Make your network analyzer like a Ridley or Venable isolated type

once upon the time I made one "isolation transformer" for a power supply guru friend of mine, who used to work at Astek at the time, the bandwidth of the transformer was DC to 500kHz, if there is substantial interest I could resurrect the project

I hope Alan Martin had his 5 Venables calibrated and tested before using them. Purchasing a Venable for $64 is only a bargain if it is properly calibrated, otherwise it will certainly cost dearly in the end. I wonder how Alan would feel if he were to get a CAT scan on a machine that the doctor picked up at a local garage sale. This is playing with fire. If you are a hobbyist, you don't have much to lose, but I certainly hope that this mentality is not being extended to the commercial world

The design of a proper transformer for a network analyzer is not an easy thing to do, nor is it inexpensive. At the other extreme, the best injection trasnformer is a Rogowski type, and those cost thousands of dollars. The injection transformer generally requires very wide bandwith. A typical PFC circuit requires measurements down to 1Hz or lower and a typical linear regulator requires measurement to several MHz. Using an improper transformer will likely yield poor results, and as a consultant, I regularly see companies lose enourmous amounts of money due to such ignorance. A Video transformer is meant for video and cannot peform at low frequencies. Part of the reason that injection transformers are so expensive is that it is a very limited market, and so the development cost must be amortized over a small quanitity of units. I assure you that "saving" money on an injection transformer is not a wise move, and statistically, most users will eventually pay it all back and then some, due to errors that go undetected as a result of improper transformer. I also see customers use off the shelf audio transformers, often of the radio shack variety. These pose the same problems. The typical power supply circuit requires measurements well below and well above the audio range or the video range

The design of a proper transformer for a network analyzer is not an easy thing to do, nor is it inexpensive. At the other extreme, the best injection trasnformer is a Rogowski type, and those cost thousands of dollars. The injection transformer generally requires very wide bandwith. A typical PFC circuit requires measurements down to 1Hz or lower and a typical linear regulator requires measurement to several MHz. Using an improper transformer will likely yield poor results, and as a consultant, I regularly see companies lose enourmous amounts of money due to such ignorance. A Video transformer is meant for video and cannot peform at low frequencies. Part of the reason that injection transformers are so expensive is that it is a very limited market, and so the development cost must be amortized over a small quanitity of units. I assure you that "saving" money on an injection transformer is not a wise move, and statistically, most users will eventually pay it all back and then some, due to errors that go undetected as a result of improper transformer. I also see customers use off the shelf audio transformers, often of the radio shack variety. These pose the same problems. The typical power supply circuit requires measurements well below and well above the audio range or the video range

Most all transformers have their low frequency performance limited by their 'volt-second' capability. That translates simply to the simple fact that you must limit volts-per-turn (on any winding) a value declining by a factor of 2 for each octave you go down in frequency. For example, a transformer rated for some input voltage V at 60 Hz will tolerate V/2 at 30 Hz, V/4 at 15 Hz, and so on. At 1 Hz, you can safely apply 2 Vrms to the primary of a 60 Hz 120 VAC power transformer. As you can imagine, DC poses a serious problem for most transformers. At the high frequency end, leakage inductance is the usual limiting factor, although core losses and winding capacitances also contribute to hf rolloff. For this reason, wideband transformers often have thin strip-wound toroidal cores and multi-filar windings.