Frequency To Voltage Converters

[miracleblue]

Hello everyone,

 

I am using LM331 Precision Voltage-to-Frequency Converters. In the data sheet, there's an application for the purpose of Frequency-to-Voltage Converters. Below is the link to the data sheet:

LM331 Data Sheet

 

Using a frequency modulated laser to send information, on the receiving end, there will be a photo-electronic sensor, a photo-transistor, LF14F1. I've tested the modulation scheme on both sending and receiving ends of transmission. They works perfectly. The square waves that is sent through the laser gets received as the original signal with all its properties (Vpeak-to-peak, frequency). However, I would like to convert the frequency on the receiving end to be in terms of voltage. Thus, adding a LM331 can make it.

 

According to the data sheet, it can take any input as high as 10 kHz FULL-SCALE, and outputs 10 V FULL-SCALE, under 15V power supply. I've tested this one, and it works like magic.

Please look at Figure 1

 

However, I need to make it to work for a minimum of 1 Hz - 2 kHz at max at the input and outputs to be anywhere between 0-5 V, since I am using a 5-V regulated power supply.

 

There's a formula to determine the Vout at the bottom of page 8 of the data sheet or to the side of Figure 1.

 

I'm new to engineering design, and I just wanted to know, which of the Resistors/Capacitors should I change? My guess is to get to lower frequency, I need to tweak the Rt and Ct values, and to change the output voltage, I should tweak RL and Rs. Please help me to confirm this.

 

I would appreciate it if anyone can help me. Thanks in advance.

[RICHARDBATTY]

To get this straight, you just want to convert a square wave into a constant volts out thats related to the frequency of the wave.

 

How much tolerance can you accept including ripple.

[miracleblue]

RICHARDBATTY, thank you for responding.

 

Here's the desired conversion factor:

sending 2 kHz ---> corresponds to 5 V

 

Example conversion:

810 Hz ---> would correspond to 2.025 V (if my calculation is not wrong)

 

Well, not to worry about the ripple. I am going to use the same one as given in the data sheet as it gives a pretty stable output. Under oscilloscope, it almost resembles a straight line. However, if you think that the filtering capacitor should change, try to make it within 10% error, including ripple. That is already pretty good.

 

Let me know if it is not possible.

[reverse]

I spose you know that a square wave put into a normal tarnsformer will be smoothed by the magnetic coupling.

and you can pick the voltage by the ratio of the windings.

 

but it still need the usual Capacitor diode stuff to get it dead flat.

 

might be to crude for you, but just in case it is some help.

[YT2095]

the problem with the transformer is that it`s impedance at a certain frequency will peak, it`s not a flat enough response

 

also, altering the freq on a square wave will still only get you a constant voltage as the Mark-Space ratio is still 50/50 regardless.

[miracleblue]

@ reverse:

No, I don't know that a transformer can do that. Thanks for sharing the knowledge. I will try it when I have the time and resource. Kinda low on budget for this project alone.

 

@ YT2095:

I'm sorry to ask this, but I have never came across the term "Mark-Space ratio" before. If you don't mind, can you point me to some useful links that explains that terminology. It seems pretty sophisticated.

 

An update on the circuit that I've tested, if the filtering capacitance change from 1 uF to 0.1 uF, the ripple output is really showing on the scope, which is expected. I've tried to use different values of resistors as well, but I can't find an obvious pattern.

[YT2095]

sure, here`s a few: http://www.electrotherapy.org/electro/pulsedshortwave/pulsed.htm

 

http://www.eleinmec.com/article.asp?28

this one even has a diagram.

 

or just do this search: http://www.google.com/search?q=%22mark+space+ratio%22&hl=en&lr=&rls=GGLD,GGLD:2004-10,GGLD:en&start=10&sa=N

 

this one will also help: http://www.google.com/search?sourceid=navclient&ie=UTF-8&rls=GGLD,GGLD:2004-10,GGLD:en&q=%22pulse+width%22

 

hope some of these help.