This design is  applicable for small or high powers, from 100 kHz (or less) to 50 Mhz,  from miliwatts to kilowatts, depending on the selected components. For  best results, use the ferrite "43" or "61" material. Materials "4C6 or  4C65" generaly painted in purple, are exactly identical to material "61"  and can also be used. Material "2" iron powder painted in red, has been  added in tables, but i do not recommand their use. 
In theory, "61" or "4C6/4C65" could be used for 144 MHz,  but i did not experimented ...
With good  electrical and mechanical design, using parts described here, and PCB  designed for, 1.8 to 30 (50) MHz directivity can be better than 35 dB,  and accuracy in power measurements in the range 1 to 2 %. Not so bad,  isn't it ... 
With  different electrical or mechanical design, or misunderstanding of the  excel sheet, please, do not complain if you obtain bad results, no or  poor directivity, or transform resistor and wires into smoke and  flammes.
 EXCEL sheet design example
Nota:  It is  better to select the schematic with one resistor R2 and mid tap  at coil, than the other scheme with 2 half values resistors. As it is  easier to have EXACTLY the same nb ot turns in coil, than having EXACTLY  the same values of the 2 resistors. Any difference in the 2 resistors  values would give errors in SWR measurements results.
Place the WIDE ground return strip(s), not a single wire,  between coaxial line, circuit board and main ground frame, as close as  possible to the toroid, to have the best accuracy in POWER measurements  over a wide range of frequencies. For SWR measurements ONLY, this is not  important, as the ratio FWD/REF is independant of absolute values of  detected voltages. 
To  have perfect balance of the bridge, with the calculated values given by  EXCEL sheet, it is important to know the " TRUE " values of the  coil(s), resistor R2, and capacitor C1. To know what is the exact Al  value of your toroid, coil 5 turns of wire into it, measure with a LC  meter, and divide value by 25. A variable C2 capacitor, as shown, is MANDATORY,  to compensate for all tolerances in components values, parasitics.  During tests, dont hesitate to modify calculated value of C2 (fixed +  adjustable value) to obtain the best balance.Tolerances and parasitics management is not an exact science.
 A variable C2 capacitor, as shown, is MANDATORY,  to compensate for all tolerances in components values, parasitics.  During tests, dont hesitate to modify calculated value of C2 (fixed +  adjustable value) to obtain the best balance.Tolerances and parasitics management is not an exact science.
Check  of coil(s) and capacitor C1 exact values can be made with a LC meter,  like the one described in pages " Technical " and proposed by DF1SR (ex  DD0SB).
Adapt  cross sectional area of wires in toroid to maximum current in R2, do  not undersize, 3 Amps by square milimeter is a maximum.
Load  on detector diodes is a first order parametre for rectified voltages  output. Do not hesitate to use operational amplifiers  LM358 or LM324 as  buffers (they  include GND in common mode range). In this case, with  very high impedance detectors load, to avoid too high time constant in  SWR readings, reduce, if necessary, the values of decoupling capacitors  in FWD & REF outputs (10 nF is OK with operational amplifiers, 100  nF or more without operational amplifiers).
Note the wide ground strap (12 x 0.5 mm) behind toroid
 Simulations of the bridge phase & balance
Bridge old revision 3 PCB & components picture
Detail of the 0.5 pF @ 4000 V capacitor ( 2 x 10 mm of copper of RG402 )
Measurement of a 0.7 pF @ 4000 V capacitor ( 2 x 14 mm of copper of RG402 )
Front plate with meter
 
