Checking attenuation numbers ...

Foundations of Amateur Radio

Before we start I should give you fair warning. There are many moving parts in what I'm about to discuss and there's lots of numbers coming. Don't stress too much about the exact numbers. In essence, what I'm attempting is to explore how we can reduce the power output from a transmitter in such a way that it doesn't blow up a receiver whilst making sure that the signal is strong enough that we can actually measure it.

With that in mind, recently I discussed the idea of adding a series of attenuators to a transmitter to reduce the power output by a known amount so you could connect it to a receiver and use that to measure output power at various frequencies. One hurdle to overcome is the need to handle enough power in order to stop magic smoke from escaping.

None of my attenuators are capable of handling more than 1 or 2 Watts of power, so I cannot use any of them as the first in line. As it happens, a good friend of mine, Glynn VK6PAW, dropped off a device that allows you to divert most of the power into a dummy load and a small amount into an external connector. In effect creating an inline attenuator capable of handling 50 Watts.

The label doesn't specify what the attenuation is, so I measured it using a NanoVNA. To make our job a little interesting, it isn't constant. Between 10 kHz and 1 GHz, the attenuation decreases from 70 dB to 10 dB. We want to measure at a base frequency on the 2m band and its second and third harmonic. The attenuation at those frequencies varies by 11 dB, which means we'll need to take that into account.

So, let's subject our currently imaginary test set-up to some sanity checking. Our receiver is capable of reading sensible numbers between a signal strength of -127 dBm and -67 dBm and we'll need to adjust accordingly.

If we transmit an actual 20 Watt carrier, that's 43 dBm. With 110 dB of attenuation, we end up at -67 dBm, which is right at the top end of what we think the receiver will handle. If we're using something like 5 Watts, or 37 dBm, we end up at -73 dBm, which is well above the minimum detectable signal. Our best harmonic measurement was around -30 dBm, which means that with 110 dB of attenuation, we end up at -140 dBm, which is 13 dB below what we think we can detect.

So, at this point you might wonder if this is still worth our while, given that we're playing at the edges and to that I say: "Remind me again why you're here?"

First we need to attenuate our 20 Watts down to something useful so we don't blow stuff up. Starting with 110 dB attenuation, we can measure our base carrier frequency and its harmonics and learn just how much actual power is coming out of the transmitter. Once we know that, we can adjust our attenuation to ensure that we end up at the maximum level for the receiver and see what we are left with.

So, let's look at some actual numbers, mind you, we're just looking at calculated numbers, these aren't coming from an actual dongle, yet. Using Glynn's dummy load as the front-end, at 146.5 MHz, the attenuation is about 30 dB. If we look at a previously measured handheld and rounding the numbers, it produced 37 dBm. That's the maximum power coming into our set-up. With 30 dB of attenuation from Glynn's dummy load, that comes down to 7 dBm. We'll need an additional 74 dB of attenuation to bring that down to -67 dBm, in all we'll need 104 dB of attenuation.

The third harmonic for that radio was measured at -26 dBm. So, with a 104 dB of attenuation that comes out at -130 dBm, which is below the minimum detectable signal supported by our receiver. However, remember that I told you that our dummy load had different attenuation for different frequencies? In our case, the attenuation at 439.5 MHz is only 19 dB, not 30, so in actual fact, we'd expect to see a reading of -119 dBm, which is above the minimum detectable signal level.

I realise that's a lot of numbers to digest, and they're specific to this particular radio and dummy load, but they tell us that this is possible and that we're potentially going to be able to measure something meaningful using our receiver. I'll also point out that if you're going to do this, it would be a good idea to take notes and prepare what numbers you might expect to see because letting the magic smoke escape might not be one of your desired outcomes.

Speaking of smoke, what happens if you consider changing the attenuation when you're measuring at another frequency, like say the second or third harmonic and you see a reading close to, or perhaps even below the detectable signal level as we've just discussed. You might be tempted to reduce the attenuation to increase the reading, but you need to remember that the transmitter is still actually transmitting at full power into your set-up, even if you're measuring elsewhere. This is why for some radios you'll see a measurement that states that the harmonics are below a certain value because the equipment used doesn't have enough range to provide an actual number.

To simplify my life, using a NanoVNA, I created a spreadsheet with 101 data points for the attenuation levels of Glynn's dummy load between 10 kHz and 1 GHz. I charted it and with the help of the in-built trend-line function determined a formula that matched the data.

I've also skipped over one aspect that needs mentioning and that's determining if the receiver you're using to do this is actually responding in the same way for every frequency. One way you might determine if that's the case is to look at what happens to the signal strength across multiple frequencies using a dummy load as the antenna. One tool, rtl_power might help in that regard.

Is this going to give you the same quality readings as a professional piece of equipment? Well, do the test and tell me what you learn.

I'm Onno VK6FLAB