Modular Power Supplies Less Efficient Say Makers
It's all about power efficiency these days.
Up until the last few years, few paid any attention to the little box that actually powers a PC--the power supply. Now with certifications like the 80-Plus, we're able to at least distinguish efficient power supplies from the rest of the pack. But even with 80-Plus, more and more power supplies are jumping on the bandwagon, so while product quality goes up overall, things are still not where manufacturers want them to be.
There's more detail to power supplies than can fit in the scope of this article, but we spoke to several power supply manufacturers at Computex and the focus customers put on their units.
Despite modular power supplies being a hot product, don't be quick to grab the up-take. The companies we spoke to told us that modular power supplies are less efficient than those that have their cables hard-wired. The reason being that the extra connection you make, adds resistance, and hence, increases heat and reduces efficiency.
Enermax, Seasonic, and Corsair all told us that if you're looking to maximize efficiency, go for a non-modular PSU.
Are the differences great? No one gave us any hard numbers, but they did indicate that the numbers are measurable. PC Power & Cooling, now part of OCZ, says the same thing:
Due to their look, convenience, and cost savings for manufacturers, modular plugs have become a popular power supply feature. Unfortunately, there has been little or no discussion of the impact of this feature on overall performance and reliability. The fact is, modular plugs limit power by adding to electrical resistance. The voltage drop can be as much as would occur in 2 feet of standard wire. Worse yet, modular plugs utilize delicate pins that can easily loosen, corrode, and burn, creating the potential for a major system failure. That's why professional system builders specify uninterrupted wire!
So in the end, you have to choose: do you go with a modular PSU for cable management and appearance, or do you spring for a hard-wired PSU?
Hearing similar claims from other manufacturers (some of which sell mainly modular units), though, give them more weight and credibility.
I'd still love to see some hard numbers, anyway.
Sorry My mistake:
R(loss) = 0.04Ohm (2 feet of 20 AWG wire + 2 feet for return ground)
Power Loss = I*I*R(loss) = 40*40*0.04 = 64Watt
This is a power loss of 10% for 650Watt PSU
Before I bought my modular PSU I did a fair amount of reading and research and in end I decided to go with a Coolermaster Silent Pro 600W. I had heard that Corsair were THE makers of modular PSUs, but in Australia you pay significantly more than the standard US to AU currency conversion due to their popularity. Anyway, in my searching for reviews I stumbled across Jonny Guru, an online reviewer who almost exclusively rates PSUs, and tests them based on many things including whether they are infact 80-Plus.
He gives the Corsairs (MX620W up to MX1000W) between 9.5 and 10, off the top of my head, and gave the CM M600W a score of 9.0, which I was really happy with.
Just google his name if you haven't heard of him.
You would think they are talking about rubbing two sticks together here!.
Have you actually mesured the difference with a good multimeter? I have, and it says there is no difference. This sounds like bs to me. IF there is any merrit, perhaps manufactures should look at trying different connectors or metals like silver/gold plating.
I think this all started because one manufacturer has the patent on a form of the modular connector and the others don't want to pay.
Why not just design one that isn't patented and standardise it.
All for modular. Cleans up the case nicely.
Second: I've been running pc's from 1999 and older. None of these ever had a corroded pin. If they did, the impedance wouldn't go up, because as you plug or unplug the plug, the corrosion gets scraped off, and newer metal will be exposed giving a good connection again.
Though it is true those mod pins are not very surdy, and could break. Especially for testing mobo's that regularly get hooked up on different hardware like PSU's.
If modular pins last 5 to 10 years, about as long as a powersupply does, then I'd say go for modular.
The chances are greater that you'd upgrade your rig with an extra DVD/blueray player, harddisk, cpu, more USB devices, or newer gaming videocard, and need the extra power; or,perhaps, buy a lower power PSU, as newer technology comes out, and smaller die ram/CPU, or newer SSD drives reduces the overall powerdraw.
"There’s been a lot of scare tactics used to convince people to not go with a modular power supply. But the reality is, even at high loads the resistance is quite minimal if the correct measures are taken."
Give us hard numbers, especially since so many modern modular PSUs have the ATX/PCIe/and primary SATA cables hardwired and only have extra peripherals/additional PCIe using the modular cables if they're needed
This in no way discourages me from the Corsair 850HX I am planning on buying as soon as my paycheck comes in. And as others have mentioned, all of the high current connectors are almost always hard wired anyways, so the loss (if any) is VERY minimized as it is.
Nahh... it's about POWER & efficiency.
-I have an Enermax Modu82+ 525W.
-It’s rated at 40A combined draw out of the 12V rails (480W).
-Two of those rails are dedicated to graphics cards.
-Those rails connect to the graphics cards via twelve pairs of 18AWG wire into what is roughly a Molex Mini-Fit Jr connector.
-I looked up the average resistance through mated connectors after 20 insertion/extraction operations from Molex. Here’s the results arranged by type of pin:
Resistance per connected pin pair
Gold plated brass = 2.62 mohms
Tin plated brass = 3.09 mohms
Tin plated phosphor bronze = 2.45 mohms
So the modular nature of my power supply adds two times the resistance listed above to each circuit. However, there are twelve circuits in parallel. So the resulting added resistance is the figure listed above divided by a factor of six, as shown below:
Total added resistance
Gold plated brass = 0.437 mohms
Tin plated brass = 0.515 mohms
Tin plated phosphor bronze = 0.408 mohms
Let’s assume I’m able to pull 100% of the total rated 12V current for my graphics cards. So here’s how much power is consumed by the pins:
Power wasted due to resistance of pins
Gold plated brass = 0.699 W
Tin plated brass = 0.824 W
Tin plated phosphor bronze = 0.653 W
Now for the nail in the coffin of their argument. Out of the imaginary 480W total power consumed in my example what percentage is wasted by the pins?
Percentage of total power wasted
Gold plated brass = 0.146%
Tin plated brass = 0.172%
Tin plated phosphor bronze = 0.136%
Yes, a modular supply is less efficient. It’s less efficient, in my example case, by a massive 0.227% on average. Completely negligible. By modular kids. Get your money's worth.