SeaSonic S12-500 problem: Can anyone explain the +5vsb uses and possible overload causes?

I have been unable to get my SeaSonic S12-500 to show any life with my ASRock 939SLI32-eSATA2 MB. The PS comes on when tested with a cheap PS tester and it worked with an old 1988 (Pentium2 ?) System. Likewise the MB worked when powered by an Antec NEO480 PS.

I contacted SeaSonic and got this reply:



I plan to try the resistor fix on the 12V rails. Technically I'm no sure what the +5vsb does and how to test.

Thanks for any help.
Rags


Problem Children:
SeaSonic S12-500
ASRock 939SLI32-eSATA2

Rest of the system:
Ever Case E4252BB
AMD Athlon 64 3000+ Venice
OCZ 2GB (2 x 1GB) 184-Pin DDR SDRAM DDR 400 (PC 3200) Unbuffered Dual Channel
eVGA 256-P2-N437-LX Geforce 7300GS
Seagate Barracuda 7200.9 ST3250824AS 250G
NEC 16X DVD±R DVD Burner
LINKSKEY LKA-CR15B 19-in-1 USB 2.0 Reader/Writer
SONY Black 1.44MB 3.5" Internal Floppy
DELL - UltraSharp 1907FP
Microsoft Windows XP Home With SP2 - OEM

I have a NEO-HE500 and MSI K8NGM2-FID and had this issue. Thanks to one of the threads in this forum I opened a case then called Richard (?) direct at Antec. I gave him the details, he said he knew about it, we swapped PSU's and I haven't had an issue since. (Yes I know my PSU is overkill for my system but I figured the PSU would last a lot longer than the rest of the system and I might need more power in the future.)

My question is, did Antec really fix the problem or did they just add an extra resistor into my HE PSU and make it a little less HE?

Hi

I have a Passive Silverstone 300W and have experienced this problem with the bootup. I thought it had something to do with the limit of the power supply beeing exeeded. So I tried swithing off some fans and it worked fine after that.. But now I read these posts and get a bit confused with the situation.

any thoughts or help?

Does anyone know where the Seasonic sleeved S12 PSUs with the startup fix are available for purchase online? The ones I've found only include 2 SATA power connectors in the description which I believe is the older version.

I'm experiencing the startup problem on my Asus A8R32-MVP motherboard with my older S12-330.

I've had a bad chain reaction of problems that had led to this problem with my HTPC:

- My M-Audio Revolution 7.1 had some crackles (not many), so I tried a Sound Blaster X-Fi Elite Pro after reading good reviews.
- The Sound Blaster X-Fi Elite Pro had even more crackles, supposedly due to an incompatibility with most of the NVidia NForce chipsets, so I got an Asus A8R32-MVP to replace my NForce4 based Gigabyte K8N-Ultra 9.
- The Asus A8R32-MVP has the intermittent startup problem with my older Seasonic S12-330.
- The Sound Blaster X-Fi Elite Pro still has some crackling even with the Asus A8R32-MVP.
- I also tried a ATI TV Wonder Elite (Theater 650 based MPEG2 encoding TV tuner) based on good reviews to replace my old TV Wonder Pro (no hardware MPEG2 encoder), but it seems to lock the machine up occasionally.

I did some research and testing on an Antec NEO-HE 430 and found a document called "ATX12V Power Supply Design Guide" at this link www.formfactors.org

It's a pdf file and there is one area of interest. It's on page 26 and 27. It describes how the motherboard tells the power supply to turn on. It essentially says that if pin 16 (on a 24 pin connector) is pulled to a low state (below 0.8V) the power supply shall turn on. When pin 16 sees anything above 2.0V it shall turn off (except for the +5VSB line, pin 9, purple which provides the initial power for the motherboard to perform the power up and suspend functions). See pages 36 and 37 for the pin diagrams.

With that info, I was able to turn on a few of my power supplies while they were not connected to a motherboard by jumping pin 16 (green) to pin 17 (black or any common pin).

The power supplies all came on and ran with no external loads and all voltages read properly with a DVM. My experiment included a NEO HE 430 (however it was the newer version A4), an older fortron 400 watt supply and an ultra xfinity 500.

