What causes my LEDs ON?

Go To Last Post
6 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hey guys, I'm facing an interesting problem.

 

The attached schema is of a simple circuit that serially clocks in data to the shift register 74HC164 and individual shift register outputs drive some NPN+PNP high-side switches. The +48V in the experiment is not connected (havent tried yet), there is only +5V. LEDs D1-D8 are there for fast visual check to see if clocked-in data are correct. If I want to use the LEDs, I close the jumper JP1. Everything works fine, LEds are signalling what they should.

 

Now the funny situation comes when I open the jumper JP1. LEDs will continue to shine, although with a reduced intensity! As there is no other connection of their cathodes to GND, I don't see what is sinking the current. I measured the resistence between cathodes and GND with the JP1 open (without power supply) and it was at least 350k (my multimeter is broken and does not measure past 350k).

 

The PCB is with SMD components so theoretically there can be an unwanted copper "bridge" but I measured the pcb and fixed is very carefully. 

 

Can you think of any reason why opening JP1 would not shut the LEDs D1-D8?

 

Any tip is appreciated :)

 

EDIT: Any other comments regarding the schematic are also welcome - if you see something that a man experienced in electronics would never do, go ahead and let me learn ;-)

Attachment(s): 

This topic has a solution.

Last Edited: Sat. Aug 1, 2015 - 08:42 PM
This reply has been marked as the solution. 
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

If there is really no error in the pcb and it's clean, then it would have to be the combined reverse leakage current of the un-selected LED's.

Actually, your 470R LED resistors are probably a bit too low value. You see that value in old circuits when LED's were all "vanilla" 20mA types. You may well have fitted much more efficient types which may only need a few mA or less to be visible. Raising the resistor value to 2k or more might well give enough light while stopping leakage being a noticeable problem.

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

That sounds very reasonable, yes, I think this could be the case and it beautifully explains the reduced light intensity. I'm waiting for SMD resistors from china, will measure leds and pick more appropriate value and will let you know! In meantime, I'll trick my software to clock-in all log.1 to see if all diodes will shut, that would confirm your theory ;-) Big thanks, man.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I would remove R5 since you want to pull the line to a logic LOW state.  That resistor is pretty high.

 

As far as the LED's remaining lit, I am not sure the HC165 is rated for driving LED's directly.  And if ALL the LED's are remaining lit I would think there is something else causing this besides leakage current.

 

Jim

I would rather attempt something great and fail, than attempt nothing and succeed - Fortune Cookie

 

"The critical shortage here is not stuff, but time." - Johan Ekdahl

 

"Step N is required before you can do step N+1!" - ka7ehk

 

"If you want a career with a known path - become an undertaker. Dead people don't sue!" - Kartman

"Why is there a "Highway to Hell" and only a "Stairway to Heaven"? A prediction of the expected traffic load?"  - Lee "theusch"

 

Speak sweetly. It makes your words easier to digest when at a later date you have to eat them ;-)  - Source Unknown

Please Read: Code-of-Conduct

Atmel Studio6.2/AS7, DipTrace, Quartus, MPLAB, RSLogix user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Any changes on the LEDs if shorting those base resistors (R2, R15-R21)? maybe the leakage is around the transistors.

How about cleaning the board with isopropyl alcohol and test again? sometimes it's all about a solder dust stuck at the wrong place.

while(!solution) {patience--;}

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thank you all for your input, guys.

 

I measured the LEDs on a breadboard - I connected in series 7k5-LED-LED-7k5, (LEDs were cathode to cathode) and with 4,5V among all four components, one of the LEDs was shining. I also measured an individual LED Volt/Amper characteristics and I learned that this type of LED has reverse current of 6,9mA (!) at 2.53V and forward current sufficient for nice brightness is 0.39mA. I'm concluding that I cannot use my circuit like this, since the reverse current would drive the NPN's open anyway. Because I don't have other types of LEDs available and I cannot tristate HC164 outputs, I'm abandoning indicating with LEDs. I would have to use additional non-LED diode in series with LED but that would impose a bit too much overhead (in terms of parts #) not worth the effect.

 

@jgmdesign - thanks for the comment, that resistance really doesn'n make much sense.

 

EDIT: I just let my MCU to clock-in 0xFF to try all LEDs ON and I confirm that the more LEDs are ON, the more dim glowing LEDs have and when all of them are ON, they become dark as there is no other "source of GND" for them left.

Last Edited: Sun. Aug 2, 2015 - 08:11 AM