Sign in to follow this  
ryeeson

Input signal for LM3915. need help.

Recommended Posts

i want to build an noise Level meter using LM3915 0dB-60dB.
i already design my noise detector and also the peramp+20w power amplifier which is work well after testing using the osciloscop.

The output from the 20w PA is connected directly to a cascaded LM3915(Figure 7 in LM3915's data sheet on application hint topic).

the probelm is, some of the LED wont light up  and only detect noise if the noise is closed enough(9-13 cm) to the mic. 
when testing with the osciloscope, the output of the 20W power amp can vary(very sensetive) according to the noise while the noise's source is within 0.7 meter from the mic.

What wrong with my LM3915 driver?? low power input signal????   

http://www.ee.washington.edu/stores/DataSheets/linear/lm3915.pdf



Share this post


Link to post
Share on other sites

Hi Ryeeson,
Years ago I made this circuit and it worked very well. I am currently improving it.

What kind of noise are you measuring, background noise or pink noise?
Do you want to see the peak level to know if the amplifier is clipping when the random peaks all add, or see the average that you can hear and feel? The meter must have enough sensitivity so that the preamp and detector have enough headroom to allow the peaks to be amplified without clipping.

Didn't you use a precision half-wave or full wave average or peak detector circuit recommended in the datasheet? Without one the LEDs will momentarily blink real-time and be very dim and a blur. The detector circuit holds the reading for a moment so that you can see it.

Did you use figure 5 or figure 6 to cascade the LM3915's? Is full-range set at 10V?

Are you operating the LM3915's in DOT mode or BAR mode? In the BAR mode the current is high and might overheat the LM3915's or overload the power supply.

Did you use a supply bypass capacitor directly at the LM3915's circuit? They will oscillate and be unstable without one.

What type of microphone, dynamic (coil and magnet) or electret (condenser with built-in FET that needs DC operating current)? 2-wire or 3-wire electret?

What microphone preamp circuit did you use?

Why use a power amplifier? Peak signals of 10V can come from the preamp.

Please attach your schematic to avoid so many questions about it. ???

Share this post


Link to post
Share on other sites

Sorry For the Late Reply.  :P(i dont have internet connection yet in my house :P)

I want to measure a noise in the library. So if the library user causing a noise, the noise level meter will vary accroding to the noise. Basiclly, i want to build an noise indicator. 

i dont use either the precision half-wave or full wave average or peak detector circuit.

--"Without one the LEDs will momentarily blink real-time and be very dim and a blur"??  ??? so do u have any suggestion which one i need to use?

"Did you use figure 5 or figure 6 to cascade the LM3915's? Is full-range set at 10V?"  So this mean, it is not suitable for a weak signal???  ??? i used figure 6.



I'm operating the LM3915 in  Bar Mode.

Where should i connect the bypass capasitor???  :-\

i used Condenser mic, 2-wire. i' not sure if it has a built-in FET.

i'll attach schematic as soon as i caome back online.

acctuly i used power amp because i was thought the signal is weak and need to be amplify.

Share this post


Link to post
Share on other sites


Here is my ugly circuit after consedering your comment. Sorry again because i use hand drawing(i dont have a software) -- poor guy here  :'(

The peramp stage is taken from some forum. they recomand me that peramp. Any comment?

Share this post


Link to post
Share on other sites

Hi Ryeeson,
I don't use special software to make or modify schematics. I just copy and paste stuff using Microsoft Paint.

Your circuit will have so much gain that it will come out of its wing-wang!
You were using Ron Elliott's mic preamp circuits for FM transmitters that have an awful lot of pre-emphasis (treble boost).
Your 1st stage had a lot of gain then your 2nd stage had not very much gain.
The low frequency response of your circuits were amplifying signals as low as 10Hz.
You don't have a separate supply resistor/bypass capacitor for the resistor feeding current to the mic so you better use a regulated supply to avoid "motorboating".

I removed the treble boost, made similar gain on both stages and reduced the value of the capacitors to give a low frequency cutoff at about 50Hz.


Your half-wave peak detector had a diode which is needed if you use a dual supply.
It had too much gain which allowed its input to go below its absolute max voltage of -0.3V.

I removed the diode and reduced the gain so that the input never goes negative.

I didn't look at your LM3915's which I assume are using the circuit in the datasheet. They are operating at a low current so the LEDs won't be very bright.
I think the opamp will need an input offset voltage adjustment pot as discussed in the datasheet. ;D

post-1706-14279142562335_thumb.png

Share this post


Link to post
Share on other sites

They dont have the MC33171. i'm going uptown.
If i dont used a negative supply, i must place resistor in series with the (+ pin) M3C33171 to ground? or i just buy the MC33172 instead?

