You really don't need that many pins, assuming the joysticks are regular analog potentiometers you only need one pin from each axis, other ones get connected to +5/+3.3 and ground, so per joystick you only need two analog pins and one digital for the switch giving a total pin count of 6, something like an arduino pro micro would do the job or you could make an even smaller custom solution.
Hey guys, i was really shy about posting this here but here we go.
https://imgur.com/luuvbkd
https://imgur.com/ewVznqP
https://imgur.com/mqWOJ2l
https://imgur.com/5hvdSw3
https://imgur.com/e25lXY6
https://imgur.com/1xIPXHo
Before you ask this is not an amp, it is actually an modular regulator with some enhancements... I wanted a high voltage DC lab power supply (150-300 volts) for experiments with vacuum tubes but could not find where to get one... So i came up with my own, by the way this isn't the whole thing this, this part right here is the main chassis...
https://imgur.com/HpZlnlK
It has two more regulators one is positive other is negative both of them are identical, both can handle around 15W the main one which i showed above can handle 60W Or 66W if you relay want to bend the tree... Main one also has features such as fast voltage switching, with relays or computer mode which is another experiment i want to try, basically to try to control regulator output with computer...
Here is the schematic of main regulator it is a little bit outdated from real thing like switches, which i thought i could fit on that piece of metal form DVD-ROM. Schematic is also full of acronyms which i will try to explain in new schematic version...
https://imgur.com/yr4tmGO
Note i still have to draw schematic of main chassis. Also i didn't put relay diodes yet because i didn't have that many on hand.
Hopefully i will complete this thing in few weeks. Also you probably have so many questions so please ask them, i'm so lonely...
That's a really neat design, from looking at it I'd say you're using the voltage regulator tubes to provide a stable cathode voltage for the 6AU6 which pulls down the grids of the 6P3S output tubes to give adjustable voltage regulation, I'm not entirely sure what the arrangement of relays and lamps is but I'd assume it's so you can get the entire 150-300V range.
I definitely approve of the construction with those old soviet style LEDs and the big connector, would not have been nearly half as nice if you used a PCB and all modern stuff.
Thanks, and yes you are mostly correct about operation except this thing doesn't use both tubes at the same time, it uses one for normal range (about 190-315 volts) and other for even lower range (about 140-230 volts). Then there is operation point adjustment, which is needed to adjust gain (only when regulator tube is switched) basically to synchronize no load output voltage with fully loaded output voltage so the regulator won't drift when load is removed ( pro tip it still drifts but only 0.1 volt in normal range and in lower range it drifts more, don't remember how much, i will have to remeasure that... and two other regulators on main chassis will drift 1.2v to 2v as i remember ). And then there is output adjustment which only adjusts the output of course, two of these controls (operation point and output) can be switched like regulator tube via relay, to adjust adjust quickly. Also i made a chart of what is what on the front...
https://imgur.com/tkFI3Op
By the way those are actually not led's those are really small light bulbs operate on 28 volts i run them at 12 volts...
https://imgur.com/lTh2gAR
https://imgur.com/1QSg8FQ
Also soviet electronics are really colorful i like using them in all of my projects if i can.
Can someone point me towards some good books for learning embedded programming? I've already read this.
https://www.amazon.com/Building-Embedded-Systems-Programmable-Hardware/dp/148421918X
I'm kind of looking for something that I can read while doing. I was thinking this one maybe?
https://www.amazon.com/Embedded-Cortex-M-Microcontrollers-Assembly-Language/dp/0982692668/
Soldering question: is a 0.5mm tip way too big to do these 3535 Neopixels, or do I just suck? Considering they have virtually no metal on the sides, they seem really difficult to solder without paste and hot air.
If it's the iron and not me, since there's no other tip sizes for it, I'm thinking the usual Hakko 888D and a 0.2mm tip might be a much better idea.
0.5mm is fine, you don't actually need to make full contact to get a good joint with surface mount, make sure that you're using a good quality flux and good leaded solder, if the pads are not tinned already do that before soldering, if it still isn't flowing correctly you may need more power particularly if there are large ground planes nearby.
