Monday, 15 December 2014

LED lighting Interference...solved, thanks to G4BVY

A couple of years ago I was convinced that swapping out the apartment lamp lighting for LED lights would save money so I bought a load of GU10 6W LEDS from Homebase and Installed them. Disaster! When any lights were turned on the noise floor on 10m and above shot up by a considerable amount. Classic FM disappeared on 100MHz. Classic FM on DAB disappeared.


Recently this situation has resulted to running some Tuesday night UKAC events by the light of a (non LED) bedside lamp, with all other lights off!


I was told LED lights had got better in the intervening years, interferencewise, so I bought some more  LED samples from Tesco, B&Q and recent samples from Homebase to see if they had. The Tesco ones were the best of the collection, but still produced a noise floor increase.

It was at this point that Roger G4BVY posted to the MHRAC reflector that he had bought some Phillips 4.5W LED in twin packs for 9.99 from Wickes from which he could detect no interference and they were quite bright. Off to Wickes in Worcester to buy a couple:-

The GOOD LED Lights


Installed them in the stairs lights listened on 144MHz turned the lights on...Nothing!  The noise floor did not increase at all!

I was intrigued how much the noise floor increased on 2m for each type, so I needed a calibrated receiver. The easiest option was to use the SDRIQ in continuum mode with my Nacton transverter for 144MHz. The increase in noise floor could be easily seen in dB.

The effect of turning on the Homebase lamp on the 2m noise floor

The overall results were Interesting:-


Homebase  6W (2012)             24dB
TCP 5W (Homebase)              26dB
TCP 4W  (B&Q)                     18dB
Tesco 5W                                 8dB
Phillips 4.5W                           0dB


The Phillips was by far the best. Interestingly Homebase don't actually sell their own brand any more, they only sell TCP so I got one of their 5W ones. haven't improved their version, It was returned as "unfit for purpose"! I was so impressed by the Phillips I went back to Wickes to get replacements for all the other Homebase lights.  Before fitting the new ones I decided to see how the Homebase LED interference varied with frequency. This time I used the SDRIQ with external Nacton transverters for 6, 4 and 2m. The noise floor increase was as follows:-

7MHz                0dB
14MHz              5dB
28MHz              7dB
50MHz              16dB
70MHz               18dB
144MHz             24dB


As a check, I also measured the results using my ANAN-10 as the receiver. The results were within +-1dB


The remainder of the Phillips LED were fitted. For the record the twin packs have the UPC 8718291789994. I looked on all the frequencies I had (even 136kHz) but couldn't find a noise floor increase. Good news all around; lower noise, decreased power consumption and Meg says that, unlike the previous LEDS the increased brightness will even support her sewing! Thanks Roger.



Thursday, 4 December 2014

Seperate Receive and transmit paths for the Elecraft K3 Internal K144XV Transverter

When using the Internal K144XV  2m transverter in my Elecraft K3, it has a single BNC connector on the back panel for receive and transmit signals. Having obtained an HA8ET 2m mast head preamp I needed to separate the two signal paths to avoid putting RF up the back of the preamp.

Elecraft in the past have said it is possible to connect a separate RX antenna to transverter but nothing has ever been documented! The only clues have been the comment  in the K144XV manual about not removing links P2 and P3 inside the transverter. Time to investigate!

I took the cover off the K3 and the K144XV transverter and found that if the link between pins 1 and 2 of P3 was removed the receiver went deaf.  Pin 3 was ground.  I connected a coax cable with centre conductor to pin 2 and the braid to pin 3 the GB3VHF beacon could be heard on my collinear. The radio still produced the normal 9W out of the BNC connector on the back panel.
I did not want to solder to the header so I found a 3 pin female header (previously used for arduino shields) and soldered a piece of RG174 with a BNC socket on the other end, which hung out of the back of the lid panel. I did look at tidying up the cabling by using the AUX RX BNC connector but this would mean taking the K3 apart and removing the sub receiver every time I wanted to change between internal and external 2m antenna switching.
For reference P2 link allows the Receive IF output to be diverted from going to the KXV3A module. P8 link enables the DC voltage on transmit on the centre pin of the Transmit port.


I have written up the "Seperate  Receive and transmit paths for the Elecraft K3  Internal K144XV Transverter" modifications with pictures at http://g4fre.com/kx144_split.htm