So much trouble, so little time...

[PlaneCrazy]

It doesn't install on this (my work) PC, so I'll try again tonight from my personal PC... "personal personal computer"? ... sorry, focusing.

How do I deal with the 4.5V span and/or the 1.5V offset during the "prototype presentation" phase, or will the LEDs light up or die down as expected when I turn the rheostat's knob?

 

[ericgibbs]

hi,
Using the 6V and the two 1K's as per your drawing, the Span is from 0v thru 2.5v, so the 10K input pot will light all the leds at ~2.5V setting.
No 27R required.
To get a Span of 4.5V you need a 1K across pins 7&8 and a 2.6K from pin8 to 0V.

To add a 1.5V lower end Offset you need 1.5V on pin4,,,, or use a 10Kpot on pin4 to 0V and adjust it to give a 1.5V offset on pin4

OK.?

 

[PlaneCrazy]

Nope, now I'm even more confused. You're saying that Pin 7 delivers 0-5V? So the 2 x 10k's will devide that into 0-2.5V each, and thus 2.5V will see all the LEDs lit? And this happens regardless of whether I apply 6V or 60V over the pots? But if I use a 1k across Pins 7&8 and 3.8k from Pin 8 as in the LM document, I'll effectively devide create a span of 5*(3.8k/4.8k)=4V?

Brain freeze would be real nice right now.

 

[ericgibbs]

hi PC,
You are getting yourself at it..
Lets step back a notch.

Have you built the circuit you posted, using the 1K's on pins 7, 8 and 0V.?
Have you removed all the resistors to the LED supply.?

Is the LM3914 connected for 6V operation.??

The voltages on pins 7 & 8 are set by the resistor values, in your case 1K's.

Is this where we are at the moment.??

EDIT:
this image shows what you should have.

 

[PlaneCrazy]

OK, so how do I get LM wired for 6V operation if my power supply is 12V?

Pin 2 is grounded.
Pin 3 is 12V (supply)
Pin 4 is grounded (0V low)
Pin 5 is the variable signal from 0V to 6V dc.
Pin 6 is shorted with (equal to) Pin 7
Pin 7 is connected to Pin 8 via 1k resistor, and to ground via 2 x 1k resistors
Pin 8 is therefore connected to ground via 1k
Pin 9 is connected to Pin 3 (bar mode)

Will 12V fry my system? If so, it may already be fried and this whole exercise will be in vain until I can order new components (but let's keep the faith and think happy thoughts). In fact, if the LEDs are connected to a back voltage of 12V, shoudn't the forward voltage (Pins 1, 10-18) be at least 9V, preferably 10V since the LEDs get fried at more than 3V? What about forward and backward current through the LEDs?

Does it matter what the amplitude/magnitude of the signal is? If so, why so?

Put my mind to rest, and I'll hook it up to the power supply.

 

[PlaneCrazy]

(I think I got "forward" and "backward" right...)

 

[ericgibbs]

This image shows the ONLY change to the circuit is a 21R resistor in the LED supply, when using a 12V supply.

If you dont have a 21R resistor to hand, change to DOT mode while you do.

OK

Edit:
Image 10, shows the DOT mode

 

[PlaneCrazy]

I don't have a 22Ω resistor. Can I power the existing circuit with 6V, and use the 9V or 12V DC supply for the signal? Otherwise, how do I go to dot mode? Did I fit the capacitor correctly?

EDIT: Dot mode = Pin 9 open? That simple?

 

[ericgibbs]

I don't have a 22Ω resistor. Can I power the existing circuit with 6V, and use the 9V or 12V DC supply for the signal? Otherwise, how do I go to dot mode? Did I fit the capacitor correctly?

EDIT: Dot mode = Pin 9 open? That simple?

Yes,,,, 6V thru 12V for the signal, cap OK.

 

[PlaneCrazy]

Thank you for your stellar efforts! I'm guessing that, if I use 12V is the signal, 6V would be the difference between signal high and V+? Even so, the circuit is now configured for dot-mode - if the LEDs don't work one by one, they won't work all together. So I'm connecting it to the 12V source and 6V signal with dot-mode, as per your pic 10.

 

[PlaneCrazy]

OK, so I've started it up and nothing seems to be happening.

Source = 12V
Signal = 6VDC

Used multimeter to confirm all LEDs are working.

Here's how it stands:

Pin 2 to ground = 0V
Pin 3 to ground = 12V
Pin 4 to ground = 0V
Pin 5 to ground = varying from 0V to 5.75V when I turn the pot
Pin 6 to ground = 2.56V
Pin 7 to ground = 2.56V
Pin 8 to ground = 1.31V
Pin 9 to ground = 0V

So the chip seems to be doing something. Why dosn't any of the LEDs light up?

 

[ericgibbs]

Quick checks,
1. Are the Led's the correct way around.?? ie: Anode to +V and Cathodes to the LM3914 pin.??

2. Are the Led Anodes connect to a +V supply.?

3. The LM3914 could be 'dead' try another.

EDIT:
4. Are the 12V and 6V supply 0V connected together.?

EDIT2:
I can see an error on your circuit, you have LM3914 pins wrong.
Compare this image to the datasheet.

 

[PlaneCrazy]

1. The LEDs are connected like this - I'm not too sure which is anode and cathode side.

2. LED to ground is at 11.65V

3. Please don't say that. I only have two, and they both went throught the same ordeal...

4.

 

[ericgibbs]

hi,
Did you see my edits.?

On your picture I dont see a link on the 0V of the 6V and 12v

EDIT:
To check the polarity of the Leds.
Power off, remove the LM3914.
Power On,
Use a 270R [approx] and connect one end of the resistor to 0V and then in turn touch the leg of the led that would have gone to the LM3914 pin with the other end of the resistor... the led should light

 

[PlaneCrazy]

OK, sorry about that, but the connections still look the same: LED1 is connected to Pin 1, LED2 is connected to Pin 18 (top row, first from left) etc.

What do you mean by "a link on the 0V of the 6V and the 12V"?

Will a 400Ω do? Too much? It's the smallest I've got... unless I use 2 of these in parallel (200Ω) Too little?

 

[ericgibbs]

hi,
The 0V lines MUST be connected together somewhere on the circuit.

A 400R will do for a test.

 

[PlaneCrazy]

I've connected the two ground bus bars together, as shown at the left of the diagram. Isn't that sufficient?

Used 2 470s in parallel - The LED flashes then dies, but lights up when I apply the multimeter, so it is undamaged.

EDIT: Likewise with just 470.

 

[ericgibbs]

I saw the 0V link on the diagram, just wanted to make sure it actually existed on the circuit, as its essential.

The led should remain lit when the 470R is used in the way I described. Keep the 470 connected for a few seconds, it should keep the led ON.

You do have the LM3914 out during this test OK.?

 

[PlaneCrazy]

I'm testing one of the LEDs that are not connected to the one LM3914. I have removed the second LM for now, will reconnect later from scratch.

 

[PlaneCrazy]

Guess what? I'm sick again... Maybe I'll try this tomorrow, maybe not until Monday. Thanks again for everything.