I want to know the principle of operation of the analog device AD534KH as ( i ) a divider and (ii) a sine of the input voltage. So can you give me an explanation on this matter?

Thank you.

Hi,

I want to know the principle of operation of the analog device AD534KH as ( i ) a divider and (ii) a sine of the input voltage. So can you give me an explanation on this matter?

Thank you.

Hi Ivan,

Thanks for the question. The basis of an analog multiplier or divider is rooted in the exponential relationship between transistor base and emitter current, and the properties of log addition and subtraction. ADI has an application guide that applies to all our analog multipliers, so rather than attempt my own 'tortured' version of the theory of operation of these interesting parts, I'm attaching a link to the best reference material we have. It was written and edited by the folks who invented the parts.

Let me know if there any questions.

Best,

js

Hi,

Thanks for your reply. In the Multiplier Application guide, they didn't write in details about AD534 as a sine function generator. I have few question to ask about this particular application of this IC:

1. Here they have written that the input is 0V to +10V. So if we use the input as the sine that has both the +ve and -ve voltage whether this IC will work or not?

2. The output is (10V)sin(theta). But theta=(pi/2)*(input/10). Now if the input is 10V,then theta will be (pi/2) and sin(theta) becomes maximum. What if the input is in the range of 0 to say 9V because theta will not become (pi/2) in this range and hence it will not reach its maximum value?

3. The 10V in front of sin(theta) and the 10V that divides the input (present in theta) whether they are the same or not?

Thank you.

Hi Ivan,The 10V located at the Y1 input is the upper end of the input range; the 10V at the OUT is a scale as shown in table 1. The scale factor is a convenience in multiplier design which ensures the range of the output voltages stays within the range of the potential input voltages.

The scale factor value is a constant built into the design for most multipliers.

Best,

js

Hi,

Thanks for your reply. In the Multiplier Application guide, they didn't write in details about AD534 as a sine function generator. I have few question to ask about this particular application of this IC:

1. Here they have written that the input is 0V to +10V. So if we use the input as the sine that has both the +ve and -ve voltage whether this IC will work or not?

2. The output is (10V)sin(theta). But theta=(pi/2)*(input/10). Now if the input is 10V,then theta will be (pi/2) and sin(theta) becomes maximum. What if the input is in the range of 0 to say 9V because theta will not become (pi/2) in this range and hence it will not reach its maximum value?

3. The 10V in front of sin(theta) and the 10V that divides the input (present in theta) whether they are the same or not?

Thank you.

Hi,

Regarding the scale factor of 10volts is fine. What about the 10volts that is present in the formula theta=(pi/2)*(input/10 volts)?

Now if the input is 10 volts then we can get a maximum output equal to 1(excluding scale factor). But when the input is say 2volts, then the output is not reaching its maximum value.... However we can see that no matter what the input(sine) is, the output will always be a sine and it reach its maximum value. If the formula is in the form of (amplitude)*sin(theta) where theta=(pi/2)*(input/input), then no matter what the input is, the output will reach its maximum value. However , this is not the formula....

Hi Johny,

That is a very large question - it all depends on your requirements, so I have to make assumptions - I assume you mean accuracy of frequency vs. input voltage.

BTW the AD639 is obsolete so I wouldn't want to spend time responding with comments on obsolete parts. This leaves just the AD534, which has a an application circuit you can use for a reference. Here the accuracy can be estimated using dc error values such as Vos, multiplier accuracy, etc. I don't have such a study to pass on to you but I suspect you get the idea.

If you want the most accurate method I would recommend a DDS (direct-to-digital synthesizer). I found a Wiki reference which looks pretty good to me, ADI does make such devices and I've attached a link to a selection guide. After you study the info from Wiki, check out the ADI product data sheets. I am not the apps person for DDS products but you can inquire as to the right person on our website.

Direct digital synthesizer - Wikipedia, the free encyclopedia

http://www.analog.com/en/products/rf-microwave/direct-digital-synthesis-modulators.html

Best,

js

Hi,

Thanks for your reply. Actually the problem that I'm having with AD534 is that it does not work according to the data sheet for higher input voltage. The input voltage according to the data sheet can have a maximum of 10volts. But in my case once I reach a voltage of even 500mVolts, the output voltage is being clipped and if I reduce the SF from 10V to 3V also,that doesn't help.(here I have taken the input voltage as sine(x) and the output voltage as sine(sine(x)). The AD534 works well when the input voltage is below 500mVpp).

Hi again,

If you send me a schematic I'll see if I can spot something.

Many if not most multiplier difficulties have to do with the input coupling - which must be dc. The inputs must have a dc path, with the dc level at 0V, otherwise the dc at the input will float high or low.

Another common error is the by-pass capacitor (or other power supply decoupling) which must be transparent (0ohms impedance) at the lowest frequency of operation. The power supply must be well regulated as well.

Probably the most common of all circuits that are not accompanied by a schematic is an error in the physical implementation of the circuit.

Yet another issue arises when people report a set of conditions which are the result of a simulation. I always respond assuming real circuitry, as simulations introduce it's own 'operating environment', with which I'm not prepared to support.

Best.

js

Hi,

Thanks for your feedback. Here I will attach a schematic diagram of AD534 as a sine function generator.

In my experiment, the input signal is a sin(x) and the output signal is the sin(sin(x)). But only with an input signal of less than 500mVpp, I'm getting a better results. Beyond this input voltage, the output voltage got clipped.

(here the input signal is ac). However, the range of the input signal which was given from the data sheet ranges from 0 to 10Volts.

Hi Ivan,

Thanks for the question. The basis of an analog multiplier or divider is rooted in the exponential relationship between transistor base and emitter current, and the properties of log addition and subtraction. ADI has an application guide that applies to all our analog multipliers, so rather than attempt my own 'tortured' version of the theory of operation of these interesting parts, I'm attaching a link to the best reference material we have. It was written and edited by the folks who invented the parts.

Let me know if there any questions.

Best,

js

http://www.analog.com/media/en/training-seminars/design-handbooks/ADI_Multiplier_Applications_Guide.pdf