Author Topic: Op Amp Output to LED Driver Issue  (Read 1167 times)

0 Members and 1 Guest are viewing this topic.

Offline skinnyTopic starter

  • Contributor
  • Posts: 35
  • Country: us
    • Skinny R&D
Op Amp Output to LED Driver Issue
« on: February 07, 2020, 05:45:21 am »
Hi Everyone,

I'm trying to put together an electric field detector. To accomplish this I looked at a bunch of different plans on the Internet and decided on 2 different designs. Each design used a different op amp and each op amp seemed to pick up on different emanation types. I wanted to combine the two designs and came up with the attached schematic.

My question has to do with the center portion. After many tests, I added the diodes. I wanted to make sure that one output wouldn't adversely affect the other. When I added the diodes to the circuit, the LED bar graph being driven by the LM3914 driver, would fully light up making the detector unusable. If I removed either diode from the path and replaced it with a wire, the whole thing would begin to operate normally again. I replaced both diodes and decided to take some measurement with a multimeter. As soon as I touched the voltmeter with one lead on ground and the other at the junction of the two diodes, the detector began to work properly again. When I removed the meter, the light bar maxed out again.

Because of this behavior, I stuck the 10MΩ resistor at the junction, and everything worked great. I figured a good meter probably has a pretty high impedance, so why not introduce a high impedance and see if that solves the problem. Here's the issue. I have no idea why this works. Admittedly I am missing some knowledge about op amps and also how this LM3914 works. It is the first time I ever used either of these particular amps or an LED driver.

Any knowledge you could share was to why I experienced this behavior would be great appreciated. Also, if you think the schematic could be helped in any other way, I'm all ears.

Thanks for your help,

Skinny
 

Online Zero999

  • Super Contributor
  • ***
  • Posts: 20364
  • Country: gb
  • 0999
Re: Op Amp Output to LED Driver Issue
« Reply #1 on: February 07, 2020, 08:01:59 pm »
Well the diodes act as an OR gate, so the higher voltage will get passed to the LM3914.

What are you trying to achieve?
 

Offline Paul Rose

  • Regular Contributor
  • *
  • Posts: 139
  • Country: us
Re: Op Amp Output to LED Driver Issue
« Reply #2 on: February 08, 2020, 12:35:20 am »
It might be an issue with the (tiny) input bias current for the lm3914 having nowhere to "go".

In that case a high value resistor (like your 10M) to ground might be just what you need.

Like Zero999 said, you'll get the higher of the 2 voltages, but I think that's what you're looking for.
 

Offline Paul Rose

  • Regular Contributor
  • *
  • Posts: 139
  • Country: us
Re: Op Amp Output to LED Driver Issue
« Reply #3 on: February 08, 2020, 12:37:39 am »
Not related to your question, but I'm a little surprised that you are seing noticable difference in electric field detection from the 2 op-amp curcuits.   Can you explain what differences you are seeing between the two?
 

Offline skinnyTopic starter

  • Contributor
  • Posts: 35
  • Country: us
    • Skinny R&D
Re: Op Amp Output to LED Driver Issue
« Reply #4 on: February 08, 2020, 04:31:28 am »
Well the diodes act as an OR gate, so the higher voltage will get passed to the LM3914.

What are you trying to achieve?

The or gate functionality is exactly what I am shooting for. Thanks for the confirmation.

It might be an issue with the (tiny) input bias current for the lm3914 having nowhere to "go".

In that case a high value resistor (like your 10M) to ground might be just what you need.

Interesting. Correct me if I'm wrong. So your saying that there needs to be some kind of dc bias current on the input pin of the 3914 and just that slight amount being hooked to ground is enough to work?

Not related to your question, but I'm a little surprised that you are seing noticable difference in electric field detection from the 2 op-amp curcuits.   Can you explain what differences you are seeing between the two?

When I was doing my initial prototyping, it seemed that the CA3130 picked up on 4G LTE energy better than the LM358. On the other hand, the LM358 seemed to respond better to lower frequency, continuous wave type energy. Instead of continuing to burn through different amps that did both well, I decided to combine them and just run with it.

Thanks to both of you for your replies!
 

Offline Paul Rose

  • Regular Contributor
  • *
  • Posts: 139
  • Country: us
Re: Op Amp Output to LED Driver Issue
« Reply #5 on: February 08, 2020, 02:20:20 pm »
Interesting. Correct me if I'm wrong. So your saying that there needs to be some kind of dc bias current on the input pin of the 3914 and just that slight amount being hooked to ground is enough to work?

From the TI datasheet for lm3914:
Definition of terms:
Input Bias Current: Current flowing out of the signal input when the input buffer is in the linear region
Electrical Charateristics
Input bias current 25 nA typical

So the 3914 wants to source a small amount of current from the input.  With the diodes, this can't flow due to the diode polarity.
The large resistor to ground gives a path for this, but isn't enough to load down the op-amp outputs.
 
The following users thanked this post: skinny

Offline skinnyTopic starter

  • Contributor
  • Posts: 35
  • Country: us
    • Skinny R&D
Re: Op Amp Output to LED Driver Issue
« Reply #6 on: February 08, 2020, 05:45:35 pm »
Thanks for the help. Pulled the datasheet and I see what you're saying. This a great news and now I understand why it's working. On a side note, I changed out the general purpose diodes for schottky diodes and the sensitivity is obviously better. 
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf