problem generating noise in a circuit

[niloy2k]

Hi I have previously posted here but did not get any replies. I wanted to know whether there is anyway of producing a noise source in a breadboard circuit. I have already tried using a zener diode in a reverse biased direction but it did not produce any results. Can anyone please help me out? Thanks

 

[rockycheng]

If you connect the power supply to that of a class-D amplifier working at high output power, don't use decoupling capacitors, it will be noisy enough

 

[niloy2k]

Hi I am not quite sure what you wanted me to do .... could you be a bit more specific please ? I have attached a picture of what I am trying and i am quite convinced something is not quite right in the circuit, could you please help me out ?

 

[rockycheng]

Could you pls tell me your idea? You want to generate a DC voltage (with noise) at node 4? Then use the LM741 to amplify the noise signal?

I think you have to block the DC voltage and add input resistor to the amplifier, otherwise the output of the amplifier will only be the negative supply voltage.

 

[zorro]

Hi niloy2k,

the circuit tries to amplify the noise generated by the reverse biased Zener diode. But in order to not saturate the OpAmp and fix the gain, a C in series with a R should be placed between the point with the mark of node 4 and the - input of the OpAmp.
The power supply for the OpAmp should be split (+ and -). You could use V1 only, but in this case the + input ot the OpAmp should be connected to a mid-pont voltage (using a voltage divider).

A very good method for generating noise is to filter a pseudo random sequence generated by a LFSR (Liner Feedback Shift Register). A limitation is the frequency band (up to about 1/10 of the clock frequency), but i guess that it is good for your case. The circuit with the OpAmp has also limited bandwidth.
Regards

Z

 

[niloy2k]

I am sorry for not being clearer earlier. I am working on my final year project which is named DESIGN AND BUILD OF A LOCK IN AMPLIFIER. My main aim of the project is to extract and amplify very weak signals buried in very high noise. I have pretty much built up the circuit for the extraction process but the only thing i am lacking now is a noise source. So basically I will have a small signal within a larger noise signal. The noise can be of any type as long as i can prove my circuit works for the extraction process. Could you people please help me? Thanks a lot for bearing with me and the previous responses

 

[zorro]

The psudo random noise generated with a LFSR has good characteristics: its distribution is very nearly gaussian if the bandwidth of the filter is low compared with the clock frequency, and the noise level and power spectral density are very predictable (from DC), making it good for measurement purposes.
Regards

Z

 

[FvM]

From your project description, I assume that you have some knowledge of analog electronics. So you should be able to understand the comments about missing AC coupling in your Z-diode noise generator. Also the Z-diode must have a suitable zener voltage within supply range, e.g. 5.6 or 6.8V, C27 can't work. But the circuit should work generally. The zener-diode supply voltage should have a good filtering, for low frequency noise, an ausxilary voltage regulator may be meaningful.

I have used similar noise generators for audio purposes. A digital LFSR (pseudo random) noise source can be used as well, it shows a periodic sequence of a defined length, e.g. 10 sec.. This property can be a disadvantage for measurements that require true randomness but may be an advantage for others.

 

[niloy2k]

doesnt the zener diode c27 has a characteristic breakdown point of 2.7v? I thought that is the case so providing a voltage source above it will cause the avalanche effect, i gues i was wrong. Also if I do use a shift register, do i need to program it since its a digital device? Can you please give me models of other zener diode that i can use? thank you so much for the help.

 

[E-design]

**broken link removed**

 

[niloy2k]

I have already tried that circuit from maxim on a breadboard.... doesnt give me any noise output when i chek on the oscilloscope ... i have also tried simulating the circuit in multisim.... but no luk

 

[E-design]

I don't think multisim have the proper noise models for the components. The noise generated will be of low amplitude so you really need a high gain wide band amp to amplify that.

 

[niloy2k]

I tried using two 741 opamps bak to bak .... each havin a gain of 100 .... so technically dat shud give me 10000 gain, please do correct me if i am wrong since i am not a pro ... was the method dat i used wrong ?

 

[FvM]

doesnt the zener diode c27 has a characteristic breakdown point of 2.7v

You have shown a MultiSim part. Check with the real part datasheet. BZX85-C2V7 would be the usual part name of a 2.7V device and -C27V for 27V. As far as I remember, higher voltage (5 - 8V) gives better noise generator performance, but you may want to try. I also think, that analog simulators as PSPICE (or MultiSim) don't model avalanche breakdown noise correctly. So you won't get meaningful results in simulation.

 

[niloy2k]

I jus wanted to say thanks a lot for the help .... the circuit is givin me proper results now .... i am gettin noise but its relatively reallly smalll .... i am only allowed to use the basic 741 op amps .... is there any other way to amplify the noise with a very high gain ? once again thanks I really appreciate the help