ws12812b splice signal wire into two different i/o pins

Hello I hv spliced the signal wire into 2 different i/o pins inorder to apply more than one pattern to a project s i know. it. is not acceptable But they don't work at the same time Each split of the signal wire connected to its own transistor as well as its own dedicated board

So when one end is on the other one is off

Also there is additional main board which has 2 out puts one. for. each. board. and each mosfet

Attaching wiring link (trying. to add attachment. doesn't work)

https://drive.google.com/file/d/1BxqlYqoYRoB2VoTP1jipmncexbKGReHa/view?usp=sharing

When testing separately both codes work as expected but when combining Someone weird happens it works for a second and stops

i*s there anyway can solve it?*

I know there is an easier way yet my programing skills are insufficient so I use more hardware

Thanks

Attaching wiring

Nop. And the link you provided will not work unless you have a google account.

There is absolutely no need to do this whatsoever. Just use one output and assign the same output to two instances of the library what ever that is because there is no information on that either.

Using a transistor for buffering the signal is hard because most cheap transistors won’t work at that sort of speed. Also the transistor will invert the signal so you have to invert it in the software.

And the code you are using is what?

Tnx for the reply

one code is designed for sound reactive leds (msgeq7 additional hardware involved) second is just something that enables. to switch among study colors ( zebra via blynk iot app) so i 'd like to be able to switch among them

btw regarding the switching using transistors i wired it in such way the transistor shots down the entire board and the second transistor turns on the second board ( not the signal wire) each split. of the signal assigned to a different board so they are biased

it wont let me add any attachments tired several formats will try. with dropbox

https://www.dropbox.com/s/4vfrhie5c3f9166/South%20wall.jpg?dl=0

seems like only Available option is to add another Led strip

it wont let me add any attachments tired several formats

Are they too big? What does the error message say when it rejects the post?

Still can not understand your explanation of what you are trying to do because you are not giving any real information.

of the signal assigned to a different board so they are biased

What are these boards of which you speak, you know but you are not telling us.

wired it in such way the transistor shots down the entire board and the second transistor turns on the second board ( not the signal wire) each split.

This makes no sense at all.

I am sorry so will try to rephrase

Equipment: Lights : Ws12812b LED strip Boards Arduino nano Node mcu Node mcu ARDUINO MKR 1000 470 resistors Moffet transistors msgeq7

Arduino Nano Has a code of sound reactive LEDs(frequency response to music)

So one end of the data cable goes there[Nano]

Arduino nano is being turn on and off by the Arduino Mkr1000 via transistor (the entire board not just the i/o)

The second board Node mcu Has a different code uploaded which just enables to change colors of the LED strip

So the second end of the signal wire goes to the node mcu

The node mcu is being tund on & off via another transistor operated by Arduino MKR1000

The 3rd board Arduino MKR basically controls the on off of the Arduino Nano and the node mcu via transistors Namely one i/o of the MKR 1000 shots on/off the nano[the board it self] and another i/o pins shots on and off the node Just like turn on /off a led , or a motor or what ever via transistor you turn on off the entire board It basically functions like a main

Only one i/o of the mkr is on so the other end of signal wire isn’t being affected

Since i want to be able to switch manually between codes manually so 2 i/o pins are required When trying to attach the wiring diagram it just says the was a problem uploading the "name of the file"

I am sorry so will try to rephrase

Well we are getting closer but it sounds even more weird.

Arduino nano is being turn on and off by the Arduino Mkr1000 via transistor (the entire board not just the i/o)

To quote a famous tennis player “you can not be serious!”
Have you taken into account that when this powers up it will go into boot mode for several seconds where it could latch up.

Have you taken into account that when this powers down it could be being phantom powered if signals are still being fed into it and therefore not being off at all?

Am I right in thinking that you have four Processor boards? Your list says so. But your description only mentions one node MCU, not two.

phrases like:-

So one end of the data cable goes there[Nano]

Don’t cut it in a circuit that no one but you has seen. We don’t know a lot of things like do all processor boards share a single ground? What are the decoupling capacitors and how are they arranged? Do you use a series resistor in the data line?

I know that it is a typical beginners response to try and use two processes to cover up the fact they can’t program, but using three or four is just being silly.

When trying to attach the wiring diagram it just says the was a problem uploading the “name of the file”

Is this a file that is on your computer? It needs to be and not a link?
Is this a jpg or png file?
Is this file smaller than 2M?

After writing the above I looked at the latest link.
Now it so happens I have a drop box account and I managed to download your image and find it is 11.1M in size. So that would be over 2M then.
So I converted it to a jpg file and that come in at 1.5M which is under 2M.
So I can post it here.

I must say that it seems not to be worth the effort, there is lots of information missing.
What FETs are these, part number?
The source and drain seem to be connected to ground?
That is never going to turn anything on and off.
You don’t seem to have ANY capacitors!
You don’t show how anything is powered.
You show a symbol for ground but also use GND as a label, are these the same points?
You show three processors but you had four in your list.

Hello I placed one capacitor 1000uf between positive and Negative of the LED strip Please let me know if additional capacitors are required as well as values and where to place them I have placed additional resistors as well for the transistors between source and gate to preventing floating current

There is a common ground for all components

Transistors RFP30N06LE 30A 60V N-Channel Power Mosfet

Transistor wiring :

gate from i/o pins

NEGATIVE of the load component to drain Source to ground I 'very tested it with a LED as well as 12v motor and it worked also in the current project it does turns on and off the boardsb ut seems with a delay

Thanks

Arrow008: Hello I placed one capacitor 1000uf between positive and Negative of the LED strip Please let me know if additional capacitors are required as well as values and where to place them I have placed additional resistors as well for the transistors between source and gate to prevent floating current

There is a common ground for all components

Transistors RFP30N06LE 30A 60V N-Channel Power Mosfet

Transistor wiring :

gate from i/o pins

NEGATIVE of the load component to drain Source to ground I 'very tested it with a LED as well as 12v motor and it worked also in the current project it does turns on and off the boardsb ut seems with a delay

Thanks

Fundamentally it is not OK to turn off the power to a device while still having active signals connected to it. It is a bad idea which will eventually destroy the device. Do not pursue it.

