To be very clear, please do not accidentally interpret anything on this web site as our approving or recommending use of ferrite devices instead of a specified isolation transformer at an FCP feedpoint. With an FCP and other deliberately non-resonant counterpoise solutions, we only and very strongly recommend use of an isolation transformer specified in this web site, or a commercially manufactured copy at the feedpoint. We do this to improve first attempt results, avoid lost time and expense, and support future improvements.
Older than radio itself, the isolation transformer was invented in the late 1800's. An isolation transformer has no metallic conducting path between the primary and secondary windings. They were first used at audio frequencies in early telephone circuits.
Isolation transformers were originally called "repeating coils", "repeating" its audio input signal on its output, while blocking, grounding or supplying DC voltages, transforming impedances, etc. Only one of its many uses, the isolation transformer paid back its invention uncountable times by making possible three circuits instead of two on two pairs of the copper wire then carrying telephony between cities.
The third circuit was called a "phantom" circuit, and made use of the same common mode current on the two pairs that today we try to eliminate from our feedlines. The common mode blocking by the isolation transformers had to be good enough to keep the conversation on the phantom circuit from being heard on the "side" (normal) circuits and vice versa.
The third circuit generated an enormous 50% long distance telephone revenue increase for two decades before telephone carrier multiplexing systems put four and then a dozen circuits on two pairs in the 1920's.
Today, isolation transformers are a category, called by many different, more specific names. Vast numbers of isolation transformers are used in general audio, telephony, power company transmission systems, RF and many other applications. More information on isolation transformers with links to related subjects can be found here at Wikipedia.
Use of a ferrite balun/choke with an FCP is considerably more work and worry than an isolation transformer, with a higher component count. To prevent heat loss in the ferrite material of baluns and common mode blocking devices, the FCP's deliberate capacitive reactance must be tuned out at center frequency by a series coil protected from the weather. Using the coil will mildly narrow the antenna system's operating bandwidth. But properly setting the coil's value requires a non-intuitive "single-ended" procedure, with the FCP and coil separated from the rest of the antenna system for tuning. Later re-adjusting that coil to reduce overall antenna SWR will UNDO the benefit.
The balun/choke used in lieu of isolation transformer needs a very high blocking value on 160 meters to reduce loss when operating away from the FCP/coil resonant frequency. As an example the ferrite-based Balun Designs 1116d is specifically designed for maximal low band blocking, delivering 5kΩ+ on 160. Click on the graph icon toward the middle of the page for it's choking impedance graph.
But in a certain unfortunate 160m configuration, with 900 volts common mode RF on an elevated, balanced feedline, even excellent ferrite devices, with 5kΩ blocking on 160, could dissipate a couple hundred watts out of 1500. A choke dissipating 200 watts in a sealed enclosure got extremely hot very fast. An isolation transformer solved that problem. W0UCE and K2AV didn't manage to set a 1116d on fire with that problem back then because it wasn't on the market back then. :>)
Run-of-the-mill commercial "all band" baluns generally have poor to no optimization for 160. 160 is usually the band with the worst blocking performance. Some ferrite balun/choke devices have displayed remarkably low blocking on 160. One nationally marketed balun marked "all band", with blocking quite useful and satisfactory on a triband yagi, measured a bare 173 ohms blocking at 1.83 MHz.
Stretching the FCP dimensions to make it self-resonant and eliminate the coil to go with the balun will unbalance the FCP's essential zero net field folding trick and increase ground losses, especially over the sometimes barely buried rubble that passes for extremely poor "earth" in urban and suburban lots.
To perform the modification which allows operating an existing 160 meter L over FCP as an 80m end-fed halfwave L, a ferrite balun/choke + tuning coil arrangement will need to be removed and replaced with a 160 m isolation transformer.
The 160+80 modification allows remote switching of the L/IsoT/FCP from 160 to 80 meters by supplying relay coil voltage from the shack. The higher voltages to be blocked at the base during 80EFHWL operation can burn ferrite and create arc paths. But those higher voltages are well within the range ** of the specificed isolation transformer: ~55 kVDC between adjacent wires in the winding and ~27.5 kVDC between wires and the core and mounting materials, or 19.5 and 9.75 kV RMS at RF respectively.
