Chip with simple program for Toy

[John Larkin]

On Sat, 3 Aug 2013 17:58:07 -0700 (PDT), Zephod Beeblebrox
<sagephotoworld@gmail.com> wrote:

The reason I want to keep ICs out of it is because ICs seem to induce delays. I encountered a very interesting audio trigger the other day.

Now, before anything else, let me describe the setup...

A gun fires making a noise (doesn't matter if it's an airgun or a cartridge)

A microphone picks up the noise and that goes to a black box. The black box controls the time delay between the noise and closing a switch.

The switch normally closed is across the contacts of a flash unit, causing the flash.

The interesting unit had a sampling interval of 100,000ms which means that if a gun fired, it would still activate because the sound is longer than 100,000ms but it would make a huge difference in the location of a pellet. If it was travelling at 500fps in 100,000ms it could travel 5 feet. That's not very helpful. I gather all ICs tend to have sampling intervals.

100,000 ms is 100 seconds. The projectile will be miles away (or on the ground)
by then.

Linear ICs don't have sampling intervals, and can process signals in
nanoseconds.

What I want to do is to have a flash unit that uses solely LEDs. Hell, I can rebuild my LED box, I don't care - time is irrelevant as I can do a few components a day if need be and I have a fancy for using the new ultrabright leds too.

I *could* build a microflash using a Model T vibrating coil, a 12v battery, a couple of ignition coils etc but the voltages are rather high and some of the parts are very expensive and hard to get - for example the 25kv diodes and the 50kv diodes etc, not to mention the mylar capacitors. Half of that stuff I'd have to source from old Soviet nuclear facilities. At least that's where most parts seem to be coming from these days. And it would be bulky.

I want to build something smaller that runs off low voltages that I'm not going to have to be so careful with. I live in Columbia SC where it is very humid and thus the likelihood of a HV accident is higher. While I like freshly fried bacon, I don't want to become freshly fried bacon.

I don't think it's possible to make a usable microsecond flash from LEDs. The
numbers just don't work.


--

John Larkin Highland Technology Inc
www.highlandtechnology.com jlarkin at highlandtechnology dot com

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators

 

[Zephod Beeblebrox]

The reason I want to keep ICs out of it is because ICs seem to induce delays. I encountered a very interesting audio trigger the other day.

Now, before anything else, let me describe the setup...

A gun fires making a noise (doesn't matter if it's an airgun or a cartridge)

A microphone picks up the noise and that goes to a black box. The black box controls the time delay between the noise and closing a switch.

The switch normally closed is across the contacts of a flash unit, causing the flash.

The interesting unit had a sampling interval of 100,000ms which means that if a gun fired, it would still activate because the sound is longer than 100,000ms but it would make a huge difference in the location of a pellet. If it was travelling at 500fps in 100,000ms it could travel 5 feet. That's not very helpful. I gather all ICs tend to have sampling intervals.

What I want to do is to have a flash unit that uses solely LEDs. Hell, I can rebuild my LED box, I don't care - time is irrelevant as I can do a few components a day if need be and I have a fancy for using the new ultrabright leds too.

I *could* build a microflash using a Model T vibrating coil, a 12v battery, a couple of ignition coils etc but the voltages are rather high and some of the parts are very expensive and hard to get - for example the 25kv diodes and the 50kv diodes etc, not to mention the mylar capacitors. Half of that stuff I'd have to source from old Soviet nuclear facilities. At least that's where most parts seem to be coming from these days. And it would be bulky.

I want to build something smaller that runs off low voltages that I'm not going to have to be so careful with. I live in Columbia SC where it is very humid and thus the likelihood of a HV accident is higher. While I like freshly fried bacon, I don't want to become freshly fried bacon.

 

[Tim Wescott]

On Sat, 03 Aug 2013 17:58:07 -0700, Zephod Beeblebrox wrote:

The reason I want to keep ICs out of it is because ICs seem to induce
delays. I encountered a very interesting audio trigger the other day.

All circuitry induces delays. Free space induces delay (see: the theory
of relativity). You can't get around that.

But the intelligent circuit designer can deal with all that.

Now, before anything else, let me describe the setup...

A gun fires making a noise (doesn't matter if it's an airgun or a
cartridge)

A microphone picks up the noise and that goes to a black box. The black
box controls the time delay between the noise and closing a switch.

And that switch isn't going to induce delays?

The switch normally closed is across the contacts of a flash unit,
causing the flash.

