Stupid design decisions

The following is more a vent than anything really constructive, but...

So for the last four days (yes, I got called in New Years day
afternoon), something has been beeping in the computer room.

Let me rephrase that.  A proper beep would have likely been little
more than an annoyance.  No, this was a single, fraction of a second,
fairly high frequency, not very loud, chirp, repeated at odd
intervals, typically in the 5-20 minute range, although with
excursions to the 30s and 2-hour-plus ranges.  And did I mention, the
volume was not particularly consistent?  And it was high enough
pitched that half the people trying to help couldn't hear it at all.

IOW, a signal with very poor directionality, in a room with a couple
of dozen computers, other electronics, A/C, and whatnot contributing
tons of background noise.

I have rarely in my life been so frustrated, as I have been trying to
locate that dang* chirp.

As it turns out, there was a device that had an embedded fan (1.5 inch
muffin), and a temperature sensor.  The fan bearing was going bad (in
a quiet area, removed from the outer enclosure, the bad bearing was
audible, but not at all in the computer room) and cooling was
impacted.  The only sign of the overheat was that dang chirp from a
little magnetic buzzer (an AX-1212-LF, if anyone cares).  Somewhat
amusingly, this is a device rated for 8-12V, and there's nothing more
than 5V headed into the portion of the device where the buzzer was
(and there are mechanically identical 5V and 3V devices in the product
line - AX-1205, AX-1203, for example).  That probably explains part of
both the volume and frequency (the device is supposed to be producing
~2300Hz, but it was at least in the 5kHz range) issues.  The
apparently minimal length of pulse used to trigger it likely explains
the rest.  And did I mention that this buzzer is inside a mostly
enclosed component inside the case of the device?

And no, the overheat does not appear to be reported anywhere (despite
considerable chat between the device and the host system).

I understand the need to not make alarms excessively obnoxious (and
there is a handy little "BUZZER DISABLE" jumper next to the buzzer),
but making them so hard to hear is just absurd.  Clearly no one ever
tried listening to this in any real environment.  There are so many
ways this could have been better.  A simple series of chirps at each
alarm would have made this infinitely easier to trace.  Perhaps having
it repeat at a reasonable interval.  Or a reasonable frequency and
duration.  They went to all this trouble to monitor temperatures, and
then have an alarm, and then made it almost unusable.

Took about 15 minutes to swap in a spare (less than half of which was
screwdriver time).  Now to find a replacement fan to refurb the old
unit.

Ah well.  My apologies for the rant.  Back to the regularly scheduled
program...


*much stronger word removed

On 1/5/2017 8:14 PM, Robert Wessel wrote:
> The following is more a vent than anything really constructive, but...
>
> So for the last four days (yes, I got called in New Years day
> afternoon), something has been beeping in the computer room.
>
> Let me rephrase that.  A proper beep would have likely been little
> more than an annoyance.  No, this was a single, fraction of a second,
> fairly high frequency, not very loud, chirp, repeated at odd
> intervals, typically in the 5-20 minute range, although with
> excursions to the 30s and 2-hour-plus ranges.  And did I mention, the
> volume was not particularly consistent?  And it was high enough
> pitched that half the people trying to help couldn't hear it at all.

The duration isn't really important depending on the pitch -- esp
if it is a *fixed* pitch.

To locate a sound source, you need a relatively large spectrum of
frequencies and, ideally, a long enough duration that you can
"focus" your hearing on the signal *as* it persists (or, is thereafter
"reissued" at some predictable interval).

Raleigh opined that time and amplitude differences are resolved in
the brain (by the /superior colliculus/) to provide direction
(elevation and azimuth) information.  The differences in the time
of the wavefront striking each ear (which are displaced by the thickness
of your head!) is a cue that is useful at low frequencies (time translating
to phase) while the difference in amplitude (caused by the "shadowing"
effect that your head represents to the "far" ear) is more pronounced
at high frequencies.

[The shape of the pinnae plays an important, though user-specific, role
as well]

With a pure tone (or, narrow band sound source), there is a locus
("cone of confusion") of vectors along which the *particular* sound
can arrive (for that set of observed stimuli).

Witness how difficult it is to resolve the direction of an approaching
ambulance at a "blind" intersection!  (their sirens tend to operate in a
very narrow band ~500-2000Hz *and* are relatively "reflective" when bounded
by the hard surfaces of building walls, etc.)

> IOW, a signal with very poor directionality, in a room with a couple
> of dozen computers, other electronics, A/C, and whatnot contributing
> tons of background noise.