This tells me they will power up with no load. The area that may be suspect in causing shutdowns seem more likely to come from either how well the motherboard circuitry "pulls down" line 16 or how well the power supply provides the +5VSB to the motherboard.

Your sample of power supplies is too limited. A good number of power supplies do run with no load, the latest A4 Neo HE among them. However, we've run into a fair number of power supplies that will not stay on without a minimum load. This was tested on our power supply tester, which uses a straight short to turn on the power supplies, just like your test. If you read through our recent power supply reviews, you should come across at least one where we reported on the problem. IIRC, it was an Enermax Noisetaker II. We've also seen other models with this issue, including older versions of the Neo HE. Invariably, boosting the load on the power supply fixes the problem.

I have some boot-up issues that I'd like to ask your opinion about. I own an Antec NSK6500 case which has a 430W "high-efficiency' PSU, my mainboard is an Asus A8R32-MVP Deluxe. (the rest: Athlon64-3800, ati x1600 pro, Audigy 2ZS, 3 HD's, one DVD-writer.
This system has a strange 30 seconds to 1 minute delay at start up, no signal from the graphics card, just a few leds light up. After that the normal procedure begins: BIOS messages, windows loadup screen. I have no problems with stability and performance is like you'd expect. Could it be possible that the delay is caused by the 1A demands described in this thread?

Unlikely. The problem described here is that the system won't boot at all. My guess is that something is not initializing properly, and the system is timing out before it boots. Check for IRQ conflicts and try pulling components out of the system one by one to see if it helps. A drive with a failing component may be the culprit — I've seen bad drives cause similar problems.

Thanks to the information here I was able to diagnose the issue I had when I upgraded the power supply in my new gateway pc. I had added a decent graphics card and needed the psu to power it properly. After adding the PSU my pc would not start. I found it had a minimum 12V load rating of 1A, so I exchanged it for one with a minimum 12v load rating of 0.4A and the problem was solved.

The guy in the shop had never heard of this issue before.

Anyways thanks for the help

Churchill

Glad you solved your problem.

Thanks for the fast reply in this relatively old thread Devonavar! I'll try to find out if a hard drive is causing the problem.

In the early days of switching power supplies the minimum load spec was simply used to define at what output load the various output voltages could be guaranteed to be within spec.

Has anyone actually read a spec that states that a power supply must power down if minimum load is not attained?

Could the underlying problem actually be that one or more outputs are overvolting at power on because the supply is underloaded?

An overvoltage transient would be a far more dangerous condition than a simple failure to sustain power up, but you'd need a scope to see it happen.

Hmm. I just had to return a seasonic S12-650 energy plus because it seemed dead on arrival. The system wouldn't boot and the fan wouldn't spin and the psu was emitting a buzzing noise. Jumpering it produced the same results. Seasonic said I should exchange it. I was trying to run an overclocked e6600, raptor, dvd drive and a regular PCI card while I wait for my 8800gtx. Would that have been enough of load or is it a non-issue with newer Seasonics?

Do these same Seasonic models have circuitry built in so that when the computer goes into 'sleep mode' and load may drop below a certain threshold, they will keep power to the computer?

Only your hairdresser knows for sure!

hey krick, I was wondering if you finally did this, cuz I don't know much about eclectronic, so if it works, I'll be glad if you could sell me on make some profit on it, i don't care, i just want my Fprtron Zen to boot the system!

What are the quietest PSU's today that will **NOT** require 1A min. current on the 12V rails? Preferably no current, or the smallest possible current.

My recently purchased P150/NeoHE 430 does have this 1A requirement.
I was going to swap the NeoHE with the ever-popular Seasonic S12-330 until I visited Seasonic's website and learned that their current S12-330 is now High Efficiency and ATX12V v. 2.2 compliant!!! (was formerly ATX12V v. 1.3 compliant).

My new system is extremely power efficient.
It's a Pentium-M based system with no video card.
The only power requirements besides feeding the MB are
an 80G notebook HD and an optical drive.

Guidance would be appreciated for this PSU.