The interesting unit had a sampling interval of 100,000ms which means
that if a gun fired, it would still activate because the sound is longer
than 100,000ms but it would make a huge difference in the location of a
pellet.

Even if you mean microsecond there's something wrong with your units
here. 100,000 microseconds is 0.1 second, and that's a loooooooong time
compared to what you're talking about.

If it was travelling at 500fps in 100,000ms it could travel 5
feet. That's not very helpful. I gather all ICs tend to have sampling
intervals.

No, 500 fps in 100,000 microseconds is 50 feet.

I'm not sure what you mean by "sampling interval" -- I think you're
inventing terminology here without having a clue about what's already
used. So I can't say whether you're correct or not given whatever
definition you've attached to the phrase.

But I very much doubt that "all ICs tend to have sampling intervals" is
an accurate statement.

What I want to do is to have a flash unit that uses solely LEDs. Hell, I
can rebuild my LED box, I don't care - time is irrelevant as I can do a
few components a day if need be and I have a fancy for using the new
ultrabright leds too.

I *could* build a microflash using a Model T vibrating coil, a 12v
battery, a couple of ignition coils etc but the voltages are rather high
and some of the parts are very expensive and hard to get - for example
the 25kv diodes and the 50kv diodes etc, not to mention the mylar
capacitors. Half of that stuff I'd have to source from old Soviet
nuclear facilities. At least that's where most parts seem to be coming
from these days. And it would be bulky.

I want to build something smaller that runs off low voltages that I'm
not going to have to be so careful with. I live in Columbia SC where it
is very humid and thus the likelihood of a HV accident is higher. While
I like freshly fried bacon, I don't want to become freshly fried bacon.

So you want to replace scary-high voltages with scary-high currents?

I think you need to learn some circuit design. You're clearly very much
in the dark about what can and cannot be done with integrated circuits,
and how fast things can be made to happen if you use the right ones.

What are you proposing to use instead of ICs that'll be faster? Vacuum
tubes?

--

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

 

[John Larkin]

On Sat, 3 Aug 2013 19:06:24 -0700 (PDT), George Herold <gherold@teachspin.com>
wrote:

On Saturday, August 3, 2013 9:29:15 PM UTC-4, John Larkin wrote:
On Sat, 3 Aug 2013 17:58:07 -0700 (PDT), Zephod Beeblebrox

sagephotoworld@gmail.com> wrote:



The reason I want to keep ICs out of it is because ICs seem to induce delays. I encountered a very interesting audio trigger the other day.



Now, before anything else, let me describe the setup...



A gun fires making a noise (doesn't matter if it's an airgun or a cartridge)



A microphone picks up the noise and that goes to a black box. The black box controls the time delay between the noise and closing a switch.



The switch normally closed is across the contacts of a flash unit, causing the flash.



The interesting unit had a sampling interval of 100,000ms which means that if a gun fired, it would still activate because the sound is longer than 100,000ms but it would make a huge difference in the location of a pellet. If it was travelling at 500fps in 100,000ms it could travel 5 feet. That's not very helpful. I gather all ICs tend to have sampling intervals.



100,000 ms is 100 seconds. The projectile will be miles away (or on the ground)

by then.



Linear ICs don't have sampling intervals, and can process signals in

nanoseconds.





What I want to do is to have a flash unit that uses solely LEDs. Hell, I can rebuild my LED box, I don't care - time is irrelevant as I can do a few components a day if need be and I have a fancy for using the new ultrabright leds too.



I *could* build a microflash using a Model T vibrating coil, a 12v battery, a couple of ignition coils etc but the voltages are rather high and some of the parts are very expensive and hard to get - for example the 25kv diodes and the 50kv diodes etc, not to mention the mylar capacitors. Half of that stuff I'd have to source from old Soviet nuclear facilities. At least that's where most parts seem to be coming from these days. And it would be bulky.



I want to build something smaller that runs off low voltages that I'm not going to have to be so careful with. I live in Columbia SC where it is very humid and thus the likelihood of a HV accident is higher. While I like freshly fried bacon, I don't want to become freshly fried bacon.



I don't think it's possible to make a usable microsecond flash from LEDs. The

numbers just don't work.

I think that's right. You can make a fast pulse for checking photodiodes or something, but usable light for photography will be harder. We made a LED flash lamp to replace an old Xenon one. This was to measure the rotation rate of a spinning ball. (I was only tangentially involved so I'm half guessing at the numbers.) We used a 100mW led and turned it on for about one milli second.
(Rotation rates ~1 -20 times per second.)

