LM3914 usage

Thread Starter

VernonLS

Joined Oct 9, 2019
42
I am trying to use the LM3914 display driver chip, but I am not sure I understand the way it works. I am using an external 1.6V voltage source on pin 6 and have grounded pins 4 & 8 and attached a 1K resistor from pin 7 to ground. My signal source is on pin 5. Pin 2 is system ground and pin 3 is +5V (relative to system ground). If I understand this correctly, my input voltage source should be dropped internally through the 10 1K resistors in steps. So, I would expect that any signal larger than 1.6V should cause all of the comparator outputs to be at +5V. As the signal source drops down I would expect pin 10 to go to ground level at about 1.44V input and pin 11 to go to ground level about 1.28, etc and for pin 1 to go to ground level at about 0.16V.

This is not happening. All of the outputs are stuck at +5 no matter what the input signal is. Am I confused about how this chip works? See the attached LM3914 data sheet. thanks, vern
 

Attachments

eetech00

Joined Jun 8, 2013
3,856
A
I am trying to use the LM3914 display driver chip, but I am not sure I understand the way it works. I am using an external 1.6V voltage source on pin 6 and have grounded pins 4 & 8 and attached a 1K resistor from pin 7 to ground. My signal source is on pin 5. Pin 2 is system ground and pin 3 is +5V (relative to system ground). If I understand this correctly, my input voltage source should be dropped internally through the 10 1K resistors in steps. So, I would expect that any signal larger than 1.6V should cause all of the comparator outputs to be at +5V. As the signal source drops down I would expect pin 10 to go to ground level at about 1.44V input and pin 11 to go to ground level about 1.28, etc and for pin 1 to go to ground level at about 0.16V.

This is not happening. All of the outputs are stuck at +5 no matter what the input signal is. Am I confused about how this chip works? See the attached LM3914 data sheet. thanks, vern
hi

a schematic is worth a thousand words. Please post yours...

eT
 

djsfantasi

Joined Apr 11, 2010
9,156
How are you trying to use the LM3914. I cannot be sure, as you did not provide a schematic (please do so ASAP), but it’s no where near the typical use schematic on the datasheet.
Grounding pin 8 (Vref) will likely cause problems, as it should be a percentage of the max signal voltage.
Pins 6&7 are typically tied together. 7 goes to ground. 6 is tied (according to you) at 1.6 V.
I don’t see how you can expect anything when it is wired in this way. All outputs at +5V is as good as any to be expected.
 

Thread Starter

VernonLS

Joined Oct 9, 2019
42
How are you trying to use the LM3914. I cannot be sure, as you did not provide a schematic (please do so ASAP), but it’s no where near the typical use schematic on the datasheet.
Grounding pin 8 (Vref) will likely cause problems, as it should be a percentage of the max signal voltage.
Pins 6&7 are typically tied together. 7 goes to ground. 6 is tied (according to you) at 1.6 V.
I don’t see how you can expect anything when it is wired in this way. All outputs at +5V is as good as any to be expected.
From the datasheet, I thought that it was not necessary to use the internal 1.25 reference voltage if you could supply your own reference voltage on pin 6 as I wished to do. Perhaps it is necessary to ONLY use the internal 1.25V source and always tie pin 7 to pin 6 as you say. All the datasheet diagrams show pin 8 grounded. I will go back and try connecting 6 to 7 and no other connection. I attach a diagram of my test circuit. All the outputs of the 3914 chip are at +5, no matter the value of the input voltage on in 6 which I have measured to be from 1.87V max to 127 mv min.

Thank you for helping.
 

