Suggestions for PSU improvements

[kc8adu]

Re: Suggestions for PSU improvements

here is a major manufacturers take on motor bearings and lubrication.
lots of other info on motors too.
http://www.fasco.com/fasfacts1.asp

[Elitist]

Re: Suggestions for PSU improvements

Useful. Fasco are well known for their industrial motors.

[999999999]

Re: Suggestions for PSU improvements

Mostly, a sleeve bearing that is wearing becomes noisy. A ball-bearing that is wearing is more likely to seize, resulting in expensive damage. Regular lubrication is a panacea in both cases. Using an MoS2 assembly compound is a precaution well worth the effort.

I'll ignore the implication that I am ignorant from anyone who clearly hasn't even opened the pages of an aeronautical design tome, but an apology would be appropriate. Bad manners tells its own story.

I'm sorry you feel this way, but I'm not going to try to twist accurate information to align to your point of view. Ignorance is not an insult, it is "not knowing". I am definitely ignorant of aeronautical design, for example.

Sleeve bearing fans are designed to run in vertical orientations. They are contraindicated for higher temp areas such as PSU exhaust. There is definite reproducible, substantial decrease in lifespan in either of these uses and these are the two possibilities in a standard PSU.

Sleeve bearing fans are more resistant to shock, it is entirely possible to present a scenario in "aeronautical design" where this factor is important, but that is not the scenario in a stationary computer PSU.

[ChaoZ]

Re: Suggestions for PSU improvements

I prefer a sleeve-bearing fan myself, as it is economical yet efficient. They are slower , at least the ones i have seen, compared to the ball-bearing(BB) ones. I still re-use some of the fans from cheapo $10 PSU, as I find suited for the home environment.

I had my days with 5W, 6W fans, and they drove me nuts! The surge when they powered up really weaken the power rails.

The shape/angle of the blades, the number of blades, the proximity of the fan mounted to grills etc. They all counts in terms of performance, noise and efficiency.

[ChaoZ]

Re: Suggestions for PSU improvements

Anyone can link me to a picture of Active PFC?

I have an old PSU with passive PFC I believe - giant transformer mounted upside down on the topside of the PSU casing.

I believe this is a passive PFC unit. GGGGrrrr... the PSU feels cheap when I handle it without the PFC. LOL!

Feels like recapping it, but it is full of cheapo caps and varistors all over the places even on the output lines.

[PCBONEZ]

Re: Suggestions for PSU improvements

I've seen cheap 'ball bearing' fans with only one ball in the bearing and those are crap.
A 'good' sleeve bearing fan will outlast a 'cheap' ball bearing and probably cost less.

Additionally, ball bearing fans are designed in more than one way by the placement of the balls and the race in the bearing. The balls can be positioned for axial thrust, radial thrust, or part way in between to accommodate both.

In short, it would be ignorant to assume a ball bearing fan is better than a sleeve bearing fan simply because it is a ball bearing fan.

How did this turn from a discussion about caps in a PSU to a lengthy discussion about fans anyway? - Oh,,, 999999 is here 'not twisting' things again.

~~

Fans are easy to replace and if you shop at *electronics supply* places (especially the surplus/overstock places) rather than PC supply retail outlets you can usually find good fans (sleeve or ball bearing) that will last several years for under $3-4. I recently picked up some new 40mm Sanyo Denki ball bearing units (for tight spots in HTPC cases) for $1.80 each.

~~

ChaoZ,
In addition to what you said the thickness of the fan and RPM play a big parts in noise and efficiency.

~~

Fans are cheap and easy to replace.
For me that's enough about fans for,,, oh,,, 6 months,,, a year..

~~

[starfury1]

Re: Suggestions for PSU improvements

OK correction to my statement, seems I've stuffed up

Redesigning for 240V only operation might be an idea
You would have to put in proper safety features, whatever that might involve and the electrical legalities.... well I am not going there...cause I wouldn't have a clue.

Looking at a couple of schematics of older psu's, seems a feed is taken from the center of the 2 caps so trying to do something like that would be more trouble then its worth if you could get it to work (one here)

So yeah scratch that idea if the above is the case.

DOH!


The orginal question asked

Is there any way whereby we can improve the PSU's performance?

So I guess that is going to cover a lot of ground in a lot of area's

[999999999]

Re: Suggestions for PSU improvements

I've seen cheap 'ball bearing' fans with only one ball in the bearing and those are crap.
A 'good' sleeve bearing fan will outlast a 'cheap' ball bearing and probably cost less.

