I've been digging through piles of 3.5" disk drives (spinning rust).
All seem to have "seals" to cover any hardware that "shouldn't be
removed". Often, the seals are intentionally inconspicuous
(like a label that is so large that it covers these fasteners
as a matter of course -- or, metallic to mimic the metal cover).
But, not always ("Gee, what's under this sticky round dot?")

I don't think anyone would intentionally remove those fasteners
thinking they could "fix" a broken disk (we remove them to decompose
the disks into recyclable parts; the drive already being deemed
"discardable").

Note that this differs from the "do not cover this port" (that
I assume is there to allow for pressure equalization?).

So, are they there to detect signs of tampering (voided
warranty)? Or, discourage folks who are too curious for
their own good?

I can't imagine they would add manufacturing steps to
cover these fasteners if there wasn't some perceived
value to doing so...

On 23/02/2024 22:49, Don Y wrote:
> I've been digging through piles of 3.5" disk drives (spinning rust).
> All seem to have "seals" to cover any hardware that "shouldn't be
> removed".  Often, the seals are intentionally inconspicuous
> (like a label that is so large that it covers these fasteners
> as a matter of course -- or, metallic to mimic the metal cover).
> But, not always ("Gee, what's under this sticky round dot?")
>
> I don't think anyone would intentionally remove those fasteners
> thinking they could "fix" a broken disk (we remove them to decompose
> the disks into recyclable parts; the drive already being deemed
> "discardable").
>
> Note that this differs from the "do not cover this port" (that
> I assume is there to allow for pressure equalization?).
>
> So, are they there to detect signs of tampering (voided
> warranty)?  Or, discourage folks who are too curious for
> their own good?
>
> I can't imagine they would add manufacturing steps to
> cover these fasteners if there wasn't some perceived
> value to doing so...

It is an anti-tamper measure so that the manufacturer can tell if the
disk has been opened up to atmosphere outside of a clean room without
having to bother with a visual examination.

Breaking the seal on the disk physical enclosure voids warrantee.

--
Martin Brown

On Fri, 23 Feb 2024 15:49:01 -0700, Don Y
<blockedofcourse@foo.invalid> wrote:

>I've been digging through piles of 3.5" disk drives (spinning rust).
>All seem to have "seals" to cover any hardware that "shouldn't be
>removed". Often, the seals are intentionally inconspicuous
>(like a label that is so large that it covers these fasteners
>as a matter of course -- or, metallic to mimic the metal cover).
>But, not always ("Gee, what's under this sticky round dot?")
>
>I don't think anyone would intentionally remove those fasteners
>thinking they could "fix" a broken disk (we remove them to decompose
>the disks into recyclable parts; the drive already being deemed
>"discardable").
>
>Note that this differs from the "do not cover this port" (that
>I assume is there to allow for pressure equalization?).
>
>So, are they there to detect signs of tampering (voided
>warranty)? Or, discourage folks who are too curious for
>their own good?
>
>I can't imagine they would add manufacturing steps to
>cover these fasteners if there wasn't some perceived
>value to doing so...

I've often wondered that, too. Anyone know?

On 25/02/2024 9:23 pm, Cursitor Doom wrote:
> On Fri, 23 Feb 2024 15:49:01 -0700, Don Y
> <blockedofcourse@foo.invalid> wrote:
>
>> I've been digging through piles of 3.5" disk drives (spinning rust).
>> All seem to have "seals" to cover any hardware that "shouldn't be
>> removed". Often, the seals are intentionally inconspicuous
>> (like a label that is so large that it covers these fasteners
>> as a matter of course -- or, metallic to mimic the metal cover).
>> But, not always ("Gee, what's under this sticky round dot?")
>>
>> I don't think anyone would intentionally remove those fasteners
>> thinking they could "fix" a broken disk (we remove them to decompose
>> the disks into recyclable parts; the drive already being deemed
>> "discardable").
>>
>> Note that this differs from the "do not cover this port" (that
>> I assume is there to allow for pressure equalization?).
>>
>> So, are they there to detect signs of tampering (voided
>> warranty)? Or, discourage folks who are too curious for
>> their own good?
>>
>> I can't imagine they would add manufacturing steps to
>> cover these fasteners if there wasn't some perceived
>> value to doing so...
>
> I've often wondered that, too. Anyone know?