Thanks. I am using lead solder but I've been relying solely on the rosin core until now, I guess I should get a flux pen then. I also tried putting dabs of solder onto all four pads before placing the LED but of course then it ends up elevated off the board, and lopsided too. I'll try wicking it away a little first.
It's really tricky (for me at least) not to melt the plastic; these things really don't like the heat. The 5050s also melt kinda easily but I managed not to pop any of those.
Smaller tips are more difficult to use, because the surface tension of the solder will make it retreat away from the tip. Use a chisel tip.
Flux pens are nice, but I would also recommend getting some liquid flux in a bottle with a brush applicator. Really useful for soldering wires.
Okay so I'm a computer engineering student at my university. Well, not for this semester as I unfortunately have to wait a little, but I was supposed to be taking Circuits I. Now I've only got some coding experience in C++ from one of my courses. Intro to Programming/C++, whatever. In that course lab we worked with an arduino board. The Apollo board from Ascension Engineering. My board is slightly different where the potentiometer knobs are horizontal. I've been pretty bummed not being able to continue this semester, and don't know how to spend my time besides working and video games. I'm always interested in so many hobbies, but I seem to have an opinion that every hobby is expensive and I don't make that much myself from a simple serving job. I'm tired of serving, honestly, and am actually working towards working for the university full time and finish my degree for, essentially, free. Salary job meaning I can budget my money better to put towards hobbies.
Anyways, I will be going to a musical festival with a bunch of friends and my roommate ordered an LED totem. This one to be exact. When he ordered it, he asked me if I think I could program some cool stuff on the LEDs. Once it arrived, I was curious how they did it. Looking at it, it looks really simple. I figured with my arduino board and a foundation for coding, sure maybe I could check it out and do some research about it. After googling around I came across a post that really shot a fire in me with inspiration.
I like LEDs, so I put 540 of them on a ball to bring to music festivals. Presenting SOL CRUSHER, the culmination of 6 months of my free time.
He provides a ton of photos of basically the entire project duration and details his thoughts about decisions.
Instantly, "Yes I want that, I want to do this shit!" Everything from the knowledge he possessed from designing that project and of course I know there is much more in the world of EE than just LEDs. Then I remember that there must be a thread on Facepunch. I've been a member here for years before actually but lurked more around SH, Polidicks, and GGD. Never kept up with the forum evolution project. I've read the OP several times and I've come to the conclusion that what's stopping me from teaching myself what I need to know towards my career even without school? And, actually, get a head start so when I go back to my uni, I'll be ready for those classes. Now I've got a couple questions.
Where do I start? What should I read first? I opened up the MIT link and it looks like that may be a good start for me? Made me also think is there some sort of "lab" I can perform here at home while I'm learning?
Speaking of lab, what materials do I need to invest in to get a good start into it? Breadboards? Solder Iron?
I had more questions but forgot them overnight, but this seems like a good start for now.
If you're already fairly proficient at coding, and wanting to get into electronics, then a good first step might be to read as much as you can about voltage, current, resistance, capacitance, inductance, etc., and how they relate to each other. From there it should be rudimentary to take off in any direction you might need, in order to better understand how to apply your coding knowledge with what you learn from your ventures into electronics (microcontrollers and SBCs/embedded systems seems like a natural transition).
Hmm honestly I don't really think you need to work for any 'certificates' besides your degree. Your university degree alone should say enough about you when you finish. Like Zero-Point said getting a good foundation in the basics of electricity is a must, and that's not hard to do with online resources. Perhaps start with a higher level circuits understanding, and work your way through getting through at least some of the basic physics behind it.
I know you're a computer engineering student, but I also encourage you to explore other subjects of EE (I consider CE to be a subset of EE) while you're at university, as my area of passion within EE is something I didn't even know existed before coming to university!
