Electrical Engineering V3 - Developers

[QUOTE=Fourier;48423022]I copy pasted this picture from the internet. Power supply is wrong yes. Maybe that R2 10 Ohm resistor needs to go to base of Q1 instead? Thanks for the tips[/QUOTE] If R2 is in series with the base Q1 will never turn on since the base would be at ground, I assume your LED strip is designed for 12V operation in which case R2 and Q1 isn't needed.

[QUOTE=Fourier;48423022]I copy pasted this picture from the internet. Power supply is wrong yes. Maybe that R2 10 Ohm resistor needs to go to base of Q1 instead? Thanks for the tips[/QUOTE] Something along the line of this would be better: [t]http://www.element14.com/community/servlet/JiveServlet/showImage/2-113322-196582/LED_drive1.jpg[/t] Then the IRF-530/IRF-540's gate will be fully saturated and thus minimal heat will be dissipated from it, not to mention no voltage drop. Ignore R1 to R3, they're part of the LED strips. The resistor value of R4 will need some tweaking, and a drain resistor between Q2's gate and GND would also be a good idea. Maybe even also something like a 100ohm resistor between Q2's gate and the microcontrollers I/O port. You'll have to factor in that the signal from your microcontroller is inverted with this circuit though.

Inverted signal is least of the problem, it can be solved with software. Ok, when weekend comes I will try to build mess with your scheme, van-man

[QUOTE=Fourier;48422778]I need to drive one big LED strip, 12V, 100Watt (7 meters long LED strip). 100Hz is a bad, I'd notice flickering. 200Hz, is that plausible? Hmm, I will go with this circuit -img- Power efficiency is not such a problem, since It will be running only about 30 minutes a day.[/QUOTE] I doubt you'd see flickering at 100hz considering the number of countries that use 50hz power systems. Lights on those systems flicker at 100hz as well (since power is the square of voltage over resistance, and a sin voltage squared doubles the frequency among other things.)

[QUOTE=Dolton;48423259]I doubt you'd see flickering at 100hz considering the number of countries that use 50hz power systems. Lights on those systems flicker at 100hz as well (since power is the square of voltage over resistance, and a sin voltage squared doubles the frequency among other things.)[/QUOTE] My country uses 60Hz and I notice flickering. (neon lamps, CFLs). But you are correct, maybe at ~120Hz I wouldn't notice it anymore.

Hi! I'm Mark Erpidos from [url]http://www.cathybarryadultstore.com/[/url]. I'm here to announce that our company has reached a break-through in dildo manufacturing through one weird trick! We take living tissue and wrap it around our special penis shafts which allows our dicks to be smooth yet natural. Product Pics: [IMG]http://www.cathybarryadultstore.com/shop/images/bendablewinky.jpg[/IMG] [IMG]http://i01.i.aliimg.com/img/pb/997/095/497/497095997_246.jpg[/IMG] [IMG]https://encrypted-tbn1.gstatic.com/images?q=tbn:ANd9GcQxCKE7Jv0QwJYY6opWwl8WjrtYb2DqJaH9ElPrxugaQP-vfBAz[/IMG] How to use it: [IMG]http://img-l3.xvideos.com/videos/thumbslll/09/8f/a7/098fa753fd1b17300b224ef76194440a/098fa753fd1b17300b224ef76194440a.22.jpg[/IMG] Sponsors: [IMG]http://search-creatives.s3.amazonaws.com/15/db/80/15db80394d1b91d7a65ac2fb9c8cb910.jpg[/IMG] [IMG]http://i2.kym-cdn.com/photos/images/facebook/000/633/851/bfa.jpg[/IMG] [IMG]https://encrypted-tbn3.gstatic.com/images?q=tbn:ANd9GcSLne2lWvL9SGhF93nrcLCeVChVYBWJrh6y4EAg0ikXW_hfg8eQ[/IMG]

So I'm glad I left my lab before refreshing the page. :tinfoil:

Got accepted into university to study electrical engineering. :toot: Super stoked, can't wait to learn all this jazz.

