Tales of the Cliff – Part 4

This is a 4 part series blog post:

We’ve spent a whole month taking about the “digital cliff”, about measuring bandwidth, wavelength and critical distance. And we’ve talked about our “safe distance” table based on the B54_CliffAheadformulas in the HD standard (SMPTE 292M). Use those safe numbers and you don’t have to think about anything, you don’t have to measure anything. Just put on some connectors and plug it in. If you get failures at these short distance, you have something very wrong, a lot more wrong than a cable a bit too long. Is everything intended for HD? Tested by each manufacturer to AT LEAST 2.25 GHz?  (Last time I mentioned how one project was stymied by using analog 75 ohm terminators!)

Read more »

Tales of the Cliff – Part 3

This is a 3 part series blog post:


For the last couple of weeks we’ve been talking about the “digital cliff” where signals no longer work.  Check out the previous installments about bandwidth, wavelength and critical distance (Part 1 and Part 2).

The critical distance is the length at which it is important to know the impedance of the cable, where the cable is now a “transmission line”. For signals like professional high-definition video (SMPTE 292M standard), the critical distance is a little over ONE INCH. That simply means that everything is critical, cable connectors, patch panels, patch cords, adaptors…..everything in that passive line between boxes.  But that still doesn’t tell you where the cliff is.

As the cable goes farther and father, there is attenuation on the cable. Bigger cable has lower attenuation than smaller cable, so if your cable is not working, you might consider changing to a bigger cable.

Read more »

Tales of the Cliff – Part 2

This is a 3 part series blog post:

We’ve been talking about digital signals and their limitations. The first step is to ascertain the bandwidth of your signal. Our example in Tales of the Cliff – Part 1 was RS-422 which has a bandwidth of 10 MHz.


Once you know the bandwidth, you can calculate the wavelength. 300 million divided by that frequency gives you the wavelength in meters. 300 million divided by 10 million equals 30 meters (about 100 feet). The critical distance is not a full wavelength but a quarter wavelength. (Think about a sine wave. Where is at its maximum effect? A quarter of a wavelength.)

So a quarter of 100 ft. is around 25 feet. And what that means is that, below 25 feet, you can use almost anything to carry RS-422. After 25 feet, it is a “transmission line” and requires that you match the impedance of the source and destination devices with the impedance of the cable. For RS-422, this impedance of the devices is 100 ohms, so the cable will also be 100 ohms. When you put connectors on your RS-422 cable (15-pin sub-D, most commonly) are they 100 ohms? No, they aren’t. But it doesn’t matter because they are less than 25 feet long. So you can use any connector. But that doesn’t tell you the MAXIMUM distance you can go.

Read more »

Impedance and Return Loss

When you deal with high frequencies, above around 100 MHz, you have entered the zone of the “transmission line.” It has to do with the wavelength of the signal, and that is a discussion we’ll have in future blogs. But what it means is that the impedance of the cable is now important and you have to match the impedance of the source and destination devices.

This also means that everything in-between must match the impedance chosen and, by “everything,” I mean cable, connectors, patch panels, patch cords, adaptors, bulkheads, feedthroughs – erything! It also means that any variation in impedance can affect the signal on the line. This is true for every transmission line, whether we’re talking about a 50kW RF signal going up to an antenna or an HD video signal going between boxes. Of course, in these two examples the impedance is different (50 ohms for that high-power line and 75 ohms for that video cable). Read more »

On Twitter