compressor efficiency
splin
bar air compressor - either with a commercial product or a home brew system?
The application I'm, thinking of is to replace the supercharger/intercooler
of a turbo charged diesel engine with a more efficient, near isothermal
compressor. The existing turbo's output would be put to good use driving an
alternator. The engine would be static so size would not be an issue and
cost would not be a critical factor (within reason).
I understand that the centrifugal compressor of a typical engine's
turbo-supercharger has a mechanical efficiency of around 70% to 75%. I would
have thought this could be significantly bettered by say a displacement
type. Also compressing isothermally should gain a further 10% over the
adiabatic compression/intercooling of the standard setup.
So does anyone know what might be achievable and what sort of compressor
would be suitable? I was thinking that perhaps a much larger engine than the
target engine might be modified into something suitable (running very slowly
so as to operate near isothermally, but I have no idea how large the
friction/pumping/leakage losses would amount to.
Perhaps some sort of Heath-Robinson-ish device - a very, very large,
multi-stage, very slow moving bellows perhaps? ;)
TIA, Splin
Bob Zawarski
somwhere>
"splin" <splin...@ntlworld.com> wrote in message
news:sEOd9.6666$0r3.1...@newsfep1-win.server.ntli.net...
dlzc@aol.com (formerly)
> The application I'm, thinking of is to replace the
supercharger/intercooler
> of a turbo charged diesel engine with a more efficient, near isothermal
> compressor.
Diesel cycle: a cycle in which the mixture is ignited by the heat generated
during the compression stroke.
isothermal compression: compression that leaves the compressed gas at the
temperature it was originally.
Do you see a conflict here? Are you going to leave your igniters running
all the time?
David A. Smith
splin
"dl...@aol.com (formerly)" <dl...@cox.net> wrote in message
news:AwSd9.70160$Fb.28...@news1.west.cox.net...
an isothermal supercharger to replace the turbocharger (which operates
adiabatically). This is simply to provide the required boost in pressure of
the intake air charge. The operating cycle in the engine is exactly the same
as before - the standard diesel cycle.
Splin
Julian Stafford
splin wrote in message ...
>Thanks David. However I think you misunderstood me; I'm talking about using
>an isothermal supercharger to replace the turbocharger (which operates
>adiabatically). This is simply to provide the required boost in pressure of
>the intake air charge. The operating cycle in the engine is exactly the
same
>as before - the standard diesel cycle.
I understand you but could you explain your reasoning for this project?
Would not a normal turbo and good intercooler amount to much the same thing?
Julian.
Steve
> What is the highest efficiency which could be practically achieved for a 1
> bar air compressor - either with a commercial product or a home brew system?
Well... there's a very easy to define upper limit. NO air compression
device can ever exceed 50% efficiency once you cool the output air back
down to the input temperature. A minimum of 1/2 the input energy goes
into raising the temperature of the compressed air.
The only way to approximate an "isothermal air compressor" (since there
is no such thing in the real world) is to compress adiabatically and
then cool the compressed air back down to inlet temperature. Which is
EXACTLY what your turbo/intercooler already does (or approximates).
You're chasing fairies and UFOs, here.
Racer X
> bar air compressor - either with a commercial product or a home brew system?
>
efficiency of around 80% over a narrow range of flow and compression.
A Lysholm twin screw type compressor (used in some mechanical
superchargers) can get up to around 85% efficiency or so over a
similarly narrow range of flow and pressure increase. I seem to
remember that positive displacement type compressors (like a roots
supercharger) are much less efficient, perhaps in the 50%
neighborhood.
I doubt any reciprocating compressor would be worth looking at. They
tend to be fairly inefficient, around 30-50% is a number that sticks
in my mind, but I could be wrong on the exact details.
Also, there's some rotary superchargers. I think Wankel actually
designed one of these superchargers back in the 1950's. I doubt you
could even find one today. I have no idea how efficient they were. The
design is similar to a wankel rotary engine, but there's an output
port where the leading spark plug would be and another intake port
where the trailing spark plug would be.
> The application I'm, thinking of is to replace the supercharger/intercooler
> of a turbo charged diesel engine with a more efficient, near isothermal
> compressor. The existing turbo's output would be put to good use driving an
> alternator. The engine would be static so size would not be an issue and
> cost would not be a critical factor (within reason).
