Medical Device Alarm

nt. To meet the requirement requires using odd and even harmonics. That s eems to form something like a saw tooth wave. Looks like the first four ha rmonics, 2, 3, 4 and 5th in a saw tooth wave will be no lower than -14 dB w hich is within the -15 dB spec, likely why they picked this number.

n getting that something more pure? Why not a square wave? I'm guessing t he harmonic content of the square wave is lower and at higher frequencies s o harder to meet the spec as written anyway. Does a square wave sound wors e? I know I've heard them plenty of times, but I can't recall at the momen t.

e accumulator which is already a sawtooth ramp.

Go to the next level up:

I saw a medical ~standard on the sounds of medical devices. The intent was to make only urgent warnings sound urgent, and so on, to combat alarm fatigue in hospitals. I cannot find the article, but Google should help. I think I mentioned on S.E.D.

Another reason to have harmonics is so one can estimate how far away the alarming unit is. We do this by the change in spectrum at the sound travels, due to the increased relative attenuation of higher harmonics.

Joe Gwinn

tent. To meet the requirement requires using odd and even harmonics. That seems to form something like a saw tooth wave. Looks like the first four harmonics, 2, 3, 4 and 5th in a saw tooth wave will be no lower than -14 dB which is within the -15 dB spec, likely why they picked this number.

ion getting that something more pure? Why not a square wave? I'm guessing the harmonic content of the square wave is lower and at higher frequencies so harder to meet the spec as written anyway. Does a square wave sound wo rse? I know I've heard them plenty of times, but I can't recall at the mom ent.

ase accumulator which is already a sawtooth ramp.

If nothing else, this group can be good for comic relief.

ent. To meet the requirement requires using odd and even harmonics. That seems to form something like a saw tooth wave. Looks like the first four h armonics, 2, 3, 4 and 5th in a saw tooth wave will be no lower than -14 dB which is within the -15 dB spec, likely why they picked this number.

on getting that something more pure? Why not a square wave? I'm guessing the harmonic content of the square wave is lower and at higher frequencies so harder to meet the spec as written anyway. Does a square wave sound wor se? I know I've heard them plenty of times, but I can't recall at the mome nt.

se accumulator which is already a sawtooth ramp.

Yes, I found many articles. It was only today I found one from Malory (an alarm maker) that indicated IEC 60601-1-8 is voluntary. It is actually ea sy to meet once the details are understood.

It seems like the sawtooth is the way to go, especially since it is simpler circuitry. I just need to identify a speaker. Not so easy at 250 Hz in a small place.

ontent. To meet the requirement requires using odd and even harmonics. Th at seems to form something like a saw tooth wave. Looks like the first fou r harmonics, 2, 3, 4 and 5th in a saw tooth wave will be no lower than -14 dB which is within the -15 dB spec, likely why they picked this number.

ntion getting that something more pure? Why not a square wave? I'm guessi ng the harmonic content of the square wave is lower and at higher frequenci es so harder to meet the spec as written anyway. Does a square wave sound worse? I know I've heard them plenty of times, but I can't recall at the m oment.

phase accumulator which is already a sawtooth ramp.

I can't find the link for the ultimate alarm. It's basically a motorised bo xing glove. Rather than a joke, someone with a sleeping disorder made it to wake in the morning.

NT

A sawtooth sounds a bit like an angry wasp, maybe we've evolved to notice it more.

A very pure sine wave sounds surprisingly restful.

A square wave sounds just about as crude and horrible as you can get.

A rough exponential sawtooth was very easy to generate and sounds alarming enough without being too annoying or distracting.

If you recall the Three Mile Island disaster the operators spent the first 15 minutes figuring out how to silence all the damned alarms and klaxons so they could hear each other over the din! Only then were they able to get on with the task in hand of trying to prevent a meltdown.

Human factors come into play where alarms and alerts are concerned.

At least the 1000 Hz sine wave is very awful indoors due to the standing waves, so you can loose orientation quite quickly.

In a room with potentially multiple alarm system, you should be able to reliably locate which device generated the alarm.

Do you actually need to have the 250 Hz fundamental? If you have multiple harmonics of the fundamental, the brain will reconstruct the fundamental. So just generate the waveform and feed it to the annucator, and see how it siounds.

Joe Gwinn

ontent. To meet the requirement requires using odd and even harmonics. Th at seems to form something like a saw tooth wave. Looks like the first fou r harmonics, 2, 3, 4 and 5th in a saw tooth wave will be no lower than -14 dB which is within the -15 dB spec, likely why they picked this number.

ntion getting that something more pure? Why not a square wave? I'm guessi ng the harmonic content of the square wave is lower and at higher frequenci es so harder to meet the spec as written anyway. Does a square wave sound worse? I know I've heard them plenty of times, but I can't recall at the m oment.

phase accumulator which is already a sawtooth ramp.

an alarm maker) that indicated IEC 60601-1-8 is voluntary. It is actually easy to meet once the details are understood.

ler circuitry. I just need to identify a speaker. Not so easy at 250 Hz i n a small place.

The spec says withing 15 dB of the fundamental, which is about 5.6x. It ca n be either direction so I suppose the fundamental can be attenuated, but t he first harmonic is about 500 which is still out of range for many small t ransducers. I guess we can leave that mounted to the chassis somehow and w ire with a connector. Then it won't impact the board design and that can p roceed.

I doubt there will be adequate resources in the FPGA, but I'd like to be ab le to make it talk as a joke.

I would think at this point in time the construction and optimization of alarm tones should be well known or is this usually done on a one time individual basis? ca 1961 I worked on a data handing device that had an error alarm that sounded like a very well tuned dripping faucet. It wasn't painfully loud but was impossible to ignore and impossible to sleep through.

content. To meet the requirement requires using odd and even harmonics. T hat seems to form something like a saw tooth wave. Looks like the first fo ur harmonics, 2, 3, 4 and 5th in a saw tooth wave will be no lower than -14 dB which is within the -15 dB spec, likely why they picked this number.

ention getting that something more pure? Why not a square wave? I'm guess ing the harmonic content of the square wave is lower and at higher frequenc ies so harder to meet the spec as written anyway. Does a square wave sound worse? I know I've heard them plenty of times, but I can't recall at the moment.

phase accumulator which is already a sawtooth ramp.

(an alarm maker) that indicated IEC 60601-1-8 is voluntary. It is actual ly easy to meet once the details are understood.

mpler circuitry. I just need to identify a speaker. Not so easy at 250 Hz in a small place.

alarm

al

rm that

but was

The IEC standard is just a recommendation and even then it doesn't directly specify all the details. Some devices use a simple fixed tone of higher f requency. Some devices use a fixed tone of a frequency used for a general alarm with a well defined cadence. Other devices use a set of tones with c adences specific to the level of severity of the alarm event. One document I read claimed the IEC standard was enforced by the customers requirement to adhere to the standard rather than a national requirement.

So there is commonality to greater or lesser degrees. It seems to not so m uch be about the basic function of the alarms, but rather about the finer p oints.

The thing that really makes this hard is knowing that if I plan an architec ture for the requirements now, I will need to redo it later when they chang e. Very few people on this project want to dig into the requirements aspec t of the matter.

I think this circuit will be table driven for both the processing and the a larm presentation. Hand shakes to coordinate with the alarm sources. It's the sort of thing that thinks about itself when you aren't thinking.