Who ever said we should buy cheap junk in either kind?

Additionally, ball bearing fans are designed in more than one way by the placement of the balls and the race in the bearing. The balls can be positioned for axial thrust, radial thrust, or part way in between to accommodate both.

Actually, almost all ball bearing fans have the same design, that they are optimal for the fan they are used in. Only remaining factor is if they are decent quality or junk. FWIW, most decent fans use NMB bearings.

In short, it would be ignorant to assume a ball bearing fan is better than a sleeve bearing fan simply because it is a ball bearing fan.

Absolutely, and it would equally be ignorant to assume a sleeve bearing fan is better, because on average, considering the real fans available in the market, there are very very few that are decent, while the majority of ball bearing fans are.

Why is this? It's because ball bearing cost more. If a fan manufacturer is only trying to build cheap junk, they'll typically choose sleeve bearings.

How did this turn from a discussion about caps in a PSU to a lengthy discussion about fans anyway? - Oh,,, 999999 is here 'not twisting' things again.

I encourage discussion, but being upset has no productive purpose?

Fans are easy to replace and if you shop at *electronics supply* places (especially the surplus/overstock places) rather than PC supply retail outlets you can usually find good fans (sleeve or ball bearing) that will last several years for under $3-4. I recently picked up some new 40mm Sanyo Denki ball bearing units (for tight spots in HTPC cases) for $1.80 each.

Sanyo Denki are indeed good quality fans, and chosen for many very high-end server PSU for this reason. It is not likely any sleeve bearing fans will come close in lifespan. Papst and Panaflo may come close, but only in vertical orientations and away from elevated temperatures.

[Elitist]

Re: Suggestions for PSU improvements

It's always a good idea for N Americans with adequate experience to switch their electronics over to 220V. Those with the necessary skills do not need explanations. Sadly, many of my friends over there seem unaware that the incoming domestic supply is actually 220V and NOT 110V! The historical reasons for choosing 110 or 220 are now recognised to be fallacious. Apart from which, the savings in copper could help offset the national debt!
In another thread, I would wholly support 9....9 's argument about Al cap types. However, since he now admits lack of knowledge about fluid dynamics, I am happy to expand the explanations just a little. It is always a bad idea to split a fluid flow unless elaborate baffling is introduced. Generally, this is either inappropriate or impossible in a PC case and certainly beyond the abilities of most. Without, the risk of introducing regions of slack flow, or even turbulence where not desired or where it causes adverse results, is too high. Therefore, the simplest expedient in practice is to adopt a single, unblown, inlet and a single, forced, outlet, giving a relatively(!) well-defined smooth flow across the board. The obvious way to provide such an arrangement is to use a high flow fan inside the PSU. If this can also be on the underside of the PSU box, adjacent to the cpu, this is obviously advantageous. Apart from this, all cosmetic and other holes in the case need to covered with tape - something most folks ignore. Not worth writing more on this subject before those with an interest read up on some basic aerodynamic principles - that includes the guy who built BX on the basis of an horrendous nightmare and utter ignorance. Stop supporting the agents of Beelzebub!

Incidentally, when ChaoZ looks at the specs of Papst sleeve bearing fans, he is going to be pleasantly surprised by their performance and longevity.

[PCBONEZ]

Re: Suggestions for PSU improvements

Damn many9's - did it snow in Hell?

We agree on something!

[999999999]

Re: Suggestions for PSU improvements

In another thread, I would wholly support 9....9 's argument about Al cap types. However, since he now admits lack of knowledge about fluid dynamics, I am happy to expand the explanations just a little.

I have to conclude you have trouble following a conversation.

That's not my loss, I'm ok with that.

[PCBONEZ]

Re: Suggestions for PSU improvements

Elitist

I use 220v in my shop for heavier machines but I think you are a little confused.

220v is not much more efficient unless it's used as multi-phase.
220v home/building general wiring is single phase. (Even where you are.)

The only significant advantage to 220v over 115v in single phase for home/building wiring is smaller diameter wire to safely pass the same power.
-
But 220v also (rightly) requires thicker insulation by the building codes.
Higher volts makes arcing more of a concern, thicker insulation takes care of this.

Unfortunately that means most buildings here would have to have the existing wiring ripped out and replaced to (legally) convert it to 220v.
- Not very practical.

There is also the fact that most small AC powered devices here are 115v only.
- It saves nothing to power them through a converter.
The converter itself would waste power lost as heat.

.