Attachments

Audioguru again

Joined Oct 21, 2019
6,671
Why use an external 1.6V source on pin 6 when the pin 7 and pin 8 with two resistors can accurately do it?
Then an input higher than the pin 6 voltage lights the 10th LED (all the LEDs if pin 9 is at the positive supply voltage) and since pin 4 is grounded then the 1st LED lights when the input is higher than 0.16V.
 

eetech00

Joined Jun 8, 2013
3,856
Hi

Here's a circuit that uses the internal reference set to 1.60v.
Each LED lights in ~160mv steps. LED current is set by R1.
If you want the LEDs to light one at a time, replace V+ with ground on ping 9.

eT
1572923842882.png
 

Thread Starter

VernonLS

Joined Oct 9, 2019
42
Hi

Here's a circuit that uses the internal reference set to 1.60v.
Each LED lights in ~160mv steps. LED current is set by R1.
If you want the LEDs to light one at a time, replace V+ with ground on ping 9.

eT
View attachment 190609
OK - I built a breadboard with this circuit. You can see the attached pictures of the breadboard. The only change was that since I had no 270 ohm resistor, I had to put a 220 ohm in series with a 51 ohm. I also only connected LEDs to pins 1, 10 and 11 since all I wanted to do was to see some evidence of lighting an LED. When I measured the signals this is what I got V+=5.02v, pin 6=1.29v, as was pin 7, pin 8=262mv. My signal input on pin 5 was 1.69v. None of the LEDs on pins 1, 10 or 11 lit. Pin 1 was 2.73v, pin 10 was 2.74v and pin 11 was 3.38v. My understanding of the circuit is that all the outputs should have switched to full on so I would have expected the outputs to be near ground level so that the LEDs would all be on. See the attached files for pictures of the board and the 'scope signal.
 

Attachments

eetech00

Joined Jun 8, 2013
3,856
OK - I built a breadboard with this circuit. You can see the attached pictures of the breadboard. The only change was that since I had no 270 ohm resistor, I had to put a 220 ohm in series with a 51 ohm. I also only connected LEDs to pins 1, 10 and 11 since all I wanted to do was to see some evidence of lighting an LED. When I measured the signals this is what I got V+=5.02v, pin 6=1.29v, as was pin 7, pin 8=262mv. My signal input on pin 5 was 1.69v. None of the LEDs on pins 1, 10 or 11 lit. Pin 1 was 2.73v, pin 10 was 2.74v and pin 11 was 3.38v. My understanding of the circuit is that all the outputs should have switched to full on so I would have expected the outputs to be near ground level so that the LEDs would all be on. See the attached files for pictures of the board and the 'scope signal.
Hi

This circuit worked perfectly on my bench as shown.

I don't see any connection to pin 5 (the signal input pin).
The voltage at pin 6/7 should be 1.60v nominal. Mine was 1.62

All LEDs on at 1.6x volts input.

Check the LED polarity is correct and there is 5 volts at the anode of each
Check the supply pins for +5 (pin3) and ground (pin 2).

eT
 

Thread Starter

VernonLS

Joined Oct 9, 2019
42
Hi

This circuit worked perfectly on my bench as shown.

I don't see any connection to pin 5 (the signal input pin).
The voltage at pin 6/7 should be 1.60v nominal. Mine was 1.62

All LEDs on at 1.6x volts input.

Check the LED polarity is correct and there is 5 volts at the anode of each
Check the supply pins for +5 (pin3) and ground (pin 2).

eT
OK, major mystery now solved. I was checking, double checking, triple checking every connection. Then, I happened to put my scope probe directly on pin 5 of the 3914 instead of the breadboard pin in the common breadboard row that pin 5 was connected to and I saw no signal. Rechecking I found that I had a signal on the breadboard, but none on pin 5 of the chip. WTF?? So, I looked carefully at the chip, saw nothing wrong, but removed it, slightly bent pin 5, plugged it back in, plugged the power in and all my LEDs lit. So, I don't know whether it was not plugged in fully, whether there was a slight defect in the breadboard or what, but pin 5 was definitely not receiving the input signal. whoddathunkit!!

And, just to verify, yes, pins 6/7 now show about 1.69v. All 3 LEDs turn on when a 1.63v signal is present at pin 5. The LED on pin 11 turns off when it goes to 1.57v and pin 10 turns off at 1.38v and pin 1 was still on at 506mv.

OK, now that I have confidence that the chip works as I expected, I can proceed.

Thank you for continuing to help me. Without that I don't know what I would have done next. My next step is to try to figure out how to set a different voltage level than 1.6v. How did you determine that?