Disk reading/writing heads fly very close to the rotating surface being
written. Atmospheric dust is bigger than the gap,and could wreck the
process.

Hard disks are assembled in clean rooms, so the internal space starts
off dust-free. Atmospheric pressure inside the unit has to match outside
air-pressure, but the two spaces are connected by a slightly porous
air-filter that lets gas in and out, but no dust.

You could build a hermitically sealed disk drive, but you'd have to make
the structure stronger if you did that, and it would cost more. And gas
molecules would diffuse in and out anyway.

--
Bill Sloman, Sydney

On 2/25/2024 3:20 AM, Martin Brown wrote:
> On 23/02/2024 22:49, Don Y wrote:
>> I've been digging through piles of 3.5" disk drives (spinning rust).
>> All seem to have "seals" to cover any hardware that "shouldn't be
>> removed".  Often, the seals are intentionally inconspicuous
>> (like a label that is so large that it covers these fasteners
>> as a matter of course -- or, metallic to mimic the metal cover).
>> But, not always ("Gee, what's under this sticky round dot?")
>>
>> I don't think anyone would intentionally remove those fasteners
>> thinking they could "fix" a broken disk (we remove them to decompose
>> the disks into recyclable parts; the drive already being deemed
>> "discardable").
>>
>> Note that this differs from the "do not cover this port" (that
>> I assume is there to allow for pressure equalization?).
>>
>> So, are they there to detect signs of tampering (voided
>> warranty)?  Or, discourage folks who are too curious for
>> their own good?
>>
>> I can't imagine they would add manufacturing steps to
>> cover these fasteners if there wasn't some perceived
>> value to doing so...
>
> It is an anti-tamper measure so that the manufacturer can tell if the disk has
> been opened up to atmosphere outside of a clean room without having to bother
> with a visual examination.
>
> Breaking the seal on the disk physical enclosure voids warrantee.

As would removing the PCB attached to the OUTSIDE of the drive enclosure.
Yet there are no similar measures "protecting" it!

Who (customer/user) would actually *think* they could repair something
MECHANICAL inside the drive? The dimensions of the mechanisms and
tolerances thereon would make any such *attempt* ludicrous.

Shirley, the cost of the drive wouldn't justify such an effort.
And, the *value* of the data would deter any half-assed attempt
at recovery (which could compromise any legitimate effort at it).

Amusingly, I see very visible tamper detection mechanisms on
some RECONDITIONED drives -- including the PCB (which is often
completely covered by a "seal" making even probing signals
impossible)

On Mon, 26 Feb 2024 02:11:31 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:

>On 25/02/2024 9:23 pm, Cursitor Doom wrote:
>> On Fri, 23 Feb 2024 15:49:01 -0700, Don Y
>> <blockedofcourse@foo.invalid> wrote:
>>
>>> I've been digging through piles of 3.5" disk drives (spinning rust).
>>> All seem to have "seals" to cover any hardware that "shouldn't be
>>> removed". Often, the seals are intentionally inconspicuous
>>> (like a label that is so large that it covers these fasteners
>>> as a matter of course -- or, metallic to mimic the metal cover).
>>> But, not always ("Gee, what's under this sticky round dot?")
>>>
>>> I don't think anyone would intentionally remove those fasteners
>>> thinking they could "fix" a broken disk (we remove them to decompose
>>> the disks into recyclable parts; the drive already being deemed
>>> "discardable").
>>>
>>> Note that this differs from the "do not cover this port" (that
>>> I assume is there to allow for pressure equalization?).
>>>
>>> So, are they there to detect signs of tampering (voided
>>> warranty)? Or, discourage folks who are too curious for
>>> their own good?
>>>
>>> I can't imagine they would add manufacturing steps to
>>> cover these fasteners if there wasn't some perceived
>>> value to doing so...
>>
>> I've often wondered that, too. Anyone know?
>
>Disk reading/writing heads fly very close to the rotating surface being
>written. Atmospheric dust is bigger than the gap,and could wreck the
>process.
>
>Hard disks are assembled in clean rooms, so the internal space starts
>off dust-free. Atmospheric pressure inside the unit has to match outside
>air-pressure, but the two spaces are connected by a slightly porous
>air-filter that lets gas in and out, but no dust.
>
>You could build a hermitically sealed disk drive, but you'd have to make
>the structure stronger if you did that, and it would cost more. And gas
>molecules would diffuse in and out anyway.

