power supplies with minimal radio interferance?

**jmullan99** · 09-17-2012, 10:07 AM

Old thread, I know. But would not putting that huge capacitor in series with the power into the M2 effectively block any conducted noise going back onto the vehicle 12V supply? In series, is it not a low-pass filter? Being so huge it cannot discharge fast enough between cycles at those high frequencies.

Not an expert so yes correct me if I am wrong.

**Mickz** · 09-17-2012, 05:28 PM

Hi, a capacitor will not pass DC and when connected in series with a supply lead it will look like an open circuit.

The larger the capacitor the LESS effect it has on higher frequencies and especially as the noise climbs into the RF spectrum. As the frequency of the noise increases, the physical size and lead length / internal construction of the cap become inductive, in other words they look like resistors in series with a small value capacitor.

Switching power supplies have noise starting at very low frequencies and running up into the UHF region. They interfere with equipment in various ways including causing mixing frequencies with other SMPS on the Motherboard and other systems to produce interference noise in the Audio chain. In reality something simple as a cap usually won’t work for SMPS interference.

IMHO heavy 0G to 4 G cable on both leads coming from the battery to the Power supply and correct grounding along with using VERY high quality double or triple shielded Audio leads between equipment will go a long way towards keeping noise to a minimum.

**jmullan99** · 09-17-2012, 05:49 PM

Thanks for that Mickz.

I too have some SMPS related noise. But the only issue I have is that the M2ATX clone impedes radio reception. Just this weekend I moved my PC out of the trunk and mounted it under my driver seat. The DirectedHD unit is in the trunk with a fender mounted antenna.

Just before moving this, I used a laptop/inverter/brick to test/see and absolutely no interference on the radio.

While the interference has been noticably reduced, it is still not gone.

Anyway, I digress. SMPS are a difficult beast to deal with.

Cheers.
John

**Mickz** · 09-17-2012, 05:57 PM

Hi John, yes that’s also a big problem with DAB and in some cases GPS as well. If you want to really kill HD and DAB Radio just try some of the newer reversing cameras mounted anywhere near the Radio aerial, some of these newer small cameras have an internal SMPS that kills everything within range.

Cheers.

**johnjore** · 12-11-2012, 05:59 AM

Old thread, but still very relevant. My DAB/FM monkeyboard radio is usless in my car with an M4-ATX powering the CarPC. Tested with a laptop and the signal strength went from 83% to somewhere between 0 and 10% when the CarPC itself was powered on. Assuming the issue is the M4-ATX. I'm therefore looking for a solution, and if possible, a less technical one, if one exists.

Current options:
1) Follow this thread and spend around 200$ on parts to build a box for the PSU to contain the interference.

2) Would an external DAB/FM car antenna bypass the issue? My empeg radio appears to work fine with the current FM car antenna and the M4-ATX powered up. (Ford Australia charges around 300$ for the parts to convert the antenna into a DAB/FM!)

3) Use a PSU from a different vendor. But which make/models are "known" to be ok? (Would an Opus 180W @ 200$ fix the issue?)

Advise appreciated.

JJ

**jmullan99** · 12-11-2012, 06:10 AM

Originally Posted by johnjore

2) Would an external DAB/FM car antenna bypass the issue? My empeg radio appears to work fine with the current FM car antenna and the M4-ATX powered up. (Ford Australia charges around 300$ for the parts to convert the antenna into a DAB/FM!)

To share my experience....

I put an external antenna on my car. I moved the PC from trunk to under the driver seat. Still get some interference, just less so. I put a couple ferrite beads on the power wires. Again, it helped a bit. Moving the power wires around shows that the noise is being transmitted (conducted?) back OUT of the power supply and radiated out from those wires.

I do not know of a "better" power supply (others might). The manufacturers need to redesign these with better filtering.

**Mickz** · 12-11-2012, 04:51 PM

Hi, the M4 is the main culprit, PC can be generate some but to a lesser degree and finally any device with an SMPS, such as a (some) cameras. It's a pain to have to go the expense and hassle of fitting the PSU into a correctly RF screened case, and even then placement and lead dressing around the PC are still important.

It's not going to be easy to find a PSU that is (truly) free from radiated interference. You really need to find four or more people with PSU brand X that are running DAB-FM and GPS and have tested reception using a laptop (PSU on and off) out of the metropolitan area where reception is not saturated with DAB-FM signal levels. I have seen people fluke an installation with an M4 and have no reception issues in their location. I have technically proved beyond any doubt that the M4 is a noise generator, but so too are most SMP supplies, it just comes down to how much they produce.

As others said, an external roof mounted aerial helps as it did in mine, but won't solve the problem when signal levels start to drop.

Like I mentioned before, a tiny 1/4" x 1/4" SMPS in a camera completely killed DAB in my vehicle and no amount of simple ferrite beads, shielding or lead layout would fix it.

**johnjore** · 12-12-2012, 12:44 AM

Thanks for the replys and confirmed its the M4-ATX by using a migrus C5 DC-DC PSU from around 2004. Signal strength on the radio: 94%. (vs. 0% to 10% with the M4-ATX). CarPC is under passanger seat with radio in dashboard and internal antenna sucked on the front window (passanger side).

I'm not looking for perfection, just something acceptable.

Guess I too will start looking for parts to house the M4-ATX as the migrus C5 states it requires 14-19V and I've fried 4x of them in the past, so its just a matter of time before it dies in the car.

JJ

**johnjore** · 12-12-2012, 06:57 AM

Ok, so based on the posts from page 14 to 22 this could work:

"The Box" Jaycar, http://www.jaycar.com.au/productView.asp?ID=HB5050 Its bearly large enough because of the internal mouting points for the lid. The circuit board is at one end of the box. The other half is empty, leaving sufficient room for the cables. Need to add a hole for a fan and 4x ventilation holes.

1x fan (of some description from a box of old fans)

14 caps: http://au.mouser.com/Search/ProductD...JaiSamRA%3d%3d

7x for output, assuming each color is a different voltage, http://img.tomshardware.com/us/2005/...max_nt_atx.jpg and no cap required for GND.

2x for input. BATT + IGN. Nothing required for GND. Bolt through the case to GND wires on other side.

2x for power button on/off

2x for USB cable as two of the 4 cables are 5V and GND.

1x for Thump. GND shared with the rest of the GND cables.

Double sided PCB as the Mouser components are for soldering, http://www.jaycar.com.au/productView.asp?ID=HP9515 will be used as per Mick's description.

JJ

**Mickz** · 12-12-2012, 05:11 PM

Make sure the box is large enough to allow the ATX lead to plug into the M4 header and also allow for easy routing of the thick cable loom connections to the feed through caps, also allow for the length that the feed-through caps penetrate into the housing. You want to be able to easily remove the M4 Board from the case if needed. That means room to unplug the ATX lead and clear the feed throughs. What feed through caps are you planning on using?