Can you tell me
 

Audioguru again

Joined Oct 21, 2019
6,671
Why not get rid of the Mickey Mouse breadboard with its intermittent contacts, rows of contacts and wires all over the place acting like antennas for interference and a lot of stray capacitance between them.

About 45 years ago I threw away my breadboard when no circuit built on it worked properly and since then I quickly laid out each circuit on a stripboard, cut all the strips to short lengths with a strip-cutter tool then soldered all the parts and a few short jumper wires. My job was designing and building a custom one-off circuit (many were fairly complicated) and my stripboard prototypes all worked perfectly and most looked good enough to be sold as the final item.

My Sound Level Indicator project uses an LM3915 logarithmic bar-graph IC similar to your LM3914, a dual opamp, a voltage regulator and two transistors. It has a microphone and shows a wide range of sound levels in my living room. It is soldered together on a stripboard and has been working perfectly and continuously for almost 15 years.
 

Attachments

Thread Starter

VernonLS

Joined Oct 9, 2019
42
Why not get rid of the Mickey Mouse breadboard with its intermittent contacts, rows of contacts and wires all over the place acting like antennas for interference and a lot of stray capacitance between them.

About 45 years ago I threw away my breadboard when no circuit built on it worked properly and since then I quickly laid out each circuit on a stripboard, cut all the strips to short lengths with a strip-cutter tool then soldered all the parts and a few short jumper wires. My job was designing and building a custom one-off circuit (many were fairly complicated) and my stripboard prototypes all worked perfectly and most looked good enough to be sold as the final item.

My Sound Level Indicator project uses an LM3915 logarithmic bar-graph IC similar to your LM3914, a dual opamp, a voltage regulator and two transistors. It has a microphone and shows a wide range of sound levels in my living room. It is soldered together on a stripboard and has been working perfectly and continuously for almost 15 years.
Thanks. I started a long project using soldered connections and after making numerous mistakes found that I needed to break the project apart into smaller segments and use a plugboard. Once I get a portion of it working (power, frequency/voltage conversion, etc.)) I wire up that part of it using a soldered type board. The current effort is just part of a larger effort. Admittedly, while I worked a long career as an electrical engineer, I am a novice at putting together circuit boards especially using the wonderful tools that have been developed since I was in the workplace. I am not familiar with these "stripboards" you mention, but I will google them and see how I might use them in my project.

So far I have made any number of dumb, novice, and just place careless mistakes, but I had forgotten how much fun electronics could be. It is hard to explain to a non-technical person, but there is a great deal of satisfaction in seeing it all work. It is much the same experience that I once had with large scale systems only without the corporate bs.

Thanks for taking an interest and I am sure I will be posting again when I hit my next stumbling block.

vern
 

eetech00

Joined Jun 8, 2013
3,856
Thanks. I started a long project using soldered connections and after making numerous mistakes found that I needed to break the project apart into smaller segments and use a plugboard. Once I get a portion of it working (power, frequency/voltage conversion, etc.)) I wire up that part of it using a soldered type board. The current effort is just part of a larger effort. Admittedly, while I worked a long career as an electrical engineer, I am a novice at putting together circuit boards especially using the wonderful tools that have been developed since I was in the workplace. I am not familiar with these "stripboards" you mention, but I will google them and see how I might use them in my project.

So far I have made any number of dumb, novice, and just place careless mistakes, but I had forgotten how much fun electronics could be. It is hard to explain to a non-technical person, but there is a great deal of satisfaction in seeing it all work. It is much the same experience that I once had with large scale systems only without the corporate bs.

Thanks for taking an interest and I am sure I will be posting again when I hit my next stumbling block.

vern
Hi
Nothing wrong with using the breadboard shown in your photo. I use them all the time. Sometimes I use multiple small breadboards separated by functional block to keep organized. They just have to be used properly to be effective.
It’s important to keep all jumpers short, flat against the board, and connections clean and tight. Use a large cap (100uf or greater) across the main power supply input connections to the board and ALWAYS use a bypass cap (0.1uf) across the supply pins of EACH IC. Always check voltages at supply pins and signal inputs. If you do these things, your breadboarding will go much smoother.

Good luck with your project.

eT
 
Top