>
OK, so it's not an automotive application, obviously. Is the operating
speed constant? Is the load constant?
Since it's a diesel, intake flow is nearly linear with crankshaft
speed. If it's a fixed speed application, you should be able to get a
centrifugal compressor or Lysholm twin screw that's designed for peak
efficiency at the required flow rate and pressure. With that, you
should be able to hit an efficiency in the 80-85% range.
If the load is variable, with wide variations, I'd go with a belt
driven supercharger. There's not much loss in a good belt drive system
and good bearings. With a variable load, there will be variations in
the volume of exhaust flow (more volume under heavier load, less
volume under lighter load). That will mean that a turbine in the
exhaust will see variable flows in, and the output speed, power
available at the turbine shaft, and the overall efficiency will vary.
Wide variations in the load would make a turbocharger more difficult
to tune for absolute peak efficiency.
If the load is constant, you might use a turbocharger and design the
turbine for the exact flow available, and the exact load presented by
the compressor. I imagine that in most situations, the turbo is going
to have an advantage from the fact that most of the energy driving its
compressor comes from combustion (and expansion of the charge) after
the exhaust valve opens. Therefore, the turbine presents much less of
a load to the pistons and crankshaft.
> I understand that the centrifugal compressor of a typical engine's
> turbo-supercharger has a mechanical efficiency of around 70% to 75%.
In a normal automotive application, the efficiency peaks around that.
But because of the wide variations in flow and pressure requirements,
actual efficiency can be all over the place in an automotive
application. Some trade off is necessary to deliver adequate
performance over a wider range, so peak efficiency might be reduced in
order to broaden the useful boost curve.
> I would
> have thought this could be significantly bettered by say a displacement
> type.
I don't know. Centrifugal compressors are pretty efficient in
comparison to the alternatives. The only thing I see that's better
would be a Lysholm twin screw type compressor.
Then, the question would be how efficiently can you turn the thing?
The advantage of an exhaust turbine to drive it is that most of the
energy driving the turbine is actually waste that the engine doesn't
capture and turn to power at the crankshaft. So, the load presented to
the crankshaft by a turbocharger is much less than the total load from
the compressor. This generally more than offsets the fact that the
turbine might not be the most efficient at extracting the energy and
turning it into rotation of the turbocharger shaft, and the fact that
the compressor itself isn't necessarily the most efficient available
to produce the required flow and pressure.
Let's say a Lysholm twin screw supercharger might need 9.5 hp to
produce the required boost and a less efficient centrifugal compressor
in a turbocharger might need 12 hp to produce the same boost. Now, the
supercharger might be driven by a belt and pulley system that's 95%
efficient, so it will take 10 hp from the crank to drive it. An
exhaust driven turbine in the turbocharger might also be around 75%
efficient, so it will take 16hp from the exhaust flow to produce the
required 12 hp at the turbine shaft. If the energy in the exhaust flow
is 62.5% from waste energy (combustion and expansion after the exhaust
valve has opened) and 37.5% energy from the pistons, then the turbine
will consume only 6hp from the crank (plus 10 hp of energy from
combustion/expansion after the exhaust valve opens, giving 16 total hp
to the 75% efficient turbine) to produce the required 12hp on the
shaft driving the compressor. As a result, there will be an additional
4hp available at the crankshaft of the turbocharged engine than on the
supercharged engine.
As an alternative, you could use an exhaust driven turbine to turn the
Lysholm twin screw compressor, and that might be slightly more
efficient still. But that would be more complex and difficult.
If you use some other method to power the compressor, then the overall
load on the engine to drive the compressor will probably increase
because the efficiency of the other parts involved won't be 100%. For
example, if you use a generator driven by the engine and a motor that
runs on that electricity, you'll have to factor in the losses in the
generator, the belts/gears that drive the generator, the wiring
between the generator and the motor, the motor itself and the belts or
gears that drive the compressor from the motor.
> Also compressing isothermally should gain a further 10% over the
> adiabatic compression/intercooling of the standard setup.
>
How do you plan to achieve isothermal compression at the pressure and
flow rate required?
