Motherboard fuses - missing?

[Jumpster Jiver]

What is it about the OP's post that you don't understand? He has
stated that all the motherboard's IO functions are operational. This
means that *all* devices are getting their power from the motherboard,
despite several fuse locations being unpopulated, ie *open*. So the
OP's question as to what function these missing fuses would have
performed is a valid and logical one.


- Franc Zabkar

While I'm sure that on this MOBO the fuse POSITIONS are open, I'd bet
that the circuit is not. Chances are that SOMEWHERE ELSE in the path of
that circuit where the fuse would have been there is a jumper. As
some other knowlegable posters stated, it may not be physically in the
same location as the missing fuse, but it is in the same circuit.
I'm sure if two boards were compared sie-by-side and only one is found
to have fuses then I'd bet that somewhere on that board it is missing a
jumper for each fuse it has.
They may be the size of a surface-mount resistor and thus very difficult
to find, or it may be a tiny solder bridge or even a full size jumper
like the 1/4 watt size, but it must be there somewhere!

 

[Conor]

And, the ports DO work.
So?

It was not a matter of omitting
fuses or jumpers for features not present if these are as
described. So, you have to have an alternate way of getting
the power to those ports if it's not traveling the same path
as (practically any) motherboard out there.

DUMB AS A FUCKING BOX OF ROCKS.

I take it you've "dabbled " in electronics as a hobby? Yes? Maybe made
your own circuits? Yes?

Whooptie friggin doo.

Now go learn about multilayered PCBS - that's what industry commonly
use. That's when one PCB has several layers of copper sandwidched
together.

http://www.thinktink.com/stack/volumes/volvi/multilyr.htm


THis company here: http://www.clarydon.com/services/multilayer.html

...does up to 24 layers on a single PCB.

Just because the tracks you CAN see go to the fuse, doesn't mean tracks
you CAN'T see don't.


--
Conor

If Pac-Man affected us as kids, we'd all be running around in darkened
rooms, munching magic pills and listening to repetitive electronic
music.

 

[Conor]

But has someone actually traced the board?

I doubt it.

Not forgetting that you need to not only trace both sides at the same
time. remembering those little holes are actually links from one side
to another, but also find out whethere there are layers in the middle
too.


--
Conor

If Pac-Man affected us as kids, we'd all be running around in darkened
rooms, munching magic pills and listening to repetitive electronic
music.

 

[Conor]

What is it about the OP's post that you don't understand? He has
stated that all the motherboard's IO functions are operational. This
means that *all* devices are getting their power from the motherboard,
despite several fuse locations being unpopulated, ie *open*. So the
OP's question as to what function these missing fuses would have
performed is a valid and logical one.

And he has had an answer. To provide flexibiity in the motherboard for
the manufacturer to use it for more than one type of product.


--
Conor

If Pac-Man affected us as kids, we'd all be running around in darkened
rooms, munching magic pills and listening to repetitive electronic
music.

 

[kony]

But has someone actually traced the board? Like I said, boards
can be laid out to permit different size parts. You do not see
jumpers across the component pads. The traces take circuitous
routes, and the parts may not be right next to the alternative
component pads.

If jumpers are needed, I suspect they'd be laid out as jumpers,
rather than expecting a machine to jam in jumpers across pads
laid out for components. And the jumper pads may not be right
next to the "missing" components.

While that seems logical in a good design, what we have in
these cases is "usually" a board layout that allows for the
various fuses, but then later the fuses are not added
(shaves a couple cents cost) but rather that fuse location
has a jumper wire, or just traced over.

Hence unless someone gets out an ohmmeter and traces the circuit
board, they can't be sure what's happening.

Michael

I agree, but still it's curious that this particular board
has this setup. He and I have both noted many boards with
the fuses there OR missing where they had used more obvious
method of continuity of the 5V supply by merely bridging the
pad(s) gap where a fuse could've gone.

 

[kony]

I doubt it.

Not forgetting that you need to not only trace both sides at the same
time. remembering those little holes are actually links from one side
to another, but also find out whethere there are layers in the middle
too.

Usually not, the 5V power plane does not extend that far and
generally singular 5V traces are not put in middle layers
for such ports. "Maybe" they did it, but if so this is
first time I (or apparently LM&C) has seen it.

 

[kony]

DUMB AS A FUCKING BOX OF ROCKS.

I take it you've "dabbled " in electronics as a hobby? Yes? Maybe made
your own circuits? Yes?

Whooptie friggin doo.

You would do to stop pretending you have vast knowledge and
gain some ACTUAL hands-on experiece with motherboards. AT
least then you'd have a slight hint at what's being
discussed and the significance of it.

Instead you guess about it, because it has some passing
similarity to some other electronics circuits? Well this
isn't "some other" circuit.

