RF Voltmeter / Projects / Forums

Forums

4hv.org :: Forums :: Projects

« Previous topic | Next topic »

RF Voltmeter

Move Thread

LAN_403

Hazmatt_(The Underdog)

Sun Aug 19 2007, 12:45AM

Registered Member #135 Joined: Sat Feb 11 2006, 12:06AM
Location: Anywhere is fine
Posts: 1735

Do you have that nasty little problem of measuring voltage at high frequency? Well I usually do. So here's a quick fix probe add-on for your DMM, its a RF Voltage Probe that will give you RMS of the sine function measured.

I plugged it in and out of the generator-scope circuit to see if it loaded the sine function and it looks to be transparent for the most part. If it is loading the source its very small, but that was with over 2V sine so it will vary with smaller amplitudes.

I didn't notice anything drastic at 10MHz, so it may be fairly flat throughout the whole range. I'll have to throw it on my digital scope to be sure, and I will do that soon. So far its a pretty awsome probe.

**************************************
Okay now is where it gets interesting!

With the high impedance probe designed we can now move on to doing something rather tricky, measuring current through an antenna or 'active' load. The loading will change as we approach the asymptote of resonance, shorting the voltage at the base of the coil and peaking the current. Using what we know about voltage measurement we can devise the current probe from the voltage drop in the resistor, simple right? Not so fast there! Metering the current is difficult because we are measuring the signal on 2 active sides of a static load, meaning we are measuring a differential voltage, so we need a differential probe! This is not possible with our common mode case grounding configuration of the probe! So what we need is an active probe, and it must have a fairly linear response to our voltage ranges of input.
So what I have done here is utilize an instrumentation amplifier to give me high input impedance and low output impedance for the meters that will be used later. The first step here is to prepare the signal to be sent to remote meters.

Why remote meters? Why go active? Why not just use insulated BNC for current transmission?
Well firstly, you need remote meters because if you get within 4 feet of your coil you capacitivily couple the coil, throwing all of your measurements off.
We need to go active so we can send that signal a ways without degredation.
We could use an insulated BNC to send the current signal, but there is just 1 problem, if anything touches the cable it couples capacitivily and it changes the reading. You could try to cut this corner, but you will soon realize that the cable is an antenna and is prone to picking up unwanted signals and noise. Going active protects your signal as well as keeps all the RF stuff inside the transmission box unit so you don't have to worry about your results being comprimised.

Later you will see a complete test suite to accompany the metering module that will improve your test bench by extending your function generator's use to incorporating voltage, current and frequency measurement on digital readouts.

Hazmatt_(The Underdog)

Fri Jan 11 2008, 03:18AM

Registered Member #135 Joined: Sat Feb 11 2006, 12:06AM
Location: Anywhere is fine
Posts: 1735

Well'p I've been working and working and working on this project trying to design a good RF probe for you guys and myself, and I kept coming up short.

The basis of this probe is a device that will be used as an instrument IN an instrument. But the initial probe design feeds RF forward into the meter, so a total redesign was necessary.

The following probe is a basic copy of the RF probe in the Boonton meters, which meter 10 KHz to 1.2 GHz with specially chosen diodes.

The probe here is designed for RMS measurement from 1 KHz to 30MHz with good accuracy within 10% accuracy, and up to about 50MHz with less accuracy because it starts diverging at that point. This probe is good enough for my work but will be extended later with the Schottky diodes and I'll see how far it can go. For now I will use the 1N34A Ge diodes. They are temprature sensitive, and not the best or really ideal these days, but will work fairly well considering I'm not requiring much of them. I just need 10 Khz to 1 MHz operation.

The biggest challenge here was designing the probe so that you could build it without needing the Boonton meters or calibrated source. The calibrated source from an equipment dealer is $700 ! Xo, the probes for the Boonton meters are $165 each, and the meters can be found, but they too range from $30 to over $500 each. One dealer wanted $4000 for the Boonton meter. So you can either take my word that the probe is decently accurate, or you can spend your money on equipment too.

The next step is to get the basic probes into a metering arrangement, and that will come soon. :)

**************** Little Update **********
Using the silicon Schottky diodes (NTE 583) the probe construction follows the Boonton meter very closely, tested to 100MHz. These diodes outperform the 1N34A easily. I am limited to 100MHz testing right now as my upper frequency limit, so I don't know how far this probe will go. More on that later on.
*************************************

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob