Originally posted by Papabravo@Mar 28 2006, 07:57 PM
The AD633 is just such a four quadrant multiplier.

[post=15550]Quoted post[/post]​

What does a Four Quadrant Multiplier mean? what other types of multipliers are there ?

 

Originally posted by karunanithi_pricol@Mar 13 2006, 04:31 AM
Hi Everyone
I have to design a frequency doubler circuit. In which the input frequency will be coming from a speed sensor(Hall effect) and i have to double the frequency with the same voltage level as the input.

* Input frequency range will be 0 to 100 Hz square wave

Can anyone suggest me a cost effective solution for this requirement?

Regards
Karunanithi.R

[post=14927]Quoted post[/post]​

Greetings karunanithi_pricol,

This may be a bit of an oversimplification from what you originally requested but have you given any thought to simply doubling the number of magnets on the rotating entity whose speed you are trying to measure? If feasable, this approach would double the frequency without having to resort to the additional electronics.

hgmjr

 

Originally posted by hgmjr@Mar 29 2006, 04:54 PM
Greetings karunanithi_pricol,

This may be a bit of an oversimplification from what you originally requested but have you given any thought to simply doubling the number of magnets on the rotating entity whose speed you are trying to measure? If feasable, this approach would double the frequency without having to resort to the additional electronics.

hgmjr

[post=15598]Quoted post[/post]​

Or if that isn't possible, two OR'ed Hall Effect sensors.

 

Originally posted by Ron H@Mar 29 2006, 06:09 PM
Or if that isn't possible, two OR'ed Hall Effect sensors.

[post=15599]Quoted post[/post]​

RonH,

I considered the possibility of "or"ing the output of two sensors but I was not clear on how I would do that with two sinudoidal waveforms.

I tend to agree with your assessment that devising a phase shift circuit that could provide a uniform 90 degree phase shift for every frequency between 1 Hz and 100 Hz would be a formidable undertaking. Not to say that it is impossible, since I have learned over the year that as soon as I say there is no way to do something, someone comes along and does it.

My experience has been that the technique of analogue multiplication to double frequency works best when the frequency being doubled is constant.

hgmjr

 

Originally posted by hgmjr@Mar 29 2006, 05:32 PM
RonH,

I considered the possibility of "or"ing the output of two sensors but I was not clear on how I would do that with two sinudoidal waveforms.
<snip>
hgmjr

[post=15600]Quoted post[/post]​

Good point. I hadn't thought of that. But, I don't see how you can get a sinusoid from one magnet on a shaft, sensed by a Hall effect device. Has Karunanithi said that this is the case? It seems to me that you would need multiple alternating magnetic poles around the circumference of the rotating device in order to get some semblance of a sine wave.

 

Originally posted by Ron H@Mar 29 2006, 07:23 PM
Good point. I hadn't thought of that. But, I don't see how you can get a sinusoid from one magnet on a shaft, sensed by a Hall effect device. Has Karunanithi said that this is the case? It seems to me that you would need multiple alternating magnetic poles around the circumference of the rotating device in order to get some semblance of a sine wave.

[post=15603]Quoted post[/post]​

Ronh,

You raise a very good point about Hall Effect Sensor output. I recall that he did say that the output was sinusoidal. Maybe karunanithi_pricol can shed light on how the Hall Effect sensor in his application is producing a sinusoidal output rather than a pulsed output.

hgmjr

 

Originally posted by Mazaag@Mar 29 2006, 04:26 PM
What does a Four Quadrant Multiplier mean? what other types of multipliers are there ?

[post=15589]Quoted post[/post]​

The four quadrants are from Cartesian Coordinates and are defined by the aritmetic signs of x an y
Quadrant 1. x is positive, y is positive
Quadrant 2. x is negative, y is positive
Quadrant 3. x is negative, y is negative
Quadrant 4. x is positive, y is negative

There is such a thing as a two quadrant multiplier.

 

Originally posted by Papabravo@Mar 30 2006, 06:43 AM
The four quadrants are from Cartesian Coordinates and are defined by the aritmetic signs of x an y
Quadrant 1. x is positive, y is positive
Quadrant 2. x is negative, y is positive
Quadrant 3. x is negative, y is negative
Quadrant 4. x is positive, y is negative

There is such a thing as a two quadrant multiplier.

[post=15605]Quoted post[/post]​

Hi all
What about this idea?

Step1: I can convert sine wave to square (using IC4069 and passives)

Step2: I can capture the square wave frequency through controller and i can
double the frequency of square wave at the controller output.

Step3: I have to convert this square wave back to sinewave. Now the output sine
wave frequency will be double of the input sine wave frequency.

will this work out?

