Searching \ for 'Air powered *, Thanks & report' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/ios.htm?key=port
Search entire site for: 'Air powered *, Thanks & report'.

Truncated match.
PICList Thread
'Air powered *, Thanks & report'
2000\04\15@163414 by Morgan Olsson

picon face
I have been working on it since I posted the question.

At the moment on the testbench is now an air powered generator delivering 12W 1W running :)

 It was really easy, actually:
I found a 12cm x 12 cm fan, 12V 5W, ball bearing, I selected this one because it felt like it had strong magnet (i turned different fans using the finger)
This was a pretty expensive one, Papst brand.
What the heck...  Really easy to open.  
And very easy to change bearings if ever needed.
Disconnected the driver electronics.
Reconnected the windngs in series, to full bridge schottky rectifier.
(tested with scope they were in phase)
Covered both sides of fan housing, air inlet tangentially, outlet center.
Nozzle 0,7mm diam and 8bar pressure give 12V 1W out at approx original fan rpm.

At bigger nozzle, double rpm, I get close to 4W output, but what about lifetime?
double rpm=4 times the radial force (mis-balance)
that and double speed make 1/8 original lifetime... I guess??
So double rpm is too high, but I can increase 50% or so to get 2W out :)


The fan has 70 000 hr L10 @40¡C when running as fan.  If I keep to fan mode rpm it will have *at least* that life.  

(L10 is the time when 10% of the units have died, contrary to mtbf which is the time where 50% have died)


So, I have reached the goal of both Service life, output power, cost, and size.

Remainding is noise level...  rubber suspension, encapsulate,...

And it consumes pretty lot of air, but the customer have not specified that... yet...

To have a turbine running efficient would requiore very high rpm = high cost, and still maybe not long enough life.

Other possibilities is using a long-life permanetn-lubricated cylinder, to drive a permanent magnet throught a coil.  But that, the big magnet, and the valves are all pretty expensive and make vibration, and I suspect it don«t have enough lifetime.

Membrane something oscillator to oscillate the magnet in r4esonance with a spring?

Air motor running very slow (for lifetime), driving a stepper motor as generator (they work at low rpm).  But how to lubricate, and how to make sure no oil is escaping in exhaust (customers customer requirement)

Piston motors; steam engine, hobby size:  lubrication, lifetime?

Ideas welcome.


 ** Some more project spec **
Air: dry, oil free, filtered, 7bar
Output: 12V min 1W
Size: max about half litre
Weight: not spec but say max 1kg
Lifetime: 10yr, 8hr/day, 5day/week, 46 week/year = 18400 hr.
NO SERVICE: no batt change, no oiling, no part switch...
No oil: not in input air, not in exhaust
Production need: about 200/yr
Air-electric conversion budget: max about $140US/unit



I know it sound strange, but not needing to electrically connect, and not needing to open the tool give valuable properties.

No, I«m not allowed to tell what it is.


Thank you for your input.

Regards
/Morgan

2000\04\15@175535 by Morgan Olsson

picon face
I wrote:

>...delivering 12W 1W running :)

Meant of course 12V 1W...

Tomorrow I will add another nozzle of same size in the straight opposite side.
This wil make approx total 2W, but brobably not increase the noise as much as it would with one big nozzle.  The fan have odd nomber of wings so only one fan blade will be hit with an airblow each time; same noise amplitude, but double frequency (if same rpm).  Butt the ssssss white noise from the nozzzzzles will be doubled of course.

/Morgan

2000\04\15@181652 by paulb

flavicon
face
Morgan Olsson wrote:

> Tomorrow I will add another nozzle of same size in the straight
> opposite side.

 You are using high pressure/ velocity air to drive a low pressure/
velocity fan.  You are probably impinging on an unnecessarily small
point on the blades.

 I would suggest arranging the nozzle inside a venturi assembly,
perhaps two inch intake diameter and ¸" just after the nozzle, then
expanding back to either two inch or the whole fan aperture.  If you can
thus entrain a much larger air volume, it should be more efficient and
may be quieter, albeit larger.

 Material?  PVC piping, heated while rolling in front of a hot air
blower, and "drawn" out like glass.
--
 Cheers,
       Paul B.

