Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Question for the Oscilloscope Experts

Status
Not open for further replies.
If you don't remember π (3.1415926...), 22/7 =3.1428571429 is a good approximation.

Some older units used were MFD for micro-farad and MMF for mico-mico Farad
 
If you don't remember π (3.1415926...), 22/7 =3.1428571429 is a good approximation.

Some older units used were MFD for micro-farad and MMF for mico-mico Farad

Thanks KISS,

Yes PI is nowadays more then the famous 3.1416........................

201203peanut-butter-and-strawberry-pie-for-pie-day.jpg
 
Hi AG,

I have been thinking about your emphasis in Frequency in regard to my circuit as you are coming from the audio ? part of the electronics.

So from my point of view, why try to filter out something that really for my setup does not bother me ? Do the eventually measured high frequencies of my circuits harm the measuring of my low frequencies ?

My guts (¿second brain? if I still have a bit :eek: Kidding) tell me that Low frequencies do make higher amplitudes (see Earthquakes) then High frequency. Not for nothing the Jericho Trumpets were low frequency trumpets so they say :happy:

Well below an image of what I am trying to achieve. First real world findings perhaps starting Tuesday 22 or Wednesday 23 ... Then we have first real world measurement on my telescope piers and can continue and exchange ideas about improving the circuit :cool:

Thanks Rainer

Vibrar_Amplitude_01.jpg
 
The abcence of a coupling cap in the first stage more mimics the app note (gain=2 or A=2). Lower frequencies probably affect the telecope more. There is circuit called a peak detector.
 
You COULD find out that for best versatility, to break the vibration into bands and then use a bar graph with say a peak detector for each band, sort of like a "color organ".

You'll easily figure that out with the FFT and some experimentation.
 
Vibrations are very low frequencies that cannot be heard (maybe a rumbling sound). Your sensor datasheet says it is good up to about 40Hz because higher frequencies cause it to resonate making a high amplitude at 75Hz that you do not want.
Some of your circuits cut low vibration frequencies that you want to see and nothing in your circuits is a lowpass filter to reduce the 75Hz resonance frequency and higher frequencies that cause the sensor to resonate.

Your new 'scope photo shows 3 cycles every 50ms which is only one frequency and it is at exactly 60Hz which might be electrical hum picked up by your wiring or caused by a synchronous AC motor driving the telescope movement.
 
Your new 'scope photo shows 3 cycles every 50ms which is only one frequency and it is at exactly 60Hz which might be electrical hum picked up by your wiring or caused by a synchronous AC motor driving the telescope movement.

Hi AG,

Well every RC filter I have tested has had as result a severe loss in sensitivity and in regard to above the sensor was and has never been up to now connected to a running scope and by the way, all telescopes mounts work with servo motors or stepper motors driven by DC ...

I will look through all messages in this topic and see where are the RC filters. I found a Japanese site **broken link removed** in which one can calculate the values of different filter types. What is your opinion about that site ?

BTW I just arrived at my Observatory and while cleaning up a bit the mess I have on my desks I found 2 Op Amp INA 122 :wideyed: which I already used once for making a Cloud sensor using a Peltier element.

As soon as I have the first tests on my columns I will come back.
 
Where is the 60Hz coming from?

The Japanese site and many other sites show how easy it is to make an RC lowpass filter and a CR highpass filter. When the filter is made properly then it has no attenuation to the frequencies you want but it cuts the frequencies you do not want. You did it wrong then it reduced your sensitivity.

An INA122 is not an opamp. Instead it is an instrumentation amp. It is a differential amplifier that you do not need.
 
The 60 Hz should really disappear or be severely attenuated when placed in a metal box and use twisted pair shielded cable to connect your sensor.
 
The Japanese site and many other sites show how easy it is to make an RC lowpass filter and a CR highpass filter. When the filter is made properly then it has no attenuation to the frequencies you want but it cuts the frequencies you do not want. You did it wrong then it reduced your sensitivity.

If I did put attention on what I have read, :confused: my RC filter should be like the image below and wired where I have it ... Not tested yet but will do

44Hz_RC_3600_1uF_1.jpg
 
One more find and I guess it is the mentioned resonance frequency of the Piezo as specified in the data sheet. Not yet applied the 44Hz RC filter

I hope I do not bore you ...

Obs_floor_01.jpg
 
Your piezo vibration sensor has an extremely high impedance. The datasheet shows loads of 10M, 100M and 1G ohms. Your low impedance 3600 ohms filter is killing its output level.
The sensor cannot drive a lowpass filter but an opamp can. The filter should be connected between the first opamp and the second opamp in the preamp.