That had a decent duty cycle, and the effective exposure was long-term. A
single, 1-microsecond shot is a tougher problem.


--

John Larkin Highland Technology Inc
www.highlandtechnology.com jlarkin at highlandtechnology dot com

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators

 

[John Larkin]

On Sat, 3 Aug 2013 19:12:10 -0700 (PDT), Zephod Beeblebrox
<sagephotoworld@gmail.com> wrote:

I'd like to have a crack at it. I know my LED panel that I built using 172 LEDs is pretty bright - brighter than most 100w light bulbs.

My flash produces a GN of 58 (metric) which I cut down to 1/128th in order to reduce the time that it's on. Thus it's actual GN is 0.45. Not sure how to convert GN to lumens though. We are talking about tiny amounts of light.

I don't think it's possible to make a usable microsecond flash from LEDs. The

numbers just don't work.

Sure - it might not work out but let's have a bash.

Well, I'm an engineer and I live by numbers. This looks like three, four orders
of magnitude from working. Against numbers like that, it's not worth trying.

Xenon and spark gap flashes dump several joules in under a microsecond,
megawatts of peak power. 100 watts isn't in the game.




--

John Larkin Highland Technology Inc
www.highlandtechnology.com jlarkin at highlandtechnology dot com

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators

 

[George Herold]

On Saturday, August 3, 2013 9:29:15 PM UTC-4, John Larkin wrote:

On Sat, 3 Aug 2013 17:58:07 -0700 (PDT), Zephod Beeblebrox

sagephotoworld@gmail.com> wrote:



The reason I want to keep ICs out of it is because ICs seem to induce delays. I encountered a very interesting audio trigger the other day.



Now, before anything else, let me describe the setup...



A gun fires making a noise (doesn't matter if it's an airgun or a cartridge)



A microphone picks up the noise and that goes to a black box. The black box controls the time delay between the noise and closing a switch.



The switch normally closed is across the contacts of a flash unit, causing the flash.



The interesting unit had a sampling interval of 100,000ms which means that if a gun fired, it would still activate because the sound is longer than 100,000ms but it would make a huge difference in the location of a pellet. If it was travelling at 500fps in 100,000ms it could travel 5 feet. That's not very helpful. I gather all ICs tend to have sampling intervals.



100,000 ms is 100 seconds. The projectile will be miles away (or on the ground)

by then.



Linear ICs don't have sampling intervals, and can process signals in

nanoseconds.





What I want to do is to have a flash unit that uses solely LEDs. Hell, I can rebuild my LED box, I don't care - time is irrelevant as I can do a few components a day if need be and I have a fancy for using the new ultrabright leds too.



I *could* build a microflash using a Model T vibrating coil, a 12v battery, a couple of ignition coils etc but the voltages are rather high and some of the parts are very expensive and hard to get - for example the 25kv diodes and the 50kv diodes etc, not to mention the mylar capacitors. Half of that stuff I'd have to source from old Soviet nuclear facilities. At least that's where most parts seem to be coming from these days. And it would be bulky.



I want to build something smaller that runs off low voltages that I'm not going to have to be so careful with. I live in Columbia SC where it is very humid and thus the likelihood of a HV accident is higher. While I like freshly fried bacon, I don't want to become freshly fried bacon.



I don't think it's possible to make a usable microsecond flash from LEDs. The

numbers just don't work.

I think that's right. You can make a fast pulse for checking photodiodes or something, but usable light for photography will be harder. We made a LED flash lamp to replace an old Xenon one. This was to measure the rotation rate of a spinning ball. (I was only tangentially involved so I'm half guessing at the numbers.) We used a 100mW led and turned it on for about one milli second.
(Rotation rates ~1 -20 times per second.)

George H.

--



John Larkin Highland Technology Inc

www.highlandtechnology.com jlarkin at highlandtechnology dot com



Precision electronic instrumentation

Picosecond-resolution Digital Delay and Pulse generators

Custom timing and laser controllers

Photonics and fiberoptic TTL data links

VME analog, thermocouple, LVDT, synchro, tachometer

Multichannel arbitrary waveform generators

 

[Zephod Beeblebrox]

I'd like to have a crack at it. I know my LED panel that I built using 172 LEDs is pretty bright - brighter than most 100w light bulbs.