> So does anyone know what might be achievable and what sort of compressor
> would be suitable? I was thinking that perhaps a much larger engine than the
> target engine might be modified into something suitable (running very slowly
> so as to operate near isothermally, but I have no idea how large the
> friction/pumping/leakage losses would amount to.
>
> Perhaps some sort of Heath-Robinson-ish device - a very, very large,
> multi-stage, very slow moving bellows perhaps? ;)
>
> TIA, Splin
It's just my opinion, but it seems like it would take a very long time
for the slight fuel savings from the slight increase in efficiency to
offset the cost and complexity of some of the things you are
considering. It is an interesting thought exercise, though.
Good luck with it, though.
Racer X
1984 VW Jetta Diesel GT
1992 Mazda Miata
1994 Caravan (OK, maybe it's a keeper, but I still want a VW Caddy)
1983 VW Rabbit GTI (ITB racer)
1988 Mazda RX-7 (Soon to be ITS racer)
1992 GMC Topkick (portable garage for racecar[s])
Bob Wilson
aware of. They accomplish this using an extremely low friction design.
Check www.vairex.com and www.dyneco.com . I don't think that they
particularly like to operate at 1 bar, but more like 2-3. I must admit
though that I'm not sure if they don't have some for ~ atmospheric pressure.
The efficiency (ideal power/actual power) is ~62%. Cost is high since
volume is low!
"splin" <splin...@ntlworld.com> wrote in message
news:sEOd9.6666$0r3.1...@newsfep1-win.server.ntli.net...
> What is the highest efficiency which could be practically achieved for a 1
> bar air compressor -
<snip>
Rick
So, you wish to use all that waste heat to charge the starting batteries
then invent some Rube Goldberg blower system in order to attempt to
improve scavenging efficiency by a percent or two ... Doesn't that seem
the least bit bizarre to you?
If the idea of an engine driven alternator is offensive, just install a
hand pumped hydraulic starting system on the thing, change all the
instrumentation to mechanical and retain the very carefully designed
original equipment to handle the scavenging chores? Just what are those
multi-thousand dollar apiece efficiency points going to do for you
assuming you can outdesign the pros at the engine manufacturer and
actually achieve any?
Rick
Rusty
As one other poster has asked, how do you propose to achieve
isothermal compression since no such thing is theoretically
possible?
The nearest you can get in practice is to have many stages of
compression with intercoolers between each. The overall efficiency
of the many stages would be very low and would negate any gains from
the extra intercooling.
Even a very large and slow moving compressor of some sort can only
compress adiabatically unless heat is leaking out during the
compression stroke, and that would be minimal. There is a good
reason that compressors generally have no more than two stages with
one intercooler and an after cooler. The complexity of adding
stages outweighs any advantage that might be gained.
dlzc@aol.com (formerly)
The end result would seem the same, but if he could compress isothermally,
then the work involved at the "turbo" is about 42% less.
David A. Smith
Christopher Green
The fly in this ointment is hiding in the "if". Isothermal compression
is one of those ideals that looks nice in theory but is darned hard to
achieve in practice.
True isothermal compression requires removing all of the heat of
compression in the compressor. This demands a very slow compression
process (such as the bellows mentioned elsewhere) or a very powerful
cooling system.
Practical sort-of-isothermal compressors are built around
compressor-intercooler stages or use water injection.
I don't see what technology is going to do much better than a
power-take-off compressor of sufficient capacity followed by an
intercooler.
--
Chris Green
ckruger
splin <splin...@ntlworld.com> wrote in message
news:sEOd9.6666$0r3.1...@newsfep1-win.server.ntli.net...
> I understand that the centrifugal compressor of a typical engine's
> turbo-supercharger has a mechanical efficiency of around 70% to 75%. I
would
> have thought this could be significantly bettered by say a displacement
> type. Also compressing isothermally should gain a further 10% over the
> adiabatic compression/intercooling of the standard setup.
>
> So does anyone know what might be achievable and what sort of compressor
> would be suitable?
I believe flooded screw compressors can be designed to get reasonably close
to isothermal compression. But I do not know whether it would fit your
application or whether the overall efficiency will be higher than
reciprocating compression with intercooling. Contact one of the vendors and
ask them.
Christiaan
splin
> "splin" <splin...@ntlworld.com> wrote in message
news:<sEOd9.6666$0r3.1...@newsfep1-win.server.ntli.net>...