[QUOTE=nutcake;48437625]Got accepted into university to study electrical engineering. :toot: Super stoked, can't wait to learn all this jazz.[/QUOTE] Better start brushing up on your calculus!

[QUOTE=chipset;48437808]Better start brushing up on your calculus![/QUOTE] nvm dropping out

[QUOTE=LoneWolf_Recon;48423586]So I'm glad I left my lab before refreshing the page. :tinfoil:[/QUOTE] [sp]I bought all of them that is why they stopped advertising[/sp] [editline]11th August 2015[/editline] [QUOTE=nutcake;48438020]nvm dropping out[/QUOTE] calculus is easy once you study it for 2 years

So over a majority of the summer (and still in progress) I and my colleague have been working on a project for our professor in our Uni's EECS dept. It applies for course credit as Electronic Instrument Design. The goal of this project is to build an Impedance Analyzer (DC to 2MHz) using only $1k of parts as most impedance analyzers of the same capability cost atleast ten times that. Through the magic of software defined radio USRPs and GNURadio (Using this [URL="http://www.ettus.com/product/details/USRPPKG"]Ettus Research USRP[/URL] in particular along with some LFTX/LFRX daughterboards to work within the frequency range specified), we've cobbled together an impedance analyzer here's the flowchart using an I/Q Quadrature algorithm (Where abs(Z_dut) = I^2 + Q^2): [IMG_THUMB]https://sites.google.com/site/lonewolfscircuits/sitefiles/System Flowchart.png[/IMG_THUMB] My main part was the hardware, thus I designed the V-I (Sinusoidal voltage-to-current source circuit to push current onto the DUT to develop a voltage across it). Here's the [URL="https://sites.google.com/site/lonewolfscircuits/sitefiles/Simulation_MC34081.asc"]Spice simulations[/URL], [URL="https://sites.google.com/site/lonewolfscircuits/sitefiles/Pic_0728_699.jpg"]some[/URL] [URL="https://sites.google.com/site/lonewolfscircuits/sitefiles/Pic_0728_700.jpg"]shots[/URL] [URL="https://sites.google.com/site/lonewolfscircuits/sitefiles/Pic_0728_701.jpg"]at[/URL] [URL="https://sites.google.com/site/lonewolfscircuits/sitefiles/Pic_0728_698.jpg"]the[/URL] prototype board I've made with some [URL="https://sites.google.com/site/lonewolfscircuits/sitefiles/Traces.zip"]O-scope traces (ZIP)[/URL] to match the Spice simulations. I built the V-I to be configurable with different RVALs as the formula goes i_out = (R5 / R1) * (v_in / R4). I had a bit of trouble for a while with previous tests as I had developed a parasitic RC pole which hampered the V-I's linearity across the frequency range. (Previously I had all of the RVALs in simulation set to 500k, thus a ~20kHz pole developed, now with 1k RVALs the pole has jumped to ~10MHz). I had to also deal with parasitic capacitance of the O-scope probe/LFTX so I just buffered the shit outta them. There's also a 32MHz LPF to filter out any system noise coming from the USRP (Which operates at 64MHz). More to come/Links to change/End-of-Post. :excited:

Okay, I got an idea. Make a giant Core frame. Technically the ferrite beads in core memory can be scaled up quite a bit. You just need more power to do the same job. I'm curious what a frame would look like if you scaled up to a 4' x 4' frame (16 x 16 bits) with 1/2" rings. This sounds like it would be an amazing art project. The X lines could be red, the Y lines could be green, the Sense lines could be blue, the rings would be yellow and the frame itself would be brown. [editline]asda[/editline] Ah fuck me. [t]https://40.media.tumblr.com/e3bf3ce91bcf516300d18ecff0975738/tumblr_mmz9lu45M21qzw5x3o1_500.jpg[/t] Someone is already doing it. :(

Throw a mattress over that and then you really have a memory foam mattress :weeb:

Just ordered my transmitter PCB's, had a bit of trouble cramming all of the components in the 5x5cm limit, but it's done! [thumb]http://i.imgur.com/w9gHq8q.png[/thumb] The circles are a visual aid for myself.

