A list of puns related to "Current sources and sinks"
My wife and I just acquired my grandmothers house (yay) and it needs some work, nothing extremely major just little things here and there. The kitchen is full of shadows, the main light source is a ceiling fan dead center of the room and the secondary light is located on the ceiling above the sink behind a 10" tall piece of molding/facade/whatever it's called that connects the cabinets on either side of the sink. That light prodivdes some light for the sink but not much and does nothing to help the brightness on the countertops.
We would like to put up some LED strip lights along the bottom of the cabinets (24" strip on each side of the sink basically) and then maybe two of those same strips above the sink where the current light is now, that way the sink is well lit and the countertop has less shadows also.
My plan is to either get a screw in plug adapter for the current socket or replace the socket entirely with an outlet (ideal solution but a little more time & money). Within the next handful of years we plan on redoing the entire kitchen so I don't want to go too crazy with hard wiring stuff just yet so I'm leaning towards the socket adapter.
Help? Ideas, opinions, etc to think about? I'm planning on just using LED strips from Walmart or Lowe's and hopefully doing this later today.
I'm trying to build an 8x8x8 LED-cube that's controlled entirely via i2c for sng's. Run of the mill concept: rotate between one floor of the cube at a time and use persistence of vision to make it look as though all 512 LEDs can be lit at once even though there's only ever really 64 at once.
In order to control the layer selection, I figure I can just use 3 output bits of the MCP23008 and feed them into a MCP74HC238A which would decode the number from binary and turn on one of eight n-channel mosfets that sink current out from the indicated layer.
Sourcing power into the LED posts proves to be a different problem altogether. I'd love to use a constant-current sourcing shift-register for simplicity but I'm not really sure how to attack that from an i2c interface.
I think the TLC59116 is worth mentioning because it's similar to what I'm looking for with two key faults; speed is a concern so I want to avoid PWM drivers to minimize i2c writes needed for setting up a layer, and the device sinks current instead of sourcing it.
Also worth mentioning is the reason that I'm trying to source current into the posts and sink current out of layers instead of the other way around. Most cheap, large packs of LEDs seem to have the longer leg as the anode, which means that connecting the LED anodes to each other vertically is easier than doing so with the cathodes. I prefer to have my LED cubes shine upwards, so that means I gotta sink current out of the layers.
Hope that's not excessive detail. Does anyone know how I can find a chip combo that will help me along?
I'm in the process of designing a PCB which I'm directly driving 48 LEDs from the I/O of an ATMEGA128A, and I'm curious whether I should have the AVR sink or source the power to the LEDs. Intuition tells me that it'll be better practice to sink current with the AVR, as it seems like it would be more forgiving with accidental overcurrent, however the problem here is that it seems like programming errors make the chip more succeptible to overcurrent by accidentally turning on all the LEDs. I am aware of the max current on all I/O and plan to overcome this by implementing my own soft PWM which allows no more than the max allowable LEDs to be on at one time.
So... Sink or Source those LEDs?
I am in a quandary trying to find an IC or circuit that will work for this application and seeing if anyone has faced a similar situation. I have a 24V supply that goes out onto long cables to a current-mode sensor which sends it's signal back on a return lead. On a short I need to limit the current to about 10mA. We have used FET based in-line current limiters, but they introduce an issue when it comes to crosstalk from adjacent cables as they have relatively high-source impedance and can't sink EMI induced current back to the source without inducing a voltage across itself which is picked up by our differential amplifiers. I have been looking into power distribution switches, but have been unable to find one for such a low current limit. I have also looked at op-amps with adjustable current limits as they can sink this current with no issues, but the few available will kill my power budget as they use almost as much current as the sensor. Kind of lost right now...
TLDR: Anyone know of a product/circuit that can be set to 10mA current limit that will sink/source current simultaneously while maintaining excellent voltage regulation?
EDIT: Added screenshot The limiting goes between V2 and R2 and is sampled between out+ and out-. The current sources are to mimic the induced xtalk, but are not true models as putting the limiter after the R2/C1 node would cancel it differentially if this model was true. Unfortunately the FET limiter alters the sink current itself on the high rail, which still creates the differential voltage signal at out+
It's the only way Bungie will change things and start actually working, because teither the 600+ exployees they have in the company litearlly do not do absolutely nothing or they have a skeleton crew of 25 people working on this game nd the rest are doing nothing, which is the same.
I know there's like 20 threads saying the same, but this sub is full of.... special people who will still buy the next season. They're gonna start teasing stuff in the next twab, because tomorrow's twab is just gonna be like this guy said:
https://www.reddit.com/r/DestinyTheGame/comments/kwuov2/to_all_those_upset_with_stasis_content_and/
This is what I currently have in my laundry room. I'm not sure if this was a kit or something at some point in time, but it's made of PVC attached to copper lines on the hot and cold side.
Image here
I would like to just convert this to copper. I notice that there is a vertical piece of PVC that is capped in the current set up, does that serve a specific function? What would be the best way to make it out of copper only?
Any and all help is appreciated, thank you!
My question is, following the attached circuit (sorry for the precariousness), the boost will take the incoming 3.7V and provide a constant 12V output at whatever demanded current. Now, in the hypothetical example of having N number of LEDs with N resistors of different values, if I want to maintain a constant current so LED brightness is constant (neglecting manufacture mismatches), the most logical idea -IMO- would be to implement a current sink. In the example, a TLC5940 is used, which to maintain the constant current will vary the voltage, which would create a conflict with the operation of the boost.
Is that right?
https://preview.redd.it/1fvr7iwbd9n61.jpg?width=1280&format=pjpg&auto=webp&s=8bce415ceff24cbb127211709c1c16a9aceaea44
I'm using cadence/catia right now as a replacement for voicemeeter banana, but I cannot for the life of me figure out how to create another jack input/output. I need something where I can have 4 inputs. One for my mic, one for discord to output to, one for my web browser to output to, and one for my games to output to. I'd also like an output for my end, one to input to discort, and one to input to a stream. ty for any help.
Edit: After messing around a bit, this translates to 4 sinks and 3 sources.
Edit 2: if I'm being honest it doesn't even need to be jack for this, if there's a simpler version that accomplishes the same thing, then I'd use that instead.
https://i.imgur.com/7FUMya6.jpg
When i use pacmd list-sinks there's only one output device so even if set it as a default in the:
deafult.pa
it will still set the speakers as default
any help would be great.
Please note that this site uses cookies to personalise content and adverts, to provide social media features, and to analyse web traffic. Click here for more information.