Bad design examples

[i4004]

Re: Bad design examples

davmax:
as pcbonez said quest for improvement is eternal, and what got humans where they are(although it's a weird lil place..hehe)

Switch mode power supplies are and can be very reliable. Why try to change for a few failures?

because it hurts so much when you lose all of your hardware and data at once?
this should not be happening to anybody, at anytime <period>.
industry should make sure this improvement happens.

"Standby Power Supply" it is.

1. The current ATX spec for 5Vstb is 2 amps. Why would manufacturers produce less?

my qurestion is this: do we need it? if we don't need it, then specs should be changed.
what is using 2A at 5V when pc in stby?
pwr on keyboard i mentioned is useless function that we don't need.
if some use this, fine, but security of most should not be compromised for comfort of few.

3.The Standby power supply also supplies the TL494 controller chip somewhere in the range of 7 - 40V (quite a tolerance) clearly it cannot be taken from 5Vstb.

for the sake of clarity here's that circuit(simplified) of psus 4 and 5 in this thread


new designs (other than those used in cheapest psus, i think) are different.
they use pwm ics on the primary of this +5v stby trafo, AND you have control loop via optocoupler, i reckon
( notice optocoupler near +5v stby trafo on my new fortron
https://www.badcaps.net/forum/attach...achmentid=7337 )
this is much better than free running oscillator in psus 4 and 5 i photographed.
but it is not bulletproof.

my suggestion of linear supply instead of either free oscialltor(which is just silly) or pwm ic there(which is much better than free oscillator, but still bad components can mess it) is bulletproof design. or as close to that as we can do with what we have today.

for example, old vcrs have much, much lower number of supply problems because they use linear supplies; new vcrs use smps and they fail more.
it is expected that something more complex fails more. so even with good caps, smps will fail more than linear supply.

5.The proposal to use a plug pack would be best applied to the controller chip power.

and for +5v stby if 2A is not needed.

these are suggestions how to improve the design; i'm not really hoping industry will change its ways.
but also it's a practical suggestion how to achieve stable +5V stby and pwm ic supply. more reliable and durable than any smps design.

again, yes, it rarely happens, but tell it to matt:
http://www.newegg.com/Product/Produc...ark=False&Page=
although it seems he was pushing it somewhat, and also didn't respect warning signs...heh...

sure such cases are extremely rare with better units, but there should be none. on either cheap or better psus.
i'm not saying psus should last forever, or something, but like i once said, i'm just asking for it not to burn my machine. is that asking a lot?
in that aspect i feel atx specs are failing, because this is still happening.
such psus should not appear on the market at all.




gonzo:

Cheap wall plugs and transformers are from my point of view not much more reliable then those if ever.

one wouldn't use supplied wall plug, but do its own soldering to cables within psu. i feel those transformers are mostly ok judging by the many devices supplied by it and working 24/7. offcourse, one wouldn't pick cheapest adapters to employ here.



pcbonez:

it would be from the vacuum tubes

godo idea there; vacuum tubes as rectifiers wouldn't let current thru when they fail!

<wink>

[gonzo0815]

Re: Bad design examples

Cheap crap will always pose the risk of burning a system. ATX spec will never protect you from a POS which manufacturer is ignoring it.

The design with 78X05 is from my point of view not in accordance with latest ATX spec, as it can never supply the required 2A €5v Sb w/o active cooling (and of course this would need an 78S05.)
The 5vSb does actually power a lot more then the keyboard. For users like me, it has to power the whole RAM too when the system is left in S3 mode.
Furthermore, USb and other stuff could be powered from 5v sb rail, depending on the mainboard manufacturers choice & creativity.

[i4004]

Re: Bad design examples

yeah, STR(S3 mode) with 2 sticks means 10W, and that's 2A right there..hmm...

that mode is sometimes working for me, sometimes not.
sometimes it leaves hdd and dvd-drives on.
sometimes it turns on by itself after few minutes...

perhaps because i use win2k.
i should try this (even though it's xp/2003 tip)
http://support.microsoft.com/kb/841858
as i indeed use usb tablet and usb mouse sometimes.

[gonzo0815]

Re: Bad design examples

s3 is always a fragile thing, and perhaps this is my first desktop systems where things are working regardless how often i put the system in s3.

I think the problem is the countless possibility of component combination. If one driver or one hw component does have some glitches, your chance of working s3 mode is gone.
That is probably one reason, why mass produced uniform systems have a lot less problems with s3 then individually build ones.