> > What is the highest efficiency which could be practically achieved for a
1
> > bar air compressor - either with a commercial product or a home brew
system?
> >
> Centrifugal compressors (like in a turbocharger) can achieve a peak
> efficiency of around 80% over a narrow range of flow and compression.
> A Lysholm twin screw type compressor (used in some mechanical
> superchargers) can get up to around 85% efficiency or so over a
> similarly narrow range of flow and pressure increase. I seem to
> remember that positive displacement type compressors (like a roots
> supercharger) are much less efficient, perhaps in the 50%
> neighborhood.
>
> I doubt any reciprocating compressor would be worth looking at. They
> tend to be fairly inefficient, around 30-50% is a number that sticks
> in my mind, but I could be wrong on the exact details.
That sounds much too low for me - I would have thought that a very large,
slow (say 100rpm or even less) converted engine could achieve much better
than that. I would have thought that 90% or better is possible, but I'm
finding it almost impossible to find *any* efficiency values on the net for
commercial recipricating compressors.
I guess it unlikely that a commercial device would achieve very high
efficiencies as the size and cost compromises would be unacceptable. In my
case the size would not be an issue. The cost is more important, but it may
be feasible (just) by modifying a large scrap engine.
>
> Also, there's some rotary superchargers. I think Wankel actually
> designed one of these superchargers back in the 1950's. I doubt you
> could even find one today. I have no idea how efficient they were. The
> design is similar to a wankel rotary engine, but there's an output
> port where the leading spark plug would be and another intake port
> where the trailing spark plug would be.
>
> > The application I'm, thinking of is to replace the
supercharger/intercooler
> > of a turbo charged diesel engine with a more efficient, near isothermal
> > compressor. The existing turbo's output would be put to good use driving
an
> > alternator. The engine would be static so size would not be an issue and
> > cost would not be a critical factor (within reason).
> >
> OK, so it's not an automotive application, obviously. Is the operating
> speed constant? Is the load constant?
>
Yes - its for a genset application running at, or near, best efficiency -
ie. constant load.
> Since it's a diesel, intake flow is nearly linear with crankshaft
> speed. If it's a fixed speed application, you should be able to get a
> centrifugal compressor or Lysholm twin screw that's designed for peak
> efficiency at the required flow rate and pressure. With that, you
> should be able to hit an efficiency in the 80-85% range.
>
> If the load is variable, with wide variations, I'd go with a belt
> driven supercharger. There's not much loss in a good belt drive system
> and good bearings. With a variable load, there will be variations in
> the volume of exhaust flow (more volume under heavier load, less
> volume under lighter load). That will mean that a turbine in the
> exhaust will see variable flows in, and the output speed, power
> available at the turbine shaft, and the overall efficiency will vary.
> Wide variations in the load would make a turbocharger more difficult
> to tune for absolute peak efficiency.
>
> If the load is constant, you might use a turbocharger and design the
> turbine for the exact flow available, and the exact load presented by
> the compressor. I imagine that in most situations, the turbo is going
> to have an advantage from the fact that most of the energy driving its
> compressor comes from combustion (and expansion of the charge) after
> the exhaust valve opens. Therefore, the turbine presents much less of
> a load to the pistons and crankshaft.
>
I suspect that it would be very difficult to match the centrigual compressor
to the actual engine operating conditions without a lot of very expensive
equipment, specialised knowledge and a lot of experimentation. In any case
since the engine I would want to use (possibly around 500kW) would already
be turbocharged, it probably wouldn't be worth the effort to try and improve
on the standard turbo by a few %. Of course, if the standard compressor were
only 60% efficient at the 'best fuel consumption' operating speed/load then
it might be worthwhile. The engine manufacturers don't seem to publish this
information though!
The efficiency of a reciprocating compressor on the other hand would I
believe be much less sensitive to the operating pressure ratio and flow
rate, allowing a lot of scope for tinkering with the engine's operating
conditions to achieve the lowest fuel consumption.
> > I understand that the centrifugal compressor of a typical engine's
> > turbo-supercharger has a mechanical efficiency of around 70% to 75%.
>
> In a normal automotive application, the efficiency peaks around that.
