how a RF connector being connected onto complex RF PCB

Y

yefj

Advanced Member level 4
Joined
Sep 12, 2019
Messages
1,220
Helped
1
Reputation
2
Reaction score
3
Trophy points
38
Activity points
7,324

Sort by date Sort by votes
Datasheet shows expected ground conductors on PCB top and bottom. They'll be connected to inner ground plane with multiple vias.

 

Upvote 0 Downvote
Using a very toxic BeCu (Beryllium Copper) as a launch pin, would make me to think twice to use this connector.
Connector, which anyway, I don't think a normal RF circuit design worth its complexity. Unless will be part of a Moon or Mars travel project.

Beryllium copper - Wikipedia

en.wikipedia.org
 

Upvote 0 Downvote
is there a practical manual on how i need physicly mount it on the PCB?
Thanks.
 

Upvote 0 Downvote
Upvote 0 Downvote
Just a tad over-reaction. It's gold plated BeCu, I believe for spring properties and precision at 35 GHz.
 

Upvote 0 Downvote
Hello , could you please point to the part where we connect to the RF ground plane?
the ground plane at my PCB is in the middle so i am i cant see where do we connect our middle gnd RF layer to the connector?
Thanks.
 

Upvote 0 Downvote
Hello FVM, i am having trouble to imagine it.
maybe the side screws needs to touch the ground pane ? why on the other layers i need to put a VOID so these screw wont touch them?
Thanks.
 

Upvote 0 Downvote
There's nothing special about the screw holes. You'll have void for non-ground layers, no isolation required on ground layers.
 

Upvote 0 Downvote
Like any end-feed SMA the pin connects on the top layer. But this 50 to 100 GHz version creates a large keep out area on the top layer of copper near the signal going out in small fan to avoid surface roughness and relies on the bottom side ground plane to maintain impedance for these millimeter wavelengths. The pin must be berryllium to have the spring properties with tension when clamped.

Solder would screw up the return loss. Don't screw it up either in your layout.

So we may assume that you are planning to use to the 8 GHz YIG Osc. to multiply up with another VCO into this frequency range using another PLL with a prescaler to mix down.

Red = Signal,, Green = slime



In order to have a 50 Ohm line as thin as shown in their test coupon board means the substrate to gnd layer must be very thin and/or high (10) Dk like Alumina.
But different models exist for various thickness boards.

ETA of working system 2030
 

Attachments

  • 1705177296242.png
    118.9 KB · Views: 37
Last edited:

Upvote 0 Downvote
You must log in or register to reply here.

Similar threads

Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more…