A smarter use of "audible annunciators" is to give each a particular
"character" (akin to picking a "ringtone").  So, instead of "something is
beeping", you can mentally *store* the (perhaps non-repeating) observation
as "something is ringing/buzzing/chiming/beeping/etc.".  Then, note what
device(s) can emit the sound having that particular characteristic.

[I use an "audio display" in my current project where users "place"
*specific* individual annunciators in space.  This allows them to
be brief and nonrepeating -- yet allows the user to "remember"
what and where he "heard something" (e.g., "it was a ringing sound
from above and to the left") so that he can analyze the sound after
it has gone.  We tend to have bad "sound memories" so need to be
able to quickly *label* a sound so we can re-CREATE the sound from
its descriptive label instead of actually "playing it back" in an
audio form in our minds.]

> I have rarely in my life been so frustrated, as I have been trying to
> locate that dang* chirp.

Try locating a cricket in a locked room!  "I know you're in here
*somewhere*... prepare to die!!"

> As it turns out, there was a device that had an embedded fan (1.5 inch
> muffin), and a temperature sensor.  The fan bearing was going bad (in
> a quiet area, removed from the outer enclosure, the bad bearing was
> audible, but not at all in the computer room) and cooling was
> impacted.  The only sign of the overheat was that dang chirp from a
> little magnetic buzzer (an AX-1212-LF, if anyone cares).  Somewhat
> amusingly, this is a device rated for 8-12V, and there's nothing more
> than 5V headed into the portion of the device where the buzzer was
> (and there are mechanically identical 5V and 3V devices in the product
> line - AX-1205, AX-1203, for example).  That probably explains part of
> both the volume and frequency (the device is supposed to be producing
> ~2300Hz, but it was at least in the 5kHz range) issues.  The

That's the most sensitive region to *hear* the sound but one of the LEAST
sensitive regions to LOCALIZE the sound!

> apparently minimal length of pulse used to trigger it likely explains
> the rest.  And did I mention that this buzzer is inside a mostly
> enclosed component inside the case of the device?
>
> And no, the overheat does not appear to be reported anywhere (despite
> considerable chat between the device and the host system).
>
> I understand the need to not make alarms excessively obnoxious (and
> there is a handy little "BUZZER DISABLE" jumper next to the buzzer),
> but making them so hard to hear is just absurd.  Clearly no one ever
> tried listening to this in any real environment.  There are so many
> ways this could have been better.  A simple series of chirps at each
> alarm would have made this infinitely easier to trace.  Perhaps having
> it repeat at a reasonable interval.  Or a reasonable frequency and
> duration.  They went to all this trouble to monitor temperatures, and
> then have an alarm, and then made it almost unusable.
>
> Took about 15 minutes to swap in a spare (less than half of which was
> screwdriver time).  Now to find a replacement fan to refurb the old
> unit.
>
> Ah well.  My apologies for the rant.  Back to the regularly scheduled
> program...
>
>
> *much stronger word removed
>

On 1/5/2017 10:14 PM, Robert Wessel wrote:
> The following is more a vent than anything really constructive, but...
>
> So for the last four days (yes, I got called in New Years day
> afternoon), something has been beeping in the computer room.
>
> Let me rephrase that.  A proper beep would have likely been little
> more than an annoyance.  No, this was a single, fraction of a second,
> fairly high frequency, not very loud, chirp, repeated at odd
> intervals, typically in the 5-20 minute range, although with
> excursions to the 30s and 2-hour-plus ranges.  And did I mention, the
> volume was not particularly consistent?  And it was high enough
> pitched that half the people trying to help couldn't hear it at all.
>
> IOW, a signal with very poor directionality, in a room with a couple
> of dozen computers, other electronics, A/C, and whatnot contributing
> tons of background noise.
>
> I have rarely in my life been so frustrated, as I have been trying to
> locate that dang* chirp.
>
> As it turns out, there was a device that had an embedded fan (1.5 inch
> muffin), and a temperature sensor.  The fan bearing was going bad (in
> a quiet area, removed from the outer enclosure, the bad bearing was
> audible, but not at all in the computer room) and cooling was
> impacted.  The only sign of the overheat was that dang chirp from a
> little magnetic buzzer (an AX-1212-LF, if anyone cares).  Somewhat
> amusingly, this is a device rated for 8-12V, and there's nothing more
> than 5V headed into the portion of the device where the buzzer was
> (and there are mechanically identical 5V and 3V devices in the product
> line - AX-1205, AX-1203, for example).  That probably explains part of
> both the volume and frequency (the device is supposed to be producing
> ~2300Hz, but it was at least in the 5kHz range) issues.  The
> apparently minimal length of pulse used to trigger it likely explains
> the rest.  And did I mention that this buzzer is inside a mostly
> enclosed component inside the case of the device?
>
> And no, the overheat does not appear to be reported anywhere (despite
> considerable chat between the device and the host system).
>
> I understand the need to not make alarms excessively obnoxious (and
> there is a handy little "BUZZER DISABLE" jumper next to the buzzer),
> but making them so hard to hear is just absurd.  Clearly no one ever
> tried listening to this in any real environment.  There are so many
> ways this could have been better.  A simple series of chirps at each
> alarm would have made this infinitely easier to trace.  Perhaps having
> it repeat at a reasonable interval.  Or a reasonable frequency and
> duration.  They went to all this trouble to monitor temperatures, and
> then have an alarm, and then made it almost unusable.
>
> Took about 15 minutes to swap in a spare (less than half of which was
> screwdriver time).  Now to find a replacement fan to refurb the old
> unit.
>
> Ah well.  My apologies for the rant.  Back to the regularly scheduled
> program...

