- Amateur Radio (Ham Radio) Community

Call Search

New to Ham Radio?
My Profile

Friends Remembered
Survey Question

DX Cluster Spots

Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement

[Articles Home]  [Add Article]  

The R-X Bridge, Tireless Workhorse

Eric P. Nichols (KL7AJ) on April 4, 2008
View comments about this article!

The R-X Bridge:

An Old Workhorse with some Good Years Left


Eric P. Nichols. KL7AJ

I'm glad to see the proliferation of some really good, surprisingly inexpensive test equipment in the ham radio domain. The MFJ line of Antenna Analyzers is a great example of something us old timers wish we had 30 years ago.

Since these devices are so convenient, however, a lot of the inner workings are hidden from us, and we lose some insights that can only be had by more primitive instruments.

One of the most venerable “tools of the trade” in radio work is the R-X bridge, sometimes called an antenna bridge, or sometimes, incorrectly, impedance bridge. Of this class of device, probably nothing was more “popular” than the General Radio GR-916 Bridge. I use the word “popular” advisedly, because these beasts, along with their required associated accessories seemed to permanently hang around the necks of us broadcast engineers like an albatross. We used them because there was NO ALTERNATIVE. Engineers by the hundreds had a love/hate relationship with their 916s. The vast majority of these devices found their retirements in ham shacks across the country.

Somewhere between the 916 bridge and the MFJ Analyzer, we had various incarnations of the “noise bridge.” For the majority of amateur radio applications, the noise bridge works pretty well, and it's certainly convenient enough. However, a typical noise bridge has somewhat limited range, as well as questionable accuracy at its extremes.

If you really want to get down and dirty, for example, measuring radiation resistances on the order of a fraction of an ohm, you really do need a good R-X bridge. A noise bridge just won't cut it. In fact, the top-of-the-line MFJ analyzer won't even cut it.

Now, one disadvantage of the 916 for amateur work is that it has an upper frequency limit of around 5 MHZ. (There were shortwave transformers available for the bridge, but these are exceedingly hard to come by. If you have a 916, most likely it's configured for A.M. Broadcasting). However, most of the places where you're going to encounter super low impedances, of the sort that only a bridge like the 916 can handle, is in very short, low band antennas. So, if you're a low band mobile operator, the 916 is your friend.

Under the Hood

The R-X bridge is exceedingly simple. It's a basic Wheatstone bridge. If we arrange the Wheatstone bridge like a baseball diamond, we have a known, variable resistance between home and 1st base. We have a known, variable capacitance between 1st and 2nd base. An r.f. signal is injected between 2nd and 3rd base. Your unknown impedance is between 3rd base and home. An r.f. detector (usually a shortwave receiver) goes between 1st and 3nd base.

Perhaps you're wondering where the variable INDUCTANCE is. Good question. In practice it's a LOT harder to build an accurate variable inductance than a variable capacitance. So the bridge uses a calibrated FIXED inductance you can switch in or out, depending on whether your antenna is inherently inductive or not. If you're using this for short antennas, you'll usually leave this inductor in circuit.

So, let's say we want to measure the impedance of an 8 foot mobile whip on 3.9 MHz. Attach the ground strap of the bridge to the frame of your car, as close to the antenna feedpoint as possible. Attach the “unknown” probe to the bottom of your whip. Set your signal generator to 3.9 Mhz. Tune your receiver to 3.9 MHz. Adjust your R and X dials for a null in your receiver. (If you're doing this right, you should EASILY be able to achieve an 80 dB null in your signal compared to totally off-tuned). Your feedpoint resistance can now be easily read off the gigantic dial. (It will probably be around 3-4 ohms). The Reactance is a little trickier, but not much. You have to be able to divide. The formula's etched right into the REACTANCE dial. Now, if you recall that bit about the fixed inductance. The 916 deals with this by asking you to note the CHANGE in reactance necessary to achieve a null. If you have to turn one way, it's inductive, the other way, capacitive. During the calibration process, you set the Reactance dial near mid-scale (it's a logarithmic device). It's a little awkward at first, but becomes intuitive quite quickly.

Oh yes, I forgot a little detail about pre-calibration. You should have a KNOWN non-inductive resistor in your accessory kit. Generally one uses a 50 ohm resistor for this task, but if you know you're going to be working with low impedance antennas all the time, you want to get (or make) a 1 ohm calibration resistor. This will make the overall calibration just a tad more accurate at very low resistances, (though the 916 is staggeringly precise anyway).

To do this, attach your 1 ohm resistor to the unknown terminals. Set the R dial of the bridge to 1 ohm. Set the X dial to mid range. Apply your signal at the frequency of interest, and adjust the R and X CALIBRATION knobs for a perfect null. Now measure your antenna.

