12V and HD problem

[kc8adu]

Re: 12V and HD problem

most drives unload before they have fully spun up.listen to a western digital some time.
and they auto park instantly when power goes off spec.
thats the click you hear when you turn it off.
bad enough power problems would just reset it.click,catch back up on speed,unload,seek fw,come ready at each disruption.

[PCBONEZ]

Re: 12V and HD problem

kc8adu
Not talking start/stop.
More like 8v-16v-8v-16v
Unload you hear is start current cutting back to run current.

i4004
I don't know how to explain it too you.
Unstable power will cause unstable mechanical movement inside the drive.
-
The only way you wouldn't get that is if you've never had a crappy PSU damage a drive. I have. I think most other people have.

Like I said: By your theory heads would never come into contact with the disk as long as the disk is spinning.
-
If that were true no hard drive would ever have disk/head damage without a physical shock. - Most people know that's not true either.

.

[i4004]

Re: 12V and HD problem

>Like I said: By your theory heads would never come into contact with the disk as long as the disk is spinning.

where did i say that? quote me.

>The only way you wouldn't get that is if you've never had a crappy PSU damage a drive. I have. I think most other people have.

how do you know for sure what damaged your drive? you dismantled it and found scrathes on the platter? umh...

we're trying to explain the modes of failures and you're just repeating stuff like
"drives go bad ergo psu destroyed them".
that's nothing.

offer technical explanation that fits the parameters.
i (and it seems kc8adu too) think you have none.

why don't you make a test in such a manner (put potentiometer(or this
http://en.wikipedia.org/wiki/Potentiometer#Rheostats ) on 12v and go 8v-16v-8v-16v) and tell us what happened. or heck, just put a switch on 12v rail, that'll vary the voltage allright...
it's your theory, you prove it.
i would try it but don't have opened up machine or drive handy....

one more thing: you had a psu producing pumping 12v that was 8/16/8v?
now i'm just having huuge troule imagining what needs to be wrong with psu in order for that to happen.
it all seems like a fairytale to me.

i'm not saying i'm completely right or anything, but at least try to give technical explanation.
again, these are polyphase servo motors driven by ics. you surely know what that means. and i'm not even going into protection modes of hdd as a whole, which kc mentioned....

[linuxguru]

Re: 12V and HD problem

> you had a psu producing pumping 12v that was 8/16/8v?
> now i'm just having huuge troule imagining what needs to be wrong with psu in order for that to happen.

I've seen numerous PSUs with multi-volt ripple on +12v. It usually happens with a single bad cap on +12v, which probably failed because of high ripple current through it after the mobo caps at the VRM input failed, which is par for the course on most Prescott boards.

+12v with +/- 4V ripple on it is par for the course, and will destroy most slim Maxtors and Deskstars fairly quickly. Put the same drives on a Delta or Zippy PSU in a case with good ventilation, and watch them run for 10 years without problems.

[i4004]

Re: 12V and HD problem

show me +/- 4v ripple on the scope, i don't believe it.

i like your broad sweeping generalization on maxtor: it was run off the enermax and it did slight above one year. TWICE!
two maxtor hdds died within 2-3 years. both powered by enermax.
maxtor was quick to die no matter the psu it was powered by. it still is.
not every single drive, but stories like "half of installed maxtors died within year" are usual.


btw. even if you had 4v ripple(which i say you won't have) you're still not explaining what exactly is happening to hdd's platter speed.

do note: i'm not contesting dirty power is destroying hdds etc., i'm contesting the claim of platter varying speed and that being the cause of bad sectors.
and i will continue to do so untill i hear explanation that makes sense.
so far i heard none...hell, not even explanation attempt...
nobody explained how would the voltage flucutation cause speed variance..
grab some driver ic pdfs and show us some proof....
you speak as if hdd is powerd by dc motor with brushes, and that it is plugged strsaight to 12v rail.
you speak as if hdd doen't monitor platter speed at all.
these are bollocks.

if you can't prove it logically, then just say you can't prove it.
i just don't buy this vary-speed bumping the heads up/down theory.
nobody here has put any real reason behind it.

[PCBONEZ]

Re: 12V and HD problem

Given a Deer PSU the question was:

>> Is it possible that, due to the bad cap on 12V rail, that HD of this PC has been damaged for the high ripple? <<

--> Is it possible <--

I say yes.
linuxguru say yes.
i4004 say no.

i4004
I barely have time just to read badcaps lately and no drives to sacrifice.

If you want to do an experiment to prove something do this.
Take Deer PSU.
Build bench P3 system (so CPU is on 5v).
Remove 12v rail caps and replace with jumpers (simulating shorted caps).
Run it for about 3 weeks with power saving turned off and to simulate drive in use format the drive 5 or 6 times a day.
You will need to do this with various (new and old) drive designs and various manufacturers. I'd say 6 to 10 different kinds of drives.

If zero develop bad sectors then you MIGHT be right.
[You will not have tested every kind of drive.]

If ONE develops bad sectors then linuxguru and I might be right.
[Drive may have been 'too used' before the test.]

....

Regardless of experiments outcome I will stand by what I said because I've had bad PSU's destroy drives by way of disk surface damage.

--> It is possible <--

I don't need to do experiments.
I've experienced the problem first hand.

.

[kc8adu]

Re: 12V and HD problem

excessive ripple is hard on everything and as we know does damage things.
and if you follow pcbonez suggestion you will only have a dead psu.
or a fireball since he suggests to do this with a deer!
a more valid test would be to simply remove the caps on +12.

