100MHz Frequency to voltage converter

[bio_man]

Hi,

Looking for possible ideas to convert high frequency signals (~100MHz) to voltage. Any suggestions?

--- Updated Feb 10, 2022 ---

Also, is there any commercial ICs can do this? I found only few with max of 1MHz! Usually, Frequency to voltage (FVC) is the in range of few tens of kHz

 

[Warpspeed]

Yes indeed, and you first divide your frequency down by say a factor of a hundred.
So 100 Mhz divides down to 1Mhz, that goes through a 1Mhz F to V converter so 1.00 Mhz = 10.00volts for example.
92.5 Mhz divides down to 925Khz then to 9.25 volts and so on.

 

[bio_man]

Yes indeed, and you first divide your frequency down by say a factor of a hundred.
So 100 Mhz divides down to 1Mhz, that goes through a 1Mhz F to V converter so 1.00 Mhz = 10.00volts for example.
92.5 Mhz divides down to 925Khz then to 9.25 volts and so on.

thanks

how I can the divider for high frequency and btw this is a sinusoidal signal.
I thought of the 555 timers or couple of flip flops but they may not be good for high frequency like 100MHz, right?

 

[danadakk]

Standard logic family dividers, or CPLD would do the job. Look for Tpd ~ < 5 nS.
If amplitude of sine too low for dividers an amp to signal condition input. Of
very fast comparator with low overdrive requirements to get speed.

Logic families (AHC) -

https://www.ti.com/lit/sg/sdyu001ab/sdyu001ab.pdf

Comparators -

https://www.ti.com/amplifier-circuit/comparators/products.html#p83=0.000225;0.005&p116=Push-pull

Example -

Very Simple PC Frequency Counter Works Up To 100MHz

We all use 74 logic in our projects as general purpose logic interfacing glue. These chips have become as ubiquitous as a general-purpose op-amp, or even as passive components. In most cases we&#82…

hackaday.com

Regards, Dana.

 

[Warpspeed]

There are a family of chips known as "prescalers" that divide high frequencies down by a fixed division ratio.
A 100:1 prescaler with a frequency rating of 100 Mhz plus would be ideal.
A quick Google search turned this up:
https://www.donberg.ie/catalogue/semiconductors/semiconductors_c-d/ds_8629n.html

That same chip or other similar or better devices should be available in your own country if you do a bit of internet searching. If you check the data sheet, most of these devices are designed to work from sine waves within a specified amplitude range.
https://pdf1.alldatasheet.com/datasheet-pdf/view/144714/NSC/DS8629N.html

 

[betwixt]

What range of frequencies are you looking to convert?
The problem with dividers is they don't only divide the frequency, they divide the difference in frequency. The voltage returned is therefore likely to be lower. That may not be an issue if you are converting a wide range of frequencies to voltages but if your range is relatively small, it may be more profitable to down-mix to a lower base frequency so the range stays the same.
For example:
range 100MHz to 101MHz is a 1MHz shift (~1% change)
down mix to remove 90MHz the range becomes 10Mhz to 11MHz, still 1MHz shift but now ~10% change.

Brian.

 

[Warpspeed]

Prescalers divide digitally with counters, they don't mix down.
The percentage shift remains the same regardless of the division ratio.
if 100Mhz = 10 .0 volts from the F/V converter 101 Mhz will give 10.1 volts output, 90Mhz 9.0 volts and so on.
Its perfectly linear.

 

[betwixt]

The point I was making is IF the range bio_man is interested is relatively narrow compared to the base frequency, it might be useful to consider down mixing. Mixing retains the original spectrum but at a lower and more manageable frequency so as a percentage of maximum, the recovered voltage might be higher.

If the full (near)DC to 100MHz is needed then dividing or using it directly is a perfectly good approach. Using your example, If the range was only 100MHz to 100.1MHz the recovered voltage would only be 0.1V but down-mixing would make it higher per Hz difference.

Brian.

 

[Warpspeed]

Quite right Brian.
An even more sensitive method, for even smaller bandwidth would be use of a phase locked loop.
But my take on this, he just requires a pretty basic wide band F/V converter.

 

[bio_man]

thanks all for your helpful comments and interesting discussion.

to clarify more, the signal of interest is changing from 40MHz to 100Mhz in about 3-5MHz range. It is very weak in amplitude (few hundreds of uV to few mV), I am thinking to use fast comp as a first stage to overcome the issue of low voltage and then use dividers and regular F/V chips.