multivibrator would come off-kilter in terms of duty cycle (As not enough current is flowing to charge C1). Thus you have a few options to work with: 1) Utilize a higher gain transistor (The 3904's have an HFE/Beta ~ 50 at a 100mA Ic load, assuming that's the current where the grain of wheat lamp is brightest) while increasing R5 to somewhere in the tens or hundreds of kilohms. I'd recommend the 2N2222A or 2N4401 any day or even better any Zetex NPN transistor (like the [URL="http://www.diodes.com/datasheets/ZTX1049A.pdf"]ZTX1049[/URL]). That way you can provide alot more I_c current for a smaller base current. 2) Make a [URL="http://electronics.stackexchange.com/questions/66940/2n3904-darlington-pair"]darlington with the 3904s[/URL] you already have and increase R5. 3) Use a MOSFET (Like a boss), and forget about R5 entirely. As MOSFETs are voltage controlled and require no current at their gates to control current flow. 2N7000s are fairly common n-channel mosfets, but depending upon the current requirements of the Grain-O-Wheat, you may want to try a higher I_d current MOSFET like the IRF840, MTP3055E.

The problem was due to OH going high impedance rather than pulling low, thus it was powered through the output PNP (not shown), R5 and R4 so it continued to operate. Simple fix: [t]http://i.imgur.com/3gJbaG3.png[/t] Woops forgot a resistor. [t]http://i.imgur.com/EiVh4Bc.png[/t]

[media]http://www.youtube.com/watch?v=k5YnOfPl_sA[/media] Most useful Make: I've seen in a while.

Connector question because I'm kind of clueless when it comes to the intricacies of wiring. My 3D printer currently uses Molex connectors to connect a few different things on the extruder head, which I pull off of the printer fairly frequently for a variety of reasons, and I potentially want to add new extruders and be able to swap them out quickly. Getting Molex pins onto the wires (I bought a crimper for it and everything) is basically the biggest pain in the ass of all time, so I want to ditch the Molex connectors entirely, preferably for something much easier to get onto the wires. Ive heard [url=http://www.amazon.com/gp/product/B00MJVKKKI/ref=oh_aui_detailpage_o00_s00?ie=UTF8&psc=1]JST plugs[/url] work well. Would the wire just slip right into these things like a breadboard or would I need to solder them in there?

[QUOTE=Timebomb575;47424619]Connector question because I'm kind of clueless when it comes to the intricacies of wiring. My 3D printer currently uses Molex connectors to connect a few different things on the extruder head, which I pull off of the printer fairly frequently for a variety of reasons, and I potentially want to add new extruders and be able to swap them out quickly. Getting Molex pins onto the wires (I bought a crimper for it and everything) is basically the biggest pain in the ass of all time, so I want to ditch the Molex connectors entirely, preferably for something much easier to get onto the wires. Ive heard [url=http://www.amazon.com/gp/product/B00MJVKKKI/ref=oh_aui_detailpage_o00_s00?ie=UTF8&psc=1]JST plugs[/url] work well. Would the wire just slip right into these things like a breadboard or would I need to solder them in there?[/QUOTE] [IMG]http://www.hobbyking.com/hobbycity/store/catalog/AM-1024B.jpg[/IMG] Crimp job?

[QUOTE=Leestons;47424640][IMG]http://www.hobbyking.com/hobbycity/store/catalog/AM-1024B.jpg[/IMG] Crimp job?[/QUOTE] [I]FUCK[/I] are there any kind of slip connectors or something that doesnt require attaching crimp pins or soldering?

[QUOTE=Timebomb575;47424670][I]FUCK[/I] are there any kind of slip connectors or something that doesnt require attaching crimp pins or soldering?[/QUOTE] There are but they're prone to un-slipping and generally falling apart. There's a reason that crimping and soldering are the two most common connection methods.

Change everything to RG-58 coax and install BNC connectors :v: Don't do that for the heating elements though, that's a hazard.

[QUOTE=Timebomb575;47424670][I]FUCK[/I] are there any kind of slip connectors or something that doesnt require attaching crimp pins or soldering?[/QUOTE] There are, but they're [I]SO MUCH[/I] shittier than soldering or crimping. Man up an befriend either your crimper or a soldering iron.