My flash produces a GN of 58 (metric) which I cut down to 1/128th in order to reduce the time that it's on. Thus it's actual GN is 0.45. Not sure how to convert GN to lumens though. We are talking about tiny amounts of light.

I don't think it's possible to make a usable microsecond flash from LEDs. The

numbers just don't work.

Sure - it might not work out but let's have a bash.

 

[Zephod Beeblebrox]

OK. Since I'm not after desperately high yields from a flash unit, thinking laterally, if I took a standard flash unit design such as this.

http://www.increa.com/reverse/dc/dcschem.gif

<img>http://www.increa.com/reverse/dc/dcschem.gif</img>

Then changed the size or size and type of the capacitor to something of less capacity that should theoretically reduce the flash duration.

I gather xenon maxes out at 1/110,000th of a second.

Does the size of the flash tube make a difference? I could perhaps put a couple of tubes wired in parallel.

 

[tuinkabouter]

On 8/4/2013 2:58 AM, Zephod Beeblebrox wrote:

The reason I want to keep ICs out of it is because ICs seem to induce delays. I encountered a very interesting audio trigger the other day.

Now, before anything else, let me describe the setup...

A gun fires making a noise (doesn't matter if it's an airgun or a cartridge)

A microphone picks up the noise and that goes to a black box. The black box controls the time delay between the noise and closing a switch.

The switch normally closed is across the contacts of a flash unit, causing the flash.

The interesting unit had a sampling interval of 100,000ms which means that if a gun fired, it would still activate because the sound is longer than 100,000ms but it would make a huge difference in the location of a pellet. If it was travelling at 500fps in 100,000ms it could travel 5 feet. That's not very helpful. I gather all ICs tend to have sampling intervals.

What I want to do is to have a flash unit that uses solely LEDs. Hell, I can rebuild my LED box, I don't care - time is irrelevant as I can do a few components a day if need be and I have a fancy for using the new ultrabright leds too.

I *could* build a microflash using a Model T vibrating coil, a 12v battery, a couple of ignition coils etc but the voltages are rather high and some of the parts are very expensive and hard to get - for example the 25kv diodes and the 50kv diodes etc, not to mention the mylar capacitors. Half of that stuff I'd have to source from old Soviet nuclear facilities. At least that's where most parts seem to be coming from these days. And it would be bulky.

I want to build something smaller that runs off low voltages that I'm not going to have to be so careful with. I live in Columbia SC where it is very humid and thus the likelihood of a HV accident is higher. While I like freshly fried bacon, I don't want to become freshly fried bacon.

Did you take the delay due the traveling speed of sound into account?
If you measure the sound 1 yard away it already have 2.6 ms delay.
One feet introduces 888 microseconds delay!

Electronics delay (in nano seconds) is negligible.

 

[Zephod Beeblebrox]

Sound travels at 1,126 fps

This is why I have a delay unit attached to my sound sensor.

All I need now is a faster flash. The fastest I have goes to 1/38,000th of a second.

 

[Tim Wescott]

On Sat, 03 Aug 2013 21:08:37 -0700, Zephod Beeblebrox wrote:

OK. Since I'm not after desperately high yields from a flash unit,
thinking laterally, if I took a standard flash unit design such as this.

http://www.increa.com/reverse/dc/dcschem.gif

img&gt;http://www.increa.com/reverse/dc/dcschem.gif&lt;/img

Then changed the size or size and type of the capacitor to something of
less capacity that should theoretically reduce the flash duration.

I gather xenon maxes out at 1/110,000th of a second.

Does the size of the flash tube make a difference? I could perhaps put a
couple of tubes wired in parallel.

Cheap flash units just turn the Xenon tube on and let the decaying
voltage in the capacitor dictate when it turns off.

Good flash units (probably like your 1/38000 second one) actively switch
off the voltage to the flash tube. The circuits that I've seen use an
IGBT or MOSFET, and a series coil (they're more designed to control
current than to strictly turn the thing off as fast as possible).

You may be able to get faster response by taking out the coil and making
a circuit that's specifically designed for one shot, that just switches
the tube off promptly.

You'll probably want to use a MOSFET for speed (IGBT's are slow), but
finding a MOSFET that can take the punishment will be a challenge. And
you will, of course, need a gate driver, the fastest of which are
integrated circuits.

--

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

 

[John Fields]

On Sun, 4 Aug 2013 06:11:25 -0700 (PDT), Zephod Beeblebrox
&lt;sagephotoworld@gmail.com&gt; wrote:

Sound travels at 1,126 fps

---
In air, yes, but only under certain conditions.