Null and Void

You may find that one of the most frustrating things about a normal R-X bridge is pickup of distant stations (and not so distant stations) that are orders of magnitude stronger than your signal generator. (This is REALLY a huge problem if you're measuring the base of a 400' broadcast tower at night, where every station on the planet comes in at 70 over S9! One of the slick ways we deal with this is to use a COHERENT null detector. It's a very simple device, you can make out of a signal splitter and a prefabricated DBM (doubly balanced mixer). You feed the signal generator into one port of your DBM, in addition to your bridge's generator port. The second input to your DBM you attach between 1st and 3rd base. The “I.F.” port of your DBM goes to a low pass filter (essentially cutoff above DC) and a milliammeter. This null detector will ONLY respond to the signal you're pumping into it. This is the one thing we broadcast engineers wonder how we ever lived without!


Now, let's look at one of the amazing things about the R-X Bridge. Our antenna, which presumably has both Resistance and reactance, is applied to only ONE port of our bridge. (In almost all cases the equivalent circuit of an antenna is a series combination of resistance and reactance). How does the bridge separate these two components? As far as what our bridge “knows,” anything we connect to the “unknown” port is nothing but a black box with two wires poking out of it, regardless of any internal complexities.

It's pretty intuitive how we could null out a third base resistance with an equal first base resistance. And we can see how we can null out a home plate inductance with a complementary second base capacitance. But how does it do that with complex impedances?

It all has to do with phase shift. No matter HOW complicated a circuit might be, at any given frequency there can only be ONE value of resistance and ONE value of phase shift. It is when these are balanced that our bridge is nulled.

In reality, an R-X bridge can't tell you whether a resistor and a capacitor are in parallel or in series. You have to have a little a-priori knowledge about your antenna (and common sense) if you're using an R-X bridge with a small tuned transmitting loop, for example.

Remote Control

If you're like me, you probably don't relish the idea of hauling a GR-916, a Collins R-390, AND a tube-type Eico signal generator up to the feedpoint of 160 meter dipole. Fortunately, you don't HAVE to—if you're careful.

If you cut a piece of transmission line PRECISELY one half electrical wavelength, (or any multiple thereof), you can use this between your antenna and your bridge/generator/null-detector setup. The impedance at the point of interest will be precisely repeated at the input of your transmission line. Interestingly enough, it doesn't matter WHAT the actual impedance of the transmission line is. (For best accuracy, you don't want a lot of LOSS in your transmission line, however. Usually this is insignificant at frequencies at which you're going to be using your 916, anyway).

Be aware that this remote measuring method only works at one frequency, however. If you're going to be sweeping over a RANGE of frequencies, you'd best have your Smith Chart handy, or some software equivalent.

Some High Points Too

The 916 will measure resistances up to about 1000 ohms, and reactances up to 5000 ohms. This makes it more than adequate for most end-fed Zepps and longwires, as well. No special rules apply other than it might be nice to build a 1K calibration resistor, as well. Speaking of which, here's something that might not have occurred to you about resistor tolerances.

Let's say you want a really accurate (1%) 1 ohm non-reactive resistor, not a real easy thing to find. You have a drawer full of 10 ohm, 10% resistors. Statistically, half those resistors will be above 10 ohms and half below ten ohms. Assume the manufacturer didn't lie to us, and they are all within the plus and minus ten percent spec. If we parallel all ten of them, we will have a one ohm resistor with ONE PERCENT tolerance! (Statistically speaking). Our chances are even better if we have 100 random 100 ohm resistors, though it might make for a rather ugly-looking arrangement. Of course, with a modern DVM, it's pretty easy to measure the things within 1%, so you should.

Since you have one anyway

Remember we said earlier that it's no problem to get an 80 dB null with a good R-X bridge? This would seem to make the bridge a really maybe not handy, but certainly effective notch filter. And indeed, it will do that quite nicely. Just attach your receiver where you'd normally put your null detector, put a 50 ohm terminator across your generator port, and attach your antenna to the “unknown.” Voila, an 80 dB notch filter, right at your antenna port! Pretty cool! Actually, except for imperfect shielding and other extraneous “blowby” there is no theoretical reason you cannot achieve an infinitely deep null with this arrangement.

Hopefully, you'll consider the R-X bridge as another important tool in your R.F. arsenal.


Eric P. Nichols, KL7AJ

Member Comments:
This article has expired. No more comments may be added.
The R-X Bridge, Tireless Workhorse  
by K9ZF on April 4, 2008 Mail this to a friend!
Great article Eric.

It may be lost on some of the Old Timers, but many of us have never used a R-X Bridge. So it is nice to know what is going on inside the little 'black box.'

K9ZF /R no budget Rover ***QRP-l #1269
Check out the Rover Resource Page at: <>
List Administrator for: InHam+grid-loc+ham-books
Ask me how to join the Indiana Ham Mailing list!

RE: The R-X Bridge, Tireless Workhorse  
by W6TH on April 4, 2008 Mail this to a friend!
Hello Eric, We old timers back in the 30's had loads of test equipment, we called it mathematics and physic's.

The science of matter and energy and of interactions between the two, grouped in traditional fields such as acoustics, optics, mechanics, thermodynamics, and electromagnetism, as well as in modern extensions including atomic and nuclear physics, cryogenics, solid-state physics, particle physics, and plasma physics.