Now go learn about multilayered PCBS - that's what industry commonly
use. That's when one PCB has several layers of copper sandwidched
together.

http://www.thinktink.com/stack/volumes/volvi/multilyr.htm

Please take the hint that the problem is not our lack of
understanding very basic circuit board concepts but yours in
not realizing what is normally seen on motherboards.

Some grand (but simplistic) thought about multiple layers
does not change the fact that no other boards that (we're)
aware of, use inner layers for port power. I welcome anyone
to come forward with examples of any boards that do.

Further, if this board had an inner layer suppling (5V), it
would make the 5V traces, fuse pads on top completely
worthless and just a waste of space to put on the board at
all... yet there they are.

THis company here: http://www.clarydon.com/services/multilayer.html

..does up to 24 layers on a single PCB.

Just because the tracks you CAN see go to the fuse, doesn't mean tracks
you CAN'T see don't.

Never claimed otherwise. Now give us even one example of a
board that does this so your theory is something more than a
wild guess.

 

[SMS]

Just thinking out loud... not only different models, but
that one board design will have been designed for use in
different parts of the world.

Perhaps outher countries have different code requirements for
things "extended from" the main chassis?

This is the reason. I used to work for a large Taiwanese motherboard
company. If the board was being sold into a channel where there was no
UL/CSA/TUV, etc. requirement for fusing the power going off the board,
then the fuses were eliminated. Motherboards sold at retail are not
going into systems that will ever need to be certified by a safety
agency, so there is a savings of a few cents per board by eliminating
the fuses.

There are other savings that are possible with boards that would never
be used in systems that would go through agency approval. You can use
cheaper PC board material. You can eliminate the double protection
against charging the on-board lithium battery. IIRC, those were the only
three things that the safety agencies looked at. For EMI, there were
other items that could be eliminated.

Wouldn't explain a missing fuse all by itself, but perhaps
one jumper is eliminated elsewhere and then the fuse is
required...

It is probable that the motherboard company changed one of the inner
layers of the board (the power layer) on the board in question, to
eliminate the need for a jumper, or a copper trace, on the top or bottom
layer. Installing a jumper, or cutting a trace, is as expensive as
putting in a fuse.

Acually UL does NOT require fuses. You have the option of providing a
keyboard and mouse where a short from power to ground, with the maximum
current that the power supply can provide, will not result in an unsafe
condition (excessive heat or fire). Never mind that the PC board trace
would be destroyed long before the maximum current condition ever
occured. Needless to say, it was easier to put in a fuse, and trying to
argue with UL is like trying to teach calculus to a cat.

 

[Conor]

You would do to stop pretending you have vast knowledge and
gain some ACTUAL hands-on experiece with motherboards.

I had hands on experience of multilayered PCBs back in 1987 working as
an engineer in the video games repair industry.

AT
least then you'd have a slight hint at what's being
discussed and the significance of it.

I have. More than you it would seem.

Please take the hint that the problem is not our lack of
understanding very basic circuit board concepts but yours in
not realizing what is normally seen on motherboards.

I know exactly what is normally seen. I also know that some functions
have ended up being integrated into ICs themselves. As well as that, my
background comes from an industry which routinely scrubbed numbers off
chips and also misnumbered them to prevent piracy.

Some grand (but simplistic) thought about multiple layers
does not change the fact that no other boards that (we're)
aware of, use inner layers for port power. I welcome anyone
to come forward with examples of any boards that do.

Why wouldn't they?

Further, if this board had an inner layer suppling (5V), it
would make the 5V traces, fuse pads on top completely
worthless and just a waste of space to put on the board at
all... yet there they are.

So they're there. SO FUCKING WHAT? I can show you 20 CB radios from one
manufacturers, all based on the same board layout. On each model, the
PCB is the same but the components differ with blanks being left for
unused sections.



--
Conor

If Pac-Man affected us as kids, we'd all be running around in darkened
rooms, munching magic pills and listening to repetitive electronic
music.

 

[Conor]

It is probable that the motherboard company changed one of the inner
layers of the board (the power layer) on the board in question, to
eliminate the need for a jumper, or a copper trace, on the top or bottom
layer. Installing a jumper, or cutting a trace, is as expensive as
putting in a fuse.

<snip>

Kony, despite two of us now saying the same thing, will claim he is
right.

--
Conor

If Pac-Man affected us as kids, we'd all be running around in darkened
rooms, munching magic pills and listening to repetitive electronic
music.

 

[James Sweet]

In this case, maybe a certain country has different safety standards
that require the keyboard, mouse, etc. to be fused, so on the models
sent to that country they are fused, and the same model marketed in
other countries leaves the fuse out and maybe puts a cheaper jumper in
to save money on components.
In large quantities the fuse may cost a penny and a jumper may cost 1/10
of a penny. Over hundreds of thousands of boards it makes a difference.

