lm3914n

Hi all.

I'm finally getting round to finishing my bar graph display that I started a
few weeks ago. The circuit that Fred Bloggs sent was a great start and I've
done a minor tweek to it, to save on resisters. You'll find the circuit
diagram at

http://www.cjdawson76.btinternet.co.uk/circuits/PowerIndicator.jpg

Basically, the tweeks that I've made to the circuit are to remove 10 * 1K
Resisters that were in series with each of the LED's. Then I've changed R14
from 1.5K to 6.8K which reduces the current drawn into the chip. Other
than that there's been no change.

Now for the problem. If I run the circuit in Bar mode (SW2 closed)
everything works perfectly, as you'd expect.
When I change to Dot mode (SW2 open) a strange thing happens. If the chip
was displaying more than one LED in bar mode then the correct LED is
illuminated as you'd expect. BUT... D10 is also partially illuminated!!!!!

How can I stop D10 more illuminating unless it is supposed to be?

Thanks in advance for the Help

Colin.

 

I sorted it!!!! I think I'm beggining to get the hang of it again
;-)

Here's what I did. First of all I changed R14 to 3K9 This has increased the
brightness. It's also well within the capability of the chip.

Next I added a 5k resister, this is connected to the 12V supply and pin 1 on
the chip (LED 1 Output). When running in bar mode, the first LED to light
will be a little dimmer than the others, but this is not noticable. When I
switch to DOT mode and one of the other LED's is lit, the chip will draw a
few hundred microamps, the resister will then come into play and use some of
that power meaning that the LED won't light up. Which is much more
astheticly pleasing overall

Regards

Colin.

 

Colin... hope you are well...

I may be missing sunnik but from the circuit you show, it is possible to
short your DCin to ground through VR1. Mal-adjustment of VR1 might allow
harm to whatever is providing your signal. You might like to put and
"endstop" resistor between the ground-end of VR1 and ground just to limit
current draw when VR1 is at zero... R11 will prevent shorting to vRef at the
other end of VR1.

just a consideration.

 

Hi, yes you are missing something....

The 5.1V, DCin and Gnd0 terminals are a seperate circuit on the Velleman
K8004 Pulse Width Modulator circuit.
These three terminals between them provide a method of manually controlling
the pulse width. If the DCIn does does get shorted to ground, it will
signal the circuit to provide a pusle width of 0%. Basically, turning the
whole thing off.

If DCin was shorted to 5.1v then the circuit would set the pulse width to
100%. I don't think there's a major issue with either scenario. I'd just
need to replace the damaged component, and nothing else will suffer from
that. It might even be possible to remove C5 and c4 completely. But I'm
not, as I'd rather have them than damage the chip.

Regards

Col.