I was tuning around the top end of 10m yesterday and noticed a lot of strong bubbly warbly noises, the unmistakable sound of switched mode power supply interference. It didn't take long to track this down to the power supply for my QNAP TS-109 Turbo Station.The TS-109 is a network attached storage device. But it is actually a small computer with a big hard drive, running Linux. I use it mainly as a backup drive for all our PCs, but it also hosts the shared documents folder so that we can easily exchange files from one computer to another. It runs a script that updates DynDNS when our IP address changes, which it does quite often at the moment. I could even run a web server on it if I wanted. So it really needs to run all the time. But the amount of interference it produces isn't acceptable. I tried adding some clamp-on ferrite suppressors but they didn't make much of a difference.
The power supply for this device is a plug mounted switched mode supply rated at 12V 3A. This is probably over-generous as the specification for the TS-109 gives the power consumption as 14.4W in operation. There are plenty of alternative 12V switched mode supplies available but I have no way of telling whether they would be better or worse than the manufacturer-supplied one as regards RF interference.
The only transformer based power supplies capable of supplying this sort of power are CB radio power supplies such as those sold by Maplin which have a 13.8V DC output. QNAP doesn't, unfortunately, specify a voltage range for the Turbo Station so I don't know if this higher voltage would be permissible. It would be very convenient if I could use the power supply I made for my QRP K2, which these days just keeps the K2's battery charged up, but the output from that is 14.2V.
Switched mode power supplies really are the bane of the radio amateur's life. I don't know how to solve this problem at the moment, except to switch off the TS-109, which is inconvenient.
8 comments:
Hi Julian, I don't know if it would work for you, but I commonly put power diodes in series to drop a voltage without blowing the regulation. May be worth a try. Many of the power diodes drop about 0.7V each but you can measure the drop on a DVMs diode test function.
Good luck.
73s
I was wondering about that, Theodore. I got a load of diodes in a bargain bag of components I bought at the Blackpool rally, but I don't know what they are. They are made by ST Semiconductor and the markings as you turn it round starting with the ST logo are: 1.5 KE 30A C807.
Unfortunately, the switch mode stuff is coming off of the AC side and ferrites on the DC side wont help any. A corcom SAC Mains filter between the switcher and its AC source with some ferrite (MIx 31) should do and be sure to bifilar the winding so you multiply the impedances.
Hi Julian,
The diode tip would work, 3 of themin series would drop your 14.2V down to 12V. I'm not sure if your 1.5KE30A diodes are suitable. Those are 1500 Watt transient voltage suppressors. I'll send you the data sheet by email.
Best of luck. Maurice.
Hi guys.
The 1.5KE30A's certainly seemed like diodes. Three of them in series dropped the voltage down to 12.9V off load and the TS-109 ran, but after a couple of minutes the diodes were hot enough to burn your finger so I didn't feel it was something I wanted to leave running 24/7.
Unfortunately the power supply is integral to the plug. It is possible to get 12V SM PSUs that don't make warbly noises - there is one powering my flat screen monitor that I have never noticed a problem with - the trouble is how to know that a particular one will be noise free before you buy it.
Recently I removed DLINK wifi router (my telco provided me with FTTH) at home. That router generated very bad white noise up to 2m. The most interesting thing was that noise disappeared completely when I pulled off ethernet cable. Sounds silly...
Hello Julian, I removed my external hard drive, because I had a lot of qrm. 73 Paul
I've heard of other people having interference from broadband routers. I don't seem to get any from mine. Its power supply is analogue.
Of course the noise level is so bad here that devices which contribute a small amount of noise won't be noticed. On 10m the background QRN is "only" about S3 so individual devices are more noticeable.
Post a Comment