I feel your pain.  I've always thought the high pitched beep is a load 
of crap to begin with.  Surely there are other ways to make a less 
annoying and more easily detected/noticed sound.  When I have a piece of 
gear in a bag making noise, good luck figuring out which bag and which 
piece of gear it is.  When it is such an intermittent thing like a low 
battery warning, well, you only get so long to locate it before it just 
quits.  Or worse, figuring out which of the two smoke detectors are 
emitting a once an hour, single chirp!  One is just above the other, two 
floors with an open balcony.  That drove me nuts for several nights.

Yeah, they really should give a bit more thought to the use cases for 
such alerts.

-- 

Rick C

On 2017-01-06, Robert Wessel <robertwessel2@yahoo.com> wrote:

> IOW, a signal with very poor directionality, in a room with a couple
> of dozen computers, other electronics, A/C, and whatnot contributing
> tons of background noise.
>
> I have rarely in my life been so frustrated, as I have been trying
> to locate that dang* chirp.

Ah, you wanted to know _where_ the problem was.  That's what the smoke
and flames are for...

-- 
Grant Edwards               grant.b.edwards        Yow! FOOLED you!  Absorb
                                  at               EGO SHATTERING impulse
                              gmail.com            rays, polyester poltroon!!

On Thu, 05 Jan 2017 21:14:17 -0600, Robert Wessel wrote:

> The following is more a vent than anything really constructive, but...
> 
> So for the last four days (yes, I got called in New Years day
> afternoon), something has been beeping in the computer room.
> 
> Let me rephrase that.  A proper beep would have likely been little more
> than an annoyance.  No, this was a single, fraction of a second, fairly
> high frequency, not very loud, chirp, repeated at odd intervals,
> typically in the 5-20 minute range, although with excursions to the 30s
> and 2-hour-plus ranges.  And did I mention, the volume was not
> particularly consistent?  And it was high enough pitched that half the
> people trying to help couldn't hear it at all.
> 
> IOW, a signal with very poor directionality, in a room with a couple of
> dozen computers, other electronics, A/C, and whatnot contributing tons
> of background noise.
> 
> I have rarely in my life been so frustrated, as I have been trying to
> locate that dang* chirp.
> 
> As it turns out, there was a device that had an embedded fan (1.5 inch
> muffin), and a temperature sensor.  The fan bearing was going bad (in a
> quiet area, removed from the outer enclosure, the bad bearing was
> audible, but not at all in the computer room) and cooling was impacted. 
> The only sign of the overheat was that dang chirp from a little magnetic
> buzzer (an AX-1212-LF, if anyone cares).  Somewhat amusingly, this is a
> device rated for 8-12V, and there's nothing more than 5V headed into the
> portion of the device where the buzzer was (and there are mechanically
> identical 5V and 3V devices in the product line - AX-1205, AX-1203, for
> example).  That probably explains part of both the volume and frequency
> (the device is supposed to be producing ~2300Hz, but it was at least in
> the 5kHz range) issues.  The apparently minimal length of pulse used to
> trigger it likely explains the rest.  And did I mention that this buzzer
> is inside a mostly enclosed component inside the case of the device?
> 
> And no, the overheat does not appear to be reported anywhere (despite
> considerable chat between the device and the host system).
> 
> I understand the need to not make alarms excessively obnoxious (and
> there is a handy little "BUZZER DISABLE" jumper next to the buzzer), but
> making them so hard to hear is just absurd.  Clearly no one ever tried
> listening to this in any real environment.  There are so many ways this
> could have been better.  A simple series of chirps at each alarm would
> have made this infinitely easier to trace.  Perhaps having it repeat at
> a reasonable interval.  Or a reasonable frequency and duration.  They
> went to all this trouble to monitor temperatures, and then have an
> alarm, and then made it almost unusable.
> 
> Took about 15 minutes to swap in a spare (less than half of which was
> screwdriver time).  Now to find a replacement fan to refurb the old
> unit.
> 
> Ah well.  My apologies for the rant.  Back to the regularly scheduled
> program...
> 
> 
> *much stronger word removed

The incorrect voltage spec may have been the brainchild of someone in 
procurement, or of an incompetent component engineer.