> But because of the wide variations in flow and pressure requirements,
> actual efficiency can be all over the place in an automotive
> application. Some trade off is necessary to deliver adequate
> performance over a wider range, so peak efficiency might be reduced in
> order to broaden the useful boost curve.
>
> > I would
> > have thought this could be significantly bettered by say a displacement
> > type.
>
> I don't know. Centrifugal compressors are pretty efficient in
> comparison to the alternatives. The only thing I see that's better
> would be a Lysholm twin screw type compressor.
>
But aren't Lysholm compressors also essentially adiabatic?
> Then, the question would be how efficiently can you turn the thing?
> The advantage of an exhaust turbine to drive it is that most of the
> energy driving the turbine is actually waste that the engine doesn't
> capture and turn to power at the crankshaft. So, the load presented to
> the crankshaft by a turbocharger is much less than the total load from
> the compressor. This generally more than offsets the fact that the
> turbine might not be the most efficient at extracting the energy and
> turning it into rotation of the turbocharger shaft, and the fact that
> the compressor itself isn't necessarily the most efficient available
> to produce the required flow and pressure.
But I would use the existing exhaust turbine to drive a high-speed
alternator (eg. such as produced by turbo-genset). I would also be
interested in adding a second turbine/alternator to extract another % or
three, but I don't know if the back-pressure would be excessive (or would
require the engine o/p to be dropped a bit and/or the boost pressure to be
raised). I know that Scania turbo-compound their 470hp truck engine but the
best SFC figure of 191kg/kWhr doesn't seem to show much if any advantage.
(Mack quote 190kg/kWhr or better for some of their engines).
> Let's say a Lysholm twin screw supercharger might need 9.5 hp to
> produce the required boost and a less efficient centrifugal compressor
> in a turbocharger might need 12 hp to produce the same boost. Now, the
> supercharger might be driven by a belt and pulley system that's 95%
> efficient, so it will take 10 hp from the crank to drive it. An
> exhaust driven turbine in the turbocharger might also be around 75%
> efficient, so it will take 16hp from the exhaust flow to produce the
> required 12 hp at the turbine shaft. If the energy in the exhaust flow
> is 62.5% from waste energy (combustion and expansion after the exhaust
> valve has opened) and 37.5% energy from the pistons, then the turbine
> will consume only 6hp from the crank (plus 10 hp of energy from
> combustion/expansion after the exhaust valve opens, giving 16 total hp
> to the 75% efficient turbine) to produce the required 12hp on the
> shaft driving the compressor. As a result, there will be an additional
> 4hp available at the crankshaft of the turbocharged engine than on the
> supercharged engine.
Interesting figures. Is the 62.5%/37.5% split for illustrative purposes or
representive for a typical diesel engine?
> As an alternative, you could use an exhaust driven turbine to turn the
> Lysholm twin screw compressor, and that might be slightly more
> efficient still. But that would be more complex and difficult.
>
> If you use some other method to power the compressor, then the overall
> load on the engine to drive the compressor will probably increase
> because the efficiency of the other parts involved won't be 100%. For
> example, if you use a generator driven by the engine and a motor that
> runs on that electricity, you'll have to factor in the losses in the
> generator, the belts/gears that drive the generator, the wiring
> between the generator and the motor, the motor itself and the belts or
> gears that drive the compressor from the motor.
Quite true, though I though chain drives or toothed belts were capable of
achieving 97%/98% efficiency.
> > Also compressing isothermally should gain a further 10% over the
> > adiabatic compression/intercooling of the standard setup.
> >
> How do you plan to achieve isothermal compression at the pressure and
> flow rate required?
Well that was my original question. Near-isothermal with sufficiently high
efficiency might well be impossible of course. I thought a large
reciproating engine (say at least 10x the displacement of the target engine)
running at say 100rpm might be reasonably close to isothermal. But on
reflection I guess that even that would still be much too fast. And it would
be enormous! A web search reveals a few commercial compressors claiming to
be near-isothermal (at least one by using water injection), but none publish
efficiency figures.
> > So does anyone know what might be achievable and what sort of compressor
> > would be suitable? I was thinking that perhaps a much larger engine than
the
> > target engine might be modified into something suitable (running very
slowly
> > so as to operate near isothermally, but I have no idea how large the
> > friction/pumping/leakage losses would amount to.
