Adding "debug" electronics to electronics to quickly diagnos

S

Skybuck Flying

Guest
Hello,

I had an idea how electronics might be diagnosed and repaired more quickly.

The idea is to add "debug leds" to electronics, and some kind of "debug
circuitry".

The designer knows what kind of inputs and outputs are expected for certain
components.

The designer then adds the debug circuitry to be able to measure if the
inputs and outputs correspond to what is expected.

The debug circuitry could be attached to special power lines or so to be
able to turn them on and off.

So during a repair session an electronics repairman could turn on the "debug
circuitry" and quickly get an idea
of how many components are potentially dead... because the debug leds don't
come on.

All good electronics would have the led on.

All bad electronics would have the led off.

Now the repairman can quickly proceed to the dead electronics and replace
them.

However there might be a problem if certain components depend on the inputs
and outputs of other electronics.

Then a cascade of "false" failures could happen.

To prevent this all electronics would have to be tested seperatedly by the
seperate "debug circuitry"...

As to be able to "debug" all electroncics individually.

So the debug electronics could have two modes:

1. "Cascade mode" where the inputs and outputs are compared to values which
corresponds with a correctly working circuitry.

2. "Individual mode" where the inputs and outputs are compared to individuel
values which correspond with a correctly working component when it's
individually powered on/tested.

Adding this kind of debug electronics would add to the cost so this should
be able to be done cheaply... with some kind of cheap debug electronics.

What do you electronics experts think of this idea, is it feasible or not
feasible ;) ? please motivate your answers ! ;) :)

Bye,
Skybuck.
 
On Tue, 11 Aug 2009 17:32:43 +0200, Skybuck Flying wrote:

Hello,

I had an idea how electronics might be diagnosed and repaired more
quickly.

The idea is to add "debug leds" to electronics, and some kind of "debug
circuitry".

The designer knows what kind of inputs and outputs are expected for
certain components.

The designer then adds the debug circuitry to be able to measure if the
inputs and outputs correspond to what is expected.

The debug circuitry could be attached to special power lines or so to be
able to turn them on and off.

So during a repair session an electronics repairman could turn on the
"debug circuitry" and quickly get an idea
of how many components are potentially dead... because the debug leds
don't come on.

All good electronics would have the led on.

All bad electronics would have the led off.

Now the repairman can quickly proceed to the dead electronics and
replace them.

However there might be a problem if certain components depend on the
inputs and outputs of other electronics.

Then a cascade of "false" failures could happen.

To prevent this all electronics would have to be tested seperatedly by
the seperate "debug circuitry"...

As to be able to "debug" all electroncics individually.

So the debug electronics could have two modes:

1. "Cascade mode" where the inputs and outputs are compared to values
which corresponds with a correctly working circuitry.

2. "Individual mode" where the inputs and outputs are compared to
individuel values which correspond with a correctly working component
when it's individually powered on/tested.

Adding this kind of debug electronics would add to the cost so this
should be able to be done cheaply... with some kind of cheap debug
electronics.

What do you electronics experts think of this idea, is it feasible or
not feasible ;) ? please motivate your answers ! ;) :)

Bye,
Skybuck.
Click to expand...
Google is your friend, even if you're a USENET troll.

Search on "built in test" or "BIT" for the military/high-end usage, and
"power on self test" or POST for PC usage.

And get a life.

--
www.wescottdesign.com
 
"Skybuck Flying" <BloodyShame@hotmail.com> wrote in message
news:7bdc8$4a818f07$d53372a9$9312@cache1.tilbu1.nb.home.nl...
Hello,

I had an idea how electronics might be diagnosed and repaired more
quickly.

The idea is to add "debug leds" to electronics, and some kind of "debug
circuitry".

The designer knows what kind of inputs and outputs are expected for
certain
components.

The designer then adds the debug circuitry to be able to measure if the
inputs and outputs correspond to what is expected.

The debug circuitry could be attached to special power lines or so to be
able to turn them on and off.

So during a repair session an electronics repairman could turn on the
"debug
circuitry" and quickly get an idea
of how many components are potentially dead... because the debug leds
don't
come on.

All good electronics would have the led on.

All bad electronics would have the led off.

Now the repairman can quickly proceed to the dead electronics and replace
them.

However there might be a problem if certain components depend on the
inputs
and outputs of other electronics.

Then a cascade of "false" failures could happen.

To prevent this all electronics would have to be tested seperatedly by the
seperate "debug circuitry"...

As to be able to "debug" all electroncics individually.

So the debug electronics could have two modes:

1. "Cascade mode" where the inputs and outputs are compared to values
which
corresponds with a correctly working circuitry.

2. "Individual mode" where the inputs and outputs are compared to
individuel
values which correspond with a correctly working component when it's
individually powered on/tested.

Adding this kind of debug electronics would add to the cost so this should
be able to be done cheaply... with some kind of cheap debug electronics.

What do you electronics experts think of this idea, is it feasible or not
feasible ;) ? please motivate your answers ! ;) :)

Bye,
Skybuck.

I think the engineers would never do this unless it would tend to make a
Click to expand...
fool of electronics experts. Very likely, bean counters would throw it out
before it got that far. In my experience, it is business as usual to
perform repairs in spite of the manufacturer rather than at the direction of
the manufacturer unless you are somehow underwritten by the sales dept.
Internal diagnostics schemes like these typically assist in manufacturing or
sales and are only found in expensive high end equipment..
 