-- 

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

I'm looking for work -- see my website!

On 06/01/17 21:22, Grant Edwards wrote:
> On 2017-01-06, Robert Wessel <robertwessel2@yahoo.com> wrote:
>
>> IOW, a signal with very poor directionality, in a room with a couple
>> of dozen computers, other electronics, A/C, and whatnot contributing
>> tons of background noise.
>>
>> I have rarely in my life been so frustrated, as I have been trying
>> to locate that dang* chirp.
>
> Ah, you wanted to know _where_ the problem was.  That's what the smoke
> and flames are for...
>

If you have a good infrared camera around (not /that/ unlikely, in this 
newsgroup) then it can be useful to spot problems when they are beeping, 
but not yet smoking.

On 1/8/2017 11:11 AM, David Brown wrote:
> On 06/01/17 21:22, Grant Edwards wrote:
>> On 2017-01-06, Robert Wessel <robertwessel2@yahoo.com> wrote:
>>
>>> IOW, a signal with very poor directionality, in a room with a couple
>>> of dozen computers, other electronics, A/C, and whatnot contributing
>>> tons of background noise.
>>>
>>> I have rarely in my life been so frustrated, as I have been trying
>>> to locate that dang* chirp.
>>
>> Ah, you wanted to know _where_ the problem was.  That's what the smoke
>> and flames are for...
>>
>
> If you have a good infrared camera around (not /that/ unlikely, in this
> newsgroup) then it can be useful to spot problems when they are beeping,
> but not yet smoking.

If you are just trying to find a hot spot, a camera is a bit of 
overkill.  A handheld remote temp sensor is great.  I tried it last 
night and found my porch was 4&#4294967295;F warmer than the yard which was 0&#4294967295;F. 
The sky was either around -56&#4294967295;F or it just said "Lo".  Maybe they should 
have used a few more letters so they could say "F****** Cold".

-- 

Rick C

On 09.1.2017 г. 00:10, rickman wrote:
> On 1/8/2017 11:11 AM, David Brown wrote:
>> On 06/01/17 21:22, Grant Edwards wrote:
>>> On 2017-01-06, Robert Wessel <robertwessel2@yahoo.com> wrote:
>>>
>>>> IOW, a signal with very poor directionality, in a room with a couple
>>>> of dozen computers, other electronics, A/C, and whatnot contributing
>>>> tons of background noise.
>>>>
>>>> I have rarely in my life been so frustrated, as I have been trying
>>>> to locate that dang* chirp.
>>>
>>> Ah, you wanted to know _where_ the problem was.  That's what the smoke
>>> and flames are for...
>>>
>>
>> If you have a good infrared camera around (not /that/ unlikely, in this
>> newsgroup) then it can be useful to spot problems when they are beeping,
>> but not yet smoking.
>
> If you are just trying to find a hot spot, a camera is a bit of
> overkill.  A handheld remote temp sensor is great.  I tried it last
> night and found my porch was 4&deg;F warmer than the yard which was 0&deg;F. The
> sky was either around -56&deg;F or it just said "Lo".  Maybe they should
> have used a few more letters so they could say "F****** Cold".
>

I use the same technique. And it is "f****** cold" here, too; -11 C
last night and was - and is expected to - be worse over the next
few days.
Those IR thermometers are cheap nowadays, mine cost me about $300
back in the day (nearly 20 years ago....), I got it to control the
PCB baking temp when I started using BGA-s. Still in use for that,
and for finding cold spots on windows, doors and walls (to detect
warmth leaks), to measure tea temp (55 C is the highest tolerable
to start drinking, how many people knew that one :), beer temp
during the summer (target being around 5C is probably known by
most people...) etc etc.