[999999999]

Re: Suggestions for PSU improvements

Agreed, abandoning all the 110V appliances, importing new 220V alternative appliances and converting circuits, is less practical and more costly than just running another 110V circuit to divide the load (and losses) if/when it would've been a significant difference. Importing 220V appliances into 110V regions is also going to incur a waste of resources and power.

Uses more copper? So what? Trashing the 110V appliances isn't exactly conservative or cost-free either. I can buy a heck of a lot of 12 ga. wire for the cost to replace a mere toaster oven, let alone anything of much value.

[Elitist]

Re: Suggestions for PSU improvements

Prejudices, prejudices!
The only reason for suggesting this simple expedient is that almost all domestic appliances are available at the same or lower cost with a little 110-220V flick preset switch, usually on the outside, but sometimes on the inside. Since everything is manufactured in Asia, what's the problem, apart from some warehouses and retailers requesting the global instead of US-only model? No waste, ripping out of wiring was suggested, certainly not lossy converters. Insulation requirements and safety are red herrings. Think spot-price of copper, think supply and extraction of exceptionally lean ores, think organised crime interfering with infrastructures by metal theft & co.
What's the average lifetime of a toaster in the USA? Notwithstanding, no-one mentioned toasters - read what was said, not offer the knee-jerk response of misplaced patriotism because the suggestion emanates from another country. Don't let irrational prejudice cloud the obvious benefits - most of the RoW lives happily with 220V. The science is clear. I was trying to be helpful to those with the capability and interest!!!!!

[ChaoZ]

Re: Suggestions for PSU improvements

Papst fans are expensive... 80mm -> $15USD alone, 120mm -> $25. I have a savaged 80mm x 38mm, still going well after 10 years.

Well, just to twist myself back into the topic. I replaced a pair of rubycon 680@200V with chemicon 680@250V. I used to wait for a whole min for this PSU to warm up.
Now, the PSU seems to come alive, it starts the system up with ease. The PSU is really silent, but i still get a slightly whine from the transformer i think. (tried the straw trick but not working )

Anyone tried one of this?
http://cgi.ebay.com/DELTA-01GEEW3E-E...QQcmdZViewItem

I always see them in more expensive PSU, like zippy. Is this the same as the EMI filter in a typical PSU but more compact version?

What about ferrite cores? I saw some designs with a BIG ring on the output wires. How does that helps? Where should they be placed?

Thanks starfury1 with your link! You rock!

<- digging for caps

[PCBONEZ]

Re: Suggestions for PSU improvements

Interesting gadget but it's only good for 1 amp.
Yes it's an EMI filter.

Ferrite cores are used mostly on DC and signal lines like printer, monitor, or usb cables to strip out RF EMI. They are not good on AC.

.

[PCBONEZ]

Re: Suggestions for PSU improvements

"Little flicks".

I checked around my house.

NO little flicks on:
Microwave, blender, lamps, coffee pot, food processor, 3 power drills, 1 small drill press, 2 vacuums, one carpet steamer, 2 TV sets, portable swamp cooler, roof swamp cooler, 5 window air conditioners, small compressor, circular saw, recip' saw, charger for my battery tools, jig saw, clocks, refrigerator, freezer, DVD player, phone chargers, roto-zip saw, dremels, ~1/2 the stereo equipment, fans, portable heaters, and, and, and, and, and...

Yes little flicks on:
1/2 the stereo equipment, 1 TV, computer gear.

220 requires a different receptacle by law so if the power cord is hard wired there is no reason for a "little flick". It cuts the manufacturing cost to not install them.

If you'd like to send me around $30,000 USD I'll start the rewiring and appliance change over trend (and haul two or three truckloads of perfectly good stuff to the landfill) but there's a little over 300,000,000 other people that will need you to send them $30,000 too. That doesn't include the business's you will need to fund too.

.

[Elitist]

Re: Suggestions for PSU improvements

So my time in the US wasn't wasted and my advice was correct. Read what I said - not what you would like me to have said so you can make a lame attempt to put down a foreigner. Your arrogance is what makes your country the least popular. Neither your country nor mine has a monopoly on perfection in design and application. Most of our stuff is aged and decaying because we were 'first'. However, after WWII, we were forced to changed our electrical installations, which makes our present designs newest and safest, bar none.
Notwithstanding, you will find that your a/c and fridge, cooker if it's electric, all run off the 220 line. Nobody is asking you to rewire houses, change your plugs, etc., but if you did adopt the British wall plug, which is the only one with a safety shutter activated by insertion of the earth pin, you could rid yourselves of those dangerous two-pin monstrosities. Cheaper-by-the-dozen, I'll gift you a box - they're only a few quid. You supply the wall receptacles.

