Toon posts:

Nieuwe PC valt uit; (Antec) voeding probleem?

Pagina: 1
Acties:

Verwijderd

Topicstarter
Zo'n drie maanden geleden heb ik een nieuwe pc samengesteld (gegevens systeem onder aan post >) welke prima liep, tot gisteren. De dag daarvoor leek het of er plots een fan begon aan te lopen / iets ging vibreren. Dit geluid was echter heel subtiel en dus heb ik er weinig aandacht aan besteed, en dacht dat het misschien stof of iets dergelijks was. Gisteren viel de pc echter plotseling uit, alsof de stekker eruit was getrokken.

Na een aantal keer herstarten bleef hetzelfde gebeuren, de ene keer viel de pc uit tijdens het opstarten van Windows, de andere keer na een minuut of 5; vrij random dus, hoewel het wel schijnt te gebeuren onder 'heavy load' van *iets* (HDD's, Grafische kaart?).

In de HardWare Monitor van de BIOS lijkt alles prima, de temperaturen zijn allemaal dik in orde, de fan speeds ook, de 3,3V, 5V, en 12V wijken allemaal niet meer dan 3 a 4% af.
Wat wel opvalt is dat de pc niet uitvalt terwijl ik in de BIOS zit, maar dus wel onder Windows.

Waarom denk ik dat het probleem bij de voeding ligt? Ten eerste; het geluidje dat ik twee dagen geleden begon te horen is aangebleven en, wanneer ik de kast open en op mijn gehoor af ga, lijkt het er sterk op dat het geluid uit de voeding komt. Ik ben 99% zeker dat het daar vandaan komt. Bovendien lijken alle andere componenten goed te werken, op een normale temperatuur. Alle fans werken ook.
Het zou kunnen zijn dat het geluidje niet zozeer van de psu fan komt maar van een condensator die kapot is. Maar dit is natuurlijk niet zeker.

Bij mijn zoektocht naar een mogelijke verklaring/oplossing, vond ik o.a. dit (link):
...
For the most part, the SmartPower 2.0 had no trouble powering the loads we placed on it. However, one of the protection circuits kicked in when we tried to go from 400W up to 444W. This happened repeatedly, every time we tried to reach full load. The power supply could be reset within ten seconds by cycling the AC power switch, which suggests that the problem was an overly sensitive protection circuit, not a failure.

The shutdown occurred with the +5V and +3.3V lines fully loaded and delivering their combined maximum of 150W. Small loads (<5W) were also placed on the -12V and +5VSB lines. A combined load of 21A (252W) on the two +12V rails could be handled without problems. However, adding a single ampere (12W) to the +12V line triggered an immediate shutdown. The same 12W increase could be achieved by adding power to the +5V or +3.3V line without causing the power supply to shut down, even though this exceeded the published specification for these lines.

It is possible (not likely, but possible) that the overcurrent protection was legitimately shutting down the power supply, if the specifications on the web site, not the box, are correct (see page 2 of this review). Adding 12W to the +12V line would have put the output power on the +3.3V, +5V, and +12V rails at 414W — a pithy four watts over the "maximum" of 410W. However, as mentioned above, adding a similar load to either the +3.3V or +5V line did not have the same effect. The other factor to consider is that because the 12V line was actually at 12.25V at full load, the real load was about 6W higher than intended. But this seems a very small "overload" to cause instant shutdown.

We contacted Antec about this problem, and they sent us a second sample to see if it would behave the same way. Fortunately, it did not. The second sample reached the maximum of 444W load without any problem, and stayed running at that output level for some time.

Because the inability to reach full power was probably the result of an overly sensitive protection circuit, not a failure, the problem that we saw with the first sample is largely trivial. No ordinary system is likely to draw more than half of the load at which the first sample shut down. A dual processor, SLI system might peak around 400W, but such a system cannot be connected to this power supply without multiple adapters.
...
Dit lijkt dus wel op wat er bij mijn pc gebeurt, maar het geluidje dat sinds twee dagen uit mijn voeding lijkt te komen doet mij vermoeden dat er gewoon iets kapot is.

Wat is nu de reden van mijn post? Het mooiste zou zijn dat er iemand is die met gelijke specificaties hetzelfde probleem tegen is gekomen en er een oplossing voor heeft gevonden. Maar alle tips, ideeën en mogelijke oplossingen zijn natuurlijk welkom. Overigens, ik ben niet van plan om mijn voeding te openen, omdat ik dan mijn recht op garantie kwijt raak. :)

Dan nog de specs van mijn pc:
- Kast: Antec Sonata II
- Mobo: Asus A8N-SLI
- CPU: AMD Athlon 64 3700+
- Graka: Asus Extreme N6800GT (PCI-E)
- RAM: 2x 512MB GeIL Ultra Platinum PC3200 400MHz
- HDD1: WD Raptor WD360GD (10.000rpm, SATA)
- HDD2: Hitachi Deskstar T7K250 (SATA II)
De (standaard) voeding in de Antec Sonata II kast:
- PSU: SmartPower 2.0 450W ATX12V v2.01, model: SP-450 P

  • Ferret
  • Registratie: November 2000
  • Laatst online: 10-02 12:01
Zie ook het topic "Piepende voeding".

Ik heb zelf een Antec Lanboy ook met een Smartpower voeding. Ik heb regelmatig BSOD's maar de voeding blijft wel gewoon doorgaan. Asusprobe geeft nette waardes voor de 3,3 v, 5 v en 12 v.
Valt jouw pc echt helemaal uit of bedoel je ook BSOD's?

Verwijderd

Topicstarter
Hoi Ferret, bedankt voor je reactie.

BSOD's krijg ik helemaal niet. Zoals ik in mijn post al aangaf, valt de pc compleet uit. Het is net alsof iemand de stekker uit de voeding trekt, of de power knop op de voeding op off zet. Alles dus in 1 keer weg.

Overigens heb ik inmiddels de voeding met een geleende voeding omgewisseld en deze werkt prima, niks aan de hand. Het probleem zit dus echt wel in de voeding. Ook weet ik nu 100% zeker dat het geluidje dat ik hoorde uit die voeding komt. Ik zal dat topic dat je noemde ook even doorkijken, hoewel ik mijn voeding denk ik maar Retour stuur en om reparatie/nieuwe voeding vraag.