power supplies with minimal radio interferance?

**wpiman** · 04-27-2009, 12:07 PM

An extra cap can't hurt. I would throw one on there.

Is the M2ATX a switching power supply? If so; they can generate quite a bit of noise. If you can figure out the frequency of the interference; you can try to filter if out.

Originally Posted by durwood

Update:

So I visited my friend who was having the static problem, this time he really had tried a desktop PSU and found that he could get reception (don't ask). Sorry for not believing the PSU was causing it. He also had another M2-ATX that we did some testing with.

1) With only the M2-ATX powered up, a little bit of static, reception got worse but was still getting stations.
2) Loading down the power supply by actually powering up the PC, reception completely killed. Makes sense, whatever is happening occurs as the supply is forced to create more power.

Now we had verified that the M2-ATX infact creates interference in the FM radio band. But, why?

My car also has same tuner and M2-ATX although my M2-ATX is older by about 2 years.

So the only two different factors are

1) Older vs newer M2-ATX
2) My car has external antenna, his has an in the window glass antenna (amplified).

He did not have a good external antenna to try, so we could not rule that out. But I did wonder about the M2-ATX versions. Digging up the changelog:

I figured that since my power supply works fine, there must be a filtering issue and low and behold cap changes and oscillating problems are noted. I looked at his power supplies, both were missing cap C15, mine has Cap C15. It is located nearby the 4 pin 12V connector. Interesting.

I'm contacting the company who makes these PSU's to see if I can get more info from them.

**dyne** · 04-27-2009, 09:06 PM

Originally Posted by durwood

great stuff

By external antenna, do you mean a mast style antenna? Think you could temporarily swap power supplies with your friend to make sure it's not your antenna thats letting you get better reception? I wonder if your cars body is providing enough shielding from interference from the power supply

**durwood** · 04-28-2009, 08:03 AM

Originally Posted by dyne

By external antenna, do you mean a mast style antenna? Think you could temporarily swap power supplies with your friend to make sure it's not your antenna thats letting you get better reception? I wonder if your cars body is providing enough shielding from interference from the power supply

We thought that might be a possibly but without a good mast antenna, we were not able to try that. I forgot to mention, when we had his extra supply plugged into my car, we took the LED power light plugged into the m2-atx (was from an alarm) and moved it outside the vehicle. With it close to or slighlty wrapped around my mast antenna, the reception worsened. IIRC, the LED power is 5V, not sure if it is the same 5V line for powering the mobo, but still interesting.

It's kind of a pain to swap out my power supply, but I'm going to see if I can solder in a cap on CAP15 for his extra PSU and give it back to him. Of course that is not easy because it is surface mount and the connector is in the way. Still waiting to hear back from ituner.

**chad** · 04-28-2009, 10:39 AM

Durwood summoned me over here for the challenge

What happens when you take the power INPUT wires for the PS and move them around? Does the noise change? AS LONG AS the chassis is sealed up well then noise SHOULD NOT get out. Does the chassis share car ground? If not, is it bad if it does? If so you could try to couple the computer chassis to the car ground with a good cap at a value that will pas 88.5-108 Mhz with little attenuation. One could also try putting ferrites on the power input wiring or even wrapping the wire or making a temp pigtail of a wire wrapped around something ferrous, if that works pull the metal and see if an air-coil will work. ALL PS's of this nature will make noise, we just want to contain it as much as possible

noise within the machine is fine, lets not let it out.

Keep in mind that many car antennas that are not long, old-skool, masts are likely pre-amplified including most glass style, sans the old Chevy "windshield dipole." Make sure the preamp is energized, distance is key, keep any and all wiring away from the antenna lead and make sure that antenna lead is GROUNDED TO CHASSIS or it will attempt the force the unbalanced coax to become balanced and reception will go to hell. Ground is key on an antenna system and remember that RF ground and DC ground can be VERY different. For example cap coupling the chassis of the computer tot he car IS RF ground, but as you can imagine it makes a damn lousy ground at DC potentials.

I'll go back and re-read but I wanted to jump in and see if we can sort this out. RF can be goofy sometimes.

**durwood** · 04-28-2009, 11:20 AM

Thanks chad,

you reminded me of other things we did try in his car. We did take a ferrite and wrap the power wires through it. No change.

Cases: Both are wood.