I would guess it depends on what the container is made from. Glass
seems to be highly effective in preventing any diffusion if thermionic
valves ('tubes') are anything to go by.

On 2/25/2024 11:26 AM, Don Y wrote:
> On 2/25/2024 3:20 AM, Martin Brown wrote:
>> On 23/02/2024 22:49, Don Y wrote:
>>> I've been digging through piles of 3.5" disk drives (spinning rust).
>>> All seem to have "seals" to cover any hardware that "shouldn't be
>>> removed".  Often, the seals are intentionally inconspicuous
>>> (like a label that is so large that it covers these fasteners
>>> as a matter of course -- or, metallic to mimic the metal cover).
>>> But, not always ("Gee, what's under this sticky round dot?")
>>>
>>> I don't think anyone would intentionally remove those fasteners
>>> thinking they could "fix" a broken disk (we remove them to decompose
>>> the disks into recyclable parts; the drive already being deemed
>>> "discardable").
>>>
>>> Note that this differs from the "do not cover this port" (that
>>> I assume is there to allow for pressure equalization?).
>>>
>>> So, are they there to detect signs of tampering (voided
>>> warranty)?  Or, discourage folks who are too curious for
>>> their own good?
>>>
>>> I can't imagine they would add manufacturing steps to
>>> cover these fasteners if there wasn't some perceived
>>> value to doing so...
>>
>> It is an anti-tamper measure so that the manufacturer can tell if the disk
>> has been opened up to atmosphere outside of a clean room without having to
>> bother with a visual examination.
>>
>> Breaking the seal on the disk physical enclosure voids warrantee.
>
> As would removing the PCB attached to the OUTSIDE of the drive enclosure.
> Yet there are no similar measures "protecting" it!
>
> Who (customer/user) would actually *think* they could repair something
> MECHANICAL inside the drive?  The dimensions of the mechanisms and
> tolerances thereon would make any such *attempt* ludicrous.

On closer inspection (on a variety of different drives), it appears the issue
may be related to the (mechanical) "security" of the head actuator mechanism.
I.e., one can break the seal of the enclosure WITHOUT touching the "sealed"
fasteners by removing the fasteners along the periphery of the drive.

None of these are "tamper proof" in any similar/obvious way.

I suspect their concern is that if you "alter the tension" on the sealed
fasteners (e.g., loosen them to something less than "spec"), the internal
mechanism may vibrate/wobble/crash.

I'll see if I have a small-ish drive, here, that I can sacrifice to satisfy
my curiosity....

On 26/02/2024 5:49 am, Cursitor Doom wrote:
> On Mon, 26 Feb 2024 02:11:31 +1100, Bill Sloman <bill.sloman@ieee.org>
> wrote:
>
>> On 25/02/2024 9:23 pm, Cursitor Doom wrote:
>>> On Fri, 23 Feb 2024 15:49:01 -0700, Don Y
>>> <blockedofcourse@foo.invalid> wrote:
>>>
>>>> I've been digging through piles of 3.5" disk drives (spinning rust).
>>>> All seem to have "seals" to cover any hardware that "shouldn't be
>>>> removed". Often, the seals are intentionally inconspicuous
>>>> (like a label that is so large that it covers these fasteners
>>>> as a matter of course -- or, metallic to mimic the metal cover).
>>>> But, not always ("Gee, what's under this sticky round dot?")
>>>>
>>>> I don't think anyone would intentionally remove those fasteners
>>>> thinking they could "fix" a broken disk (we remove them to decompose
>>>> the disks into recyclable parts; the drive already being deemed
>>>> "discardable").
>>>>
>>>> Note that this differs from the "do not cover this port" (that
>>>> I assume is there to allow for pressure equalization?).
>>>>
>>>> So, are they there to detect signs of tampering (voided
>>>> warranty)? Or, discourage folks who are too curious for
>>>> their own good?
>>>>
>>>> I can't imagine they would add manufacturing steps to
>>>> cover these fasteners if there wasn't some perceived
>>>> value to doing so...
>>>
>>> I've often wondered that, too. Anyone know?
>>
>> Disk reading/writing heads fly very close to the rotating surface being
>> written. Atmospheric dust is bigger than the gap,and could wreck the
>> process.
>>
>> Hard disks are assembled in clean rooms, so the internal space starts
>> off dust-free. Atmospheric pressure inside the unit has to match outside
>> air-pressure, but the two spaces are connected by a slightly porous
>> air-filter that lets gas in and out, but no dust.
>>
>> You could build a hermitically sealed disk drive, but you'd have to make
>> the structure stronger if you did that, and it would cost more. And gas
>> molecules would diffuse in and out anyway.
>
> I would guess it depends on what the container is made from. Glass
> seems to be highly effective in preventing any diffusion if thermionic
> valves ('tubes') are anything to go by.