It's reasonably common to design a board to accomodate two or more styles of
certain components so either can be used depending on which is most
available/least expensive at the time of manufacture. For example one of the
boards made where I work had pads in place for two different package styles
of voltage regulator, the pads are wired in parallel but only one location
is stuffed with the part.

 

[James Sweet]

I had hands on experience of multilayered PCBs back in 1987 working as
an engineer in the video games repair industry.

I have. More than you it would seem.
I know exactly what is normally seen. I also know that some functions
have ended up being integrated into ICs themselves. As well as that, my
background comes from an industry which routinely scrubbed numbers off
chips and also misnumbered them to prevent piracy.

Why wouldn't they?

So they're there. SO FUCKING WHAT? I can show you 20 CB radios from one
manufacturers, all based on the same board layout. On each model, the
PCB is the same but the components differ with blanks being left for
unused sections.

Regardless of your level or lack of knowledge, your attitude certainly
leaves much to be desired.

 

[Conor]

Regardless of your level or lack of knowledge, your attitude certainly
leaves much to be desired.

I guess you're not that old?

--
Conor

If Pac-Man affected us as kids, we'd all be running around in darkened
rooms, munching magic pills and listening to repetitive electronic
music.

 

[kony]

<snip>

Kony, despite two of us now saying the same thing, will claim he is
right.

You offer a general theory on how it would be possible to do
it. That theory is of course obvious (that is, "possible in
theory so long as we only consider that detail) to anyone
who knows about multi-layer boards, BUT you are ignoring a
few things.

1) We have no, zero, evidence that _ANY_ motherboards do
this. Do you carefully examine modern motherboards? I
think not... LM&C, and I, do. "IF" this particular board
uses an inner layer to connect to the 5V rail and/or 5VSB,
this is the first evidence we have of it, evidence meaning
an actual board that does so... or at least nobody has
observed and/or mentioned it.

2) I continue to question the idea of implementation of the
surface-mount jumpers or fuses as optional based on using
one board layout for multiple featured (or zoned) products.
It is questioned because to implement the fuse ever (else it
would not be layed out on the board at all) the inner layer
cannot be connected in that implementation. Therefore, when
some boards used the fuse, they would have a more laborous
modification needed, to rework the inner layer before
fabbing the PCB rather than choosing which surface-mount
pads to populate.

You can't have it both ways, if the inner layer trace(s)
suppy power then there is no purpose to surface-mount pads
at all, they are a complete waste of space and never
useable, UNLESS they reworked the inner layer every time. I
don't think they do that, but again I'll welcome any real
evidence of it rather than some theory that it "could be
done" without consideration of whether it ever is.

There is still another possiblity, that the board can
accomodate other I/O ports/features, that the missing
fuses/jumpers do not actually serve to supply the ports that
are implmented on this specimen of the board, nor do the
ports on the board have inner layer traces, but that the
ports on the board ALSO have surface-layer traces suppling
power, yet another 5V supply line for them.

 

[kony]

I had hands on experience of multilayered PCBs back in 1987 working as
an engineer in the video games repair industry.

Good for you, now would you feel it's reasonable for anyone
else to insist that they "know" for certain how one of those
1987 PCBs was designed based on their observation of a
modern PC motherboard, when every single board you've ever
seen from '87 was NOT like that "anyone" insists they are?

Generaly theory that multiple layers can allow a trace not
seen, is not evidence that this IS how and why LM&C observed
what he did.

I have. More than you it would seem.

Again you fail to grasp the obvious... that you would have
to actually examine modern motherboards to have any
applicable evidence. A theory about what "could" be, is not
the same thing as evidence of it actually being true.

I know exactly what is normally seen. I also know that some functions
have ended up being integrated into ICs themselves. As well as that, my
background comes from an industry which routinely scrubbed numbers off
chips and also misnumbered them to prevent piracy.

No, you clearly have no clue about what is normally seen.
Look on the motherboard in the system you're using RIGHT
NOW. Dig a few dozen out of a closet and look at them.
Look online at good pictures. Your "opinion" is worth
little if it is not consistent with actual boards.

Would you agree than when the surface-layer's pads ARE
populated with a fuse (or a jumper, inductor, 0-Ohm
resistor, or trace closing the two fuse pads) that it would
be pointless when there is also a parallel power trace in an
underlying layer? Certainly anyone with as much circuit
board experience as you claim would at least recongnize this
basic electrical fact, that one cannot fuse one of two
parallel supply lines and have the fuse be useful.
Therefore there is no point to having surface-mount pads
unless the inner layer is reworked.

Is this what you claim, that they redesign and remanufacture
boards with a different inner layer every time? That would
be quite the opposite of this "universal" board concept you
claimed where they can use one board with ommision of some
surface-mount components based on need.