1uF is a fairly high value but the signal to it is AC so it cannot be an electrolytic type, it should be a film type. Why 1uF? Why not use 0.1uF if you change the resistor value also 10 times.
A 0.1uF film capacitor is small and inexpensive and connect it with a 36k resistor. It is the simplest lowpass filter called first-order. At 44Hz the level will be 0.707 times the very low frequencies levels. But the 75Hz resonance level is 4 times the levels of very low frequencies so maybe you need a second-order active filter instead. With a cutoff frequency of 37.5Hz (half of the 75Hz resonant frequency), a Butterworth second-order active lowpass filter will have an output level 0.707 times at 37.5Hz and a level of 0.25 times at the 75Hz resonant frequency. Then the resonance will not peak the response. The filter can be a Sallen and Key type.
 
1uF is a fairly high value but the signal to it is AC so it cannot be an electrolytic type, it should be a film type. Why 1uF? Why not use 0.1uF if you change the resistor value also 10 times.
A 0.1uF film capacitor is small and inexpensive and connect it with a 36k resistor. It is the simplest lowpass filter called first-order. At 44Hz the level will be 0.707 times the very low frequencies levels.

Thanks, that is what differentiates a Pro = You from a Dummy = Me :confused: and where is the limit and how do I decide what values to take. The next step could be a 0.1uF and a 360k resistor but what means that for the 1st stage to the 2nd stage ...

OK, I made a test with ~ 34Hz cut off using a 100KΩ resistor and a 47nF capacitor (not film = ceramic) because that is what I have here Sorry

Below some images of the outcome, circuit, oscilloscope with FFT at ~20.5Hz but that peak is not always there(purple line) and calculation. On my table I have integrated 2 fans in order to pull out the heat of the PCs.

RCFilter_100K_47nF_02.png
RCFilter_100K_47nF_01.png
RCFilter_100K_47nF_03.png

Edited at 20:15 hours. The preamp has now a gain of 21 as if I would have left the gain of 2 there was signal at the output ...
 
Last edited:
Your 34Hz single-order lowpass filter is too simple since it cuts the frequencies you want (34Hz to 40Hz) but does not cut the 75Hz sensor resonance enough. It has a response that is -3dB (0.707 times) at 34hz and is -6dB (0.5 times) at 68Hz. Then 75Hz is about -7dB (0.45 times) but you need at least -12dB (0.25 times).

The datasheet for the sensor shows that its load resistance cuts its low frequencies making a highpass filter. Your 4.7M produces a cutoff frequency of 138Hz so all your vibration frequencies have their levels reduced. You need a load of 65M for a 10Hz cutoff frequency. I have never seen a preamp with such a high input resistance.
 
If you don't look at anything else here **broken link removed**
Look at the table on PDF page 3, figures 1.1 and 1,8

Also see here **broken link removed** for how to do a Salen-Key design.
 
Hi AG and KISS,

Thanks. Did not answer before as I needed some time to assimilate the Filter question. I even dreamed about Low and High Pass filters ...

I am dreaming of a white Christmas ... or Low Pass Filters :arghh: :arghh: :arghh:

So in an easy way, If I chain 4 of my LP filters using the 100k and 47nF caps I get a 4th order LP filter, but with what sort of problems ?

I have not yet soldered anything onto a perforated board as day by day this circuit is changing

:confused: :confused: :confused:
 
Last edited:
Question:

If I use the Piezo on a long distance. Where should I put the Resistor. Near the piezo or on the board ?

Thanks
 
If I chain 4 of my LP filters using the 100k and 47nF caps I get a 4th order LP filter, but with what sort of problems ?
That will be a horrible filter. Its output will be -12db (0.25 times) at the cutoff frequency and it will be very droopy and still reduce almost all vibration frequencies. A Sallen and Key filter uses an opamp to boost the response at the cutoff frequency so the cutoff is very sharp above it and it does not attenuate lower frequencies.
 
Thanks AG. I have been taking a small rest and doing what I think I can do a bit better then electronic circuits :happy:

Imaging the night Sky like this one from yesterday. A Mosaik from the Moon made out of 12 images and each image made out of a video of 250 images each
 
Last edited:
Amazing. You can actually see the topography of the meteor craters at the top and down the left side :woot:.

That's quite a system you've got there, rsfoto.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top