westom

Members
  • Content count

    7
  • Joined

  • Last visited

Everything posted by westom

  1. Every posted number was wrong; not just the purple wire. If the meter is working, then directions were violated. Those numbers imply a probe connecting to something. BTW, all measurements must be made without making any changes or disconnecting anything. Implied and hopefully not used in any measuring was that PSU tester. If a probe is too fat, then push a tiny screwdriver or sewing needle in. Touch / attach the probe to that or some other thin metal rod. No reason exists to spend time and money sending a PSU to Seasonic. Since 1) it is a power 'system'. A PSU is only one part of that system. You are not sending the entire 'system' to Seasonic.. 2) The meter, if used as described, means a PSU and other 'system' parts are defined good or bad - without doubts. BTW, static discharges to a computer is how to identify a manufacturing / assembly defect. Humidity only cured a symptom; did not solve the defect. A defect was not identified and elminated.
  2. First, a PSU tester is wasted money. It can identify some defects. But cannot report any power supply as good. What some consider 'in tolerance' can still be defective. That will be apparent once some good three digit numbers for each wire are obtained. A power 'system' is more than just a PSU. The entire 'system' must be tested with all components interconnected. A tester is only testing one component; not the many other 'system' components. A solution is complicated if any connector is disconnected. Second, a purple wire cannot be 0.7 volts. As noted earlier, it must be somewhere around 5 volts. Something is seriously wrong since a computer cannot even power up if its purple wire is 0.7. Put the black and red meter leads to a nine volt (or 1.5 volt) battery. If meter does not report somewhere around 9 volts (or 1.5), then the meter needs fixing. If it does measure about 9, then we will move on to discover why every posted number makes no sense, Hopefully no connectors inside the computer were disconnected. Instructions did not request that. And changes only complicate a solution. Third, once the actual static defect is identified and fixed, then those other anti--static solutions were to be recommended. Curing symptoms now does not fix an actual defect. A computer must suffer the most powerful static shock and not crash. Only then are anti-static solutions implemented as part of a 'layered' solution.
  3. When driest - where static discharges are worst - a properly constructed computer should be static shocked and not crash. Deal with each problem separatedly. For example, ignore line voltage, static electricity, and the battery backup. Get numbers from six wires using the meter as described earlier. That task should take maybe one or two minutes. That is one anomaly. Nothing more about that anomaly can be posted unitl three digit numbers from six wires are obtained. Do static discharge tests to create a discharge from one spot across the chassis to another. An incoming (finger) to outgoing (wire) path that causes most crashes must be identified. That is irrelevant to a completely different anomaly identified by using the meter. Low voltage to a computer is something below 85 volts. Meter measured voltages (from six wires) should not change when line voltage is 90 volts or 130 volts. Because this potential anomaly is separate and different from others. Deal with each separately. Information for each anomaly was described. Perform experiments to step towards a solution for each. Purpose of those facts will not be apparent until much later. Nothing more for each anomaly can be discussed until those facts are provided. Do both experiments, at this point, without knowing why or knowing what it will solve.
  4. Are you saying static electricity is cause crashes? Then fix the computer. A most painful shock to the computer case or keyboard should not cause any crash. IOW use static electricity to find an internal defect. For example, if off the floor and static discharges are less painful. Then attach a wire from one corner of the computer the floor. Static discharge to other corners. Which ones cause a computer crash? IOW determine the path from finger to other corner that causes crashes. A defect is in that path. Same applies to the keyboard. Static discharges should cause no crash. Determine any paths from finger to wire that causes crashes. Only then can we move on to finding the defect. Then later eliminate it.
  5. By shocked, I assume you refer to static electric discharge. Understand the electric circuit. Static discharge is a connection from charges in your body to charges beneath shoes. How good is that connection? Things you consider not conductive will conduct that discharge current. A stronger (more painful) discharge indicates a better connection. Discharge to various wall switch and receptacle plates. A better safety ground means the discharge should be more painful - makes a more conductive connection to charges beneath your feet. Appreciate that even wood, some wall paints, concrete, and linoleum tile are electrical conductors. Insufficiently conductive to properly safety ground the chassis. But conductive enough to discharge a finger (with less pain).
