Cascading LM3914 Bar/Dot displays

[Wannabe EE]

Greetings everyone! This is my first post. I have questions about cascading LM3914 Bar/Dot drivers.

First, it may help for you to know a little about my background....

I am a Mechanical Engineer, but I have dabbled quite a bit in electronics. I know and understand the very basics like ohms law, and I have some knowledge of more advanced concepts as well. What I have found is that I have gaps in my understanding of many intermediate concepts.

Years ago, I worked with a friend to develop a 8052 based engine gauge for homebuilt aircraft. I did much of the programming (with a compiled BASIC), PCB layout and made the Gerber files (the hard way from AutoCAD layout), assembled and tested the prototypes.

I am also a skilled machinist and have retrofitted an obsolete CNC Bridgeport with a Windoze based system. I designed and built my own BreakOut Board to interface between the computer and machine.

Much of what I do in my consulting business involves PCB layout and prototype assembly of a few ATTINY based circuits. My programming skills are rudimentary so "C", Assembly, and writing ISRs is beyond by capabilities.

I find it is easier to make things with discrete logic ICs than to learn how to program. Many things I find I want to do, require fairly advanced programming skill. I have had others do the difficult programming for me.

So now I am using LM3914s to make a bar graph display. I don't understand the data sheet's explanation of calculating the resistor values for cascading several ICs.

I have a need for several displays ranging from 10 LEDs to 60 LEDs.

I understand voltage dividers, but there must be something very general or ambiguous about the explanation that has me confused. Could someone walk me through the process of figuring the values? I understand the LED brightness is also involved, as well as linearity and voltage range.

Thank you everyone! I have spent hours reading through various posts here, and haven't found the answer I am looking for.

 

[ljcox]

What data sheet do you have?

I have the National Semiconductor one dated January 2000.

In order to save my time, could you please post extracts from your data sheet of the parts you don't understand.

 

[ericgibbs]

hi Wanna.
This link has a dual LM3914 set up guide.

https://www.electro-tech-online.com/blogs/ericgibbs/136-single-dual-lm3914-v3-0-calculator.html

Download and install as explained, it will give you a good idea for the required component values.

 

[Wannabe EE]

Download and install as explained

Thanks Eric. When I run them I get an error:
"MSVBVM50.DLL was not found"
I have WinXP 64 Pro with the .NET framework installed, so I don't know what's going on.

 

[ericgibbs]

Thanks Eric. When I run them I get an error:
"MSVBVM50.DLL was not found"
I have WinXP 64 Pro with the .NET framework installed, so I don't know what's going on.

hi,
Have you installed the Microsoft VB5 runtime files to your Windows/System folder as the notes explain?

 

[Wannabe EE]

What data sheet do you have?

February 2003

please post extracts from your data sheet of the parts you don't understand.

How do I determine the total resistance of the stack?
Why is R5 different than the rest?

 

[Wannabe EE]

hi,
Have you installed the Microsoft VB5 runtime files to your Windows/System folder as the notes explain?

The notes say "The Dual LM3914_CalcV3.0. exe file requires the Visual Basic 5 or 6 runtime files."

I have the .NET framework (as I mentioned) because I loaded VB6. If It requires VB5 then the note should be more specific.

-EDIT-

I now have the required dll and the program is running correctly. THANKS!

 

[Wannabe EE]

I'm having a hard time connecting the dots!

Eric, the resistor calculator you provided has five resistors for each LM3914. The data sheet (as shown in my previous post) has a completely different resistor arrangement!

This excerpt from the data sheet, shows even a different setup!!!

Why are there so many differences???

 

[ericgibbs]

hi wanna.

The link is for a Tool that can be used to configure 'most' projects, so it has all the options.

I would suggest we work out way thru your project, lets consider just a two LM3914 project.

A few questions.

1. What are your supply voltages for the project.
2. Colour and number of LED's, connected in series on each LM3914 output pin.
3. The voltage range you wish to measure and display on the LM3914's.
4. Bar or Single LED Mode.?

That should get us started.

E.

 

[Wannabe EE]

1.) 5 Volts
2.) Green (3.3 V fwd voltage), Yellow (2.0 V), Orange (2.1 V), Red (1.95 V). Only one LED per pin.
3.) 0 to 4.8 V (full scale)
4.) Bar

I will condition the various signals with an op amp before the LM3914. The voltages are from the LED data sheets. There will be various colors in each bar. These are fairly bright LEDs and only need between 5 ma and 8 ma for my needs.

 

[ericgibbs]

1.) 5 Volts
2.) Green (3.3 V fwd voltage), Yellow (2.0 V), Orange (2.1 V), Red (1.95 V). Only one LED per pin.
3.) 0 to 4.8 V (full scale)
4.) Bar

I will condition the various signals with an op amp before the LM3914. The voltages are from the LED data sheets. There will be various colors in each bar. These are fairly bright LEDs and only need between 7 ma and 10 ma for my needs.