[i4004]

Re: 12V and HD problem

Given a Deer PSU the question was:

>> Is it possible that, due to the bad cap on 12V rail, that HD of this PC has been damaged for the high ripple? <<

--> Is it possible <--

I say yes.
linuxguru say yes.
i4004 say no.

nope. that's not what i said at all. you obviosuly didn't read the whole thread, and you're just avoiding to answer my question.
i said bad sectors are not caused by platter speed changing.
you said they are but you never offered explanation, otoh kc and i offered some explanations why this should not be possible.

i never said dirty power can't damage the drive. i said i don't see how can it produce bad sectors. nobody here offered viable explanation how can dirty power cause bad sectors. nobody.

so here's challenge for you: offer good technical explanation, and i'll send you 5 wd hdds. each 160gb. i'll buy them here and send them over to you.
is that fair enough offer?
you can then experiment with them.

Regardless of experiments outcome I will stand by what I said because I've had bad PSU's destroy drives by way of disk surface damage.

that is the claim you cannot prove. are you aware of that?
you never did experiment with a new drive and bad psu to establish that. you got used and beaten up machines with bad sectors and crappy psus so you connected the two.
you never did any logging about who had how many bad sectors at what time, were they getting worse, or stopped progressing once you put the good psu in.

right?
and that means you don't have experience to say what you're saying.

can dirty power damage the hdd. sure. i'm saying that from the beginning.
what i'm disputing is the explanation on how does this happen.
(why am i saying that? because op mentioned bad sectors. later it turned out he had tons of them, which points to
http://en.wikipedia.org/wiki/Head_crash
and that TOO is not caused by wrong platter speed)

i'm not disputing linuxguru's explanation(that one makes sense to me), but one you offered. lg never mentioned platter speed.
if you were in the field of data recovery i would say "fine, man knows what he's talking about", but you're not in that field, AND you're offering no explanations.

[Drive may have been 'too used' before the test.]

swell!
so now age comes into play too?
age can cause bad sectors(wow!) and not the dirty power alone?
how weird.

i like this one too:

If zero develop bad sectors then you MIGHT be right.
[You will not have tested every kind of drive.]

so you're talking about possibility that some drives are, and others are not affected by dirty psu power? ie i can be testing 3-4 different makes/models never to experience bad sectors caused by bad psu.

well, you've surely find an easy way out of this conundrum...huh

i'll be waiting your explanation for which you will be rewarded with 5hdds.
if you offer none, you were just babbling without reading what i really said.

you don't have to do any test, just offer explanation how dirty power slows down the platter.

[i4004]

Re: 12V and HD problem

and if you follow pcbonez suggestion you will only have a dead psu.
or a fireball since he suggests to do this with a deer!

yeah, that too.
i bet he can explain how would psu work with short circuit on +12v rail.
but in a same way he'll explain platter speed thing.

[PCBONEZ]

Re: 12V and HD problem

i said bad sectors are not caused by platter speed changing.

That is the claim you cannot prove. are you aware of that?

The fact that you don't see how can it produce bad sectors is not a proof.

You say in can never EVER happen.
I say there are conditions where it could.

~~

If you refer back to post #1...
- DEER is the PSU over which this thread was opened.
It is the logical choice.

~~

I work really hard (12-16 hour days) for 3,4 months and then take a vacation for 2,3.
I'm in the work part now and I don't have time to play.

If you have all those resources to buy drives just for that then you do the experiment. Though I'd love to, I flat don't have time.

.

[arneson]

Re: 12V and HD problem

Decided to monitor a drive platter with a strobe under the influence of unregulated power fluctuation at varied frequency.
I discovered that at a quantum level the heads ride higher while writing and lower when reading.
Then taking into account the platter continuum and the motors momentitude with a normal amount of sticktitution I found a higgs field effect that drives the aluminium molecules to increase thier ionizatioal half life.
At that time the drive window popped off and the platter flew out.

[i4004]

Re: 12V and HD problem

That is the claim you cannot prove. are you aware of that?

The fact that you don't see how can it produce bad sectors is not a proof.

You say in can never EVER happen.
I say there are conditions where it could.

it was your claim. and you never proved it. i didn't ask for material proof, but just an explanation.

http://en.wikipedia.org/wiki/Burden_...%28rhetoric%29
http://en.wikipedia.org/wiki/Negative_proof

when you present something as a possible cause it's upon you to prove it. at very least that would be honest thing to do.

it kinda reminds me of "does god exist" discussions: it's not up to me to prove he doesn't, it's up to you to prove he does, because you made that claim.


and i made that extra effort by explaninig why it should not be possible.
so, your burden, not mine..you carry it.

The fact that you don't see how can it produce bad sectors

and this makes 2 of us not seeing it, you and me...hehe...

i really don't understand this stubborness: if i can't prove something i'll just give up and say my premise was wrong.
and the only explanation you gave was that "it is possible".
no it's not untill you explain how and why, and to some detail.

so untill then...
(plaease PM me if you have explanation, without it i'm not coming back here...)

[PCBONEZ]

Re: 12V and HD problem

No:
I gave a possible cause and you chimed in with 'that can't be'.
As far as I'm concerned the burden of proof is on you.

What I said would make perfect sense to anyone the understands the dynamics of mechanical assemblies in motion.

The need for clean 12v power to drives is expressed in their specifications which general say something like within +/-5% (to 10%) INCLUDING RIPPLE.
That wouldn't be there if dirty power didn't cause problems, and as 12v is the motor power, what do you think the problem they are worrying about is?

I suggest you read up on aerodynamics and maybe you'll eventually get it.
I don't have time to teach you right now... Sorry.

.

[PCBONEZ]

Re: 12V and HD problem

https://www.badcaps.net/forum/showpo...8&postcount=27