The filament does act a "getter", and most tubes included an explicitly
introduced getter.

https://en.wikipedia.org/wiki/Getter

People who have worked with vacuum systems do tend to know about them.

--
Bill Sloman, Sydney

In article <urjv00$324kk$1@dont-email.me>, bill.sloman@ieee.org says...
> > molecules would diffuse in and out anyway.
> >
> > I would guess it depends on what the container is made from. Glass
> > seems to be highly effective in preventing any diffusion if thermionic
> > valves ('tubes') are anything to go by.
>
> The filament does act a "getter", and most tubes included an explicitly
> introduced getter.
>
> https://en.wikipedia.org/wiki/Getter
>
> People who have worked with vacuum systems do tend to know about them.
>
>
>

The filiament does not act as the getter in vacuum tubes. It is the
plate that acts as the getter. In most tubes that use the plate it must
be heated to red hot.

The heater in the Wike artical is a heater in the vacuum pump system,
totally different than the vavuum tube.

On 28/02/2024 1:35 am, Ralph Mowery wrote:
> In article <urjv00$324kk$1@dont-email.me>, bill.sloman@ieee.org says...
>>> molecules would diffuse in and out anyway.
>>>
>>> I would guess it depends on what the container is made from. Glass
>>> seems to be highly effective in preventing any diffusion if thermionic
>>> valves ('tubes') are anything to go by.
>>
>> The filament does act a "getter", and most tubes included an explicitly
>> introduced getter.
>>
>> https://en.wikipedia.org/wiki/Getter
>>
>> People who have worked with vacuum systems do tend to know about them.
>
> The filament does not act as the getter in vacuum tubes. It is the
> plate that acts as the getter. In most tubes that use the plate it must
> be heated to red hot.

The filament runs hot enough to evaporate a few atoms. When they
condense on cooler surfaces inside the tube they do provide some getting
action.
> The heater in the Wiki article is a heater in the vacuum pump system,
> totally different than the vacuum tube.

Heat is heat wherever it is generated. Classic getters generate a lot of
condensation close to the getter itself, and trap an appreciable amount
of residual gas when they are activated. Getter action to deal with the
slow diffusion through the envelope can be slower, but it is the same
mechanism.

--
Bill Sloman, Sydney

On 27/02/2024 10:11 pm, darius wrote:
> The idiot Bill Sloman <bill.sloman@ieee.org> persisting in being an Off-topic troll...

Darius the Extremely Dumb posted this in response to a thoroughly
on-topic post, which he has snipped.

">>> Disk reading/writing heads fly very close to the rotating surface
being written. Atmospheric dust is bigger than the gap,and could wreck
the process.
>>>
>>> Hard disks are assembled in clean rooms, so the internal space
starts off dust-free. Atmospheric pressure inside the unit has to match
outside air-pressure, but the two spaces are connected by a slightly
porous air-filter that lets gas in and out, but no dust."

This is not only on-topic, but makes an entirely relevant and quite
useful point.

Darius the extremely dumb seems to think that "off-topic" actually means
"incomprehensible to Darius" and he does not seem to be able to
comprehend very much at all.

