# Need verification on schematic ...

**URL:** <https://community.sparkfun.com/t/need-verification-on-schematic/28223>\
**Category:** Projects\
**Created:** [February 6, 2014, 7:31am UTC](https://community.sparkfun.com/t/need-verification-on-schematic/28223 "2014-02-06T07:31:23Z")\
**Posts on this page:** 5\
**Page:** 5

<div class="post-metadata">

**Author:** ![kirash4](https://avatars.discourse-cdn.com/v4/letter/k/f1d935/32.png) [@kirash4](https://community.sparkfun.com/u/kirash4)\
**Post date:** [February 20, 2014, 1:00am UTC](https://community.sparkfun.com/t/need-verification-on-schematic/28223/81 "2014-02-20T01:00:06Z")

</div>

Ah, gotcha. Yeah, I don’t know if that’s the case, though it would not surprise me.

---

<div class="post-metadata">

**Author:** ![markaren1](https://avatars.discourse-cdn.com/v4/letter/m/ac91a4/32.png) [@markaren1](https://community.sparkfun.com/u/markaren1)\
**Post date:** [February 20, 2014, 4:05am UTC](https://community.sparkfun.com/t/need-verification-on-schematic/28223/82 "2014-02-20T04:05:31Z")

</div>

Yup, so before you remove power to the LED chain, you make the data and clock outputs from the CPU as inputs.

And after you apply power to the LED you make data and clock outputs once more.

-mark

---

<div class="post-metadata">

**Author:** ![kirash4](https://avatars.discourse-cdn.com/v4/letter/k/f1d935/32.png) [@kirash4](https://community.sparkfun.com/u/kirash4)\
**Post date:** [February 20, 2014, 4:26am UTC](https://community.sparkfun.com/t/need-verification-on-schematic/28223/83 "2014-02-20T04:26:03Z")

</div>

> markaren1:  
> Yup, so before you remove power to the LED chain, you make the data and clock outputs from the CPU as inputs.
> 
> And after you apply power to the LED you make data and clock outputs once more.

No, that would defeat the purpose of doing it the way I am with the FETs. I need to send an RGB::Black to the string \*before\* I apply power to them. It's how it's working now. If I apply power to the strings before doing anything, some of the drivers tend to randomly light up. So I'll end up with random LEDs along any of the four strings lit up before the controller kicks in. By keeping VCC off, and letting the MCU turn it on, I'm able to send an RGB::Black first, then apply power and all is well. It's a hack of a solution, but there's nothing else I can do.

---

<div class="post-metadata">

**Author:** ![markaren1](https://avatars.discourse-cdn.com/v4/letter/m/ac91a4/32.png) [@markaren1](https://community.sparkfun.com/u/markaren1)\
**Post date:** [February 20, 2014, 8:22am UTC](https://community.sparkfun.com/t/need-verification-on-schematic/28223/84 "2014-02-20T08:22:45Z")

</div>

If it works, fine. But you are sending data to ICs that do not have power applied 🙂

---

<div class="post-metadata">

**Author:** ![kirash4](https://avatars.discourse-cdn.com/v4/letter/k/f1d935/32.png) [@kirash4](https://community.sparkfun.com/u/kirash4)\
**Post date:** [February 20, 2014, 9:57am UTC](https://community.sparkfun.com/t/need-verification-on-schematic/28223/85 "2014-02-20T09:57:08Z")

</div>

> markaren1:  
> If it works, fine. But you are sending data to ICs that do not have power applied 🙂

I'm well aware of that, however I believe what n1ist is suggesting to be true. I may be sending data to ICs that aren't powered through the normal VCC line, but they are still receiving the data sent to them, and when power is applied, that data is reflected.

[Previous page](https://community.sparkfun.com/t/need-verification-on-schematic/28223.md?page=4)