[QUOTE=pentium;48442486]Okay, I got an idea. Make a giant Core frame. Technically the ferrite beads in core memory can be scaled up quite a bit. You just need more power to do the same job. I'm curious what a frame would look like if you scaled up to a 4' x 4' frame (16 x 16 bits) with 1/2" rings. This sounds like it would be an amazing art project. The X lines could be red, the Y lines could be green, the Sense lines could be blue, the rings would be yellow and the frame itself would be brown. [editline]asda[/editline] Ah fuck me. [t]https://40.media.tumblr.com/e3bf3ce91bcf516300d18ecff0975738/tumblr_mmz9lu45M21qzw5x3o1_500.jpg[/t] Someone is already doing it. :([/QUOTE] Make yours prettier, make them regret their decision of not being more colourful.

[QUOTE=helifreak;48448752]Make yours prettier, make them regret their decision of not being more colourful.[/QUOTE] LEDs that pulse the state of a particular core when sensed?

So as you guys probably remember I have that large collection of relays. Well, I want to use one of them to control power to my desk. I've got a momentary rocker switch that goes up and down and I'd like to make up toggle the relay on and down reset it to off. Now normally this would be easy, use a NO switch for the on and a NC switch to cut the power to the relay. Problem is, the circuit for my switch looks like this: [img]https://i.gyazo.com/928ed66254a0b1aa554ecc3d668b58d6.png[/img] So there's no normally closed option. What are other ways to accomplish this without just buying a latching relay? The other complication is that the coil voltage on these things is also 120VAC so I can't use any logic on it.

[QUOTE=LoneWolf_Recon;48449344]LEDs that pulse the state of a particular core when sensed?[/QUOTE] I'm still doing the math on how the hell you even drive it. You will need a pretty serious voltage considering you are scaling upwards of something like 300x. That and I got two other framing projects on the go. The Pachinko cabinet and the back for an old sign. I'll come back to this perhaps in a few months.

[t]https://dl.pushbulletusercontent.com/OzLfK9rYB8BEYbuWoKyRIe5OacmJG9v4/IMG_20150814_010914.jpg[/t] Just some 1am high frequency inductor winding.

[QUOTE=papkee;48453917]So as you guys probably remember I have that large collection of relays. Well, I want to use one of them to control power to my desk. I've got a momentary rocker switch that goes up and down and I'd like to make up toggle the relay on and down reset it to off. Now normally this would be easy, use a NO switch for the on and a NC switch to cut the power to the relay. Problem is, the circuit for my switch looks like this: [img]https://i.gyazo.com/928ed66254a0b1aa554ecc3d668b58d6.png[/img] So there's no normally closed option. What are other ways to accomplish this without just buying a latching relay? The other complication is that the coil voltage on these things is also 120VAC so I can't use any logic on it.[/QUOTE] You can use logic just fine, just get a solid-state relay or a lower-voltage relay.

[QUOTE=Zero-Point;48456988]You can use logic just fine, just get a solid-state relay or a lower-voltage relay.[/QUOTE] Nah the point was to use the relays I already have.

So I have a Casio SA-2 that I want to convert into a MIDI keyboard. I figured I would try just soldering wires to the pins on the IC that the keys are wired up to so I can plug them into an Arduino, and I kind of ended up with this disaster [img]https://i.imgur.com/CzUSMdV.jpg[/img] The wires snap off if I look at them funny and I don't even want to think about trying to put the case back together. There really has to be a better way to do this (thinner, more flexible wire maybe? Even when closed, the switch resistance is like 3kΩ so I can probably get away with that). I'm pretty sure the keys make contact with the other side of the circuit board, so I'm reluctant to try to shove those wires into the other side of the IC. Is there at least something I can do to cram all of these wires back into the case that has about 1cm of clearance without ruining everything, and with minimal risk of anything happening when I actually start mashing buttons on the keyboard and moving around with it?

