C2550D4I/C2750D4I Power Supply

[yakideo]

Hi,

I've heard several reports that the ASRock motherboards C2550D4I and C2750D4I don't work very well with power supplies that don't support no load operation across all rails. Someone explained that since those boards draw very little power some power supplies might begin to oscillate if the minimal ampere rating for each rail isn't met, by which the motherboards takes damage.

Are there any compatibility problems with Silverstone's SST-ST30SF? It's minimum ampere rating is 0.1A (+3.3V) 0.2A(+5V) 0.5A(+12V).

My own C2550D4I broken down only after a couple of days of operation. Could it be a bad batch or could it be the PSU that is outputting unclean power? To bad I don't own an oscilloscope to measure the ripple during low load operation.

[parsec]

Re: C2550D4I/C2750D4I Power Supply

Your description of the power supply problem with these SOC boards sounds somewhat similar to the Intel Haswell processor incompatibility with some power supplies. That is, little to zero load on the 12V rail can cause its voltage to go beyond the ATX specification, and cause the OVP to shut off the PSU.

Oscillation may not be the same thing as an increase in voltage, unless it is the output voltage that is oscillating above its normal value. I'm not sure if ripple on a voltage rail qualifies as the oscillation you referred to.

I can point out that Silverstone includes the SFX ST30SF in its list of Haswell compatible power supplies:

http://www.silverstonetek.com/downlo...PU_support.pdf

The Avoton processor in this board does not seem to be a full blown Haswell family CPU, it does not support the new C7 C State and does not have onboard VRMs. It is a 22nm processor, but may be an Ivy Bridge family CPU, I can't tell from a quick scan of Intel's datasheet.

Regardless, Haswell compatible PSUs need to be able to withstand low to no load levels on the 12V rail only AFAIK, and the ASRock boards you listed don't have 12V CPU power connectors, besides the 12V pins on the 24 pin connector. But the usual reaction to the low load situation is the PSU shutting off due to its OVP feature activating.

If you have HDDs and a 12V fan or two connected to the PSU, then the 12V rail will have a load on it constantly, so that should not be an issue.

The board you had may have been bad, or the PSU could be bad as well. I'd give the board another chance, since the 3.3V and 5V rails have such a low minimum voltage spec, so my guess would be a bad board.

[yakideo]

Re: C2550D4I/C2750D4I Power Supply

Your description of the power supply problem with these SOC boards sounds somewhat similar to the Intel Haswell processor incompatibility with some power supplies. That is, little to zero load on the 12V rail can cause its voltage to go beyond the ATX specification, and cause the OVP to shut off the PSU.

Oscillation may not be the same thing as an increase in voltage, unless it is the output voltage that is oscillating above its normal value. I'm not sure if ripple on a voltage rail qualifies as the oscillation you referred to.

I can point out that Silverstone includes the SFX ST30SF in its list of Haswell compatible power supplies:

http://www.silverstonetek.com/downloads/PSU/Haswell_CPU_support.pdf

The Avoton processor in this board does not seem to be a full blown Haswell family CPU, it does not support the new C7 C State and does not have onboard VRMs. It is a 22nm processor, but may be an Ivy Bridge family CPU, I can't tell from a quick scan of Intel's datasheet.

Regardless, Haswell compatible PSUs need to be able to withstand low to no load levels on the 12V rail only AFAIK, and the ASRock boards you listed don't have 12V CPU power connectors, besides the 12V pins on the 24 pin connector. But the usual reaction to the low load situation is the PSU shutting off due to its OVP feature activating.

If you have HDDs and a 12V fan or two connected to the PSU, then the 12V rail will have a load on it constantly, so that should not be an issue.

The board you had may have been bad, or the PSU could be bad as well. I'd give the board another chance, since the 3.3V and 5V rails have such a low minimum voltage spec, so my guess would be a bad board.

I wouldn't trust the OVP. The ATX specification dictate that the max OVP voltage for the 12V rails is 15.6 V, way beyond anything the motherboard circuitry could possibly handle. I wonder what ASRock's takes is on this one. I hesitate to contact them since last time I did that I didn't even get a replay.

[Retteren]

Re: C2550D4I/C2750D4I Power Supply

FWIW, I have the C2750D4I running on the Corsair RM450 PSU without issues so far, but I do have 4 disks loading the +12V rail (it's my new NAS setup).