Hardware Insights Forum

Wester547 wrote:>5mV

I think that's supposed to be <5mV.

Yes, typo on my part, the <4-5mV range.

That NetVista is a DDR1 board so 2.5V caps on the RAM power would be running right at their rating. They would be fine for DDR2 or newer, on the other hand.

c_hegge wrote:2. They were different units and both were used. My guess is that the other one was older and had more worn out caps.

I don't mean to necropost, but even in light of the worn out capacitors, you reviewed the HP-D3057F3H after you recapped the failed Teapos on +5VSB with Rubycon MBZ and Nichicon HE, and the ripple voltage on +5VSB, especially at the high loads, was much, much higher than it was even in the "before" polymod test (IE 3mV vs. 38mV...)... I didn't think the flyback section of the PSU, a completely separate part of the PSU, could be affected ripple wise by worn capacitors in the SMPS part of the PSU (the output of the forward converter) ? Or could it, since both sections of the PSU have different feedback?

EDIT: That's right, MBZ has a blue sleeve and MCZ has a black one.

It was an MBZ, not MCZ (if you look carefully, the sleeve is violet in colour). The only reason I can think of for the higher ripple even there was if the MBZ was worn-out too (after all, even Topcat can't say they're perfect).

Yeah, the older (and cheaper) Hantek oscilloscope always seemed to give higher readings than the stingray does. I'm more inclined to believe the Stingray's results, though. Mainly because they agree with others like when testing similar units. For instance, compare my ripple results of the In Win IP-P600CQ3-2 and Antec HCG-750 with those of Hardware Secrets, and you'll see what I mean. They are not exactly the same, but close enough

So you switched to the Stingray oscilloscope in 2012? I thought the oscilloscope might have been the reason. I wonder if the Hantek one might have exaggerated the results for the ATX-300-12Z as well, along with the SU-430 and Coolermaster 460W.

That's Possible. I'll only know if I get the chance to re-test them.

Revised the text a bit. At least, if memory serves TMZ are related to KZG and KZJ (and note the black sleeves with gold print).