Re: LDMOS mounting inductance confusion.

Robin Szemeti - G1YFG

Quite so, and that matches my understanding, and I would always do that, but clearly there are some builds out there with poor or no contact at that point and I would have expected far more serious degradation that there is, which is why I ask ...

On Mon, 27 Jul 2020 at 10:08, Ian White <gm3sek@...> wrote:

Bearing in mind that RF currents only flow on metal surfaces due to the skin effect, and trying hard to avoid those ambiguous words "ground" and "groundplane"...

The return path for the gate drive current current begins on the top surface of the heat spreader (to which the LDMOS device is soldered). Normally that path is very short before it meets the edge of the copper backplane on the PC board. At that point the current path transfers to the inside surface of the backplane, and so completes the return path into the stripline matching circuit.

Using paper packing between the underside copper of the PCB and the topside of the heat spreader is deliberately inserting a short section of very low impedance transmission line in series with the current return path from the LDMOS gate. The electrical length of this section of line will depend on the locations of the board mounting screws and the quality of contact at those points, which will also affect the conductive losses. Added to that will be the dielectric loss due to the paper.

Well, it clearly works and the lossy components may even be doing some good - but it's messy, with an awful lot of unknowns.

Am I missing something, or what's wrong with bolting the PCB flat down onto the spreader and filling the gap between the gate leads and the topside of the board with honest, reliable metal?

73 from Ian GM3SEK

On 27/07/2020 08:41, geoffrey pike via wrote:
Hi Robin,
Yes i agree seems very strange notion, how many mounting screws does Jim use to hold the pcb to the copper spreader?
Also ask Jim directly as i'am sure he would give you an answer
seems pointless going to all the bother of soldering direct in the first place!

On Monday, 27 July 2020, 00:21:34 BST, Robin Szemeti - G1YFG <robin@...> wrote:

The 13cm PA notes are interesting and remind me of a question I had from when  I put the W6PQL 23cm boards together.

Here's my confusion:  The source of the device is connected to the flange. The matching circuit that is a transmission line is made from PCB, with the groundplane and the top conductor forming the transmission line transformer.

My understanding would be that the  groundplane end of that transmission line has to terminate at the source, with minimal inductance, just as the "inner" has to terminate on the gate connection.

The SP5MXU documents talk about the 13cm boards oscillating if the connection between the groundplane and the source is not sufficiently good or close enough. This sounds reasonable and as I would expect, a good grund contact is important.

My confusion is that W6PQL mentions getting the board height just right for the devices, if necessary, adding thin sheets of paper between the board and the copper spreader ... obviously in this arrangement, the source to groundplane contact is somewhat convoluted and distant ... yet, apparently many have been built like this with no ill effects.

Surely that cannot be right? Did I miss something? Is there some "well, the groundplane and the spreader are basically in the same path it doesnt have to be actually electrically in contact" thing that I missed?

Best regards,

Robin Szemeti

Redpoint Consulting Limited 

E: robin@... 
T: +44 (0) 1299 405028 
M: +44 (0) 7971 883371 

The information contained in this e-mail is intended only for the
confidential use of the above named recipient. If you are not the
intended recipient or person responsible for delivering it to the
intended recipient, you have received this communication in error 
and must not distribute or copy it. 
Please accept the sender's apologies, notify the sender immediately 
by return e-mail and delete this communication. 

Thank you. 

Join to automatically receive all group messages.