[999999999]

Re: Suggestions for PSU improvements

Anyone tried one of this?
http://cgi.ebay.com/DELTA-01GEEW3E-E...QQcmdZViewItem

I always see them in more expensive PSU, like zippy. Is this the same as the EMI filter in a typical PSU but more compact version?

There are two reasons a PSU might use these.
1) PSU is high enough quality it helps to get a few parts off the main PCB by putting them in this type of module. I mean they ran out of room on the PCB so it was this or a separate daughterboard.
2) PSU manufacturer does enough volume or has strategic alliances with module manufacturer to get a really low price so it's more cost effective.

Essentially I'm saying that many PSU have the equivalent function components on the main PCB instead of inside these metal modules. If yours already has these components soldered to the AC socket discretely or on the main PCB it wouldn't be a solution rather being redundant.

FWIW, 1A is undersized for most PSU, you can get larger modules at various places online or cannibalized from other equipment. one place with reasonable prices is excesssolutions.com

What about ferrite cores? I saw some designs with a BIG ring on the output wires. How does that helps? Where should they be placed?

It reduces high frequency pickup. It's one of those last 1% tweaks, not needed to remain within proper operational thresholds for most devices but if you were building a (nearly) no-expense barred PSU, it'd have a few where convenient, which would be right were the wires exit PSU so noise doesnt' get into the feedback subcircuit, and 2ndly right where the wires plug into the powered equipment so noise doesnt' get into them either. For practical purposes, it is seldom placed next to powered equipment socket as that is bulky and could interfere with other parts, but is more often seen on extrenal power socketed equipment where it is known there is no interferenece with other parts (and being outside of a grounded chassis, is susceptible to external noise pickup.

[999999999]

Re: Suggestions for PSU improvements

So my time in the US wasn't wasted and my advice was correct.

Usually not. Most equipment in large regions like the US isn't going to have a dual voltage switch with exception of some switching PSU. It would not be economically feasible to import these items, and the power savings is trivial at most. A hypothetical ideal has to be weighed against other real-world factors, in significance, cost, time, and other alternatives.

The other alternative is usually what I'd already mentioned, to use the equipment available and if the load is high enough that there is any significant difference (which usually, there is not!) then put in another circuit.

A length of wire is no more precious than all these devices you'd have to run to use enough current that it'd matter.

Read what I said - not what you would like me to have said so you can make a lame attempt to put down a foreigner.

Perhaps as a foreigner you don't have exposure to devices imported into large 110V-centric regions like the US, are assuming there are switches for dual voltage when usually there are not. PCBonz checked a lot of equipment. I have as well over the years and in addition to that list, most US stereo equipment and TVs do not have a 110/220V switch. Why? It would cost more to add parts to support both voltages, when there are sufficient units sold to any region to make it more cost effective for a manufacturer to make two versions. In cases where the cost difference is not much, yes they'll tend to make a dual voltage version, but instead they usually design a PCB to accomdate either option or use a separate PSU module and populate the device as required.

Your arrogance is what makes your country the least popular. Neither your country nor mine has a monopoly on perfection in design and application. Most of our stuff is aged and decaying because we were 'first'. However, after WWII, we were forced to changed our electrical installations, which makes our present designs newest and safest, bar none.

We don't seem to have any problem, it is your idea we "should" think it's important that is a problem to you. Who brought up this idea of switching things to align with what you feel is important? Wouldn't THAT be the larger arrogance? The world is really supposed to endure additional cost, time, and waste of existing products to align with this idea? Good luck with that campaign.

Notwithstanding, you will find that your a/c and fridge, cooker if it's electric, all run off the 220 line. Nobody is asking you to rewire houses, change your plugs, etc., but if you did adopt the British wall plug, which is the only one with a safety shutter activated by insertion of the earth pin, you could rid yourselves of those dangerous two-pin monstrosities. Cheaper-by-the-dozen, I'll gift you a box - they're only a few quid. You supply the wall receptacles.

Most people in the US find the only problem to be when the power goes out, due to a storm or drunk driver hitting an above-ground supply, not the things you suggest. They're not realizing any danager, manage to get by just fine. What are the odds of harm from this supposed danager? Far less than most things in life. Focus on what matters, the rest is just trivia. If your wall sockets give you excitment and a feeling of superiority, that is subjective, not our concern.