Hello All, I've been playing around with the LM3914 IC and on it's datasheet listed under "Other Applications" is this...

"Moving Hole Display - Indicator LED is dark, Rest of Bar Lit"

I have googled and searched everywhere I can think of but can find no example of this project? Has anyone ever seen this or know how it would be done?

Thank You.

 

How does it "move" actually. Can u explain it a bit more

 

I have built the inverse circuit, i.e., a "moving lit LED" that corresponds to the temperature. My guess is that the inclusion on the datasheet is an afterthought circuit that is just an accidental capability of the moving lit LED function.

But I may not have understood your question.

 

How does it "move" actually. Can u explain it a bit more

On the data sheet it shows some typical applications for the LM3914 IC... Then at the end of that section it lists "Other Applications", in which they do not show examples for...

One of the Other applications is listed a "Moving Hole Display". The way I read it, it would be like a Dot mode display, except that the active LED indiacator would be dark (not lit) and all the other LEDs on the bank would be lit. Think of normal Dot mode but reverse the lighting actions of the LEDs...


See the second to last bullet of page 17 here... http://www.national.com/ds/LM/LM3914.pdf


Would this be done by simply reversing the + and - supply to the IC?

--

 

Would this be done by simply reversing the + and - supply to the IC?--

I don't think that will work. More likely, the LED's are put in reversed and the common cathodes would be connected to ground. I haven't tried this and it may be wrong.

 

You could use a transistor or Darlington array (like a ULN2803) plus a couple of discrete (Darlington) transistors to invert the output signal.

Each of the LM3914 outputs would need to have a pull-up resistor to Vcc, and each output connected to a Darlington channel input (base).

The Darlington output (open collector) would sink current from the LED cathodes. You would need to use current limiting resistors between the LED anodes and Vcc.

[eta]

See the attached schematic. Just the LED and driver portion is shown. You could use discrete transistors instead of the ULN2803A if you wished, but the resistor array would need to be reduced somewhat; perhaps down to 1k to 2k depending on your +V.

I didn't show all of the LEDs to keep the schematic less cluttered.

Note that I used SIP (single inline pin) resistor arrays. You can use discrete resistors if you'd like, but the SIPs will save wiring and board space.

 

Yup..as Sgt said.
One LM3914 and one driver for each ring.
Inverted or not is possible.

 

Thanks All. Although Sgts response would surely work... I was trying to find out what exactly what the data sheet was referring to... I thought maybe you would wire the IC differently to get this effect. Since I've found no direct example of this "Moving hole" application, I'm thinking it's not something easily done by just using the IC in a different configuration. Thank you everyone for your input!

 

You should be able to use the LM3914 outputs as shunt switches, instead of series switches, as they are usually used. See the attachment.
Sorry for the sideways printing. I didn't want to redraw the whole danged thing.
Leave pin 9 floating for dot mode.
This should work for LED currents up to about 20mA. Output saturation voltage is typically about 1V at 20mA, so if your LED will glow with 1V across it, the circuit may not be satisfactory without some modification, such as a diode (1N400X) in series with the common ground connection of the LEDs.
Pay attention to resistor power dissipation ratings.