2000\04\15@190928 by Yves Chamberland

picon face
Good innovative pneumatic Project !
 Imagine.Next you could add an aeolian (Kites) system regulated with an
inlet (min/max)pressure(input) valve wich would have only to pick the
"kinetic air" from the wind to a "potential lung baloons tank"
(output). Then to calibrate it to (8)bars with a Feedback loop
between two input/output variable AREA pneumatic valves.
And thus offerring an illimitated renewable Power system.
                                        <`0)   Best Whistling !.
{Original Message removed}

2000\04\15@221403 by David VanHorn

flavicon
face
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

You might do better by trimming the blades down to minimum size (both ways)
and cutting out the drag of the unused blade surface. This sounds like you
need a relatively high airspeed for it to run, but not necessarily a lot of
volume.

Good job though, I didn't expect you'd get that much power out of it.


- --
Are you an ISP?  Tired of spam?
http://www.spamwhack.com  A pre-emptive strike against spam!

Where's Dave? http://www.findu.com/cgi-bin/find.cgi?kc6ete-9

-----BEGIN PGP SIGNATURE-----
Version: PGPfreeware 6.5.2 for non-commercial use <http://www.pgp.com>

iQA/AwUBOPk7iIFlGDz1l6VWEQLlYQCgoI5hvB3aQE62OGKDjWUQZbnzydQAoIXZ
cYxBgkuY5RujxFLvCsAb9RYK
=pXM5
-----END PGP SIGNATURE-----

2000\04\16@052109 by Russell McMahon

picon face
Morgan,

Important conclusion at end so I'll copy it up here -

*************************************************************
My point is, using the low pressure fan you are using now,  if I'm correct,
or even vaguely to correct, then you can get very very very low airflows -
maybe 10 or 20 cc per second IF you convert the air correctly to match the
impeller characteristics.
Look at velocity, volume and head of original fan.
The head will be in the 10 kPa or 1 psi region (if that)
The velocity will be modest  The volume will be high (litres/second)
*************************************************************

Interested? - read on.

Sounds like lots of fun and good to see that you have something with an
output at about what you need.
However, the efficiency sounds very very very bad if it needs "pretty lots
of air".

Considering someone's suggestion of an air entrainment venturi to "amplify"
the airflow may well be very worth considering. All this does is use the low
volume high pressure source to generate a high volume low pressure flow.

Very easy, fully static, covered in text books and on the web.

My air requirement caluclations for a 100% efficient unit go like this.

Power = watts = NM/sec.
If a piston of area A is displaced distance D in 1 second it seeeps out
volume V
Assume constant Pressure P throughout
Force = P x A
Work = Force x Distance = P x A x D
But A x D = V
so Work = V x P

Volume in metres^3
P in Pascal (= N/m^2)

When the pressure changes across a linear sweep the mean pressure for
calculation is Po x ln (Pmax/Po)

Here Pmax = 7 bar
Po = 1 bar
ln (7/1) is about 2 (close enough for us)

Say Patmospheric = 100,000,Pascal
So Pmean is 2bar = 200,000 Pascal

Power = V x P x ln(Pm/Po)
or V = Power/ (Po x  ln(Pm/Po))

For 1 watt
V = volume of air at atmospheric per second
= 1/(100,000 x 2) = 1/200,000th m^3/sec
= 1/200 litre/second
= 1/30 l per minute
5cc/second
!

Scale this up by a few times (even 10 times) for various loss factors and
you still have minimal air-flow

Check my figures by all means - I have but I have been known to miss out the
odd multiple of 1000 from calculations like this :-)

Does this sound sensible?
Lets plug rough figures back in -
5cc x 2 bar = 5/1000000 x 200000 = 1 watt
Doesn't prove much but ...

1 watt isn't very much power really!

My point is, using the low pressure fan you are using now, if I'm correct,
or even vaguely to correct, then you can get very very very low airflows -
maybe 10 or 20 cc per second IF you convert the air correctly to match the
impeller characteristics.
Look at velocity, volume and head of original fan.
The head will be in the 10 kPa or 1 psi region (if that)
The velocity will be modest  The volume will be high (litres/second)









     Russell McMahon
_____________________________

- http://www.easttimor.com
      Updated regularly:
      100,000 refugees STILL in  West Timor face starvation!

- http://www.sudan.com
   And you think Kosovo and Chechnya are bad!