Thanks!

I'm getting pretty frustrated trying to build energy efficient, low wattage computers!

A few months ago I had the familiar Antec NeoHE problem were it wouldn't boot any of my boards (all Asus).

Now I have a ENP 5130GH from Enhance and it will not boot my Asus M2A-VM setup. Actually, it always boots when first plugged in, but will not reboot after shutdowns without switching PSU off and back on.

I replaced with another, less efficient FSP power supply and the system works like a charm. However, my wattage at idle jumps from 41W with the 80 PLUS ENP3150 to 47W with the other power supply (just checked - it's 400W so that explains some of the increase).

I can't believe this problem still exists. Any recommendations on highly efficient 300W (or close) power supplies that WILL start on Asus boards??

EDIT: I just pulled an Antec EarthWatts 380 from another rig and it works wonderfully. Guess I will stick with EarthWatts for now since they seem to work great on all of my mobos.

Hello,

You might want to try the Fortron Source "Green" power supplies?

http://www.silentpcreview.com/forums/viewtopic.php?t=38885
http://www.silentpcreview.com/article280-page1.html

Hey guys, just to let you know, that I've found some pretty good quality resistor in an aluminium housing, at an online store that will ship to Canada.

Here's the link. They ship Purolator Air, it tooks twelve hours from Detroit to Montreal for my package to arrive.

http://www.alliedelec.com/Search/Produc ... 8022A9E17F

hi, may i know the revision of your EarthWatts?

I have a Rev A1 and it cant boot up my system. An older FSP (ATX v2.0) has no problems.

I have an old board, Asus CUV4X, socket 370, p3 1Ghz, and it looks like it's got the same problem, so it's not just recent mb's.
Seasonic S12-600.

Got it originally to replace the antec 430 in my sonata case on an athlon 64 3000+ DFI nforce4-ultra system, but it wouldn't power up, so I switched it back. I don't believe I hooked all the hd's (all 5 of them) up to it though, so now I think I have to try again..

anyway, on the CUV4X, 1dvdrw and 1 hd connected, it won't start the hd. fanmate controlled (lowest) zalman fan spins fine, so does the psu-fan. but you can hear soft clicks from the hd, and it doesn't power on. Power off doesn't work immediately through the button up front, have to keep it pressed for a few seconds.

adding 1 or 2 hd's doesn't help.

adding a cdrw does help, system starts fine, beeps, bios etc. fine. turns off immediatelu with front button as well.

adding a 12V case fan to a molex works too. a 5V modded fan doesn't... But revving up the fanmate to full together with the 5V fan does work again.

I have some quiet nexus fans coming in, will see how quiet they are at full load, so I can run this thing without losing a bay with a useless cdrw..

looking at this behaviour, would it power on the DFI Lanparty UT nF4 SLI-D with some extra load in the form of a fan or cdrw?

Sorry to dig up this old thread. I bought an Antec NeoHE380 for use with an ASUS M2NPV-VM, one ODD and one HD, one case fan, and a VFD. I had seen this thread before, but was under the impression that Antec had fixed the problem, and assumed that this much later only newer revision PSUs would be in the supply chain. My bad. When installed, the motherboard LED would light up, but when I turned the system on the fans would spin for a split second, then everything would shut down.

I recalled this thread and figured this sounded like a similar problem. I plugged in a fan and when I ramped it up to a pretty high speed, the system POSTed. The Antec was an A3.

I opened a ticket with Antec but didn't feel like spending the money to ship it back to them.

Fortunately, my store took the PSU back. Because of the physical constraints of my case (low profile HTPC), I need a PSU with straight-through air flow. The only other PSU my store had was a 500W Antec which seemed like overkill for this system. My case is cramped so I liked the Neo because of the modular cables, high efficiency and reputed quietness.

My questions:
- do all rev A4 Antecs run with low load now ? Does sleep mode work properly on all of them ?
- do the revision #s hold between different sized Antec PSUs ? i.e., would a rev A4 380W have the same fixes that an A4 430W would ? I prefer to stay with a 380W, would even go down in size, but if I have to, I am willing to move up in size to fix this.
- is there any way to tell what revision a PSU without opening the box ?
- alternatively, can some kind sould recommend a different PSU that will work with a low power seutp, has straight-through airflow AND is quiet (and preferably highly efficient)?

thanks !