However mine has a perforated mesh cover, not sure if that would do some blocking of sorts. Cap coupling the computer chassis to the car chassis is going to be hard.

In my car, the antenna is not old school mast 33" style, but a shortly with a wire wrapped around it. Mine is not amplified because there is no power to it and I still get reception.

His car is a civic, and you know how noisy those civics are.

**chad** · 04-28-2009, 11:58 AM

Ouch, wood case, regardless the PS should be metal right? Is that chassis ground?

It's tough to screw up FM to be honest, really easy to screw up AM, what's likely happening is that the interference is strong enough to de-sensitize the tuner!

FM works by frequency deviation, the amount of deviation of the frequency determines the audio, in fact modulation level is determined by the amount of deviation, whereas AM is simple amplitude modulation and modulation level is determined by the amount of signal strength over the carrier. So in order to screw up FM you have to either desens the tuner OR have SO MUCH noise that's broad banded that it completely wipes out any intended deviation that's getting in there. and if the latter is the case, that's pretty amazing, beyond noisy civic amazing

**chad** · 04-28-2009, 12:03 PM

As for antenna amplification, does the stock stereo supply DC along the coax in a "phantom power" type of scheme? Have you ever had the antenna apart or have anyone from forums that can tell you if there are active electronics in the antenna base?

**durwood** · 04-28-2009, 12:38 PM

Originally Posted by chad

Ouch, wood case, regardless the PS should be metal right? Is that chassis ground?

Open frame/no frame design. See here

So in order to screw up FM you have to either desens the tuner OR have SO MUCH noise that's broad banded that it completely wipes out any intended deviation that's getting in there. and if the latter is the case, that's pretty amazing, beyond noisy civic amazing

Now you feel me/us.

Originally Posted by chad

As for antenna amplification, does the stock stereo supply DC along the coax in a "phantom power" type of scheme? Have you ever had the antenna apart or have anyone from forums that can tell you if there are active electronics in the antenna base?

Hmmm, well maybe I am making assumptions because I am assuming that power is not typically applied to the antenna unless the factory radio is on, which in my case I do not have the factory radio turned on.

Well, let's assume we both have amplified antennas then that takes another piece out of the equation.

**chad** · 04-28-2009, 02:50 PM

Originally Posted by durwood

Open frame/no frame design. See here

Wow, well spank me with the newb noodle!

I'd try shielding that bad-boy somehow to be honest. I have PWM battery chargers in plastic cases that will tear up my RF gear MUCH farther away than from the trunk to roof! not to mention all the wiring around it that becomes excellent inductive antennas![/QUOTE]

Originally Posted by durwood

Now you feel me/us.

Well the civic was not FM reception noisy but it sure would tear up 2M ham radio which is not TOO far away at 144Meg, it also tears up 10M but that's AM.

Originally Posted by durwood

Hmmm, well maybe I am making assumptions because I am assuming that power is not typically applied to the antenna unless the factory radio is on, which in my case I do not have the factory radio turned on.

Well, let's assume we both have amplified antennas then that takes another piece out of the equation.

I think I'm lost now. IF it's amplified and it's NOT receiving power then your reception will flat out suck, maybe not in Chicagoland but it won't be pretty. AM will REALLY suck. Does the civic antenna look like this? (sans the XM antenna)

**durwood** · 04-29-2009, 10:19 AM

The civic antenna is part of the window disguised as defroster lines near the top of the window. Mine is about that length of yours but also has a wire wrapped around the mast. For the civic, there is a box in the passenger side c-pillar, two wires going from antenna to the box and then a two wire plug plugged into the other end. Box appears to be shielded and grounded via the mounting bolt.

More info:

We swapped M2-ATX's mine for his and even though mine has the CAP15, it still caused interference, so that isn't the problem either. I did not check mine because the HDradio was disconnected, but let's assume it would have worked fine as normal in my ride.

So get this though, he found an old school mast antenna in his shop, we used that instead of his window antenna no change UNTIL I was also touching the antenna portion, then loud and clear. :music:

I told him I could be his antenna at a small cost.

So it appears this is the problem-> combo of vehicle and power supply. I guess all the stars aligned to really screw this up. This is still no good, however there is hope if we can get an old school antenna to work. Any more thoughts? Bad grounding scheme of PSU? Bad grounding scheme for antenna (we tried a new ground point)?