Other areas you may consider just exploring if at most:
Photonics, Electronics, and Nanodevices (Learning how electronic devices are designed, Materials Science, The physics behind it all the way down to the electron level)
Systems (Notoriously difficult, but my personal favorite , the study of different systems and signals: i.e. Communications, Digital Signal Processing, Control Systems, Robotics, Computer Vision)
ASIC synthesis is a fickle mistress
always @ (posedge clk or negedge rst_n) begin
if (!rst_n) begin
// It's best to ask nicely:
// synthesis please_on
for (i = 0; i < N_REGS; i = i + 1) begin
mem[i] <= {W_DATA{1'b0}};
end
// synthesis please_off
https://files.facepunch.com/forum/upload/108652/7842f77a-d7f1-44be-982a-721a538b8711/20180321_003153.jpg
It's a ghetto switch circuit i taped together from components i found lying around. I tend to use car batteries as power supplies and this comes in handy. You can turn it off with the red button and turn it on with the yellow button. When it's on the green LED lights up. If you accidentally short circuit the output it automatically turns off and you can turn it on again.
Here's the code you can import in this simulator Circuit Simulator Applet
$ 1 0.000005 10.20027730826997 50 5 43
g 144 320 144 352 0
f 352 192 352 144 9 5 0.02
f 400 272 352 272 0 5 0.02
s 224 320 224 256 0 1 true
R 160 144 112 144 0 0 40 12 0 0 0.5
r 288 144 288 256 0 13000
w 288 256 352 256 0
w 352 256 352 192 0
w 288 144 336 144 0
w 400 272 400 144 0
w 368 144 400 144 0
w 400 144 480 144 0
w 224 144 288 144 0
w 224 256 288 256 0
w 352 288 352 320 0
s 224 144 224 256 0 1 true
g 480 240 480 304 0
r 144 320 224 320 0 750
r 224 144 160 144 0 0.1
r 480 144 480 240 0 1000
s 528 144 528 240 0 1 true
w 480 240 528 240 0
w 480 144 528 144 0
w 224 320 352 320 0
o 7 64 0 4099 20 0.00009765625 0 2 7 3
o 9 64 0 4099 20 0.00009765625 1 2 9 3
Edit:
I can't find the proper place to post this, where is the new EE thread?
That's awesome. Kinda want to make one myself even though I have no use for it.
Care to explain how it works to an idiot like me?
(Top transistor is P type, button is N type)
http://carp.cf/$/KntYGk.png
In the off state, the P mosfet is turned off because it has +12V at the gate, the drain (and therefore the the N mosfet gate) is pulled to the ground. No current flows and nothing happens. The P mosfet gate and N mosfet drain are kept pulled high with the 13k resistor.
http://carp.cf/$/3v0Ish.png
When you press the ON switch, the P mosfet gate gets pulled to ground and therefore it turns on. The drain is now at +12V and the N mosfet also turns on and keeps the P mosfet gate pulled to the ground.
http://carp.cf/$/5aw8FN.png
If a short circuit happens or too much current gets pulled. The voltage at the N mosfet gate will drop below a certain value and turn it off. The gate on the P mosfet will get pulled high over the 13k resistor and turn it off. Similar thing happens when you press the OFF button, except the P mosfet gets turned off first which in turn turns off the N mosfet.
Thank you! I couldn't get the code to work in the simulator. I was just getting a blank circuit.
It's such a weird thing, looking for L298n motor drivers on eBay, and seeing all these boards that are specifically labeled as L298n drivers, but the numbers have been completely sanded off in the pictures.
Like c'mon man, either it's a genuine L298n and scratching the numbers off is pointless because the pin-outs and characteristics are literally everywhere, or you're being sheisty.
Hi, I know almost nothing about electronics. I want a microcontroller that can control 2,640 (44 feet of 60LED/m) SMD 5730 LEDs (12v). Can you have a low-rated Arduino control a high amperage strip of LEDs safely via a transistor or something without frying your arduino? Also, what should my power supply be rated for for the LEDs, do you think?