1. impossible to say; whether sleep mode works has nothing to do with the PSU -- it's a motherboard function.
2. yes, afaik
3. I don't think so... but if the store has them, surely they'd be willing to open up the box before you take it home.
4. there are no others that really meet all the criteria, at least not among the PSUs we've tested. The retail market has gone almost entirely to 120mm fans.

Thanks Mike. The problem is the store is a little one and they had to order in the first NeoHE for me. I got the feeling they are reluctant to special order and then let me open some unknown # of PSUs ... I might go to a bigger store and see if they'll open boxes for me, but I was hoping to patronize the first because of the great service they give me.

Thanks for the great site.

Hello,

I have posted a request for help:
http://www.silentpcreview.com/forums/vi ... hp?t=41792

Can you please help me? I want to keep the Zen PSU

Thank you,
Lucian.

Here's a mystery that I came across a couple of weeks ago which I hope someone here can shed some light on:

My Silverstone ML02's built-in Enhance END-0512 PSU could not power on a Biostar TA690G-AM2 mATX mobo. On power up, the CPU fan just spun for a second and then stopped. But it had no problem powering up another mobo similar in features, the Gigabyte GA-MA69GM-S2H. I know both mobos were in working order as I could power both without any problem using picoPSU120 powered by a 80W brick.

With the Enhance powering the Biostar, if I had not connected the 4-pin P4 12V AUX connector, the CPU fan would stay on (of course the system would not POST since there was no power to the CPU). Also, if I had the P4 12V AUX connected, it appeared the PSU shut down only the 12V rail as my voltmeter was showing normal readings on the other rails. At first I thought perhaps the CPU was overloading the 12V rail. This was later proven not to be the case when I used a low power BE-2350 and the system still would not power up.

When I stumbled on this thread, I thought I found my answer. But when I tried to overload the PSU with an additional 7200rpm drive and a PCIE video card, I was still getting the same results.

Before I resort to the passive resistor solution, I wanted to make sure I'm dealing with the same issue mentioned in this thread.

Your thoughts will be appreciated.

Hello,

I intend to buy a S12II-380, but my PC system has minimum power on the 12V line. For example, with the fanless Forton Zen my PC doesn't boot (only when I add additional components...)
Do you know if the Seasonic S12II-380 has an implemented solution to fix this issue so that my system will boot without adding additional components?

Thank you,
Lucian.

I'm surprised nobody thought of this:

Take a large value (4700mfd+) 16v capacitor (~$2USD) and connect it to your 12v line.

As your system starts, the PSU sees a heavy load across the 12v line for a very short period of time. Then as the capacitor charges, it virtually disappears from the circuit which means ZERO additional current draw or heat once the system powers up.

When you turn the system off, your other 12v devices will drain the charge from the capacitor so the next time you turn the machine on, that load on the 12v line will be there again. If 4700mfd doesn't do it, just go higher. If you want to smooth out the current spike (draw less current over a greater period of time), use a small value resistor in series with the capacitor (1-2ohm).

If for some reason the cap trick doesn't work and you don't want to mess with complicated circuits or manual switches, try this:

Have a large resistor connected to the 12v supply in series with the NC terminals on a relay. Attach the coil of the relay to the 12v supply on your motherboard (fan header, etc) or any other delayed turn-on voltage. When the system is off, the relay is closed, so the PSU will see the resistor. After the system stabilizes and the delayed 12v supply to the relay turns on, the relay will open, disconnecting the resistor from the PSU so you don't have to worry about wasted power or additional heat. Because the resistor will be on for such a limited time, it probably doesn't have to be rated for the full amount of current. A 2w ceramic or metal film should do the trick without smoking.

Do current Corsair 520HX power supplies have this problem?

I purchased a Corsair 520HX PSU to use with an Intel DP35DP motherboard, and I am experiencing the "does not always turn on" problem.

Thanks!