What can one man* do?
Help the hungry at no cost to yourself!
at  http://www.thehungersite.com/

(* - or woman, child or internet enabled intelligent entity :-))


{Original Message removed}

2000\04\16@075736 by Morgan Olsson

picon face
David VanHorn wrote:

>You might do better by trimming the blades down to minimum size (both ways)

Been thinking of that.
But not decrease diameter, as stated previously we must maximize torque, thus diameter.

By decreasing the axial length of the fan I can decrease the axial length of the housing, thus decrease the friction air-housing

>and cutting out the drag of the unused blade surface.

As long as the air is following the rotation of the fan, it is not the fan blades that is the braking; it is the friction between the movig air and the housig.

I have a plate closeing the fan inlet, except for a center hole for air outlet, and one plate between the fan and the motor suspension, so all surfaces to the rotating air are flat.

>This sounds like you
>need a relatively high airspeed for it to run,

The airspeed of a 7 bar air nozzle is well above speed of sound.  No normal fan can come close to that speed...

>  but not necessarily a lot of
>volume.

The higher peripheral speed (blade tip) I can reach, the better efficiency, and the less peripheral thrust is needed.

>Good job though, I didn't expect you'd get that much power out of it.

Neither did I...
First I had 200mA schottky rectifiers, and 1,3W 12V zener as load.

The Zener got overheated, and also melted the plastic part of test probe i had connected on it :/

I changed the schottkys to 16 A types (had no 3 A) to decrease the drop over each rectifier to only just above 0,3 V.  Meaning slightly lower rpm, longer life, and more electric output for same air input.

Now I have two 5,6V 1W big metal body old style zeners plus one 1N4001, all in series, as load.



Thinking of instead of schottkys use a MOSFET bridge and control it, making a drop of below 100mV total.
Maybe use the already present magnetic sensor to control it!
The protection diodes will power it up during start.
Ideas on how to make correct intermediate time (where no mosfet is conducting) to not power the coils backwards??
PLL? (in the PIC of corse)
Problem is the intermediate time is dependant on output voltage, speed, and thrust (induction make that)
Probably let the magnetic sensor control the two bottom fets without intermediate time, and have OPAMP to turn on top mosfets whan respective input is higher than output.
Sounds reasonable.

Lower loss mean i can increase power without overspeeding :)
More efficient electronics increase system life and reduce air consumption, so it seem good idea to spend time and money in the electronics.


Next step is to find a higher rpm fan for increased efficiency.
if very high rpm i can go with smaller fan too to save space.
I see Papst have a 6W 5000 rpm 80x80m mm 70 000 h L10 fan :)


/Morgan

2000\04\16@102019 by Morgan Olsson

picon face
Russell McMahon wrote:

>Interested? - read on.

of course :)

>Sounds like lots of fun and good to see that you have something with an
>output at about what you need.


Actually more output than the initial goal... :)


>However, the efficiency sounds very very very bad if it needs "pretty lots
>of air".

I knew It would be very very bad...

Well I have no flow meter.
Do somebody know how to calculate the mass flow from:
nozzle: cylindric hole, 0,7 mm diam, 3 mm long
(drilled in brass, not very low friction, btw...)
Supply: air, 8 bar presently (will use 6-7 in product)
Outside: 1 bar (atmospheric)


>Considering someone's suggestion of an air entrainment venturi to "amplify"
>the airflow may well be very worth considering.


Tested, no change in output, but more silent


>All this does is use the low
>volume high pressure source to generate a high volume low pressure flow.


It seems like the impulse power is kept, but energy is lost in turbulence.


>Very easy, fully static, covered in text books and on the web.


Do you happen to have a URL handy?


>My air requirement caluclations for a 100% efficient unit go like this.

Thanks, I«ll dig into it later.
(now I have fun testing, and this week I must do tax paper work (AARGH))

I did not se you calcualting the losses?
My present theory is that the impulse (thrust) is the same with or without the venturi, and the loss is in form of turbulence (making little heat, at the same time more than consumed by the expanding and therefor cooling gas)

I made this experiment:
Nozzle blowing free in room
I hold different pipes, different diameter, conic etc so the nozzle blos inside or from outside through the venturi.

If the ventury should increase the thrust (for making more torque in the fan), then I would have felt a force from my hand held pipe against the nozzle flow.  I never did that; the good pipe gave no force, bad ones made force in the wrong direction.