  6. The paragraph discussed surge outlet verses wall receptacle. They are electrically same. It said nothing about battery backup outlets. Battery backup outlets typically connect a computer directly to AC mains when not in battery backup mode. IOW, as if connected directly to a wall receptacle; operates electrically as if connected directly. Most of what was posted is probably completely new. And sometimes contradicts popular beliefs. Rereads may be necessary. Moving on. Start by collecting some essential facts. Set the meter to 20 VDC scale. Black probe attached to chassis. Computer is off but connected to a wall receptacle. Locate where the PSU's purple wire connects to motherboard. Push the red probe into that nylon connector to read purple wire voltage. It should be about 5 volts. But all three digits are necessary. Next touch the green wire. It will read something less than 5. Again, all three digits report useful information. Press the power button to monitor what happens on the green wire (how fast it changes). It should read a number between 0 and 0.7. Report that number. Repeat same for the gray wire. It will start somewhere near zero. Report how fast it rises when the power buttton is pressed. And its final voltage. Finally, setup the computer to execute to all components. IOW download from the Internet, while searching disks, while executing complex graphs (ie a movie), while reading a CD player, while playing sounds, while accessing USB devices, etc. Now hardware is ready to report more numbers. Measure any one red, orange, and yellow wires. Report those numbers to three digits. Information is embedded in those numbers. Provided will be why it is significant, and, most important, how a power 'system' is relevant to intermittents.
  7. Reading so many replies is a classic example of knowledge only from observation. For example, numbers should have been provided for what is a normal 120 VAC voltage. Good power for any computer is voltage from 85 to 130 VAC. 112 to 116 is well within that range. If measuring receptacle voltages properly, then the neutral to ground measurement should not have been 0 volts. Either a wiring defect exists, the meter was not set to a low enough range, or measurements were being made without something consuming power from that circuit. Do lights dim or brighten when a major appliance power cycles? That implies a house wiring problem. Moving on. A UPS does not do power conditioning. In fact, some of the 'dirtiest' power seen by a computer comes from a UPS in battery backup mode. For example, this 120 volt sine wave UPS outputs 200 volt square waves with a spike of up to 270 volts. Those square waves and spikes are created by a sum of pure sine waves. So the naive assume is it pure power. Meanwhile, superior powre conditioning inside all computers makes a 'dirtiest' UPS power into ideal perfect and stable DC voltages. Best conditioner should be inside a computer so that even 'dirty' UPS power causes no problem. The not battery backup outlets on a UPS are no different than connecting the computer directly to that duplex receptacle. UPS surge outlets or connected directly to any other receptacle is, electrically, the same connection. A better safety ground means a static discharge to a wall receptacle or switch is more painful. Static is made irrelevant by protection provided by a case and keyboard of better computers. However we never rely on one layer of protection. That means a room should be at least 40% relative humidity so that larger (but usually not harmful) static discharges do not exist. Meanwhile even where screws are locating inside the case can make a system crash due to static discharge. Like every paragraph here, details are too long to discuss in one post. But static discharge is a tool that may even identify an internal system problem I'm hardly sure where to begin. But you should again ask relevant questions for answers provided by some who actually do this stuff. How do you separate hearsay from knowledge? Knowledge also says why and provides perspective (ie numbers). No numbers is a first indication of hearsay from advertising and urban myths. Some good info has been posted. For example the WD Lifeguard diagnostics and CrystalDiskInfor provide hard facts necessary to avoid the 'it could be this or might be that' speculation. Polarity tester or a PSU tester are wasted money. Will not report something relelvant; cannot report a PSU as good. An outlet teseter cannot even say safety ground is good. It can only identify some safety ground defects. Your computer is intermittent? First step starts with numbers from six wires using a multimeter. Since a defective 'power system' component will still boot a computer and cause strange, intermittent failures. Numbers using a meter can even identify a supply that will cause failures months later. BIOS readings are mostly useless until calibrated with a meter. To some, 12.3 volts is good. However that same number can identify a defect - if numbers from the six wires are provided AND with proper loading. Again, the point. Useful answers need or are based in specific numbers.