If you are hell bent on running one strip with the code in two separate processors you can not connect outputs together. What you need to do is signal routing. In other words to switch the signal being received by the strip from one processor to the other. The way you do this is by using multiplexer chips.

There are lots of different types available but the one I would recommend is a 74HTC157. This is described as a quad 2 input to 1 line data selector / multiplexer. So quad means there are four in the one package. Each one has two inputs ( the data outputs from the two processors you want to drive the strip with ) and one output, which you connect through a resistor to your LED strip. The 74HTC157 has four of these, you only need to use one. The input that gets routed through is controlled by one "select" line.

Other chips you might try are the 74HTC153, 74HTC155 and 74HTC156. I would not use the analogue multiplexers because your MKR 1000 and node MCU produces a 3V3 signal that is not always sufficient to drive a 5V WS2812, it is under the minimum guaranteed voltage to drive it, but sometimes it works, but why take the risk when it is so easy to use the right chips.

The other thing I would do is to implement some sort of signaling that indicates when the processor driving the data on the chip out and only switch over the multiplexers when it is not driving the output. The master processor could also signal when the driving processor should suspend operations. Be aware you can't just send a 5V signal from the nano into the 3V3 system inputs of the other chip, it needs to be cut down and a simple potential divider of 510R and 1K should be used.

Much as I cannot figure what on earth is going on here, there should be no need to use multiplexers.

Each processor has a 330 Ohm resistor in series with its output (outputs? Who knows?) and the output is shut down (switched to INPUT or INPUT_PULLUP) when not driving the display. If you fail to do this, you get a garbled display but that is your own fault in not coding it correctly; no damage is done. :grinning:

If we are talking about WS2812 strips, that resistor should be there in any case.

Each processor has a 330 Ohm resistor in series with its outpu

How do you make that out? Do you mean the 33R internal output impedance?

and the output is shut down (switched to INPUT or INPUT_PULLUP) when not driving the display.

The OP is crap at programming which is why he is using such a bazaar arrangement in the first place. There is no synchronisation between the two processors driving the LEDs other than him trying to turn one of them off. Then if he does there will be phantom powering problems because two outputs are connected directly together and one of them will have no power.

Using a multiplexer is the safest option, and closest to what the OP wants to achieve.

I am not describing how any one should do this at all as there should be only one processor driving the LEDs and if that does not create too much dead time for the Wi-Fi to cope with there should only be one processor in whole thing.

Grumpy_Mike: How do you make that out? Do you mean the 33R internal output impedance?

As best as I can figure out the situation here, the subject line suggests the OP is driving a WS2812 strip.

The string of WS2812s retains its present pattern unless and until a new pattern is loaded. While there may be the possibility of loading only part (always from the start of the strip), the entire pattern must usually be re-loaded in one go.

The proposition here appears to be that there are two processors that are intended to load a pattern into the strip. That is to say, two alternate patterns. If we accept that as the intention, then this should be reasonably straightforward. Whichever processor is intended to load a pattern can turn its out put on (pinMode as OUTPUT) and send data using the appropriate library. Having done so it can yield to the other processor by setting pinMode as INPUT.

Now as Mike is always keen to point out, there is supposed to be a 330 Ohm (or may be 220 or 470) resistor in series with the data input of the WS2812 strip. If two processors are driving the strip, this is terribly convenient as each processor can have its 330 Ohm series resistor which now fulfils the additional function of protecting against contention between the two outputs if they are incorrectly synchronised and both attempt to drive the display at once. The data to the display will necessarily be garbled, but at least the processors are protected.

The other suggestion seems to be that one or other processor might be powered down which might cause "phantom powering". It is not clear from where this proposal originates or why, but the 330 Ohm resistors provide some protection against this; if it was to be a serous concern, a Schottky diode from the output pin to supply rail would complete the protection.

And if one processor is indeed an ESP8266, then it will also require a 74HC(T)14 to buffer its output to the proper levels, though a 74AHC(T)132 would provide a perfect solution to combining the two data sources as well.

Whichever processor is intended to load a pattern can turn its out put on (pinMode as OUTPUT) and send data using the appropriate library.

But there as the schematic stands there is no way of one processor knowing when it is its turn to generate or yield.

So if you can be bothered to read reply 4 the OP says:-

Arduino nano is being turn on and off by the Arduino Mkr1000 via transistor (the entire board not just the i/o)

The why is because he admits to being bad at code and wants to wire two outputs together.

The proposition here appears to be that there are two processors that are intended to load a pattern into the strip. That is to say, two alternate patterns.

Two alternate patterns but not alternating patterns.

The other suggestion seems to be that one or other processor might be powered down which might cause "phantom powering". It is not clear from where this proposal originates or why, but the 330 Ohm resistors provide some protection against this;

Well first of all that is not what he had, second it is not what he needs and third, it doesn't stop parasitic powering, it just limits the current. Have you actually bothered to look at the FETs in the "schematic" that are doing the switching off the power?

It would be good to hear of your positive suggestions about this post instead of constantly misunderstanding things.

You seem to be ignoring the last line of my last post which I will say again:-

I am not describing how any one should do this at all as there should be only one processor driving the LEDs and if that does not create too much dead time for the Wi-Fi to cope with there should only be one processor in whole thing.

I am trying to tell the OP how to do what is as close to his original design as possible without making a huge mess of things.