开云体育

I was not very satisfied with the quality of the calibration standards I ordered off of ebay, so I made my own out of PCB edge mount male SMA connectors as shown here:



The picture is attached. It is simple. One of them is unmodified for the open, you solder a wire between the center conductor and two PCB edge mounts for the short, and use two 0805 or 1206 100 ohm resistors, soldering the resistors from the center pin to the edge mounts.

A pack of 5 of the connectors cost me under $2 on ebaay.

Why not use ordinary BNC hardware with an adapter?? Maybe not good all the way to 900 MHz but fine for casual use.
Bob

The BNC calibration standards are not nearly as good as the SMA ones.
There is more stray capacitance and series inductance in BNC because of the
bigger size. I made BNC calibration standards and they work ok, but the
error is definitely much greater above 400 to 500 MHz.

On Sun, Nov 24, 2019 at 1:55 PM Bob Albert via Groups.Io <bob91343=
[email protected]> wrote:

Why not use ordinary BNC hardware with an adapter? Maybe not good all
the way to 900 MHz but fine for casual use.
Bob
On Sunday, November 24, 2019, 10:35:34 AM PST, Daniel Marks <
profdc9@...> wrote:

I was not very satisfied with the quality of the calibration standards I
ordered off of ebay, so I made my own out of PCB edge mount male SMA
connectors as shown here:




The picture is attached. It is simple. One of them is unmodified for the
open, you solder a wire between the center conductor and two PCB edge
mounts for the short, and use two 0805 or 1206 100 ohm resistors, soldering
the resistors from the center pin to the edge mounts.

A pack of 5 of the connectors cost me under $2 on ebaay.

Only non-inductive resistors are needed around the GHz, which must be soldered to the connector and cut off excess feet. The calibration on the images is appropriate for the shortwave. Calibrations made in China are tools that can be used for example ebay sellers are likely to have fallen out of quality control, so sell for that and that.
The measuring point should always be the same as the calibration point.


Gyula, HA3HZ

Your standards could be significantly improved as follows:

Short: Don't use wires, too much added inductance. instead use braid or a thin piece of sheet copper (with hole) pushed down flush against connector body and soldered all around.

Open: Long center pin makes for added uncertainty of reference plane location. Instead, cut off the center pin flush with the connector body.

Load: Resistors floating off the connector body adds inductance. Instead, trim center pin to short stub (~1 mm), and solder down the resistors laying right against the connector body, preferably turned over with the coated side facing the connector body.

Dana (K8YUM)