Just a passing note, with some unrealated info for how weird I am.
While it was a long time ago and adding to my prev. post I just thought I'd throw this one in. The bad power supply I got with a case and have heard from others was Sparkle power supplies. I swore I'd never go near them again but times do change and I saw to my amazement that they even made some PC Power supplies at one time. Boy was that a shock but I do owe them for making one terrible one that got me interested long ago (it was for a Cyrix 686 I think that was it's number as it was long long ago).. lol..
So I put them in my bad list mostly due to habit and because it was a pain finding out it was the source of my erratic acting computer. The 3.3v line was so noisy and dropped a bit when loaded so the computer worked fine under mild use but got unstable under various loads. What a nail biter that was.
But I took it as a good excuse to educate myself about power supplies because of them so I do owe Sparkle for that. Laughs...
I am kind of not impressed by the EarthWatts lineup. They say multi rail as the guy that did this said about false rails and are cheap to buy and doubt the multi rail part but I don't have any actual experience with them other than buying one for a disaster backup computer if my main one fails. I also bucked my standing trend of buying only PC Power by getting my Seasonic 850W Power supply.
I know it was expensive like was posted but they do make fantastic power supplies and are getting great reviews from sites that do take the time to do proper tests and check out many area's of power supplies that a lot of sites don't like noise and voltage stability when loaded. I also don't disagree but I question the remark of testing at room temp. I've seen a lot of product testing over the years and testing power supplies in colder or hotter than normal is very important because it tests all the components quality and some computers do work in high room temp or very cold room temps and it's is important to know they will be able to handle it. With electronic parts each different type of parts react differently to temps and if they are in critical circuit area such as feedback loops where any offset error is compounded it's important to know this. Though most don't need to go as far as millspec. testing it's still very important to know it will work in the worst of conditions and weed out the weak spots of a design. In power supplies where they are boxed in even with good fans they will still have hot spots and can run up temps pretty fast under heavy loads.
I Hope/pray the starter of this topic understands I'm not being critical or nasty. It's the last thing I want to do is even imply a insult which is not where I'm heading with this so please don't get mad at me. The world and the universe wants to kill me with a passion and I sure don't want to add people into the mix as well. Fate has me pegged as a target already it seems..
It's just the way I was trained to look at things when I was in school. And your more intelligent than I am. And I used to be a LOT smarter than I am now but like I said the world hates me and wants me out of the way.
But I got a lot of this working for Intel at their Fab4/production plant and Fab5 R&D plant next door in Beaverton Or. Man was that the worst job I've ever suffered. I leaned a lot of neat things but mainly kept the production equipment on line and felt more like a Maytag repairman that had to wear a bunny-suit. This is what we called the clean room attire we had to wear while in the production area.
When I got out of there and went to work for BTS,
Business Telephone Systems and repaired the phones and KSU that typical office buildings use that handle up to 50 phones off of the one KSU I saw some weird and supprising things. Like phones from Texas filled with dead cockroaches. Man what a stink that was and people would talk on these things.. gross...
But what I really want to get at is when a IC chip got wacked by lightning you would think that any chip with the same part number would work but this was seldom true. Either it's slew rate was a micro second off or it had a slight ring in the square wave front or back but we couldn't put,say a TI (texas inst.) brand chip in, we had to use another brand no matter if it should say it works in the TI parts data manual that gives you all it's tested specs but you read the total specs and you see a slightly different graph over a given temp range or current or fan out specs and all the good timing specs and as far as the repair goes you are working with a chip from Mars as far as it's concerned. In designing every element is usually tested at 10% beyond normal if possible to make sure of reliability in the field.
Here is a weird example of murphy's law in action. We had to use this brand chip for this particular circuit or their chip or another brand but some would work and some wouldn't even though everything said they should work for the most part everything was equal. In other area's of the circuit it would work but in come area's they wouldn't. Because of the circuit design some aspects of a design are more important than others. In feedback circuits it was super critical that everything worked to perfection and this is a good example of why they test to beyond normal conditions. And with parts now coming from China where they are still on a learning curve it makes it even more important to test their stuff harshly.
But anyway, back to power supplies... I do drift don't I..lol..
I sometimes treat power supplies differently for good brand name ones I like to think you get what you pay for a lot of the times and I prefer to buy knowing I'm getting quality components.
Without tearing into them like I used to and seeing caps and resistors with 1% tolerance rated parts instead of the cheaper 5% was the most common used spec tolerance or worse yet they used the 10% parts that goes into cheaper made power supplies.
It can be very important to get parts like these and worth the extra cost.. They have a lot better chance of working right the first time coming off the production line. But they put variable caps or resistors or inductors in to tune out these problems a lot of the time but they can't do it for everything and when you add more variables the problems get worse. At least Power Supplies are simple designs but they still need good quality parts like any other circuit.
At least to me now being stuck on disability and sometimes have to fight it out with my cat on who gets to eat each day it really shows how much importance I put into cost/reliability/length of use ratios. While I'm worried about costs a lot I also look at the long term use that can offset costs down the road. This is important. I will spend a bit more if I buy one that lasts 5 to 8 years instead of having to buy one every 2 to 4 years because it was a bit cheaper and not so well designed or has caps that won't dry out or explode... booom.. lol Does this make logical sense to you?