10
$\begingroup$

I would like to build my own HF (1.8 - 30 MHz) SWR meter, however I'm not sure how I can calibrate the meter without an accurate power source. I'm targeting a power capacity of 150 W so I can use it with my HF radio. However, my test equipment at hand does not include a directional wattmeter, and I wouldn't trust the power meter on my radio when the radio is turned off.

I have a number of precision RF sources, but most of those don't exceed 1W in output power. I can get a good 100W+ 50Ω load, spectrum analyzers, and oscilloscopes, but nothing that I would really consider "power RF".

Can I calibrate or validate my meter without an accurate power source or accurate directional wattmeter?

Improve this question
$\endgroup$
4
  • $\begingroup$ If you just need SWR, you don't need an accurate power source, just a consistent one, because SWR is a ratio. $\endgroup$
    – Phil Frost - W8II
    Commented Mar 25, 2014 at 13:04
  • $\begingroup$ Also I don't have time for a full answer, but here's a hint: $P=I^2 R$, and $P=V^2/R$. R is probably 50Ω, and between the scope and the spectrum analyzer I bet you can find a way to measure V or I. $\endgroup$
    – Phil Frost - W8II
    Commented Mar 25, 2014 at 13:06
  • $\begingroup$ @PhilFrost So then I'll know that my reading is good when the antenna is matched. Should I be worried about the scope adding error to the measurement by mismatching the system? How would I verify that it's measuring power correctly when the SWR isn't 1:1 (e.g. what about coupling between forward and reverse power measurements)? What mistakes might I make by (perhaps naively) hooking up a scope directly across a matched load? $\endgroup$
    – W5VO
    Commented Mar 25, 2014 at 18:02
  • $\begingroup$ You can also put an unmatched load on it. If you don't have a 25 ohm dummy load, you can make one by putting two in parallel. As long as you make the connections with wires much shorter than a wavelength (easy at HF), then you won't be introducing additional mismatches of significance. $\endgroup$
    – Phil Frost - W8II
    Commented Mar 26, 2014 at 13:56

2 Answers 2

Reset to default
4
$\begingroup$

There are many ways you could calibrate. For measuring SWR, try this:

Attach a 50Ω dummy load to your transmitter. The SWR should be 1:1. Then, attach two 50Ω dummy loads in parallel, giving you an effectively 25Ω load. The SWR should be 2.

It seems you also want to measure power, and not just SWR.

If you have some attenuators available, you can put them on your transmitter's output until you are within the acceptable range for your spectrum analyzer. Then measure the power for that, and add to the measured power the power absorbed by the attenuators.

Or, you could transmit into the dummy load, then measure the voltage with the scope (provided it has sufficient bandwidth), and calculate the power from $P=V_{RMS}^2/50\Omega$. If you use a high impedance probe ("high" meaning anything $\gg 50\Omega$) then the scope won't affect the measurement much. 50Ω in parallel with 10kΩ is still 50Ω, within your measurement error.

Improve this answer
$\endgroup$
2
$\begingroup$

Mike Bryce covered a 3:1 dummy load in QST, in Feb 2013, this will NOT give you an accurate input power reading, but WILL generate a known SWR.

The technique uses a 150 ohm resistive load creating an SWR of 3:1, with enough current/watt capacity, and an adequate heat-sink, so you do not overload the dummy load's thermal limits, when you transmit from the desired transmitter. Your SWR meter can be calibrated while you have your transmitter keyed and attached to this 3:1 dummy load. I believe there is no reactance or inductance involved in this dummy load, it's purely a resistive load, but at the "wrong" resistance, which SHOULD creating a 3.0 readout on a properly calibrated SWR meter. I would think even a tiny amount of transmitter power (say 0.5 to 1.0 watts at HF) should work admirably.

http://connection.ebscohost.com/c/articles/85802751/three-one-dummy-load

Improve this answer
$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .