Question for the Oscilloscope Experts

[KeepItSimpleStupid]

Probe is fine. Just remember in x1 the input Z is 1M and in X10 it's 10 M.

3000 to 4000 seems like a weird frequency, Remember to take into account the possible doubling if you look at your FW ourput.
Do look at the power supply pins at each IC and at what stage it appears.
That first stage where the input Z is high, is suspecttable. Thus, it makes sense that it e a straight x1 gain amplifier that will amplify DC including the amplifier DC offsets.

 

[rsfoto]

First thought is a mechanical source, such as an HVAC fan or heat/AC motor. Sympathetic harmonic vibrations from the 50/60Hz powered hardware in your home/office/neighborhood.

Don't forget, your piezo sensor is very sensitive across a wide range of vibration frequencies. You may end up having to insert some filters into your circuit to "dump" (or block) the unwanted noise.

Just to muddy the waters even more, here's an interesting, recent ETO thread of note: https://www.electro-tech-online.com/threads/mysterious-sounds.146595/

Hi cowboybob,

See my message #219 before yours and now looks a bit better. No such things as you mentioned near here and my probes do work nicely, at least I think so, Look at my image of the calibration and the noise the Oscilloscope is measuring during calibration message # 213

Will try to understand my circuit and see where can I filter things without jeopardizing the sensitivity.

 

[rsfoto]

Thus, it makes sense that it e a straight x1 gain amplifier that will amplify DC including the amplifier DC offsets.

Hi Kiss,

Did I understand correctly and use the Unit A of the Amplifier Op Amp as a Voltage Follower ?

Do I need a resistor between Output Pin and Inverting Pin when using as Voltage follower ?

 

[rsfoto]

About my prior question I found this

**broken link removed**

 

[rsfoto]

Here is the latest version which I am testing.



and below the test setup



A video of how the vibration is measured by the piezo.

**broken link removed**

 

[audioguru]

Your piezo sensor works only at low frequencies, it does not pickup high frequencies. Your C3 was shorting the output of the opamp causing it to oscillate.
If you want a lowpass filter then it should be like this:

10k and 1uF cut vibration frequencies below 16Hz that maybe you want. 1uF is a big expensive film capacitor and since it gets AC then it MUST be a non-polarized type (not a polarized electrolytic). Then increase the value of R2 and R3 so that the capacitors can be small inexpensive film ones.

I see that your active rectifier produces a fullwave output but it is still wrong. It should be like this:

 

[audioguru]

Why did you kill the gain of the preamp by changing R5 to only 100 ohms?

 

[KeepItSimpleStupid]

R5 of 100 has to be a typo. Please short out C1.

To answer your question about voltage followers a bit. Generally you see the output returned to the - input and the plus input to ground.

That turns out "not EXACTLY right". The best scenario is the same value of what the inverting input sees. It's a real world thing. Using a non-zero value helps cancel the input bias current.

Sometimes you need a resistor from the output of an op amp to the input to say another follower. This won't work without some resistor value between them. It just can't be zero.

Shorting out C1 gives you a gain of 1 + Rf/Rin. Big deal at this point. You don't have to change the circuit to get a gain of 1. (A true voltage follower).

That cap in the first stage is TOO early in the signal chain to have a high pass filter. High pass ::= defined as > DC.

 

[audioguru]

I agree that C1 should be a piece of wire since the first opamp has a gain of only 2 times then it amplifies the input offset voltage only 2 times.

 

[cowboybob]

... Look at my image of the calibration and the noise the Oscilloscope is measuring during calibration message # 213 ...

(My emphasis)

Good job calibrating your 'scope's probe. Noted edges are nice and square, as they should be. I am curious about the noise (most evident on the positive peaks). Wondering where that's coming from, i.e., internal to the 'scope or external into the 'scope's input probe/coax. I also wonder if that noise is worming its way into your circuit.

I might point out, having been in the precision instrumentation calibration field, that there is a diminishing return from worrying overly about how "clean" a circuit's output signal varies from "perfect", given the ultimate (or at least initial) goal: as you noted in post #11 -

... The signal form of the output does not matter ...

and was simply being applied to an LED ladder circuit to indicate the amplitude(s) of the vibration of telescope's support system.

