Dual stage Op Amp Gain distribution question

[gvi70000]

Hello,
I have an LMV 722M op amp that i use as an amplifier for an microphone and photo diode.
On this link https://www.grozeaion.com/hsp/high-speed-photography
i need to modify the circuits (Sound Trigger and Light/Dark Trigger) so that the second stage of the LMV 722M will not be used as "set" threshold level. It will be used as an second stage amplifier.
These sensor should detect low/high sounds (water drops/ popping balloons) and low/high light level.
I'm thinking that first stage must have a constant value and the second one variable.

Can anyone suggest me how to distribute the gain between the two stages?

Thanks

PS: The Vcc of the circuits will be 5V(it is provided by an digital output of the ATmega328) and the output of the sensor will be the input for one analog input of the same microcontroler

 

[keith1200rs]

Can anyone suggest me how to distribute the gain between the two stages?

That is tricky until you know the range of signals you are going to get. If you needed a gain of say 40dB to 80dB then I would put 40dB on the first stage, or maybe a little less, and then the next stage could be 0dB to 40dB variable.

The idea would be to get the largest signal out of the first stage without overloading it, then use the second stage to boost that as required.

Depending on the actual gains require, that may not make sense. For example, if the gain required was 60dB to 90dB, I would probably make the first stage 45dB and the next 15dB to 45dB. In general, putting half the gain in each stage at maximum gain.

Keith.

 

[gvi70000]

What do you think about this one?

 

[keith1200rs]

That's not going to work too well. Your connections to C3 are wrong. You have no input resistor on the non-inverting input of IC1A. The two opamps cannot share a capacitor C3.

Something like this would be an option.

Keith.

 

[gvi70000]

It took me 15 min to see what do you man by

You have no input resistor on the non-inverting input of IC1A.]You have no input resistor on the non-inverting input of IC1A.

:|
I do have one question. Can you explain why i can't use the same capacitor?

 

[keith1200rs]

You will get feedback between the two stages. See that attached simulation. The green trace is the circuit I showed earlier, the green one connected sharing the capacitor.

Keith.

By the way, I'm sorry I didn't make myself clearer earlier - I am try to help, honest!

 

[keith1200rs]

One more thing - if you try simulating it, don't use National Semiconductor's LMV721 model - they have got the connections wrong. I spent ages tracking that down for someone else on here a while ago. b@$$t@rd$

Keith.

 

[gvi70000]

Thank you!
No need to excuse yourself, i am still new in this field..i am the one who ask "stupid" questions

I think i will try on each stage a variable resistor to set the gain because i can't find in any store an 3 way small switch.

PS: What software do you use for simulation?

 

[keith1200rs]

I use SIMetrix. There is a demo version available which is fully functional - only limited by circuit size. https://www.simetrix.co.uk/

Keith.

 

[gvi70000]

If until next week i will find a 3 position switch i will use the "switch gain" variant, else...i will use the one with variable resistors.
Meanwhile i also modified the photodiode sensor, so do you think that it will work?

Thanks

 

[keith1200rs]

You have a huge gain there compared to the original circuit. The original circuit simply triggered relative to a 1.5V bias point. You are simply amplifying the light. The other circuit reverse biased the photodiode and triggered against a fixed threshold (1.5V) using the opamp like a comparator. You have no comparator or "trigger" - simply an amplifier.

Using a transimpedance amplifier with 1M and 1nF will make it pretty slow.

With the other circuit, I wouldn't worry about a switch - simply use a pot, as you have drawn.

Keith.

 

[Mad I.D.]

It took me 15 min to see what do you man by

You have no input resistor on the non-inverting input of IC1A.]You have no input resistor on the non-inverting input of IC1A.

:|
I do have one question. Can you explain why i can't use the same capacitor?

You have connected a very large capacitor directly to the inverting input. That means you are adding a phase shift of 90 deg !! You cannot do that! The circuit would oscillate. If you add the resistor like the nice man said, you are creating a zero to compensate for the pole, and the system works

enjoy

 

[keith1200rs]

If you add the resistor like the nice man said, you are creating a zero to compensate for the pole, and the system works

enjoy

But you still cannot share the capacitor between the two stages.

Keith

 

[gvi70000]

Hello
The output of the both preamplifier's will be the input to an ATmega328 microchip (the brain of my new remote) thats why i made the second stage of the circuit as amplifiers and not as comparator (not needed anymore)

 

[keith1200rs]

A couple of issues:

1. you have AC coupled the second gain stage with C2 - this is a DC stage so you need to connect R5 directly to ground, unless you are deliberately trying to only couple a pulse through the second stage.

2. you have a lot of missing junction points on your schematics. That doesn't matter unless you are planning to use it to make the PCB - the software will think they aren't connected. If you run an ERC it will probably report the missing connections as wire overlaps, I think.

Keith.

 

[gvi70000]

I am nit trying to couple a pulse through the second stage, so i will follow your advice and remove the C2.

Using a transimpedance amplifier with 1M and 1nF will make it pretty slow.

With the light sensor i want to capture lightning, fireworks and when a laser beam falling on the photodiode is interrupted i want to capture water drops.
How small should be C1(1nF) ? From what i was reading regarding the speed of similar sensor they operate at high frequencies and i don't believe this is may case.
This is a wrong assumption?

Regarding the value of R1 - 1000KOhm, should i decrease the valu to 500k and increase the gain in the second stage?

 

[keith1200rs]

C1 needs to be large enough for stability. However, you also need to watch out for saturation. Until you actually build it and try it, it is difficult to know how much gain you will need. I would be tempted to make the second stage have a lot more gain - maybe 100x.

With 1nF and 1M ohm for the transimpedance amplifier your bandwidth will be 160Hz. which seems rather low for "high speed photography".

In theory you probably only need 0.2pF for stability, but then things are going to be dictated by stray capacitance which isn't good. I would suggest using something like 4.7pF. That will give a reasonable bandwidth of 34kHz.

Keith