This is a troll-like but legitimate response - he does deserve it.

https://en.wikipedia.org/wiki/Troll_(slang)

--
Bill Sloman, Sydney

On 2024-02-23, Don Y <blockedofcourse@foo.invalid> wrote:
> I've been digging through piles of 3.5" disk drives (spinning rust).
> All seem to have "seals" to cover any hardware that "shouldn't be
> removed". Often, the seals are intentionally inconspicuous
> (like a label that is so large that it covers these fasteners
> as a matter of course -- or, metallic to mimic the metal cover).
> But, not always ("Gee, what's under this sticky round dot?")
>
> I don't think anyone would intentionally remove those fasteners
> thinking they could "fix" a broken disk (we remove them to decompose
> the disks into recyclable parts; the drive already being deemed
> "discardable").

The basic rules of drive repair are only open in a clean room, and
thus the seals are to indicate actions that would actually void the
warranty. Unlike the warranty stickers seen on appliances that are
just there to scare off the curious. if you're in a clean room
disassembling a hard drive the cost of the voided warranty is lost in
the noise of all the other costs.

> So, are they there to detect signs of tampering (voided
> warranty)? Or, discourage folks who are too curious for
> their own good?

Certainly the first, I can't think of anything inside a
drive that would be a danger to someone competent enough
to open it, unless they going to eat the magnets or
something dumb like that.

> I can't imagine they would add manufacturing steps to
> cover these fasteners if there wasn't some perceived
> value to doing so...

The value is avoiding unneccesary warranty returns.

--
Jasen.
🇺🇦 Слава Україні

On 3/3/2024 2:32 AM, Jasen Betts wrote:
> On 2024-02-23, Don Y <blockedofcourse@foo.invalid> wrote:
>> I've been digging through piles of 3.5" disk drives (spinning rust).
>> All seem to have "seals" to cover any hardware that "shouldn't be
>> removed". Often, the seals are intentionally inconspicuous
>> (like a label that is so large that it covers these fasteners
>> as a matter of course -- or, metallic to mimic the metal cover).
>> But, not always ("Gee, what's under this sticky round dot?")
>>
>> I don't think anyone would intentionally remove those fasteners
>> thinking they could "fix" a broken disk (we remove them to decompose
>> the disks into recyclable parts; the drive already being deemed
>> "discardable").
>
> The basic rules of drive repair are only open in a clean room, and
> thus the seals are to indicate actions that would actually void the
> warranty. Unlike the warranty stickers seen on appliances that are
> just there to scare off the curious. if you're in a clean room
> disassembling a hard drive the cost of the voided warranty is lost in
> the noise of all the other costs.

Superficially, that *seems* correct. But, if you look
at an actual disk, you will realize that you can break the
"atmospheric seal" without tampering with any of the
fasteners that are *under* seal. E.g., there are typically
6 torx screws on the periphery of the drive cover that
are not "sealed". Loosen/remove each of them and you've
surely compromised the drive. Yet, if you can do so -- and
later replace them without marring their "heads" -- there
is no way of visibly inspecting the drive to ascertain this
"violation" (the "seals" provide such a visible indication
for OTHER fasteners)

You can also remove (and reinstall) the PCB and, thus, "void
the warranty. Again, nothing on the fasteners for the PCB
that would give this indication.

There is something "special" about the fasteners that are
thus protected (associated with the head actuator) that
merits being able to determine if the fasteners have been
altered from their current "settings", in any way.

>> So, are they there to detect signs of tampering (voided
>> warranty)? Or, discourage folks who are too curious for
>> their own good?
>
> Certainly the first, I can't think of anything inside a
> drive that would be a danger to someone competent enough
> to open it, unless they going to eat the magnets or
> something dumb like that.
>
>> I can't imagine they would add manufacturing steps to
>> cover these fasteners if there wasn't some perceived
>> value to doing so...
>
> The value is avoiding unneccesary warranty returns.

They can't avoid the returns; they can avoid *paying* for
them (covering them) with a mechanism that, presumably, is
indisputable (though the seals don't appear to leave "VOID"
behind when removed, like many others do).

It is informative to see how other refurbished drives are
"sealed": the entire surface of the PCB is covered with
an adhesive seal (no probing, no access to the fasteners)
and a seal ties the top cover to the frame on each side
(so removing it via the screws I mentioned above) would
be obvious.