Now, of course, if your intent is the pursuit of gaining experience and knowledge about instrumentation signal conditioning, then by all means continue that pursuit. I'm sure most of ETO's membership have experienced your exact learning curve. And I'll bet you're having fun!

I like your bench setup. Keep in mind that a breadboard, such as you're using, has inherent connection issues that might be aggravated (compromised) by the vibration of the toothbrush. I'd suggest putting the board on a foam mat.

 

[rsfoto]

I like your bench setup. Keep in mind that a breadboard, such as you're using, has inherent connection issues that might be aggravated (compromised) by the vibration of the toothbrush. I'd suggest putting the board on a foam mat.

Thanks for the Tip

Will put the board on foam.

 

[rsfoto]

What is your opinion about the following FW rectifier ?

If I look at that I think I could even use a LM358 with single supply ...

What about taking away the Voltage buffer or is it necessary ?

https://www.linear.com/solutions/1608

 

[rsfoto]

I was searching like Mad for the circuit I am using as FW rectifier and finally fund it here in this forum

https://www.electro-tech-online.com/threads/precision-rectifier-opamp-help.135041/ message # 5

 

[cowboybob]

If I look at that I think I could even use a LM358 with single supply ...

One of ericgibbs' designs - his stuff is pretty bulletproof:

Single supply LM358s. NO idea of current capability, but your stated use would be a voltage driven situation, for which this circuit will work just fine.

I made a minor change to the load resistor to allow the LM358's ability to respond a little easier.

<EDIT>Just replaced the graphic with one having the correct diodes ... Note cumulative losses across the diodes.

 

[rsfoto]

One of ericgibbs' designs - his stuff is pretty bulletproof:
View attachment 96168
Single supply LM358s. NO idea of current capability, but your stated use would be a voltage driven situation, for which this circuit will work just fine.

I made a minor change to the load resistor to allow the LM358's ability to respond a little easier.

<EDIT>Just replaced the graphic with one having the correct diodes ... Note cumulative losses across the diodes.

If I understood correctly the datasheet of the LM358, it can output 40mA

http://datasheets.datasheetpro.com/STMICROELECTRONICS/70013564.pdf

Thanks a lot for that. At the moment I am using Erics circuit with dual voltage supply as well as a TL082CP

Tha 10μF capacitor as well as the 10K resistor I assume is to get a DC voltage and not a alternating DC voltage. That is worth thinking about it because that in my case, it would mean that I get on the LM3914 a fading signal and not a flickering signal of the LED bar.

I do not need current capability as this circuit will drive a LM3914 and there I will apply the supply voltage which is powering the whole enchilada.

The 1N4148 diode is very very similar to the specified 1N914. Checked it already using the page **broken link removed**

 

[KeepItSimpleStupid]

This https://www.vishay.com/docs/81857/1n4148.pdf datasheet confirms the equivelency of the 1N914 and 1N4148 which I knew anyway.

The buffer is kinda necessary, because you don't want the diode to pass current. e.g. No effects from Rds ON.

 

[audioguru]

I like the fullwave rectifier using a single supply opamp. I tweaked it.
The input of an LM3914 is a high resistance so the 22k output resistor of Eric's circuit can be much higher, as high as 1M ohms. Then the filter capacitor value can be much smaller, try 0.1uF to 0.33uF.
Cowboy Bob's circuit has the second diode outside the negative feedback loop of the second opamp, use Eric's circuit that has the diode inside so that its voltage drop is cancelled.
I added a resistor so the input of the second opamp is not floating, then the output level is half the input level.
Of course it needs an input coupling capacitor if the preamp also uses a single supply.

 

[rsfoto]

Why did you kill the gain of the preamp by changing R5 to only 100 ohms?

Typo when redrawing the circuit Sorry

 

[cowboybob]

... Cowboy Bob's circuit has the second diode outside the negative feedback loop of the second opamp, use Eric's circuit that has the diode inside so that its voltage drop is cancelled. ...

He's right. Corrected schematic:

My error, although in my defense, it was a wiring "typo" .

 

[audioguru]

I think the 10uF output capacitor will take too long a time to charge. If the 10k output resistor is changed to 1M then the capacitor value can be reduced so that it charges much quicker.