> >
> > Perhaps some sort of Heath-Robinson-ish device - a very, very large,
> > multi-stage, very slow moving bellows perhaps? ;)
> >
> > TIA, Splin
>
> It's just my opinion, but it seems like it would take a very long time
> for the slight fuel savings from the slight increase in efficiency to
> offset the cost and complexity of some of the things you are
> considering. It is an interesting thought exercise, though.
Well that was my expectation, but I though it worth investigating (and
interesting).
> Good luck with it, though.
>
> Racer X
Thanks for your detailed response - much appreciated, Splin
splin
"Christopher Green" <cj.g...@worldnet.att.net> wrote in message
news:c31fa7b1.02090...@posting.google.com...
> "dl...@aol.com \(formerly\)" <dl...@cox.net> wrote in message
news:<7KPe9.91273$Fb.37...@news1.west.cox.net>...
> > Dear "Julian Stafford":
> >
> > > >Thanks David. However I think you misunderstood me; I'm talking about
> > using
> > > >an isothermal supercharger to replace the turbocharger (which
operates
> > > >adiabatically). This is simply to provide the required boost in
pressure
> > of
> > > >the intake air charge. The operating cycle in the engine is exactly
the
> > same
> > > >as before - the standard diesel cycle.
> > >
> > > I understand you but could you explain your reasoning for this
project?
> > > Would not a normal turbo and good intercooler amount to much the same
> > thing?
> >
> > The end result would seem the same, but if he could compress
isothermally,
> > then the work involved at the "turbo" is about 42% less.
I believe that the efficiency advantage of isothermal compression over
adiabatic would be more like 10 to 15% for a pressure ratio of 2 (ie. 1 bar
boost).
> >
> > David A. Smith
>
> The fly in this ointment is hiding in the "if". Isothermal compression
> is one of those ideals that looks nice in theory but is darned hard to
> achieve in practice.
>
> True isothermal compression requires removing all of the heat of
> compression in the compressor. This demands a very slow compression
> process (such as the bellows mentioned elsewhere) or a very powerful
> cooling system.
>
> Practical sort-of-isothermal compressors are built around
> compressor-intercooler stages or use water injection.
>
> I don't see what technology is going to do much better than a
> power-take-off compressor of sufficient capacity followed by an
> intercooler.
It probably isn't practical. I was particularly interested in learning what
efficiencies are being achieved by current compressor designs - its not easy
to find any such figures on the web, except for automotive after-market
compressors including centrifugal, twin-screw and roots type compressors.
Thanks, Splin
splin
"Steve" <n...@spam.thanks> wrote in message
news:3D78F9E7...@spam.thanks...
> splin wrote:
> > What is the highest efficiency which could be practically achieved for a
1
> > bar air compressor - either with a commercial product or a home brew
system?
>
> Well... there's a very easy to define upper limit. NO air compression
> device can ever exceed 50% efficiency once you cool the output air back
> down to the input temperature. A minimum of 1/2 the input energy goes
> into raising the temperature of the compressed air.
This definitely isn't true - do a search on "isothermal compression" to find
many explanations. Thermodynamics isn't an especially easy subject though as
I am discovering.
> The only way to approximate an "isothermal air compressor" (since there
> is no such thing in the real world) is to compress adiabatically and
> then cool the compressed air back down to inlet temperature. Which is
> EXACTLY what your turbo/intercooler already does (or approximates).
I'm sure this is not correct. With a low pressure ratio their isn't much
difference - but for a ratio of 2 or more the difference starts to get
significant. Cooling the air after adiabatic compression doesn't alter the
fact that it was compressed adiabatically and thus rather less efficiently
than a near isothermal compressor. Several commercial compressors claim to
operate near-isothermally (search on "near isothermal" for examples). At
least one sprays water into the inlet air stream using the latent heat of
vaporisation of the water to keep the air relatively cool during
compression. It doesn't quote the mechanical efficiency though
> You're chasing fairies and UFOs, here.
On this point you're very probably correct though.
Thanks, Splin
splin
"Bob Wilson" <rj...@comcast.net> wrote in message
news:Rgee9.487787$2p2.19...@bin4.nnrp.aus1.giganews.com...
> Fuel cell system compressors are the most efficient that I am currently
> aware of. They accomplish this using an extremely low friction design.