What value a the baypass capasitor do you recomand for the LM3915?

btw, i will use the 1N4148 for the half wave detector. Is there is any problem with it?



Share this post


Link to post
Share on other sites

If you use positive and negative supplies for the half wave detector, you can use any opamp but the circuit will need the 2nd diode you had. A resistor from the half wave detector opamp's + input pin to ground will reduce input offset voltage but not eliminate it enough for two LM3915's that show an input voltage as low as 1mV. You need a pot connected to the offset adjustment pins of a single opamp.
An MC33172 is a dual MC33171 and doesn't have offset adjustment pins.

I always use a 100uF supply bypass capacitor for IC circuits and the datasheet for the LM3915 says to use at least 2.2uF at the LEDs if a resistor feeds the LEDs.

The 1N4148 or 1N914 are fine for the half wave detector.

Share this post


Link to post
Share on other sites

Hai Audioguru,

I might try your peramp design if my current peramp could not met the required level.

I went to my local farnell today only to find out that they out of stock for the MC33171N and it quite expensive. My order will only avilable on friday next week.
Until that time, i cannot show you the result.

i am Newbie in electronic project (this is my 1st project).

I plan to connect the LM3915 to a PIC16F84 Microcontroller. Let say if i set 50dB is the limit of allowed noise level inside the library. the pin (at 50+ dB) on LM3915 is connected to one of the interrupt pin on PIC16F84. When the noise level exceed the allowed level, the LM3915 will tigger the interrupt and the PIC now will run a program to blinking an warning LED to the user library 8).

this is a part of my project  ;D

Share this post


Link to post
Share on other sites

Hi Ryeeson,
Farnell in America is called Newarkinone, and an MC33171N costs only $.68US. They have only 30 in stock.

50dB is a low sound level. If your project uses the peak detector then it will cause an alarm when people walk on a bare floor or if a book is dropped. Maybe you should make the averaging detector so there are fewer false alarms.

Do you have a sound level meter to calibrate your project?

Share this post


Link to post
Share on other sites

Your circuit with the LM3915 will be a sound level meter after it's mic sensitivity and circuit's gain is calibrated with a real one.
The LM3915 is just a voltmeter with a logarithmic scale.

Go to a sound system installation company and set yours to read the same as theirs.
RadioShack also has a sound level meter.

Share this post


Link to post
Share on other sites

Thanks again. Another question is, can we change the Mic's sensitivity?
when come into calibrate topic, i know nothing..  :( do you mean i need to go to sound system company and ask them to calibrate my circuit?  ???

Share this post


Link to post
Share on other sites

Another question is, can we change the Mic's sensitivity?

All circuits that I showed have a "volume control" to set the circuit's sensitivity.

when come into calibrate topic, i know nothing..  :( do you mean i need to go to sound system company and ask them to calibrate my circuit?  ???

Yes, but you adjust the sensitivity of your circuit to match the accurate reading of their sound level meter when a sound is playing, and your mic is the same distance and location from the sound source as theirs. Set theirs so it doesn't use its audio filters.

My project uses a single LM3915 but has a slow automatic adjustment to keep its display showing a wide range of sound levels. It reduces loud sound measurements so they don't cause the display to show continuously at max. It boosts quiet sounds so it shows them instead of nothing. I have my LEDs operating at 26mA so they are really bright. I haven't installed a power jack yet so I'm using a rechargable 9V battery. It discharges in about 1 hour when the LM3915 is in the bar mode, about 6 hours in the dot mode. ;D

Share this post


Link to post
Share on other sites

Thanks...  ;D ;D ;D

I just need to calibrate my circuit using the 2 volume control.  ;D ;D

i try to calibrate it first by myself before go to a company. may be i can use the computer microphone since there is a built in software for displying the noise level??
I only can calibrate after i finished the circuit.. damm have to wait until 30 dec.

btw, i dont need very accurate Sound Level Meter.

Merry Christmas pal..

Share this post


Link to post
Share on other sites

You have the power supply pins connected correctly. Add a supply bypass capacitor.
The same power supply can be used by all ICs or use an MC33172 dual opamp or MC33174 quad opamp. They have different pins.
The MC3317x series operate from a supply voltage from 3V to 44V.
Your circuit doesn't have a DC reference voltage nor negative feedback. When the input signal swings negative, the opamp's input will break.

Share this post


Link to post
Share on other sites

I already test the cricuit but its wont working. The circuit can't detect any signal  :(
what wrong here?? i thinks there wasnt have enough gain.

i used DC power supply(singgle supply + and ground). i revert back to the original circuit ant its working. but the second stage of the peramp seem dont have any benefit.

can u help me? may be by showing your draw that you used before??

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this