Phase matched Coaxial Cables Assemblies: How to make this?

phase matched cables

Hi,

I need to make semirigid coaxial assemblies matched to +/- 0.5 degrees (at about 1GHz).

Does anybody know where I can find a tutorial/book?

Does anybody can send a tip?

Thanks!

how to make sma cable

There are two things to keep in mind. One is to make them from the same spool of coax so that part is identical. The other is to be very precise when cutting them to length and applying the connectors.

You may have to make a test fixture to measure them by comparing the phase of each of them against a single coax and labeling them with the difference. Then select pairs with the same difference labels.

This can be very expensive and time consuming. An alternate approach is to have a phase trim in the final product. You could use a series L shunt variable C in the signal path and trim the C value. Select the LC values to have a low pass filter with a 40 GHz cutoff so there is just a few degrees of shift at your 1 GHz operating frequency. The temperature coefficient of this network will be another source of error.

making sma cable

Hi Flatulent,

Thank you for your response.

I have very long cable lengths to cut it with enough precision.
Also, I am not allowed to insert discret components.

I am trying to use, preferently, a method of tuning the cable while one of the connectors has been soldered. After the tuning, I'll solder the other connector.

Do you know any way to use such a method?

Thanks!

how make sma cable

The discrete component idea was from a system level decision. They would be in the boxes that the cables go between.

It is very tedious, but making slight changes to the connector placement is probably the only thing. Deforming the cable will change the delay, but it will also make reflections which will have other effects on your system.

A further way is to use software to make a table of errors to correct the readings.

how to phase match cables

After i read the precedent posts on this topic, I'would add my personal opinion.
I' think there is an another way.
If your semirigid is .141 and if your connector is the SMA that use the no central pin (a type of SMA that use the central cable to do the male pin).

1) connect one end of the calbles to the connectors.
2) trim, as accurate as possible the cables lenght
3) mount the SMA "pin free" on the other hand but don't solder it (or don't crimp it).
4) do a delay measurement
5) gently trim the longer
6) fix the connectors with SnPb or crimp
The job is now finished.

But are you sure that long cables will remain stable during the time, or during the bending, or during the Temperature change?

0.5°@ 1 GHz means 0.42 mm (in vacuum). What the TCE of semirigid? I don't know what cable are you using, i suppose i.e. TCE may be 100 ppm/°C .
In this case, for every meter (in vacuum), 1°C of temperature change will produce a difference on delay of 0.1mm. Is it acceptable?

coax phase trim

Hello people.
I would like to add another idea that I had successfuly used in the past.
I don't know the lengh you will be using. In my case it was about 20 centimeters.
You make the cables the best you can to get the best match possible, for example using the techniques already proposed by our colegues.
After the cables are made you can, with pliers, slightly smash them. This will give you some phase variation that can go, maybe, to 0.5 degrees( I don't remember how much I got at those old times).
I my case I had no noticeable changes in S11 but I could get small delay variations in S21 enough to make the cables match perfectly.
Just make some tests with a small cable to see if this method can be usefull in your case.
Greetings to all.
S.