Announcement

Collapse
No announcement yet.

Protecting from EM Radiation

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Protecting from EM Radiation

    Whenever I drive in specific places (e.g. near an army radio tower on the way to work) the voltage to my Car PC spikes and the M4 ATX trips the shut down sequence. I was wondering how everyone else deals with this?

    I've looked into voltage regulators I could put right before the PSU but I can only find 12V ones at around 1A - I need about 20A. Is there anything I should be doing extra to shield my power cables to prevent voltage spikes?

  • #2
    What are your computer specs? needing 20amp seems very high to me. All of my car pc's/test boxes very rarely go over 4 amps under full load. SNO

    Comment


    • #3
      Originally posted by SNOtwistR View Post
      What are your computer specs? needing 20amp seems very high to me. All of my car pc's/test boxes very rarely go over 4 amps under full load. SNO
      Sorry I just took that figure from the PSU (it's a 250W PSU), 7-8A is probably closer to right but most of the 12V voltage regulators I've seen are 1A.

      Specs are:
      Intel DH77DF
      Core i3 3220T
      40GB Intel x25-V
      Asus Xonar Essence STX

      + two USB powered LCD screens (1A each @ 5V), a USB charging Wifi modem (900mA @ 5V), potentially two mobile phones at the same rate (900mA @ 5V each).

      Comment


      • #4
        How do you know the input voltage is the problem? (And if so, why not a filter?)

        Isn't it signal injection into the M4 etc? (Especially the M4 - it's a known noise emitter.)

        It's probably a need for shielding which IS th protection from EMR.


        And you'd need a dc-dc converter rather than a regulator, but that too would require shielding.

        Comment


        • #5
          Originally posted by OldSpark View Post
          How do you know the input voltage is the problem? (And if so, why not a filter?)

          Isn't it signal injection into the M4 etc? (Especially the M4 - it's a known noise emitter.)

          It's probably a need for shielding which IS th protection from EMR.


          And you'd need a dc-dc converter rather than a regulator, but that too would require shielding.
          I've tested the input line with a volt meter, it shoots up to 30V in some places which trips the shutdown sequence. I've managed to block that by having my car PC software make the shutdown button do nothing when travelling at speeds greater than 10km/h but at one specific place the voltage shoots up to about 50V and the system crashes anyway.

          So how do you suggest I shield it? The M4 ATX is in a steel PC case which is grounded to the chassis. If it's emitting noise itself would there be a problem housing it in the same case as the rest of the PC? Do I need to shield the power cables too?

          Comment


          • #6
            Are you blowing bulbs and other components in the car that are not used to that voltage? This makes no sense that kind of voltage spiking without other damage. Waiting for OldSparks reply Just my 2 cents SNO

            Comment


            • #7
              There is a thread on this system that deals with RF shielding the very noisy M4-ATX. Not sure where it is but this might be your solution. Also make sure your input wires are large enough. Depending on what else you are powering and how far you are going you are likely going to need 8 gauge power wires.

              You MAY want to look at another power supply for your computer because of the known RF issues with the M4-ATX supply but check out the thread where the posted does a lot of work to remove the RF from the power supply.

              Comment


              • #8
                Originally posted by SNOtwistR View Post
                Are you blowing bulbs and other components in the car that are not used to that voltage? This makes no sense that kind of voltage spiking without other damage. Waiting for OldSparks reply Just my 2 cents SNO
                The computer seems to be the only think that's being affected. The car's volt meter goes up less than a volt - my thoughts are that the power cable to the computer is picking up the EM radiation like an antenna.

                Originally posted by redheadedrod View Post
                There is a thread on this system that deals with RF shielding the very noisy M4-ATX. Not sure where it is but this might be your solution. Also make sure your input wires are large enough. Depending on what else you are powering and how far you are going you are likely going to need 8 gauge power wires.

                You MAY want to look at another power supply for your computer because of the known RF issues with the M4-ATX supply but check out the thread where the posted does a lot of work to remove the RF from the power supply.
                I used jumper leads (without the clips) for my input wires - they're very thick. Is the M4-ATX really that bad? Are there any good alternatives that have the same features (auto turn on, auto delayed turn off, etc)?

                Comment


                • #9
                  Originally posted by wprime View Post
                  I've tested the input line with a volt meter, it shoots up to 30V...
                  Which input - 12V to the M4, or from M4 to PC?

                  If it's at the M4 input, does it shoot up with the M4 removed or off?

                  If it's between the M4 & PC, I'd try another brand of PSU unless some simple bypass capacitors work (eg, 0.1uF or 0.01uF or smaller across the M4 output; maybe input, and signal wires), else as per redheadedrod.

                  Also, what type of voltmeter - DMM? Is the DMM shielded? AC volts or DC volts?

                  Comment


                  • #10
                    My guess is you've got insufficient current paths between the M4 and the battery. The car battery should shunt any large voltage spikes to the point that it wouldn't be measurable by the M4. If, however, you have a high resistance somewhere in between the M4 and the battery, then you can certainly get some spikes shooting back. After all, a wire without an easy current path is the definition of an antenna. This needs to be sufficient low resistance paths on hot AND ground.
                    2000 Ford Expedition Eddie Bauer - Bi-Fuel Gasoline/CNG
                    Intel D945GCLF2 w/512MB RAM, CL Audigy w/KxProject, M2-ATX, Lilliput EBY701

                    Comment


                    • #11
                      But if that resistance were significant, the M4 would not operate.

                      Comment


                      • #12
                        Originally posted by OldSpark View Post
                        But if that resistance were significant, the M4 would not operate.
                        Not entirely true - a switching power supply can handle a very wide input range and not affect it's output regulation. The fact is, it's pretty much impossible for a standard RF broadcast to induce enough current on a car battery to cause the voltage to jump more than double the nominal battery voltage. The induced current would be shunted right through he battery, and the voltage would never be allowed to climb that high.

                        What he is describing is one of two things - either there is something preventing that induced current from shunting without generating a voltage spike (high resistance), or he is not really getting a voltage spike on the line, but his meter is.
                        2000 Ford Expedition Eddie Bauer - Bi-Fuel Gasoline/CNG
                        Intel D945GCLF2 w/512MB RAM, CL Audigy w/KxProject, M2-ATX, Lilliput EBY701

                        Comment


                        • #13
                          Yes, it is well known that power lines can pick up EMI, but that is not path resistance. Impedance maybe, but not resistance.
                          Batteries are BIG capacitors and hence snub AC (as in the typical relatively low frequency of alternators - eg, up to 1kHz etc) though that might not mean high frequencies (I'm not sure, but electrolytic caps do not snub/suppress/quench HF noise...).

                          But I think you and I are in agreement. I too wonder if the voltmeter is picking up HF voltage (hence the "rubbish" reading) as well as EMI injecting the M4 to cause shutdown since IMO they are the most likely, and hence my replies.

                          I was going to ask what sort of radio tower - presumably not ULF Navy, but maybe HF or UHF and higher comms & radar.
                          But I'd suspect the M4 since they may be uncommon - ie, I expect PCs to be more common and hence if interference from that radio tower was a problem, it would be reasonably well known.

                          But I am not familiar with M4 over-voltage shutdown, nor whether the OP's PC shuts down due to overvoltage (in my experience many do not, but that may well have changed - after all, there enough supervisory chips in common usage).

                          Comment


                          • #14
                            Old thread I know, but clip-on ferrite beads (#31 material works well) every half foot or so on the wires coming from and going to the PS may help a bit.

                            Comment

                            Working...
                            X