Armed with an e-mail from Graham Wiseman and a PM downconverter I went to work.
Here's what Graham had to say.

Hello Kerry
After cutting the thing open, it became obvious that the connector screws into a threaded section of the mount/heatsink, and that there was a pin
soldered to the board that connects to the Type F Through-Connector inner. I clamped the connector in a strong pair of pliers using some rubber to
protect the thread, and simply unscrewed it. (Leather would probably have been better).

It seems obvious that this method is the best dis-assembly method instead of cutting the tube open as I did.  The mounting screw nearest the antenna also
needs to be unscrewed. When these are unscrewed, the plastic housing (the actual unit) should simply slide out of the metal tube.

Dis-assembly of the plastic housing is a different matter, I tried the methods I normally try, with limited success.  I ended up slitting it with a sharp knife, taking
care not to damage any internal components (there is no internal metal box for protection). Another factor is the increased brittleness caused by the UV exposure.

When dis-assembled, a pair of PCBs are revealled, appearing to be one for the vco/synthesizer, one for the RF/Mixer (which extends to form the printed antenna).

There are two small sheet metal shields, one on each side of the board. Under the one on the RF side of the board is a printed stripline filter.  If this is in the RF
path, it could probably be cut down to raise the operating frequency to the 13cM satellite bands. Hope this information will be of assistance.

Regards, Graham.

This is a photo of the unit in a Dodgy-Brothers test rig. It is plugged into a power feed box and then directly fed to  an RF Detector. The RF Detector drives the vertical axis of a cro and the sweep signal from the microwave sweep generator drives the horizontal axis. The funny antenna is a crude but effective broadband Bowtie Antenna connected directly to the sweeper output.
 
 

After Removing the unit from the tube I Split the case using a new stanley knife blade. I rested  the unit on a vice and then placed the blade along the seam. With a small tack hammer I hit the top of the blade quite firmly. Move around the seam untill it opens.
 

You can see the threaded hole at the top where the connector screws in at the other end is a center fed dipole with some passive reflector elements. Maximum radiation is in the direction of the Grill that this thing mounts in. It has a 7808 Regulator chip so it should run on 11 volts up.

Here you can see what Graham identified as a Microstrip filter section right in the middle. Sounds tricky retuning this. Mabey if I had about ten spare ones to play with :)  Worth a try though.
 

This sweep shows two bands just as found with the connifer. There are no gain details as evereything here is an indication not a measurement I will provide estimates of bandwidth in the next day or so. The critical factor is that the dropoff of the upper band is right on 2401Mhz. It's hard to do real measurements without real test gear. but my impression from what I have done is that it will  just scrape it in with reasonable gain.
 
 
 

This Information is provided by Kerry Richens VK1TKR in the pursuit of Amaetur Radio and the free exchange of knowledge and ideas.