You should check out the data sheet for the LED strip, it should mention the power consumption somewhere. You can control it with a single power MOSFET and stack multiple of them in parallel depending on a single MOSFET power rating and LED strip power rating (2 10A MOSFETS for 20A of LED power draw). IRFZ44Ns are ones that i use for basically everything and they can be toggled just fine with the arduino and can pull a ridiculous amount of power when you tape a passive radiator on it.
How many color channels do you have on your strip and what kind of strip is it? Common anode or cathode?
If a single LED is rated at X Watts, can I assume that 60 of those LEDs will consume 60 * X Watts? Or does it not work that way
The LEDs will be RGB, and I don't know what the difference is between those two (Anode/Cathode). As is turns out, I can't seems to find 5630/5730 SMDs in RGB strips, so I'm settling with 5050s instead.
Here's what I had in mind:
https://www.aliexpress.com/item/Not-Waterproof-DC12V-5M-RGB-LED-Strip-Light-SMD-5050-5630-2835-3528-Led-string-Ribbon/32704479405.html?spm=2114.search0104.3.8.5b2c2fe4Sq5OOF&ws_ab_test=searchweb0_0,searchweb201602_4_10152_10151_10065_10344_10130_10068_10324_10547_10342_10325_10546_10343_10340_10548_10341_10545_10696_10084_10083_10618_10307_10059_10534_100031_10103_441_10624_10623_10622_10621_10620,searchweb201603_25,ppcSwitch_5&algo_expid=dd9eac63-8bd1-45e4-8356-688446dc3786-1&algo_pvid=dd9eac63-8bd1-45e4-8356-688446dc3786&transAbTest=ae803_4&priceBeautifyAB=0
It says 4 W/m, which leads me to the following calculation for my powersupply:
44 feet = 13.4112m
13.4112m * 4W/M = 53.6448W
P = IV -> 53.6448 = I * 12 -> I = 4.4704
which means I'll need a Power supply rated for 12v @ ~5A
Does this sound about right?
Cathode is negative (-), Anode is positive (+). When you have a common cathode RGB led for instance, this means all the cathodes are tied together. So you'll have the single cathode pin always connected to ground, and three anode pins that you can use to actually drive the three different colors individually.
Note that this usually isn't relevant for LED strips that are individually addressable, since these will use a data pin that lets you send in data for each LED
Pretty much, you should be able to power all of them using 3 IRFZ44N's (one for each channel, PWM them to dim them and mix colors), at full brightness (50W) each individual transistor would only heat up about 10°C above ambient temperature according to my calculations. You won't even need a heatsink.
I think it's common anode. The three pins go to ground, iirc
http://carp.cf/$/Rdlnmx.png
Excuse the shitty schematic.
So, going back to my original question, can I safely control the transistors using an arduino or raspberry pi, without frying it? Also do you think these transistors need a heatsink?
I am not a baby
An Arduino or Pi can only source < 30mA or so from their pins, and a Pi will probably do even less than that, so a transistor or other switching mechanism (relays wouldn't work so well for PWM ) is pretty much mandatory if you're going to switch something that uses 4W/m. Speaking of which, not sure if that 4W/m is for the RGB or just the white, as the "details" only lists the white colors. There could be a difference, but AliExpress is about as useful as most Hong Kong eBay sellers with providing details (took some detective work to find the mounting dimensions for the mounting brackets my gear-motors came with, as the batteries in my calipers are deader 'n shit ), so I'd look into finding out what the current draw for your specific strips are, then look for a MOSFET sized accordingly. (N-channel would be preferred here)
As for heat-sinks, might not be necessary, but it wouldn't hurt it, either.
Well RGB LED strips are just the same as White LED strips but they simply alternate Red Green and Blue, so it's still 60 LED/m. However, I don't know if each individual LED draws the same as a white LED (I'd imagine it would draw less).
I'm trying to prototype something for my new apartment where I can control the LED lights in a room using a node server (hence a raspberry pi) running on a local network (yay IoT) or maybe in the future a tocuhscreen panel
Sorry, you need to Log In to post a reply to this thread.
Electrical Engineering V3