Can some one help to convert square wave to sine wave?

Regards
Karunanithi.R

 

Some sort of low-pass filter. A square wave consists of many sine waves.

 

Hi,
I was originally thinking of the full wave bridge rectifier idea, too. The nice thing about it is that you get half sinusoidal waveforms which you should then be able to filter after passing through an amplifier for buffering. Then you can try running that waveform through a gain controlled amplifier. It needs to have automatic gain control because the higher the frequency, the weaker the input. Dunno if it would work.

 

Originally posted by karunanithi_pricol@Apr 8 2006, 10:21 AM
Hi all
What about this idea?

Step1: I can convert sine wave to square (using IC4069 and passives)

Step2: I can capture the square wave frequency through controller and i can
double the frequency of square wave at the controller output.

Step3: I have to convert this square wave back to sinewave. Now the output sine
wave frequency will be double of the input sine wave frequency.

will this work out?

Can some one help to convert square wave to sine wave?

Regards
Karunanithi.R

[post=15968]Quoted post[/post]​

You could use an active bandpass filter.

http://www.captain.at/electronics/active-filter/

 

I saw recently on the CD4047B data sheet that it can be used for frequency doubling. I think it is because it can trigger on both positive-going and negative-going waveform transitions. The output would be taken at the oscillator output pin instead of Q or Q not.

 

Originally posted by Papabravo@Mar 28 2006, 06:26 PM
Then the analog multiplier, such as an Analog Devices AD633 is the ticket you need.

[post=15525]Quoted post[/post]​

Hi Everyone,
I have tried with AD633 for my application ; this circuit doubles the I/P frequency from (10-100Hz) to (20-200Hz) but the problem is I/P sine wave amplitude is constant 5Vp-p and the output sine wave amplitude is getting varied with response to the input frequency variation.
Can anyone suggest me a solution for this? ie how to maintain the amplitude of output waveform constant irrespective of I/P frequency variation?


Thanks & Regards
Karunanithi.R

 

Originally posted by karunanithi_pricol@Jun 2 2006, 08:33 PM
Hi Everyone,
I have tried with AD633 for my application ; this circuit doubles the I/P frequency from (10-100Hz) to (20-200Hz) but the problem is I/P sine wave amplitude is constant 5Vp-p and the output sine wave amplitude is getting varied with response to the input frequency variation.
Can anyone suggest me a solution for this? ie how to maintain the amplitude of output waveform constant irrespective of I/P frequency variation?
Thanks & Regards
Karunanithi.R

[post=17476]Quoted post[/post]​

Would it be possible to post the schematic of your circuit?

hgmjr

 

Originally posted by hgmjr@Jun 4 2006, 01:37 AM
Would it be possible to post the schematic of your circuit?

hgmjr

[post=17480]Quoted post[/post]​

Hi hgmjr
Thanks for your response, pls find the attached pdf contains the schematic which is taken from the AD633 data sheet. i have used 10K for 'R' and 0.1uFD for 'C'. I put a potential divider between pin7 and ground.


Regards
Karunanithi.R

 

Thanks for posting the schematic.

In looking at the application information on the AD633 when used in a frequency doubling circuit, I belive that the example provided is based on a constant frequency on the input.

The reason that you are experiencing a variation in the amplitude of the frequency doubled output is due to the fact that the RC network being used to introduce the necessary 90 degree phase shift between the x-input and the y-input produces a phase shift that varies with the input frequency. An RC integrator produces a 90 degree phase shift at one frequency only. This frequency is determined by the value of RC.

To obtain an output that has little or no amplitude variation will require a circuit that will that provides a constant phase shift of around 90 degrees that is independent of the input frequency. That way you will be able to feed your reference signal into the x-input and the 90 degree phase shifted version of the reference input into y-input.

I can't think of a circuit that has that feature right off. Maybe one of the other members can suggest one. In the meantime I will need to do a little research into such a circuit and get back to you.

hgmjr

 

Rather than attempting to maintain a phase shift of 90deg across frequency (which is difficult), why not just amplify the output with some sort of agc/limiter to maintain the required 5V level? This is certainly an easier solution.

Why do you need to maintain a constant output voltage?

 

Originally posted by n9352527@Jun 5 2006, 03:46 PM
Rather than attempting to maintain a phase shift of 90deg across frequency (which is difficult), why not just amplify the output with some sort of agc/limiter to maintain the required 5V level? This is certainly an easier solution.

Why do you need to maintain a constant output voltage?

[post=17521]Quoted post[/post]​

This signal is going to be processed by controllers and it requires constant amplitude. Can anyone suggest me a agc/limiter circuit for my application?

Karunanithi.R