Ambient temperature, atmospheric pressure, and humidity all need to be
defined in order to determine sound velocity under those constraints.

--
JF

 

[Jasen Betts]

On 2013-08-03, sagephotoworld@gmail.com &lt;sagephotoworld@gmail.com&gt; wrote:

OK.

I want to build a lower-power LED microflash. You're probably familiar with Edgerton's microflash that produced a 1/1,000,000 second flash by arcing 50,000v across an air gap that had been ionized by 25,000v and is commonly used to photograph bullets in flight?

The fastest flash available from xenon tubes is 1/110,000 second and a flash such as a Canon 580EX2 runs at a maximum of 1/38.000th second.

What I want to do is to build an LED version of a flash since LEDs though lower in power, will pulse very quickly. I'm not planning to build the ultimate LED microflash. I have an LED light panel that I built with about 170 white LEDs in parallel. That runs happily off 2 x C cells.

If I could pulse light from those LEDs at 1/50,000th second I would be happy. If I could get down to 1/100,000 second I would be ecstatic.

I realise it probably wouldn't be very bright and that I might well have to crank my ISO upto 1600 but I'd like to have a go and play with it; maybe try it with different LEDs and different voltages later.

it will be extremely non-bright. Take your LED panel and do some
normal photos (or just take light readings) using only it for
illumination (in a darkened room) Look at the minimum shutter speed
you can use

That should give you some idea of how bad this will be.



--
⚂⚃ 100% natural

--- news://freenews.netfront.net/ - complaints: news@netfront.net ---

 

[Phil Allison]

"John Fields"
Zephod Beeblebrox

Sound travels at 1,126 fps

---
In air, yes, but only under certain conditions.

Ambient temperature,

** Proportional to the sq.rt of the air temp in K.

atmospheric pressure,

** Practically not at all.

humidity

** Somewhat, a thick fog makes a difference.




.... Phil

 

[Zephod Beeblebrox]

Yes... humidity, altitude and temperature all play a part

On Sunday, August 4, 2013 4:32:14 PM UTC-4, John Fields wrote:

On Sun, 4 Aug 2013 06:11:25 -0700 (PDT), Zephod Beeblebrox

sagephotoworld@gmail.com&gt; wrote:



Sound travels at 1,126 fps



---

In air, yes, but only under certain conditions.



Ambient temperature, atmospheric pressure, and humidity all need to be

defined in order to determine sound velocity under those constraints.



--

JF

 

[Zephod Beeblebrox]

Well, if we can time accurately then we can simply switch the current on/off from the capacitor. It's not a terribly high voltage. Usually somewhere around 250v - 300v. My thought is simply to adjust the timing that the cap is emitting. Somewhere upwards of 1/111,000th of a second.

On Sunday, August 4, 2013 12:47:14 PM UTC-4, Tim Wescott wrote:

On Sat, 03 Aug 2013 21:08:37 -0700, Zephod Beeblebrox wrote:



OK. Since I'm not after desperately high yields from a flash unit,

thinking laterally, if I took a standard flash unit design such as this..



http://www.increa.com/reverse/dc/dcschem.gif



img&gt;http://www.increa.com/reverse/dc/dcschem.gif&lt;/img



Then changed the size or size and type of the capacitor to something of

less capacity that should theoretically reduce the flash duration.



I gather xenon maxes out at 1/110,000th of a second.



Does the size of the flash tube make a difference? I could perhaps put a

couple of tubes wired in parallel.



Cheap flash units just turn the Xenon tube on and let the decaying

voltage in the capacitor dictate when it turns off.



Good flash units (probably like your 1/38000 second one) actively switch

off the voltage to the flash tube. The circuits that I've seen use an

IGBT or MOSFET, and a series coil (they're more designed to control

current than to strictly turn the thing off as fast as possible).



You may be able to get faster response by taking out the coil and making

a circuit that's specifically designed for one shot, that just switches

the tube off promptly.



You'll probably want to use a MOSFET for speed (IGBT's are slow), but

finding a MOSFET that can take the punishment will be a challenge. And

you will, of course, need a gate driver, the fastest of which are

integrated circuits.



--



Tim Wescott

Wescott Design Services

http://www.wescottdesign.com

 

[Phil Allison]

"Zephod Beeblebrox"

Sound travels at 1,126 fps

This is why I have a delay unit attached to my sound sensor.

All I need now is a faster flash. The fastest I have goes to 1/38,000th of
a second.