Hey guys. Not really a frequenter here, but I'm thinking I might need a hand with an upcoming project (instrument gauge for a motorcycle). I'm looking for some sort of sensor for a tachometer on a bike that never had one; I was thinking of doing a Hall effect sensor on the camshaft lobe and doing the math after, but that seems like a dirty and error prone method. Instead, would it be possible to use a Hall sensor capable of fast updates on the ignition coil/spark plug to pick up each pulse? If I haven't already made it clear I really know very little about circuitry and all this shenanigannery. If you guys can help me, that'd be extremely cool.

[QUOTE=halofreak472;48463132]So I have a Casio SA-2 that I want to convert into a MIDI keyboard. I figured I would try just soldering wires to the pins on the IC that the keys are wired up to so I can plug them into an Arduino, and I kind of ended up with this disaster The wires snap off if I look at them funny and I don't even want to think about trying to put the case back together. There really has to be a better way to do this (thinner, more flexible wire maybe? Even when closed, the switch resistance is like 3kΩ so I can probably get away with that). I'm pretty sure the keys make contact with the other side of the circuit board, so I'm reluctant to try to shove those wires into the other side of the IC. Is there at least something I can do to cram all of these wires back into the case that has about 1cm of clearance without ruining everything, and with minimal risk of anything happening when I actually start mashing buttons on the keyboard and moving around with it?[/QUOTE] Could try desoldering the chip, sticking on a DIP socket and plugging in both the chip and associated wires. That might be pushing the 1cm budget, but that's what I can think of.. [QUOTE=ramirez!;48463786]Hey guys. Not really a frequenter here, but I'm thinking I might need a hand with an upcoming project (instrument gauge for a motorcycle). I'm looking for some sort of sensor for a tachometer on a bike that never had one; I was thinking of doing a Hall effect sensor on the camshaft lobe and doing the math after, but that seems like a dirty and error prone method. Instead, would it be possible to use a Hall sensor capable of fast updates on the ignition coil/spark plug to pick up each pulse? If I haven't already made it clear I really know very little about circuitry and all this shenanigannery. If you guys can help me, that'd be extremely cool.[/QUOTE] Electrically it'd be more reliable to read off each pulse from the ignition coil instead of the tachometer yes. I assume motorcycles use 12V as their main battery as well (My automotive knowledge is pretty bad besides changing tires, filters and fluids :v ) A hall effect would [URL="http://www.scienceshareware.com/how-to-measure-AC-DC-current-with-a-hall-effect-clamp-.htm"]definitely work[/URL] and for each pulse I'd probably see a spike in the output voltage (As the current is being dumped from the condenser to the ignition coil): [IMG]http://www.w6rec.com/duane/bmw/points/BMWPointsDiagram.jpg[/IMG] Thus clamp on that sensor around the low voltage side, and you'd probably be able to get away with measuring the output voltage with an Arduino's analog input. Doing some math and outputing the RPMs to an LCD or whatever display you'd like to use.

[QUOTE=halofreak472;48463132] The wires snap off if I look at them funny and I don't even want to think about trying to put the case back together. There really has to be a better way to do this (thinner, more flexible wire maybe? Even when closed, the switch resistance is like 3kΩ so I can probably get away with that). I'm pretty sure the keys make contact with the other side of the circuit board, so I'm reluctant to try to shove those wires into the other side of the IC. Is there at least something I can do to cram all of these wires back into the case that has about 1cm of clearance without ruining everything, and with minimal risk of anything happening when I actually start mashing buttons on the keyboard and moving around with it?[/QUOTE] How exactly are you soldering those wires on, if you're putting solder on the tip and then on to the joint you'll never get a good result since the flux will burn off the tip long before it can be used, you need to have the wire held in place then feed the solder on or coat the DIP lead and wire with liquid flux before applying the solder.

