> Is there a rule of thumb or a formula for calculating the number of
> twists/inch, depending on the data rate, length of the cable, etc?

Last bit 1st.
I just saw who was asking this. I wasn't trying to be smart or 
unhelpful - I think this is a reasonably good answer. I know that some 
of my answers may feel more confrontational or whimsical than 
apposite. I'll try a bit harder not to make it so :-)

______________


I believe that cable impedance and loss is the key issue. The twists 
per unit length are a means to an end in achieving this and to 
achieving rejection of induced unbalanced signals. ie I suspect it's 
safe to a first approximation to assume that the cable maker have done 
their job in meeting an agreed specification and this leaves you free 
to look at the necessarily interacting factors of loss, impedance, 
resistance, capacitance and inductance per unit length etc.

Only if you have a super critical application should you need to look 
beyond the spec to specific implementations.

ie Properly terminated cable meeting a standard will have specified 
performance limits across the required frequency range and it's just a 
matter of using a provided table or plugging in standard formulae.

All that said you can single at 300 baud with a 5v monopolar signal 
across something close to wet string over almost any distance :-).





            Russell


-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist