Twisted pair transmission line impedance question

[g579]

twisted pair impedance

Can sombody help me with some graphs or formulas to calculate or find the Zo of a twisted pair transmission line, depending on wire diameter, wire isolation thickness and number of turns/length unit ? (for Zo range of 25 to 150 ohms, wire diameter range of 0.1 to 1mm)
Thanks for any links or info .
g579

 

[unkarc]

twisted pair transmission line

Hi,

One of the best infos on your problem is a book from Jerry Sevick:
Transmission Line Transformers, published by ARRL (www.arrl.org).
ISBN: 0-87259-296-0 but there are newer editions already then this 2nd Edition. See also Noble Publishing.
If you cannot get it from local ham/amateur radio experimenters or friends in your country, then try to turn to any public library and use their inter-library service. This way you may have it in 2-3 weeks's time in your hand.

See also Motorola AN749 app note (**broken link removed**) where on its last page there is a good Reference nr 8 to an IEEE article from 1971. Here I also recommend the library service.

In the meantime I recommend to borrow/obtain a good inductance meter in the some ten/hundred nanoHenry range and a good capacitance meter in the some ten/hundred picoFarad range.
This way you could check your own hand made bifilar wire twists from enameled Cu wires as follows:
1)make at least half a meter long twist from two enameled Cu wires, try to keep uniform twists for a unit-length. Rule of thumb here is that the higher the number of twists, the lower the trl impedance becomes for a given wire diameter. Also, the bigger wire diameter you use (practicality), the lower the trl impedance becomes.
2)measure the capacitance at one end of the wire-pair (leave the other end open!)
3)measure the inductance at one end of the wire-pair (short-circuit the other end!)
4)calculate the impedance by dividing the measured L value with the measured C value and extract the square root of the division. Be careful with turning the nanohenries and picofarads into Henries and Farads to get Ohm for the impedance.

I hope these are of help

rgds,
unkarc

 

[g579]

zo of twisted pairs + calculate

Hi unkarc,

Thanks for your help.
I find the practical way you proposed in detail really useful. I'll try it.
About the transmission line transformer book. I'll try library, but it may take a month. Could you perhaps scan a few pages and attach to this board. Perhaps other peoples are also interested.
By the way does this book exist in electronic format ? Perhaps somebody can give a link to it.

Thanks for your help
g579

 

[unkarc]

twisted pair number of turns

Hi,

Sorry, I do not have facilities to scan the relevant pages


Maybe other members here could be of help in this respect.

You can gain good experiment by checking some 1-2 meter long coax cabel pieces with known 50 or 70 ohm impedances with the proposed measurements. This way you will have insight to the accuracy of your L meter which usually is more problematic in the some tens of nanohenry range. The C meters are usually more accurate and several DMMs include them.

If you happen to have access to a return loss bridge or other impedance measuring instrument then you may do the following:
1)connect one end of your trl to be measured to the bridge, the frequency should be much less then the 1/4 wavelength of your trl
2)terminate the other end of your trl or twisted wires with a 100 ohm variable potmeter (though it is rather difficult to find a good quality carbon potmeter, maybe a range of chip resistors can be used instead)
3)try to balance the bridge with the potmeter/resistors; make sure to achive the best balance possible. The indication you have found the correct termination (i.e. you terminated the trl with its own proper impedance) is that by changing the measurement frequency of the bridge it remains in balance in a wide freq range. Of course these freqs should be much lower than the 1/4 wavelength of your piece of twisted wires' mechanical length (so that the impedance transforming properties of the trl should not come into effect).
Again, with these bridge measurements you can gain experiment by testing some known coax cable pieces first.

rgds,
unkarc