Question Lowest EMI ATX PSU

Jun 18, 2019
7
0
10
0
What is the best ATX PSU replacement that emits the least amount of EMI?
Dell Optiplex XE2, i5-4570S with Geforce GTX 1050.
 
Jun 18, 2019
7
0
10
0
Working at my PC drains my energy levels, get headaches, tension..

I do have found some good PSU's by looking at a lot of TomsHardware EMC test reports.

  • Seasonic Focus only SGX 650W model (needs ATX adapter)
  • Corsair RM850/1000 (not the x or v2 version)
 
Jun 18, 2019
7
0
10
0
Yes could be, but I do have a 4K screen with blue light filter.

Most ATX cases are made from steel, most higher freq. radiation goes right through it.
Unfortunately I can't find images showing all metals and frequency shielding capabilities.
 
Most ATX cases are made from steel, most higher freq. radiation goes right through it.
Unfortunately I can't find images showing all metals and frequency shielding capabilities.
If by "higher frequency" you mean X rays or gamma rays, sure. That's way higher frequency than you're going to see in PSU EMC testing though.

You are surrounded by EM radiation all the time. Wifi, cell phone, AM/FM radio, etc., and of course visible light. Your motherboard and graphics card also having switching regulators similar to the PSU, so they're also generating some EMI.
 
Jun 18, 2019
7
0
10
0
Yes, I don't know exactly what the problem is (maybe me), but I have read that steel allows radiofrequencies through. Logical, because iron transformers are only suitable around 50/60Hz.
The only way I can find out is to buy a RF meter with KHz range and see if something comes through and how it compares to GSM 900 radiation for instance.
I am looking for the 50KHz to 960Mhz RF Explorer meter. I have a meter that goes from 50MHz to 8GHz and it shows no leaked radiation.

So I wonder if I should buy that RF Explorer for 199,- or invest it in a new PSU or do nothing.

Visible light is no problem of course. As is AM/FM radio, because it is not pulsed radiation. Unless you live next to it.

Wifi Beacons (standby signal) are sent at 100 ms beacon interval. That's 10 Hz. Most routers still transmit 2,4 GHz beacons at 1 Mbit (for 802.11b compatibility), taking up 3% airtime per SSID.

If you look 10 Hz up in the RIFE frequencies, you find that it is associated with boosting adrenalin.
That shows it is like a drug. And causes people to get dumber theoretically.

I have gotten dumber, too. And that why I want to solve this negative spiral.

What does your avatar mean?
 
No, radio frequency radiation cannot pass through steel. Magnetic fields (present in transformers) are not the same as EM radiation (although all EM radiation has a magnetic field component).

Why is whether or not the radiation is pulsed important?

I have never heard of Rife machine/frequencies before, but from a quick search it appears to be pseudoscience.

My avatar is the mascot/logo for the engineering faculty where I went to university (electrical engineering).
 
Last edited:
Reactions: anort3
Jun 18, 2019
7
0
10
0
I can only tell that some "famous" dutch anti-radiation experts claim that about pulsed radiation.
For instance Charles Claessens (www.milieuziektes.nl).. Unfortunately he passed away at age 80 and his website is down.
He had a lot of spectrum analyzers and graphs online.

So the noise that was picked up in TomsHardware EMC tests, is emitted by the power cord/mains?
I do have shortened the power cord to my PC and at the powerstrip I have a shielded capacitor (Calmspace Pro unit).
 
Jun 18, 2019
7
0
10
0
I had a magnetic probe from LessEMF that plugins to a speaker, and it clearly picked up all kind of audible noises next to my PC back then. How is that possible?

https://www.rfmentor.com/content/dbm-dbw-dbuv-calculators
How dBuV compares to dBm values.
I see that the PSU output compares to wifi dBm values, although the bandwidth is way bigger.

https://www.electronicproducts.com/Power_Products/AC_DC_Power_Supplies/EMI_test_considerations_for_power_supplies.aspx
Those PSU's fail EMC tests at 230VAC but succeed at 120VAC.
 
Last edited:
What exactly is this probe you have, can you link to it?

PC cases and PSU cases aren't 100% sealed metal boxes. They have holes/gaps in the metal. If the frequency is high enough it can get through these gaps. And you may have EMI coming from things connected to your PC as well.

What did you use for system impedance did you use for converting to dBm? For air it would be 377. What makes you say the bandwidth is way bigger?
 
Last edited:
So the noise that was picked up in TomsHardware EMC tests, is emitted by the power cord/mains?
The Tomshardware test I looked at measured conducted EMC. This means they measure signals travelling along power cable, they're not even looking at radiated emissions. It looks like your other EMI testing link is doing the same based on the bandwidth they're looking at (<30 MHZ).

"FCC rules decree that any spurious signal greater than 10 KHz be subject to these regulations. The FCC further specifies the frequency bands in which these spurious emissions must be controlled according to the type of emission.
[...]
Conducted emissions, i.e., those RF signals contained within the ac power bus, must be controlled in the frequency band between 0.45 MHz and 30 MHz."
https://www.cui.com/catalog/resource/emi-considerations-for-switching-power-supplies.pdf
 
Last edited:
Jun 18, 2019
7
0
10
0
What exactly is this probe you have, can you link to it?
http://www.lessemf.com/probe.html
Magnetic Pickup (Cat. #A175) …........…. $4.95

I choosed 50 Ohm as specific impedance. What to choose for a power cord? I read that it can be 50 to 200.
I was wrong about the way bigger bandwidth. It is about the same as a single wifi channel.

Example:
55 dBuV, 50 Ohm, correlates with 8 dBm.

If I set my wifi to 1dBm I have a range of at least 15feet/5m.
So 8dBm ≃ 4x stronger. Double the range.
I was wrong about the bigger bandwith. Wifi channel is 20/40/80/160 Mhz.
from 150Khz to 30 Mhz is, 29.850 Mhz bandwidth.

So with 55 dBuV spikes / 50 Ohm within this bandwidth even at low PSU load, they emit from the mains wiring as far as 30feet/10m.

I taped aluminium tape around my power cable going to the shielded capacitor box. Does that help, or needs it to be grounded?
The range of the plug-in capacitor box is said to be from 4KHz to 2Mhz. Then my APC powerstrip has a filter from 100kHz to 10Mhz or so.
 
Last edited:
As I said above, the dBuV values that are being measured exist within the cable only. We do not have enough information to know how much energy is actually being radiated. So your calculation of 8 dBm/30 feet is not valid.

I don't know what the intrinsic impedance of the cable would be. A power cable isn't meant for RF applications so there probably isn't a lot of info on what their impedance would be.

That probe measures (alternating) magnetic fields, not actual EM radiation intensity.
 

ASK THE COMMUNITY

TRENDING THREADS

Latest posts