OK,
The first point is that we will have to scale down the 0 to 4.8V input, when using 5V supply , you should allow a 1.5V to 2V overhead on the supply against the voltage being input.

You could do this by a simple resistive divider on the 0 to 4.8V.

 

[ericgibbs]

OK,
The first point is that we will have to scale down the 0 to 4.8V input, when using 5V supply , you should allow a 1.5V to 2V overhead on the supply against the voltage being input.

You could do this by a simple resistive divider on the 0 to 4.8V.

EDIT:
Set up the LM3914 exactly the same as this image and then use the signla input slider on the bottom LM3914 to see the effect

 

[Wannabe EE]

OK,
The first point is that we will have to scale down the 0 to 4.8V input, when using 5V supply , you should allow a 1.5V to 2V overhead on the supply against the voltage being input.

I have a bunch of LMC6494 Rail to Rail Op Amps that I will use. I used them in the afore mentioned Engine Gauge with 4.8 V full scale.

I'll play with the simulator... Thanks!

 

[Wannabe EE]

So now I have experimented with the LM3914 simulator. How does this relate to the actual circuit. The circuits from the data sheet that I have posted are different.

The simulator has sliders R3 and R4, implying that they are resistors. But the numbers in the boxes next to these sliders do not show any units. I don't know what these numbers really are.

 

[Wannabe EE]

The first point is that we will have to scale down the 0 to 4.8V input, when using 5V supply , you should allow a 1.5V to 2V overhead on the supply against the voltage being input.

OOPS! I had assumed that you said this to allow overhead for the op amp I plan to use for the input signal. But I have now found that the data sheet says this overhead is needed for the LM3914.

The simulator is valuable for getting a sense of the relationship between R1 and R2, along with their influence on the LED current.

I understand the RLo of the second LM3914 is the same as RHi of the first. And that RHi of the second LM3914 needs to be 3 volts. The data sheet isn't clear to me how to provide the proper resistor value or configuration.

 

[ericgibbs]

Look as this from the LM3914 datasheet.

Set input for 0 to 2.4V,

 

[Wannabe EE]

Thank you Eric.

I may be looking farther down the road here. I have a need for at least 23 displays of around 10 different input signals (temperature transducers, pressure transducers, analog position transducers, etc.). These displays will vary between 10 and 60 LEDs.

The example you show above, has an input range of 0 to 2.4 V. The latest image I posted has an input range of 0 to 3 V. I can see your example is quite simple and uses a minimum of parts. But it seems to me that my example has the potential of having a 25% better signal to noise ratio. I get that it may not matter for a 20 step display. But when I get 60 steps, I would rather have that extra margin. Especially if it only requires a few carefully placed resistors. Am I missing something?

I think I now have enough understanding to "brute force" the reference voltages on a stack of LM3914s, but I want to understand more of the inner workings of the chip and understand what the datasheet is showing with the voltage divider stack and how the selection and placement of resistors will allow my last image (from your simulator) to function.

If I am mistaken, and the IC won't work as I have modeled, then please let me know, and if possible, explain why.

I queried in post #14:

The simulator has sliders R3 and R4, implying that they are resistors. But the numbers in the boxes next to these sliders do not show any units. I don't know what these numbers really are.

And in post #15:

I understand the RLo of the second LM3914 is the same as RHi of the first. And that RHi of the second LM3914 needs to be 3 volts. The data sheet isn't clear to me how to provide the proper resistor value or configuration.

This is the crux of what would help me understand more.

I am wanting to get the "big picture" here. If you give a man a fish, he has a feast for a night. If you teach a man to fish, he will eat the rest of his life. If I can understand the overall system, I won't need to keep asking for help on each different circuit I will need. I expect the datasheet has all the info, but with the holes in my understanding of electronics, I am unable to decipher the key points. I have said many times, I know enough about electronics to be dangerous. You may be noticing that right now.

I thank you for all your guidance, and the simulator program you wrote is VERY NICE! I have read this post a couple times and I certainly do NOT want to bite the hand that feeds me.

 

[Wannabe EE]

a wonderful resource!

I found a pdf online that answers many of my questions!

https://www.electro-tech-online.com...ng20the20LM3914-620LED20Bargraph20Drivers.pdf

It is good explanation of the workings of the LM3914. It is in an easy to understand prose rather than that of the cryptic and arcane datasheet. After reading much of this, I can see, in retrospect, that the information is in the datasheet. Although in the datasheet, isn't explained the same way (or at all).

Thank you Eric for your responses! It was seeing your participation and comments that originally prompted me to join this forum.

 

[chemelec]

Here is my Article and Projects for Cascading 2 or 3 LM3914's.
It includes the PCB's for them.

**broken link removed**

 

[audioguru]

How are you doing, Gary? We haven't heard from you for ages.
Did you have a nice trip to SA?