We just had different names mentioned than today, took a little longer to calculate, even to calculate the RX bridge.


RE: The R-X Bridge, Tireless Workhorse  
by W1ITT on April 4, 2008 Mail this to a friend!
The manual on my 916 says it's good up to 60 mc (it was cycles in those days), but I haven't used it above 10 mhz. My favorite method is to excite it with a broadband noise source (I use my old Omega-T noise bridge) and use my FT-817 as the sensitive, frequency selective detector, leaving the microphone far away so I don't inadvertently transmit into the bridge. It works well, I can run the whole setup out in the field running off batteries and I don't own the original generator/detector stuff anyway.
Accuracy is plenty good enough for practical antenna work, but it's single frequency, as opposed to swept measurements. I also have a HP 8714C network analyzer for swept measurements, Smith chart displays, etc., but the old 916 can still hold its own. They don't linger long at flea markets.
The R-X Bridge, Tireless Workhorse  
by K3GAU on April 4, 2008 Mail this to a friend!
I have a old 916 that I have used. There seems to be two issues I have with it.

1. Finding a WELL sheilded generator so the RF doesn't leak into the antenna and thus change the null.

2. A dirty "X" (I think) readout control.

Does anyone know how to clean them to remove the 'scratchiness' that age seem to bring on?

Otherwise it is a fine piece of equipment.

Dave K3GAU
RE: The R-X Bridge, Tireless Workhorse  
by WA1RNE on April 4, 2008 Mail this to a friend!

So Eric, when's the new book coming out?

RE: The R-X Bridge, Tireless Workhorse  
by KL7AJ on April 4, 2008 Mail this to a friend!

Couple of chapters left to go...then I have to work on all the graphics and such. Still looking for artwork, cartoons, jokes, stories, and other sidebar material. The more we get, the sooner it happens!

I think I might have also landed a new literary agent. I've been trying to market Steel Stonehenge, the prequel to Plasma Dreams for some time. I think I might see a glimmer of hope!

Wish me the best. :)

RE: The R-X Bridge, Tireless Workhorse  
by KL7AJ on April 4, 2008 Mail this to a friend!

You will need lots of time, a jeweler's screwdriver, and containers to hold about three hundred screws. (I only partially exaggerate!)

The capacitors in that puppy are TRIPLE SHIELDED. Once you have the shields off, you can spritz the capacitor wipers with a good contact cleaner/lubricant. Don't forget the calibration capacitor either.
I did it once to mine about 15 years ago, and it seems to have "held".

The R-X Bridge, Tireless Workhorse  
by AE5I on April 4, 2008 Mail this to a friend!
A very fine article!

I still have a General Radio 1606A and it's my standard for impedance measurements. The quality of design and construction is second to none. It's what we used for yearly measurements at the AM radio station I worked at back in the 70s and a 1606A is as good now as it was then, or when it was made decades earlier.

The only maintenance I've ever had to do on mine was to clean the wiper on one of the "inital null" capacitors.

It and its companion Unit Oscillator won't be leaving my shack until I do for the last time! ;-)

Thanks again for a fine article!


Tom AE5I
RE: The R-X Bridge, Tireless Workhorse  
by W4VR on April 4, 2008 Mail this to a friend!
A Delta Electronics OIB-3 (or an older OIB-1) operating impedance bridge is a better piece of test gear. Every ham who does serious phased array work on 160 and 80 meters should have one.
RE: The R-X Bridge, Tireless Workhorse  
by KL7AJ on April 4, 2008 Mail this to a friend!
VR et al:

Delta Electronics kindly released the design of the OIB to the amateur community (with the proviso of no commercial intent, of course!) I think it's still in the latest ARRL Antenna Books.

The R-X Bridge, Tireless Workhorse  
by AI8H on April 5, 2008 Mail this to a friend!
Super post, Eric.

Have an oldie James Millen 90672 'Antenna Bridge' with coils and an also oldie Millen 90652 dip meter.

Both are fun to work with and simple to use.

The basics are always there.

RE: The R-X Bridge, Tireless Workhorse  
by G3RZP on April 7, 2008 Mail this to a friend!
They are invaluable for measuring components such as transmission line chokes, RF transformers and baluns, too. Some people swear by Network ('Notwork') Analysers for this, but their accuracy degrades rapidly when you are a long way from 50 ohms. I have two bridges: one cost me $10 (Wayne Kerr, covers up to 10K resistive and from 0.1pF shunt C up to 250pF, plus and minus capacity and frequencies up to 100 MHz)and the GR 1616 cost $50. Both cheap at twice the price.
RE: The R-X Bridge, Tireless Workhorse  
by K8AI on April 9, 2008 Mail this to a friend!
I've got a Boonton 250A (I think that's the number)
R-X meter in fair shape. I think it's still usable.
Send me an e-mail if interested. k8ai at biblebelieverschurch dot net
Email Subscription
You are not subscribed to discussions on this article.

My Subscriptions
Subscriptions Help

Other Editorial Articles
My Oscilloscope Affair: A Love Story
To Amp or Not to Amp