Dimiter

On 1/9/2017 4:30 AM, Dimiter_Popoff wrote:
> On 09.1.2017 г. 00:10, rickman wrote:
>> On 1/8/2017 11:11 AM, David Brown wrote:
>>> On 06/01/17 21:22, Grant Edwards wrote:
>>>> On 2017-01-06, Robert Wessel <robertwessel2@yahoo.com> wrote:
>>>>
>>>>> IOW, a signal with very poor directionality, in a room with a couple
>>>>> of dozen computers, other electronics, A/C, and whatnot contributing
>>>>> tons of background noise.
>>>>>
>>>>> I have rarely in my life been so frustrated, as I have been trying
>>>>> to locate that dang* chirp.
>>>>
>>>> Ah, you wanted to know _where_ the problem was.  That's what the smoke
>>>> and flames are for...
>>>>
>>>
>>> If you have a good infrared camera around (not /that/ unlikely, in this
>>> newsgroup) then it can be useful to spot problems when they are beeping,
>>> but not yet smoking.
>>
>> If you are just trying to find a hot spot, a camera is a bit of
>> overkill.  A handheld remote temp sensor is great.  I tried it last
>> night and found my porch was 4&deg;F warmer than the yard which was 0&deg;F. The
>> sky was either around -56&deg;F or it just said "Lo".  Maybe they should
>> have used a few more letters so they could say "F****** Cold".
>>
>
> I use the same technique. And it is "f****** cold" here, too; -11 C
> last night and was - and is expected to - be worse over the next
> few days.
> Those IR thermometers are cheap nowadays, mine cost me about $300
> back in the day (nearly 20 years ago....), I got it to control the
> PCB baking temp when I started using BGA-s. Still in use for that,
> and for finding cold spots on windows, doors and walls (to detect
> warmth leaks), to measure tea temp (55 C is the highest tolerable
> to start drinking, how many people knew that one :), beer temp
> during the summer (target being around 5C is probably known by
> most people...) etc etc.

I don't know where you are, but I thought most of Europe drank warm 
beer.  I know they do in England.  A friend was working here and 
couldn't stand drinking cold beer.  He'd ask the bartender to find a 
case in the hallway that wasn't refrigerated.

-- 

Rick C

On 09.1.2017 г. 11:58, rickman wrote:
> On 1/9/2017 4:30 AM, Dimiter_Popoff wrote:
>> On 09.1.2017 г. 00:10, rickman wrote:
>>> On 1/8/2017 11:11 AM, David Brown wrote:
>>>> On 06/01/17 21:22, Grant Edwards wrote:
>>>>> On 2017-01-06, Robert Wessel <robertwessel2@yahoo.com> wrote:
>>>>>
>>>>>> IOW, a signal with very poor directionality, in a room with a couple
>>>>>> of dozen computers, other electronics, A/C, and whatnot contributing
>>>>>> tons of background noise.
>>>>>>
>>>>>> I have rarely in my life been so frustrated, as I have been trying
>>>>>> to locate that dang* chirp.
>>>>>
>>>>> Ah, you wanted to know _where_ the problem was.  That's what the smoke
>>>>> and flames are for...
>>>>>
>>>>
>>>> If you have a good infrared camera around (not /that/ unlikely, in this
>>>> newsgroup) then it can be useful to spot problems when they are
>>>> beeping,
>>>> but not yet smoking.
>>>
>>> If you are just trying to find a hot spot, a camera is a bit of
>>> overkill.  A handheld remote temp sensor is great.  I tried it last
>>> night and found my porch was 4&deg;F warmer than the yard which was 0&deg;F. The
>>> sky was either around -56&deg;F or it just said "Lo".  Maybe they should
>>> have used a few more letters so they could say "F****** Cold".
>>>
>>
>> I use the same technique. And it is "f****** cold" here, too; -11 C
>> last night and was - and is expected to - be worse over the next
>> few days.
>> Those IR thermometers are cheap nowadays, mine cost me about $300
>> back in the day (nearly 20 years ago....), I got it to control the
>> PCB baking temp when I started using BGA-s. Still in use for that,
>> and for finding cold spots on windows, doors and walls (to detect
>> warmth leaks), to measure tea temp (55 C is the highest tolerable
>> to start drinking, how many people knew that one :), beer temp
>> during the summer (target being around 5C is probably known by
>> most people...) etc etc.
>
> I don't know where you are, but I thought most of Europe drank warm
> beer.  I know they do in England.  A friend was working here and
> couldn't stand drinking cold beer.  He'd ask the bartender to find a
> case in the hallway that wasn't refrigerated.
>

Oh no, this is not the case in Europe; I am surprised about England,
never been there.
But where I have lived, in Bulgaria and Germany, and I think all the
countries between, beer is supposed to be around 5C. In fact in
Germany I have been told at the bar (must have been in the morning)
"we don't have beer yet, it is still warm".

Dimiter