Finally got around to swapping out the 220R 1206 resistor in the input pad of my Kuhne 10GHz G2 transverter kit. The new 39R is the unmarked green 1206 below the BSP171. Been in a bag in my junk box since 1983.

Still a bit of flux and some fluff and muck to remove with the very wonderful Techspray E-Line defluxer pump-spray thingy before the lid goes back on. http://uk.farnell.com/techspray/1621-400s/flux-remover-398ml-spray-bottle/dp/2501335

I can now drive the 3cm board past the 1dB compression point with 18dBm out of my G4DDK Anglian 144MHz transverter. It looks like the 3rd-order IMD products are acceptable up to about 17.7dBm output with a pair of tones 1.4kHz apart at about 14dBm per tone.  Hard to see in the plot because of all the phase noise from the external LO.

The CW output at that setting is 21.2dBm, cable loss in the lead to the analyser is 2.6dB.

It remains to be seen how good the IMD/ACPR result will be when I get he proper local oscillator built.

http://cache.nxp.com/files/rf_if/doc/data_sheet/A2I25H060N.pdf

Internally-matched MMICs which can do up to 30W with a bit of compression on 2.3GHz, running on 28V at a PAE of around 40%. Reasonably linear at 10W, ACPR quoted at about -34dBc

I had two problems with my brand new Kuhne MKU 23 G4 transverter.  Both were my fault.  I blew the PTT input protection diode, so that needs replacing.  I’ve removed the original and at the moment I’m using an external diode.  Second issue is that I can’t get enough output when driving it with my DDK Anglian 2m transverter, so I need to change the input pad resistors. As I have already invalidated the warranty, I might as well go ahead.

Anyway, here is what it looks like inside.  Note the missing diode…

Nicely CNC milled cover

You *can* see that missing diode, can’t you?

OK, here’s a clue. It is near to the 16 other diodes at the top left, near the Atmel CPU.

To fix the second issue, I need to change two 1.2k resistors next to the relay and TX level pot.

Just hope I have something suitable in the junk box.  I found a 39R 1206 in there and used it to increase the gain of my Kuhne 3cm transverter kit just now, so I can get full power out of that now. For reference, the coding on SMD resistors is here :  http://www.resistorguide.com/resistor-smd-code/

Finally, a gratuitous shot of the PA stage and input amplifier

The finished DB6NT G2 10GHz transverter kit under test. I can get +19dBm out of it as the input pad has too much attenuation, don’t seem to have any 47 ohm 1206 resistors in my junk box.

Phase noise performance is truly terrible with the E4421B signal generator as the 106.5MHz LO source.  As it is a x96 mult LO chain, it is hardly surprising.  Finished system will have a proper GPSDO-locked low-noise LO.

Aftermath of the build process.  Time for a major tidy-up of the SMD assembly area.

Quick snapshot of the 2m activity session this evening. 7 QSOs at full ragchew status, beaming 210 degrees from IO93NR08.  Couple of others in other directions.  High quality of chat, proper ham radio.

Mondays and Fridays 8pm-10pm local, Wednesdays 10am-noon.  Note that my transverter has a 14MHz IF, this is not 20 metres!

Making steady progress with building the DB6NT 10GHz transverter board.  Hand-soldering it under the microscope rather than reflowing with solder paste.  Just the protection diode and some jumper wires to fit, then a good clean-up to remove the flux residues and a good drying-out and it will be time for the smoke test!  This version takes a synth LO input at 106.5MHz, there is no crystal.  Should make about 23dBm out and have a noise figure around 1.2dB.  I forgot to change the values of the input pad, so I have no idea if my G4DDK Anglian transverter will have enough output to drive it to full steam.  Maybe have to revisit the board and do some radical surgery on the IF input stage once I get it aligned.

Just received a nicely-made aluminium spigot for the socket on top of my SCAM12 from http://www.ebay.co.uk/usr/martinmpsracing

It is about 9 inches long and weighs 1.15kg including the clamp ring and Allen screws.  The section with the flat is a sliding fit inside a 2 inch/10SWG aluminium tube mast section.

I needed a few more 10MHz GPSDO connections, so I bought one of Dave’s excellent DA4-HL kits and made it up last night.  I had to add in a 4dB pad to get the levels right, but I left in the LPF for this one.  Originally I was going to use it to distribute a locked 100MHz ref, but now I am going to a multiple-output DDS for the SHF LOs instead.

Heavy rain over S Yorks/Derbys/Staffs this evening, GB3ZEM signal was enhanced by 3-5dB but also had 25Hz of doppler spread, which I think must be rain scatter.  That spread would come from a few metres per second effective path length change.  At 2.3GHz, large drops of 5-6mm start to be large enough to cause significant scattering.  Interesting that the aircraft scatter was not obviously affected by the RS.  Looks like a couple of big radar splats across the whole spectrum as well.

Later this evening, the rain had gone and we are back just about to normal.  Still lots of aircraft scatter as usual

GB3BSS on 2320.924923 from IO81SR32 (Stroud, Glos) at 206.3 degrees from here in IO93NR08, 248km, 40W ERP

Pretty much normal condx, just above noise by 2-3dB.  Path says 212dB loss.  Slight wobble on the signal

More obvious wobble with each CWID, only 4Hz or so drift over 10-15s, then back.