need 5V at enable pin of a register if the arduino is offline

I guess it's a simple question but i am confused.

How do i get a high value to the enable pin of a shift register, if the arduino is "offline"? It enables if the pin is low (Which i set with the arduino IO). But i want to be sure, that there is no low value to the pin if the arduino is not up yet to prevent from damage to some hardware. The hardware got it'S own powersupply so i can add 5V to the pin without the arduino.

Regards
BennX

Pull it up to its +5 with a resistor, 10kΩ.

I tried with 1 2k,10k and a 18k but it does not work as expected. It still is enabled.

Is it enabled with a HI (E) or a LO (/E)?

Have you tried a pull-down?

it enables at LO. I noticed if i cut of the line (enableing line) between the arduino and the Hardware the enableing works with a pull up. But if i do not cut it the pull up does not work. (Arduino has no power!)
(Get the low signal thrue the pin of the arduino from the gnd connection from the arduino to the hardware)

They are just connected by the gnd and 4 lines for the SPI to the shift register.

The Arduino I/O pins are Inputs by default, high impedance.
So, if you did nothing else with it, the Enable pin should stay HI.
At what point does it go LO?

You could pull it up "harder", with 2.2kΩ or 4.7kΩ

See my edit. If i cut of the gnd connection between the arduino and the hardware it works. So i think it get a LO thrue the gnd connection to the arduino thrue the pins back to the shiftregister. Even a 2k resistor does not pull it up to high.

Added a picture. Sorry for the litle chaos on this. exchanged the resistor with 2k 10k and 18k. All combinations does not works.

It doesn't look like Pin 49 is mapped ... I would try using another pin.
ATmega2560-Arduino Pin Mapping

dlloyd:
It doesn't look like Pin 49 is mapped ... I would try using another pin.
ATmega2560-Arduino Pin Mapping

PLO(ICP4) pin 35 of the Mega. And if it wouldn't be mapped i would have noticed it because my whole effect wouldn't work without the disabling of the register.

Solved the with some other help: electronics.stackexchange.com/questions/136344/pull-up-does-not-work/136350#136350

BennX:
I guess it's a simple question but I am confused.

Yes, it is a very simple question, but somehow you managed to get everyone else confused!

When the Arduino is powered down, the protective diodes to each and every pin pull the pins down to Vcc and if Vcc is clearly zero, then all pins are pulled down to (nearly) ground by definition. You need to invert the sense of OE with another gate such as a 74HC14 (one of the six inverters).

You could do this with a transistor and pull-up however I warn you that the simple circuit with just collector resistor, base resistor and pull-up does not in general switch fast enough for logic circuits. For OE to control a LED, this does not matter much.

In fact, your design is very strange insofar as you are controlling both individual LEDs in the RGB and the common cathode, and that you are controlling OE in any case. This is clearly unnecessary, why would you want to do it?

This is clearly unnecessary, why would you want to do it?

Yes he has been told that before in another thread.

The point here is that you should never power only part of a circuit.

If you absolutely must do this then I would use opto isolators for all connections between the Arduino and the rest of the circuit.

Paul__B:

BennX:
I guess it's a simple question but I am confused.

Yes, it is a very simple question, but somehow you managed to get everyone else confused!

When the Arduino is powered down, the protective diodes to each and every pin pull the pins down to Vcc and if Vcc is clearly zero, then all pins are pulled down to (nearly) ground by definition. You need to invert the sense of OE with another gate such as a 74HC14 (one of the six inverters).

You could do this with a transistor and pull-up however I warn you that the simple circuit with just collector resistor, base resistor and pull-up does not in general switch fast enough for logic circuits. For OE to control a LED, this does not matter much.

In fact, your design is very strange insofar as you are controlling both individual LEDs in the RGB and the common cathode, and that you are controlling OE in any case. This is clearly unnecessary, why would you want to do it?

Thanks for your reply. I think that the 2N3904 circuid is fast enough for what i am doing. I multiplex at 8Khz and it has a max time of 300Mhz so it should do fine.

Yes it seems to be strange.

I do multiplex 125 rgb Leds by the GND (N Mosfet) and PWM them by the Vcc with the PNP by pushing several 0/1 thrue it. I do have 25 Leds per level which are all connected thrue 3 wires to a PNP to pwm. (75 lines in total) To now enable that i can handle all 125 Leds i multiplex those 5 levels with the n Mosfet. In conclusion I have a 5x5x5 Cube where i can set every single LED and have different colors. (I could post the code here and you'll understand whats happening)

To be sure that i dont have any flickering while setting the next level i disable the outputs which shuts down the cube. Then i set the next 75 color bits for the next level and enable the ouputs. (Well we can't see it) I even push the next 80 bit in it after this i clock one time the shiftregister data to the storage register data and then enable the registers again. (Just to be sure)

Grumpy_Mike:

This is clearly unnecessary, why would you want to do it?

Yes he has been told that before in another thread.

The point here is that you should never power only part of a circuit.

If you absolutely must do this then I would use opto isolators for all connections between the Arduino and the rest of the circuit.

Yes i know that this should not happen but in fact that it meight could happen because of the user (me) i want to be safe that i dont get in trouble. The Powersupply i am going to use has 2,5A and the whole cube (125*3 leds at 20mA) would be to much for it. But a single level shouldnt be a problem. I want to prevent the system from damage here.

Regards

If you want to prevent the system from damage then power it all off one power supply.
125 * 3 * .02 = 2.35 Amps
The Arduino takes about 50mA so yes there is enough to power it all.

Grumpy_Mike:
If you want to prevent the system from damage then power it all off one power supply.
125 * 3 * .02 = 2.35 Amps
The Arduino takes about 50mA so yes there is enough to power it all.

But i cant power the arduino from a 5V supply or can i?

Note that you can power an Arduino happily from 5V straight into its 5V pin if that helps; you'll be bypassing the Arduino's voltage regulator.

(Don't try plugging 5V in via the barrel connector, though; the Arduino's voltage regulator needs a few volts more than 5V to work correctly)

Yes you can just feed the power into the 5V pin, I do that all the time.

For added safty wire the Vin and 5V pins on the arduino together.

howiem:
Note that you can power an Arduino happily from 5V straight into its 5V pin if that helps; you'll be bypassing the Arduino's voltage regulator.

(Don't try plugging 5V in via the barrel connector, though; the Arduino's voltage regulator needs a few volts more than 5V to work correctly)

Good to know. Thanks alot for that!
Just tried it and it works.yeay

But i wonder if i want to programm it, i need to earase the wire i think. Since i would power it from usb like this

Grumpy_Mike:
Yes you can just feed the power into the 5V pin, I do that all the time.

For added safty wire the Vin and 5V pins on the arduino together.

So i should add the Vcc from my board to the Vin and 5+?

I also noticed that the Pullup works if its connected like this. I got no undefined state before the arduino is booted.

I've never had a problem with powering an Arduino from 5V (straight into its 5V pin) while programming it with USB.

Don't think it should be a problem (though it'd be nice to get confirmation from someone smarter than me :slight_smile:

No plugging a USB I to the arduino when powered by 5V is no problem.