This thread has just reminded me of something I'm really shit at, soldering wires onto things (at least stranded wire), I can solder wires together, if it has a hole for me to put the wire through (like some motors, switches etc) I'm fine, but if I need to solder a wire onto a flat pad I'm fucked. The wire just tends to sit on top of the solder.

[QUOTE=LoneWolf_Recon;48464605] Electrically it'd be more reliable to read off each pulse from the ignition coil instead of the tachometer yes. I assume motorcycles use 12V as their main battery as well (My automotive knowledge is pretty bad besides changing tires, filters and fluids :v ) A hall effect would [URL="http://www.scienceshareware.com/how-to-measure-AC-DC-current-with-a-hall-effect-clamp-.htm"]definitely work[/URL] and for each pulse I'd probably see a spike in the output voltage (As the current is being dumped from the condenser to the ignition coil): [t]http://www.w6rec.com/duane/bmw/points/BMWPointsDiagram.jpg[/t] Thus clamp on that sensor around the low voltage side, and you'd probably be able to get away with measuring the output voltage with an Arduino's analog input. Doing some math and outputing the RPMs to an LCD or whatever display you'd like to use.[/QUOTE] That's more or less what I was thinking. I imagine there would be a lot of signal 'noise' from the ignition coil, or whatever the correct term would be, so it would have to be cleaned up somehow. I'm sure I can find a filter or something to do that. I hate to ask, but could you help me find a suitable Hall Effect sensor? I found one on Adafruit which seems like it'll do the trick, but I need one to pick up on the ignition coil wave and another one that can hover near the axle and pick on a ring of six or so magnets to measure speed. Both would have to be able to survive temperature ranges of probably 0-160 Fahrenheit, as well as being generally durable so they don't fail after a month.

[QUOTE=Leestons;48466676]This thread has just reminded me of something I'm really shit at, soldering wires onto things (at least stranded wire), I can solder wires together, if it has a hole for me to put the wire through (like some motors, switches etc) I'm fine, but if I need to solder a wire onto a flat pad I'm fucked. The wire just tends to sit on top of the solder.[/QUOTE] Tin the wire and pad, then put some flux on the pad and hold the wire down on the pad. Put some solder on your iron then press it to the pad and wire. That should give a nice solder joint, if you can't hold the wire close enough to the pad use a pair of tweezers.

[QUOTE=ramirez!;48466714]That's more or less what I was thinking. I imagine there would be a lot of signal 'noise' from the ignition coil, or whatever the correct term would be, so it would have to be cleaned up somehow. I'm sure I can find a filter or something to do that. I hate to ask, but could you help me find a suitable Hall Effect sensor? I found one on Adafruit which seems like it'll do the trick, but I need one to pick up on the ignition coil wave and another one that can hover near the axle and pick on a ring of six or so magnets to measure speed. Both would have to be able to survive temperature ranges of probably 0-160 Fahrenheit, as well as being generally durable so they don't fail after a month.[/QUOTE] The sensors I've found on Adafruit/Sparkfun honestly wouldn't live up to the abuse of being exposed like that on a motorcycle. Especially not the temperatures it could feel being that close to the engine. I'd recommend [URL="http://www.mouser.com/ProductDetail/Honeywell/CSLT6B100/?qs=sGAEpiMZZMsPYxdnOoPGenym%2fAi0vPAnO%2fHHFcU07E0%3d"]this[/URL] ([URL="http://sensing.honeywell.com/honeywell-sensing-cslt-series-product-sheet-005862-2-en.pdf?name=CSLT6B100"]datasheet[/URL]), you shouldn't need a split voltage supply (as its DC current) but its worth and it's generally cheaper and more robust than the other sensors on Adafruit/Sparkfun.