What does the noise sound like, and where is it coming from?

"'Allo? 'Allo?"

The noise is generated by the switch mode power supply. If the radio is used with a linear power supply there is no additional noise on the frequency I am tuned to. However when the switch mode power supply is used the "hissing" sound increase to such a level that only the very strong stations can still be heard over the noise. IE if the noise level is at S4 when not using the SMPS, it goes up to S7/ S8 when using that SMPS. How does one get rid of that emi noise?

Well then, you may not be able to operate a switch-mode supply with your equipment. These things generate RF as a result of their high operating frequencies. Try complying with the tips that I suggested earlier. You may try moving the supply to a remote location far away from the transceiver and transmission lines.

Rorschach,

Nice simple test and explanation. You get a GA from me for that.

A few things I would like to add to help the OP.

Look inside the switcher for the power inductor, if it appears to be a solenoid like inductor (coil wrapped around a ferrite rod) this can be considered the switching supply's radiating antenna. For the magnetic field return path for this inductor is free space itself. To remedy this you have two options. First, slightly alter the original switching supply design by using either a gapped pot core (preferred) or a toroidal inductor where the magnetic field path need not travel through free space. Naturally you will have to determine the inductance value of the original and match this with the replacement. I've also seen people fabricate a small steel C clamp on the ends of the rod to close this inductor's magnetic loop. But in doing this, they used a small piece of an insulator (G10 circuit board chip) to provide a gap to store the magnetic energy. Second, use a steel box around the supply to steer the free space magnetic field generated by the inductor.

You may also find that the higher frequency noise being radiated by this supply occurs on the AC line cord. To remove this, an EMI filter between your line cord and supply will greatly attenuate the high frequency noise while easily passing your low frequency AC power.

Namjam,

You might try some ferrites on your DC power supply lines to see if you can eliminate the noise, something like this:

Or you may need a full-blown DC EMI filter placed inline with your DC supply lines. These types of filtering will only work if the noise is being conducted down the DC power lines, there is also the possibility that the switching supply is radiating the noise, in that case, you need to move the supply further from your transceiver (as stated by TropicalSpeed).

Do some Google searches for DC EMI filter and EMI ferrites. Keep in mind you are probably dealing with some large DC currents when you go to transmit, so if you are looking for an inline DC filter, it has to be rated for the maximum current your transceiver is going to pull when you transmit.

As a way to figure out if your noise is being conducted down the DC power line, or radiated through the air, try running your rig off a different supply, then power up the switching supply to see if the noise appears. This test is not ideal though, because your switching supply will have a different noise signature when it's not loaded.

Do some experimenting, and get back to us, let us know how you're doing.

Tom

It is true that the switcher creates rf noise as a byproduct of its operation. However, it is not necessarily true that the switcher (henceforward smps - switched mode power supply) is incompatible with the use of your radio set.

The first thing you have to do is figure out if the noise path into your set is conducted or radiated. If it is conducted, you ought to be able to put it on a long extension cord, but leave it plugged in where you use it, and physically move it around and it should make little difference. If that is the case, the question is how is the rf conducted into your radio. It's not clear from your post what the function of the smps is. If the smps output (12 Vdc, other?) does not directly connect to your radio set, then the conducted energy s getting in through the mains connection. You need to build a filter box. You can buy lots of high quality but cheap EMI filters on ebay, install them in a grounded metal box with input and output mains connectors, and be done with it.

If the smps dc output directly powers your radio set (usually 12 Vdc for amateur use), then you do the same thing on the dc side between the smps and your radio set's 12 Vdc input. Here you will need much better filtering; more of the same you might have used on the ac mains side, or filters actually designed for dc use, such as -48 Vdc used in telecom work (again available off ebay - search "EMI filter" or "rfi filter" under electrical and test equipment under business and industrial - 129 hits today under "EMI filter").

Finally, if radiation off the ac mains or dc cable is getting into your radio set via radiation into the antenna, then you still need the filtering, but you will be better off the closer the filters are to the smps, and you will need to shield the cable between smps and filter, using 360 degree shield termination at both ends. Meaning you terminate the shield the way you pull a heater hose up over the radiator circular opening, then hose clamp it. As opposed to a shield pigtail, which isn't as effective at these frequencies.

If you are a "real ham" and want to build your own filters, we can talk about that, too. SMPS emit two types of conducted noise, differential and common mode, and you have to build and install filter elements appropriately based on mode. It would take some real test equipment to figure all that out, but you might be able to use your radio set and some simple tools to get by.

Don't give up. An smps is not inherently incompatible with radio reception. Every PC has a 200 Watt or higher rated smps installed. How close or far is your PC from your radio set, and its antenna? If you are compatible with a similarly placed PC, you can make your kilowatt smps work, too.