[** See double polyimide coated wire and teflon tubing specifications. Typical "heavy" or "double" Polyimide (14.6 kV), plus #12 standard wall teflon tubing (800V/mil times .016 inches or 12.8 kV) calculates 27.4 kVDC rating for the combination. With tolerances, this rounds to ~55 kVDC between wires in the winding and ~27.5 kVDC insulation between the wire and the core and environment.]
Many correspondents have asked about using a ferrite balun/choke with their FCP. Others had already tried it with unexpected and confusing poor results. Understandably they hadn't wanted to spend time or money on a new device.
Not at all fair to Jim, others heard "K9YC said we don't need an isolation transformer," with no further qualification.
With no fact-checking at the source to discover killer details from K9YC like the series coil Jim requires, AND just exactly what Jim means by a good balun for 160, the admonition suffered further classic rumor-game deterioration: "Jim said just use a balun, forget the isolation transformer." This produced the double whammy of the awful-on-160 leftover balun made even worse for the application by the missing series tuning coil.
Some included the coil but then tuned it for lowest SWR, defeating its purpose.
Of course, any resulting difficulties were blamed on the FCP.
Some of these unfortunate installers got in touch, and with explanations were directed back to an isolation transformer, solving the problem in a single stroke.
Some of these unfortunate installers took the FCP down and told others that the FCP didn't work.
The owner of the earlier-mentioned balun with 173 ohms blocking on 160 meters, ran 100 watts on 160. He used it without the series coil on an FCP.
NEC 4.2 double precision calculations estimate no more than 39 watts went to his antenna. 2 watts were dissipated via normal FCP loss and at least 59 watts were dissipated in the balun and ground beneath the feedline. That was a 4 dB reduction from what should have been roughly 5 watts loss in the FCP and FCP-induced surroundings, a watt or two lost in the antenna wire, and 93 watts radiated abroad.
The owner correctly observed the lack of improvement (or possible worsening) of results. But he refused to try an isolation transformer saying that "an authority" said it was not necessary, and removed the FCP.
It is ironic that as used, running 1500 watts instead of 100 would have dissipated 500 to 900 watts in the balun, almost certainly destroying it, proving its unsuitability to the application and removing it from the leftover electronics box. This was a reality K2AV and W0UCE discovered themselves the hard way multiple times back in the comedic what-did-you-burn-up-or-set-on-fire-today pre-publish history of the FCP.
In helping correspondents clean up lossy and detuning interaction from nearby antennas, we specifically refer them to K9YC's excellent reference work on ferrite devices for methods on how to isolate a vertically running feedline shield from the dipole/vee it feeds, render the common mode non-resonant, and isolate it from the part of the feedline shield that runs along the ground.
But, with all our respect and gratitude for Jim's work, we simply have too many stories from FCP installers about things going wrong or lossy when they attempted to use a ferrite balun or current choke where we specify an isolation transformer. Using ferrite devices to feed an aerial wire over FCP clearly seems a procedure that defeats all but advanced users with advanced test gear. Our extensive work, the weight of hundreds of correspondent's results, and over a century of general use in all areas electric and electronic, all validate the isolation transformer.
We have to publish a single procedure that we know will work anywhere with anyone, novice, oldtimer, career engineer, with minimal test equipment, a procedure that we have confidence we can support and improve going forward, fitting into a single work of documentation to maintain. We cannot and will not publish or pass along modified or alternate methods that fail for some or many users. Experimental and Beta methods will be identified as such.
If you inquire, we have to advise you to start your remedy with an isolation transformer. There is no feedpoint device that we are positively convinced of, and willing to vouch for regardless, other than a specified isolation transformer.
Therefore with an FCP and other deliberately non-resonant counterpoise solutions, we will continue to only and very strongly recommend use of an isolation transformer specified on this web site, or commercial versions manufactured to these specifications. We do this to improve initial results, avoid lost time and expense on flawed no-isolation-transformer advice, and support future improvements.
This admonition includes the widely used ON4UN 4 times 1/8 wave elevated radial scheme, where many installers use the series coil to adjust SWR and by so doing defeat the coil's purpose. With the isolation transformer you can harmlessly use the coil as one method to center SWR and use the ferrite choke/balun appropriately to good effect elsewhere.