What determines the power level of a SWR meter?

<< < (2/3) > >>

Dale Hunt:
You need to have the right ferrite mix in the toroid - your -2 core is powdered iron, and
probably wouldn't be suitable because the permeability is much lower so it would require
several times the number of turns.

Silicon diodes will work, but the readings will be non-linear at low power levels.  This
may be an issue if you tune your antenna up at, say, 10 watts, then want to monitor
the SWR at full output.  You may be able to add a voltage divider on the RF side of
the diode to lower the voltage across it.

Also, make sure you are using a good RF diode.  Many of the diodes with higher voltage
ratings are power supply diodes with a relatively high capacitance and may not work well.

A good start would be to measure or calculate the voltage across the diodes currently
in the circuit at 100W and see what the corresponding rating for 600W would be.

In the end you're just going to have to experiment.  As Steve indicated, you don't
know where the manufacturer cut corners or otherwise built an inadequate design
to start with, so you might just have to keep blowing it up and repairing whatever
broke until it works for you.

Nelson Derks:
Quote from: KD4SBY on January 12, 2012, 10:41:22 AM

They all use geranium diodes.

If you decide to homebrew... One of the more robust diodes you can use is an HP 1N5711 Schottky: http://datasheetcatalog.com/datasheets_pdf/1/N/5/7/1N5711.shtml

Most SWR meters use a 1N34A or some generic low-tech equivalent, but you'll see better sensitivity and more reverse voltage headroom with a 1N270... Probably the best low cost general purpose Germanium diode for small signal applications. http://www.datasheetcatalog.com/datasheets_pdf/1/N/2/7/1N270.shtml

Dale Hunt:
Quote from: KD4SBY

...They all use geranium diodes...

Geraniums don't handle power very well at all.  The flowers wilt and the leaves get
singed.  Besides, you have to keep them watered, which isn't convenient for most
SWR meters.

Germanium usually works better.

Pat Bailey:
Shouldn't the plural of "geranium" be "gerania"?

Tom Rauch:
Quote from: KD4SBY on January 12, 2012, 10:41:22 AM

Thanks fellows. Some details, I have SWR Meters of different designs. Two of them have a stripline etched on a piece of PCB of abt 2 inches. Then I have two other ones of which one uses a 0.5" toroid with  a few  inches of bare wire going through the middle, and one that has a 3" piece of coax (RG-58)going through a somewhat bigger toroid. I think that they were all intended for CB use.

The strip line type uses one or two small transmission lines near the main transmission line to sample voltage and current.

The ratio of voltage to current is correct when the small transmission line is terminated in the right impedance, and the resulting signal is detected on the end opposite the termination.

The toroid is a current transformer, and it is summed with a signal from a voltage divider. For current flowing one way, the signal phases are such that the voltage and current adds. For current flowing the other direction, the signals from current and voltage samples subtract.

When the line voltage and line current is in some ideal ratio, there is either perfect addition or perfect subtraction. There is usually a small trimmer capacitor to set the ideal ratio between signal levels from the current sample and the voltage sample.


I was thinking of replacing the latter with a larger toroid, like a 106-2 size (I have one of those), use the same amount of windings and using different resistors to accept larger power.

Magnetic characteristics of the core are far more important than size. I have 0.1 inch thick 1/2 inch diameter cores that can handle over 15 amperes of RF, enough current for 10,000 watts. Magnetic properties are critical, size is not so long as it fits over the wire.


I Googled some designs which show the general construction of a SWR meter with a toroid for the 100 W range, with no mentioning how to change it for higher power. They all use geranium diodes.

Many things control upper and lower power limits, and frequency response. The termination resistor on the current transformer, the number of turns, the core material, the voltage rating of components in the voltage divider, and so on.

The diodes also might set a limit, but usually not.


Incidentally, designers all warn that silicon diodes should not be used as detectors due to their higher threshold voltages. That may be true for low power, but how about high power? It really should not matter then (?), and silicon diodes are much easier to get with high reverse voltage ratings.
Any comments?

Schottky diodes are best. If you use enough detector output voltage and very low current, a silicon diode is perfectly fine. Just be sure the diode is low capacitance, and very fast switching.

Usually there is resistor for forward and a resistor for reflected that sets sensitivity. Perhaps all you need to do is change them.

As you double overall resistance, full scale power reading increases almost four times. It won't often be exact, so you will have to recalibrate.

73 Tom


[0] Message Index

[#] Next page

[*] Previous page