** You have only the slimmest chance of success and none at all with your
present approach.

I decided to Google the famous Edgerton "Microflash" device from the 1960s.

http://quickblink.com/2011/07/egg-microflash/

It produced an arc in air with few Joules being dumped in a microsecond or
two - with *just* enough light output for a film camera (with decent lens
and 200 ASA colour film) ) to catch a pic.

Modern digital cameras are far more light sensitive and do not suffer from "
repricosity failure" like film does with very short exposure . So - IF you
can concentrate the light from your 170 LEDs on a small area, get the slug
in that area and snap it - you may get something usable.

Forget sound triggering and go for a physical switch - like a pair of
close spaced metal foils the slug simply shorts together as it passes
through them.

This event can trigger the gate of an SCR to dump a charged capacitor into
the LED array with a very small delay and a current pulse duration in the
order of 10 microseconds.

The camera lens needs to be close too - like 50 mm away.

Guessing that a regular white 5mm LED can stand a single 10uS pulse of 200mA
once in a while - your peak current is 34 amps. So the load resistance is
gonna be in the order of 5/34 = 0.15 ohms.

For a time constant of 10uS, that equates to using a cap of 6.8uF - it has
to a metallised film type for low ESR .

A low voltage, 10 amp SCR will do fine, sensitive gate type preferred.

Maybe you can make a hole on the middle of your LED array that fits the
camera lens and aim each LED at the same spot 50 or 100 mm in front of it.

You will need a dark room, luck and lotsa patience.

But SFA cash investment - so why not try it ?


..... Phil

 

[Phil Hobbs]

On 8/4/2013 11:34 PM, Zephod Beeblebrox wrote:

On Sunday, August 4, 2013 4:32:14 PM UTC-4, John Fields wrote:
On Sun, 4 Aug 2013 06:11:25 -0700 (PDT), Zephod Beeblebrox

sagephotoworld@gmail.com&gt; wrote:



Sound travels at 1,126 fps



---

In air, yes, but only under certain conditions.



Ambient temperature, atmospheric pressure, and humidity all need to be

defined in order to determine sound velocity under those constraints.

Yes... humidity, altitude and temperature all play a part

(Top posting corrected)

Just temperature and humidity. Barometric pressure doesn't change the
sound velocity.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510 USA
+1 845 480 2058

hobbs at electrooptical dot net
http://electrooptical.net

 

[Phil Allison]

"Phil Hobbs"

Sound travels at 1,126 fps

---

In air, yes, but only under certain conditions.



Ambient temperature, atmospheric pressure, and humidity all need to be

defined in order to determine sound velocity under those constraints.

Yes... humidity, altitude and temperature all play a part


(Top posting corrected)

Just temperature and humidity. Barometric pressure doesn't change the
sound velocity.

Cheers

Phil Hobbs

** Tell 'em why Phil - tell 'em why .....

The physics behind that fact is cute.




.... Phil

 

[Phil Hobbs]

On 8/5/2013 9:56 AM, Phil Allison wrote:

"Phil Hobbs"

Sound travels at 1,126 fps

---

In air, yes, but only under certain conditions.



Ambient temperature, atmospheric pressure, and humidity all need to be

defined in order to determine sound velocity under those constraints.

Yes... humidity, altitude and temperature all play a part


(Top posting corrected)

Just temperature and humidity. Barometric pressure doesn't change the
sound velocity.

Cheers

Phil Hobbs


** Tell 'em why Phil - tell 'em why .....

The physics behind that fact is cute.




... Phil

Sound is transmitted via molecular collisions. Since there are many
collisions per acoustic cycle, and each collision takes negligible time,
it's only the mean molecular speed that matters.

From classical equipartition, the energy per degree of freedom is kT/2,
i.e. 3/2 kT altogether. Since KE = 1/2 mv**2, the mean velocity is

v_bar = sqrt(3kT/m),

where m is the mean molecular mass. The sound velocity is proportional
to v_bar. (H2O is lighter than O2 and N2, so increasing the humidity
makes the speed of sound go up a little.)

For thermodynamic reasons that I'd have to go remind my self of but
which involve internal degrees of freedom, the speed of sound isn't just
1/sqrt(3) times the molecular speed, as you'd expect, but is less by a
factor of about 0.7ish.

But for a gas of non-interacting molecules, it depends only on the gas
composition and the temperature.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510 USA
+1 845 480 2058

hobbs at electrooptical dot net
http://electrooptical.net