/Morgan

2000\04\16@114224 by David VanHorn

flavicon
face
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1


>The airspeed of a 7 bar air nozzle is well above speed of sound.  No
normal fan can come close to that speed...

What you're doing here is a modified savonius rotor, which should be able
to get to .6-.8 of the airstream speed.

>I changed the schottkys to 16 A types (had no 3 A) to decrease the drop
over each rectifier to only just above 0,3 V.  Meaning slightly lower rpm,
longer life, and more electric output for same air input.

You might add a supercap to help catch whatever energy is available when
available.


- --
Are you an ISP?  Tired of spam?
http://www.spamwhack.com  A pre-emptive strike against spam!

Where's Dave? http://www.findu.com/cgi-bin/find.cgi?kc6ete-9

-----BEGIN PGP SIGNATURE-----
Version: PGPfreeware 6.5.2 for non-commercial use <http://www.pgp.com>

iQA/AwUBOPn5UYFlGDz1l6VWEQIBWwCg393arcvmusXpbImPZfwDjohAZhMAoOT+
fOUPRE1zQf5tDLbo506y/IKc
=nuiJ
-----END PGP SIGNATURE-----

2000\04\16@123302 by Morgan Olsson

picon face
David VanHorn wrote:

> >The airspeed of a 7 bar air nozzle is well above speed of sound.  No
>normal fan can come close to that speed...
>
>What you're doing here is a modified savonius rotor, which should be able
>to get to .6-.8 of the airstream speed.


Theoretically, yes, byt if the blate tips reach MACH 2 I would not want to be anywhere near it...

There is also the bearing service life...

> >I changed the schottkys to 16 A types (had no 3 A) to decrease the drop
>over each rectifier to only just above 0,3 V.  Meaning slightly lower rpm,
>longer life, and more electric output for same air input.
>
>You might add a supercap to help catch whatever energy is available when
>available.

Yep.  Or large electrolyte.  (12V system) Need to save to manouver two 2W solenouds, but in very short pulses.

Regards
/Morgan

2000\04\16@173122 by Grif\ w. keith griffith

flavicon
face
<x-flowed>>
>*************************************************************
>My point is, using the low pressure fan you are using now,  if I'm correct,
>or even vaguely to correct, then you can get very very very low airflows -


Methinks most of us were over thinking the fan at hi pressure.  I'd bet
he's only using a very small area of the fan blade, probably making it look
more like a hi pressure turbine with a flat blade.  I'd expect that any
figures derived from the designed operation characteristics of the fan mean
very little with the airflow descirbed.

What ya all think?


'Grif'   N7IVS

</x-flowed>

2000\04\16@175651 by Morgan Olsson

picon face
Russell McMahon wrote:

snip:

{Quote hidden}

Thanks for the formula and nice example :)

Practical measurement:
I put the nozzle which make the fan deliver 1W at 7 bar, in a 5 litre water tank, kept the opening of the tank down below water, opened the 7 bar air supply, and measured 9,8 s for the air to fill the tank.

That is about a half litre per second atmospheric pressure air through the nozzle.

According to Russels calculations above that make I have an efficiency of roughly 1%.

Well the tangential speed of the fan blades at this present 2700 rpm is only 16 m/s, so no turbine like efficiency is likely to be seen when it is hit by a 7 bar air beam...

Let«ts see total efficiency... the air compressor IIRC is a 3kW electric input, 600L/minute intake air, so it compress 10 Litres atmosphere air per second, 20times more than my generator need at 1W output.  So the compressor consumes 3000/20=150W in order to supply 1W to my electronics in the end.

Well, the remaining 149 watts are generated as heat and the workshop need heating anyway... most of the year.

Bedtime
/Morgan

2000\04\17@115408 by Don Hyde

flavicon
face
I'm no aerodynamicist, but I've been trying to absorb a little while
building a kitplane.

>From what I've been reading on engine cooling, what you need to do is to
convert the low-volume high-velocity airflow from your nozzle to high-volume
low-velocity for the fan.  The ideal shape for this is something similar to
an exponential horn that would feed that expanding air (to get from 8 bar to
1 it's going to get 8 times as big) to the whole fan.  It would also help to
round the upstream end of the nozzle hole.  That should improve both
efficiency and noise.

> {Original Message removed}

More... (looser matching)
- Last day of these posts
- In 2000 , 2001 only
- Today
- New search...