"Skybuck Flying" <BloodyShame@hotmail.com> wrote in message
news:7bdc8$4a818f07$d53372a9$9312@cache1.tilbu1.nb.home.nl...
Hello,

I had an idea how electronics might be diagnosed and repaired more
quickly.

The idea is to add "debug leds" to electronics, and some kind of "debug
circuitry".

The designer knows what kind of inputs and outputs are expected for
certain components.

The designer then adds the debug circuitry to be able to measure if the
inputs and outputs correspond to what is expected.

The debug circuitry could be attached to special power lines or so to be
able to turn them on and off.

So during a repair session an electronics repairman could turn on the
"debug circuitry" and quickly get an idea
of how many components are potentially dead... because the debug leds
don't come on.

All good electronics would have the led on.

All bad electronics would have the led off.

Now the repairman can quickly proceed to the dead electronics and replace
them.

However there might be a problem if certain components depend on the
inputs and outputs of other electronics.

Then a cascade of "false" failures could happen.

To prevent this all electronics would have to be tested seperatedly by the
seperate "debug circuitry"...

As to be able to "debug" all electroncics individually.

So the debug electronics could have two modes:

1. "Cascade mode" where the inputs and outputs are compared to values
which corresponds with a correctly working circuitry.

2. "Individual mode" where the inputs and outputs are compared to
individuel values which correspond with a correctly working component when
it's individually powered on/tested.

Adding this kind of debug electronics would add to the cost so this should
be able to be done cheaply... with some kind of cheap debug electronics.

What do you electronics experts think of this idea, is it feasible or not
feasible ;) ? please motivate your answers ! ;) :)

Bye,
Skybuck.
Click to expand...
Many pieces of electronic equipment already have 'debug' modes. Almost
everything from toasters to your car now have anything between 1 and 50
supervisory processor ICs in them, and it is a comparitively simple job for
the programmer to build monitoring and testing routines into his code. The
results of these may be accessible to the service engineer via several
different routes, including some kind of 'standard' serial computer
interface - RS232 or USB etc - or the unit's user display, or in its
simplest form, by different flash sequences of the standby LED.

Yamaha AV amplifiers, for instance, have a very comprehensive monitoring and
diagnostic tool built in, complete with memory to tell you what events have
happened leading up to the failure. Your car also has very comprehensive
monitoring of just about everything electrical and electronic on the car,
and also most if not all of the electromechanical systems, again, complete
with storage to tell the mechanic just exactly what went wrong. Most (decent
makes of) TV sets have had the flashing LED type diagnostics for many years
now.

Doing it by way of the system control processors is cheap for the
manufacturers, as there is only a one-time investment in code writing
overhead. No, or very minimal, additional hardware support, is needed.

Arfa
 
I was thinking about "simple" hardware like audio hardware (simple?).

Where there are many little simple components ?

Bye,
Skybuck.
 
Yes but how usefull are current systems for actually repairing hardware ?

Time-wise too.

Repairman should be able to very quickly spot the defect and replace it...

I am even thinking about easy to remove and plug back components so that
even
non-eletronics experts could repair electronics by simply going to a store
buying the necessary little components and simply stick them into place
without any soldering... that would be cool ! ;)

Bye,
Skybuck.
 
On 8/11/2009 1:01 PM Skybuck Flying spake thus:

Yes but how usefull are current systems for actually repairing hardware ?

Time-wise too.

Repairman should be able to very quickly spot the defect and replace it...
Click to expand...
You're assuming that manufacturers of such devices would actually want
to make them easily repairable.

The opposite is in fact the case. Almost everything today is made to
just be shitcanned and replaced if it stops working. Repair? What's that?


--
Found--the gene that causes belief in genetic determinism
 
"Skybuck Flying" <BloodyShame@hotmail.com> wrote in message
news:b2ee8$4a81ce01$d53372a9$26366@cache4.tilbu1.nb.home.nl...
Yes but how usefull are current systems for actually repairing hardware ?
Click to expand...
Hardware "repair" these days usually just means board swapping -- it's rather
more expensive (in most cases) to perform component level
troubleshooting/repair anymore, so few companies bother.

I am even thinking about easy to remove and plug back components so that
even
non-eletronics experts could repair electronics by simply going to a store
buying the necessary little components and simply stick them into place
without any soldering... that would be cool ! ;)
Click to expand...
Ask your grandparents about their old tube radios some time... they
occasionally came with replacement parts provided inside. I'm young enough
that while I grew up seeing the ubiquitous "No user serviceable parts inside"
disclaimer on everything I bought new (and would wonder, "Why do they even
print that when it's the same for everything?"), one day I was astonished to
come across an old radio of my mother's that said something like, "Serviceable
parts inside: Two fuses, alginment tool, etc..."

Besides just the cost factor, another factor that led to electronics no being
user serviceable include the vast improvement in reliability that come as IC
manufacturing processes matured -- in many cases by the time the electronics
in a widget fail, the widget will be obsolete anyway and the customer will
just use the failure as a reason to upgrade to something newer/fancier.
Indeed, other than "infant death" and "abuse," most remaining electronic
failures are due to poor designs (see, e.g., the Xbox "ring of death"
debacle... or Jan's post about his USB plugs) and not transistor/IC technology
relability problems.

---Joel
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top