74HC595 uses less power when enabling 8 pins compare to 7 pins

Thread Starter

Shun Takahashi

Joined Jan 31, 2016
4
Does shift registers consume less power when enabling 8 output pins altogether than enabling 7 output pins? I was measuring the current at Vcc pin of 74HC595 shift register with an LED connected to each output pins, and noticed that enabling all output pins drives IC more efficiently than enabling individual pins say seven out of eight. (Enabling seven LEDs consumes 44.31mA vs 33.87mA for eight)

I presume enabling all pins bypasses certain logics inside the chip but could not find related information in the datasheet. I am new to electronics and would like to hear from experts in this forum to see if this is the typical behavior of shift-registers (not just for NXP one, or for my testing specific).

# of enabled pins - Voltage on LED on PIN0 - Current at Vcc
1 - 2.96v - 7.56mA
2 - 2.91v - 14.75mA
3 - 2.89v - 22.12mA
4 - 2.88v - 28.25mA
5 - 2.86v - 34.23mA
6 - 2.84v - 39.54mA
7 - 2.82v - 44.31mA
8 - 2.86v - 33.87mA
* I am supplying 3.28v to the chip.
 

dl324

Joined Mar 30, 2015
16,840
Welcome to AAC!

Please post a schematic of how you have the shift register connected.

The outputs are controlled by a single enable, so are either all enabled or 3-stated.

There is no pin 0...
 

AnalogKid

Joined Aug 1, 2013
10,986
You are over-currenting the device. Notice that the current per output steadily decreases as you connect more loads. The maximum test current on any pin on pages 7 and 8 is 6 mA. What is the value of the LED current limiting resistor for each output?

ak
 

hp1729

Joined Nov 23, 2015
2,304
Does shift registers consume less power when enabling 8 output pins altogether than enabling 7 output pins? I was measuring the current at Vcc pin of 74HC595 shift register with an LED connected to each output pins, and noticed that enabling all output pins drives IC more efficiently than enabling individual pins say seven out of eight. (Enabling seven LEDs consumes 44.31mA vs 33.87mA for eight)

I presume enabling all pins bypasses certain logics inside the chip but could not find related information in the datasheet. I am new to electronics and would like to hear from experts in this forum to see if this is the typical behavior of shift-registers (not just for NXP one, or for my testing specific).

# of enabled pins - Voltage on LED on PIN0 - Current at Vcc
1 - 2.96v - 7.56mA
2 - 2.91v - 14.75mA
3 - 2.89v - 22.12mA
4 - 2.88v - 28.25mA
5 - 2.86v - 34.23mA
6 - 2.84v - 39.54mA
7 - 2.82v - 44.31mA
8 - 2.86v - 33.87mA
* I am supplying 3.28v to the chip.
I couldn't confirm your observation. Running from 5 V (I don't know if it would make a difference). I grounded all the inputs giving clear and enabling the outputs. On all 9 outputs I put a red LED and a 560 ohm resistor to VCC. Connecting them one at a time current ramped up incrementally to about 45 mA. Did you do much different? My meter was on the ground side.
 

Thread Starter

Shun Takahashi

Joined Jan 31, 2016
4
Thanks everyone!

>DL324
Please see attached (overly simplified) schematic. I have an arduino supplying 3.3v to 74hc595n. Each output pins has an LED with 47ohms resister. And wrt pin0, I apology for the confusion. I meant to say QA as the pin is always enabled.

I am measuring current between 3.3Vout on Arduino and VCC on 74hc595n to see how much current the chip withdraws. Also for the reference, I am now measuring current between resister and the anode of LED at QA.

> Thanks AnalogKid for pointing it out. Yes you are right and I increased current limiting resisters to 47ohms and now each pins only withdraws less than 5mA but my chip behaves the same. (enabling 7 pins would consume 35mA on Vcc and then drop to 22mA when activate 8th pin)

> hp1729,

Yes, I am suspecting it too. And that's why i want to see if this behavior is specific to my setup (or the way i set up is not right) or specific to the chip i am using or common for 74hc595 chip in general...
 

Attachments

Thread Starter

Shun Takahashi

Joined Jan 31, 2016
4
Thanks everyone! After changing my measuring point to the ground between the Arduino and ground of the circuit itself, then this behavior disappeared and now my chips consumption. I am presuming some of my output pins hooking to the chip's latch/data/clock pins are supplying current...

# I should have measured the entire circuit not the chip... thanks again hp1729 for the crew!
 

Thread Starter

Shun Takahashi

Joined Jan 31, 2016
4
And that was my data pin (Pin 14). I accidentally put the pin state to unknown after enabled 8th pin and somehow 74hc595 chip withdraw current from that pin...

Thanks again everyone for the lights!
 

hp1729

Joined Nov 23, 2015
2,304
Thanks everyone!

>DL324
Please see attached (overly simplified) schematic. I have an arduino supplying 3.3v to 74hc595n. Each output pins has an LED with 47ohms resister. And wrt pin0, I apology for the confusion. I meant to say QA as the pin is always enabled.

I am measuring current between 3.3Vout on Arduino and VCC on 74hc595n to see how much current the chip withdraws. Also for the reference, I am now measuring current between resister and the anode of LED at QA.

> Thanks AnalogKid for pointing it out. Yes you are right and I increased current limiting resisters to 47ohms and now each pins only withdraws less than 5mA but my chip behaves the same. (enabling 7 pins would consume 35mA on Vcc and then drop to 22mA when activate 8th pin)

> hp1729,

Yes, I am suspecting it too. And that's why i want to see if this behavior is specific to my setup (or the way i set up is not right) or specific to the chip i am using or common for 74hc595 chip in general...
Amazing. Doncha love a mystery?
 

atferrari

Joined Jan 6, 2004
4,764
For an anomaly like that, a fully detailed schematic is a must. A waste of time otherwise.

Discarding possible mistakes is part of the process.
 
Top