Why wouldn't they?

For one, because they have them on the top layer. I've
already mentioned the issues surrounding parallel lines and
inner layer rework for supposed universal boards.

So they're there. SO FUCKING WHAT?

So you're not using evidence or any thorough evaluation of
whether your idea is reasonable, merely making a passing
comment about internal layers is just an educated guess...
but not all that educated because you are not using ANY
modern motherboards as examples, while there are absolutely
zero reports of any boards (thus far) that actually employ
what you're guessing they do.

I can show you 20 CB radios from one
manufacturers, all based on the same board layout. On each model, the
PCB is the same but the components differ with blanks being left for
unused sections.

You have fallen victim to a common human delusion, that if
someone has a lot of experience in something loosely
related, they suddenly become an expert on that other topic
too.

Yes, the PCB can have unused pads filled as needed. The
WHOLE point of doing this is to avoid having to rework the
circuit traces every time. If there are inner layers which
short parallel paths to those available on the top layer,
there is no point to having the top layer traces in ANY
implementation of features (or lack thereof).

 

[kony]

I guess you're not that old?

LOL.

So you're making the "I'm a cranky old fart" argument as a
defense?

 

[Sparex]

What is it about the OP's post that you don't understand? He has
stated that all the motherboard's IO functions are operational. This
means that *all* devices are getting their power from the motherboard,
despite several fuse locations being unpopulated, ie *open*. So the
OP's question as to what function these missing fuses would have
performed is a valid and logical one.

And the answer it seems from me reading this entire series of rants and
logical converstaions is that the only function that would have served
is to allow the board to be Certified for use in a safty controled
enviroment. Otherwise there is probally no other function to the fuse..
there or not.. its all a matter of certification and that seems to be
all, unless I see the 95% similar model with the fuses in place, no
cert, and new features that the other didn't have due to a small change
of 1 fuse... i dont think there would be a point to attempt to find out
further.

 

[kony]

It's reasonably common to design a board to accomodate two or more styles of
certain components so either can be used depending on which is most
available/least expensive at the time of manufacture.

True.

HOWEVER, it is not common for them to do so when it requires
reworking the inner layer. If they do not rework the inner
layer, there is no point to having these surface mount pads,
as the inner layer is always a closed (sub)circuit and
nothing done (whether open or closed, populated) with these
surface mount supply lines makes a difference.

For example one of the
boards made where I work had pads in place for two different package styles
of voltage regulator, the pads are wired in parallel but only one location
is stuffed with the part.

Agreed, that's quite true and common. It does not, however,
mean that inner layer traces are hardwired which would be
necessary if the upper layer pad population is entirely
optional... else as I mentioned briefly in another post,
there may be entirely separate traces for those features
actually implemented. Thus the key issue is whether these
empty, and presumed open circuit pads the OP observed are
actually those for supply to implemented features.

 

[Conor]

1) We have no, zero, evidence that _ANY_ motherboards do
this.

Nah - only years of examples of cost cutting in industry.

Do you carefully examine modern motherboards? I
think not... LM&C, and I, do. "IF" this particular board
uses an inner layer to connect to the 5V rail and/or 5VSB,
this is the first evidence we have of it, evidence meaning
an actual board that does so... or at least nobody has
observed and/or mentioned it.

You wouldn't even be able to tell.

2) I continue to question the idea of implementation of the
surface-mount jumpers or fuses as optional based on using
one board layout for multiple featured (or zoned) products.

Why? Industry standard practice.

It is questioned because to implement the fuse ever (else it
would not be layed out on the board at all) the inner layer
cannot be connected in that implementation.

Eh? And you're trying to tell ME that it can't be done? Clue: Design
feature 1 - solder runs through the holes connecting the layers.

--
Conor

If Pac-Man affected us as kids, we'd all be running around in darkened
rooms, munching magic pills and listening to repetitive electronic
music.

 

[Conor]

Good for you, now would you feel it's reasonable for anyone
else to insist that they "know" for certain how one of those
1987 PCBs was designed based on their observation of a
modern PC motherboard, when every single board you've ever
seen from '87 was NOT like that "anyone" insists they are?

**** sake. You are incredibly dense. THe same cost cutting techniques
used then are also used now.
you it would seem.

Again you fail to grasp the obvious... that you would have
to actually examine modern motherboards to have any
applicable evidence.

I don't need to. Its out ther in the interweb. Others have already done
it.

No, you clearly have no clue about what is normally seen.

I'm going to end this bullshit right here.

Sorry but unless you're a qualified electronics engineer with
experience of working to component level - which your stupid
statements clearly show you aren't - then your comments are worth
precisely **** all.

Good day.

--
Conor

If Pac-Man affected us as kids, we'd all be running around in darkened
rooms, munching magic pills and listening to repetitive electronic
music.