Is anyone aware of a circuit that will take an RF input of as low as 10µV and provide a reasonably accurate (3%) linear output of .1 or 1 VDC? My searches have not yielded any usable results. Part of the search problem is probably trying to describe what I am thinking of.

Maximum input of 1 volt would be good but lower would work. A flat response from 10 kc to 150 Mc would be very desirable. Using a range switch to accomplish a flat response would be just fine. My search did not yield any results that would seem to work. Voltage would be measured across a 50, 75 or 300Ω load. It could be useful to measure the output of things like signal generators into their specified output load.

An additional good thing would be if it had its own calibration voltage. If the frequency response was low enough, a 60-cycle signal with divider could be used for calibration.

To sum it up I want to use my VTVM (or DMM) as an accurate, very sensitive RF voltmeter. Ideas?

I just built three probes you see them on Building a VTVM. One probe was for 250 Mhz

Dave

My Probes are made from a old wind-chime the tube size is 12mm or just under ½"

Here is a RF schematic on I used


This one is Eico

Question: Can solid state diode detectors actually work with microvolt level inputs? Don't they have thresholds of a few hundred millivolts before they conduct?

Solid state diodes start conducting at zero volts. You can verify this by checking the current-voltage graphs from any diode data sheet.

The "turn-on voltage" premise is an approximation that's used for high level signals, but inaccurate for low level signals.

The product datasheets I've looked at normally never graph down to near zero applied volts. I had this chart in my mind.

Yup, you got it a simple detector probe will not provide an accurate linear output since the voltages I wish to measure are well below the “turn on voltage�. Ideally the module will turn the 1.5 VDC scale on my VTVM into a 150µV RF scale. My thinking is it will need to be a RF voltage amplifier with a linear DC output.

Thanks everyone for your replies!

Maybe you need an FET preamp in your probe, ahead of the detector. If that is too delicate, get a miniature or subminiature vacuum tube. A remote-cutoff tube would be non-linear, you need a sharp-cutoff tube. Look at television tuner circuits, too, those tubes are cheap now.

The words "compressed scale" and "expanded scale" should be the clue that that chart is not remotely to scale. On a diode V-I graph when drawn at a constant scale, the line will be smooth and it will pass through the origin (zero current at zero volts), current going negative to the left, and positive to the right.

Do you have schematic of the FET per amp probe ?

Dave

You won't find anything cheap, and I doubt you'll be able to get anything that's wideband and also usable down to 10 µV. The current-production Boonton 9240 RF Voltmeter is only good down to 200 µV. You'd do better with a selective level meter, though I don't know if you can get one that'll cover up to 150 MHz (I only know about the vintage instruments, which usually top out at 20 or 30 MHz).

You could build something out of an Analog Devices AD8307 log amp that'll get you down to 40 µV, but the accuracy will be poor.
http://www.k9ivb.net/RF_Power_Meter/fil ... AD8307.pdf

https://pa3csg.nl/wp-content/uploads/20 ... -meter.pdf

The 1950's and 1960's VTVM could goto to 250Mhz.
You can find tube type probes to 500Mhz.

The FED type goes down to 20pA.

When see uV I think of Audio equipment meters

Dave

I'm sorry, I don't have any schematics for an FET Preamp. For my radios, I just replace bad components and it either works or becomes parts for another radio.

Looks like I am trying to find something that does not exist, yet. Just trying to build a poor man's direct reading RF µV meter. As noted, the AD 8307 accuracy is poor at low inputs, I have a nice little log detector using the AD 8307 already. It does play nicely with the silly scope.

The QST article is interesting, it looks like I need to study it a bit more. The Fig. 5 plot appears to provide a fairly linear output from around 71 µV (-70dBm) on up. No big deal to use a chart instead of direct reading on the VTVM, or DMM. Maybe a few extra components with my AD 8307 will get me close to what I need. You do not always get what you want but if you try sometimes you get what you need.

Hey folks keep those cards and letters coming.

Thanks to all!

The only chip I have use was the 555 when first cameout.
Just last month I try a LF353 work great. Now have a bag of LF353N.
Before that it was all transistor or tube.

Dave

The specifications that you have in mind (150 MHz wideband response at 10 uV sensitivity into 300 Ohms) cannot be met in a room temperature environment. This is because the thermal noise voltage at room temperature when using a 300 Ohm system is 27 uV. Even if your system were 50 Ohms, the noise voltage is 11 uV. See Electronics Notes Thermal Noise Calculator. This thermal noise voltage is part of the signals that you are trying to detect, so there is no way to improve the performance of the probe, no matter how low-noise the RF preamplifier circuitry in it might be.

To measure microvolt-level signals, you would need to "cheat" by having some RF selectivity in the probe that narrows the frequency response to a band of interest. Your probe would essentially take on the form of a radio with a very precise signal strength meter (which could be a voltage sent to a multimeter for display). A selective level meter works in this way.

Excellent information, seems my specifications are unreasonable to say the least! This was just something that popped into my head this week and I thought would be fun to explore the possibilities. Need to get back the slow lane where I belong. Shucks I get nervous if the radio has those new-fangled 7 and 9 pin tubes. Now I will return to my regularly scheduled program. But I am still thinking!

There is a slow lane?

Dave

This out of the ALLR Book on AC ,& RF PROBES.


Dave

The compressed & expanded axes did not affect the upper right quadrant which was the area of interest. Also, the general shape of the curves matches what I recalled from experiments back in the physics classes.