Good luck, and let us know how you progress.

This kind of thing can be fun.

A switch mode power supply operates square wave pulses that vary in duration (pulse width modulation) at 50 KHZ to 200KHZ in various power supplies.

If you recall your fourier analysis, a square wave is the infinite sum of odd order harmonics. If your primary is 1000 watts at 100 Khz, then the third is 300Khz, fifth is 500Kgz and so on well up into the hundreds of megahertz, getting smaller and smaller as you increase in frequency, but with good grade receiver having a .1 microvolt sensitivity, then can be heard all the way up into many bands.

How to eliminate? receiver makers and transmitter maker take precautions to round the edges of the square waves a little. This reduces the high end hash from the power supplies. They also build the power supply in double Faraday shielded containers with low pass filters on the power in and power out. These low pass filters have inductive blocking of highs and capacitive bypassing of sharp transitions, and they can be very effective.

sadly, people making general purpose supplies do not do this as well as they could, it costs extra.

You can do it by making a filtered box for the power in and out, with sealed RF gaskets on the aaccess lids etc and a dual set of the filters mentioned above. Some hints can be found by drilling down this search.

http://www.google.ca/search?hl=en&q=%22PWM+power+supply%22+%2Bfiltering&btnG=Google+Search&meta=&aq=f&oq=

#1 Get rid of that switching mode power supply. FCC regulations require that your power supply be adequately filtered. That noise you hear is going out on your transmitted signal and is probably worse since during transmit the PS is far more heavily loaded.

#2 buy a "Radio Amatuers Handbook" from the ARRL (American Radio Relay League) and build a proper supply for your linear.

#3 Actually read part 97 of the FCC regulations, maybe this should be #1...

Guest,

very sensible and practical comment.

Namjam's remote commnicators are likely not to want hear his call sign!

Unles it of course the basic frequency of the smps is very high which may not get througthe transmit circuits except through non linearities.

GA awarded.

Good Luck Namjam

Sleepy

Considering the power levels involved it may not be practical or possible to filter or shield the noise away. Return the supply stating the problem with it and either go back to a linear supply or find a switcher designed to be used in the same environment with an RF receiver.

Good day to all,

Thanks for all the sugestions and help. I found some ferrites and that already helped a lot. I will get more and of the correct size to fit.

Just for interest - the other stations I spoke to did not hear the noise my tranceiver picked up from the smps.

Thanks again.

Namjam

Turn your rig on with a linear power supply and turn the SMPS on separately with some dummy load. (There must not be connection between the rig and the SMPS.) If you can hear the noise it's coming through the air so you can get your SMPS rid off (or put it into a closed shielding box).

If you cannot hear the noise a proper filter on the DC cable can help. Maybe a ferrire ring can be enough but maybe you will need an LC "ladder".

Brilliant suggestions to fix the problems, but my suggestion is simply to return it, get your money back as somewhere in that unit, quality simply went missing.....

Almost all the small 'wall wart' power supplies are now PWM switchers and they all emit broadband RF hash you can hear on any AM radio, this distributed hash from countless millions of these devices probably ups the noise floor of all the AM/SSB bands.

Where is the FCC on this? silent, I expect, because it does not bother most people

RAF? ex-RAF? (Per Ardua Ad Astra).

Aim High!

RCAF? - sorry for the slip!

Yes, Indeed, it was arduous, but I persevered and they hit me on the head = stars

Cool! Ex USAF here. Avionics.

And, I know what you mean about the perseverance thing.

I would start with a large capacitor in parallel with the output. If that is not enough, then you may need a C-L-C (pi-network) to filter it. Disconnect your antenna when you do this. When the noise is gone then connect back the antenna. If the sound comes back, you will need a Faraday cage (metal box around the supply) with it connected to earth ground and possibly with a line filter. You may also need to connect it to your radio case.

Good Post.

You have also given me a couple of ideas, we have all assumed that the OP has installed his PS correctly on a 3 wire connection, what if only used a 2 wire???? When it should have been 3 wire....???

It could explain a few of his problems......

One new thought; you have a clean transmit signal, so you only have a receive problem.

You did not detail the SMPS capabilities, if it has a logic enable/disable input (or remote sense or voltage control) you can run receieve mode from a linear supply and get peace and quiet by a signal from the push to talk to enable/ disable the SMPS. The SMPS power line rectifier will be running all the time and switch-on should be some 10ms. There should be no harm with plain parallel connection of the two supplies.

To minimise costs (if you follow the SMPS powering during receive way) a cheap enclosure is a junked micro-wave oven. You may well find a suitable power line filter in situ.

Good luck,

Zaf