Replacing the stock Gigivac G41C relays with Jennings RJ1A's

This modification was performed on my 9500 amplifier with the sole objective of achieving less vacuum relay noise. Another objective was to be able to return the amplifier to its original configuration with ease. The mod was done on 11/15/2012 in collaboration with Jim W8ZR and in his incredible workshop.

Jim had several dozen Jennings RJ1A's (Figure A) on hand, and we picked 2 that were noticeably quieter on the bench than the others. In contrast, the Gigivac G41C's were much noisier on the bench than any of the RJ1A's, and have a characteristic high frequency "twang" that is most annoying during 9500 full QSK or faster semi-QSK operation.

That said, G41C's are rated at higher power/current levels than the RJ1A's, but we reasoned that the latter have been used in previous Alpha amplifiers with consistently good results over many years of operation. The two stock G41C's are shown installed on and parallel to the back side of the controller board in Figure B. Because of the relays' very different physical configurations, mounting the RJ1A's perpendicular to the controller board and inside the board's existing cutout seemed logical. There is a RCA (red)plug/jack that takes space on a diagonal within that cutout (Figures C & E) which dictates the RJ!As' orientation on opposite corners - the only way they will fit.

We decided that the SPDT contact side of the relays should be on the back side of the board (Figure D) where the stock Gigivac's had been installed. An attempt was made to carefully measure the clearances of the modified configuration but they were too close to call. We proceeded with the modification BUT to be on the safe side, thin Teflon sheeting (kind of like kitchen plastic wrap; no adhesive, but it clings) was applied to the rear bulkhead on the controller enclosure. The Teflon material is rated at 10KV.

Thinking about noise mitigation, appropriately sized rubber grommets were located. The inside groove fit around the relay "collars" (Figure A below the metal band with blue lettering) but not too snugly. The outside grove accommodated the thickness of the controller board in the corners of the cutout (Figures C & D).

Teflon wire was used exclusively in the modification, although less critical for the low voltage coils than for the high voltage contacts.

The relays needed a little more support, so we initially added a diagonal foam strip on the back side and later a narrow rectangular piece of PC board on the front side of the controller board. The latter caught the outer groves of both grommets and was secured using a drop of RTV on each side. RTV was also used in the corners of the cutout at the grommet. Finally, for additional noise mitigation, foam was added around the relays within the controller board enclosure but not so much as to significantly impeded air flow.

The amp has been in constant use and on all bands since the modification in Nov 2012 with no problems noted. Despite being amazingly quiet on the bench, the controller/amp enclosure seems to act like a sound board that amplifies noise within it. We had expected the end result to be quieter but, as it is, the new relays make a significant improvement in the T/R noise of the amplifier, as well as in the frequency content of the noise; i.e. that annoying "twang" is gone!

73,

John DePrimo, K1JD

Santa Fe, NM

Figure A   Jennings RJ1A  - The relay's physical configuration is completely different than that of the stock Gigivac Relays, Figure B

Figure B (Before modification) - Two stock Gigivac G41C's installed parallel to the controller board on its back side

(Note that the back side of the controller board actually faces forward towards the amp's front panel)

Figures C, D and E (After modification) - RJ!A's installed perpendicular to and through the controller board

                Controller board front side Figure C, left                               Controller board back side Figure D, right

 Figure E - Modified controller board reinstalled in the 9500