> Check www.vairex.com and www.dyneco.com . I don't think that they
> particularly like to operate at 1 bar, but more like 2-3. I must admit
> though that I'm not sure if they don't have some for ~ atmospheric
pressure.
> The efficiency (ideal power/actual power) is ~62%. Cost is high since
> volume is low!
Interesting. One of their (variex) graphs shows "overall adiabatic
efficiency" reaching 80%. Unfortunately, after reading the 'compressor
efficiency definitions' page about definition of terms I'm now thoroughly
confused. Isothermal compression apparantly implies a mechanical efficiency
of 0 which seems counter-intuitive to me. Clearly I need to spend some more
time trying to get to grips with thermodynamics (I'm a novice if you hadn't
spotted it)!
Thanks, Splin
splin
"Rusty" <rto...@oanet.com> wrote in message
news:3d7b939c...@nntp.oanet.com...
> As one other poster has asked, how do you propose to achieve
> isothermal compression since no such thing is theoretically
> possible?
>
> The nearest you can get in practice is to have many stages of
> compression with intercoolers between each. The overall efficiency
> of the many stages would be very low and would negate any gains from
> the extra intercooling.
>
> Even a very large and slow moving compressor of some sort can only
> compress adiabatically unless heat is leaking out during the
> compression stroke, and that would be minimal. There is a good
> reason that compressors generally have no more than two stages with
> one intercooler and an after cooler. The complexity of adding
> stages outweighs any advantage that might be gained.
I'm sure you are right; multiple stages would very likely be too
inefficient. Water injection seems to be one way of achieving
near-isothermal compression, but the efficiency of such compressors may not
be enough to make the scheme worthwhile.
Thanks, Splin
splin
"Rick" <tu...@earthlink.net> wrote in message
news:3D7ABED2...@earthlink.net...
Sorry Rick, I guess I didn't make myself very clear. The application is for
a gen-set; the output of the turbo driven altenator would be added to that
produced by the main synchronous generator. The altenator would be a high
speed device such as those produced (to be produced?) by turbogenset.
Its not a question of outdesigning the engine's skilled design engineers;
rather that they clearly wouldn't be interested in the design build and
marketing costs of a product with a likely customer demand of 1 (or less -
it would almost certainly be too expensive for me!). Bizarre is likely
correct though.
Thanks, Splin.
carl mciver
turbo whose shaft was tied through a gearbox to the engine, further
increasing the efficiency of the engine. This way more of the energy
heading out the tailpipe can be recycled into useful work.
dlzc@aol.com (formerly)
> turbo whose shaft was tied through a gearbox to the engine, further
> increasing the efficiency of the engine. This way more of the energy
> heading out the tailpipe can be recycled into useful work.
Get more energy by pulling a vaccum in the crankcase...
David A. Smith
Duncan Woodbury
"carl mciver" <cmc...@mindspring.com> wrote in message
news:aloumf$sgs$1...@nntp9.atl.mindspring.net...
> I seem to recall seeing somewhere that some engine manufacturers had a
> turbo whose shaft was tied through a gearbox to the engine, further
> increasing the efficiency of the engine. This way more of the energy
> heading out the tailpipe can be recycled into useful work.
That sounds like a nifty idea....... Somewhere in the rational part of my
brain a little voice is telling me it will produce very little if any
performance increase though. I'd love to read more about that though.
Duncan
Rick
exhaust energy directly. The turbine shaft was geared back to the cam drive.
Have a look at the Wright 3350. It used, IIRC, about 6 PRT's (power
recovery turbines) which produced a considerable increase in power
output. The development was not without interesting moments, the PRT's
tended to act like centrifuges and deposits from the lube oil caused
problems as did torsional vibration problems.
Rick
splin
news:OyTf9.472448$Aw4.19...@bin2.nnrp.aus1.giganews.com...
Scania use it on their 470hp diesel truck engine -see
http://www.scania.com/ms/events/010112/photos/html/b.htm for illustrations
and http://www.scania.com/ms/events/010112/press/p01102en.htm for a typical
press release.
MAN B&W also make a turbocompound system (TCS) for their very large 2 stroke
diesels, claiming up to 4% gain in efficiency. See
http://www.manbw.dk/documents/p291_0108.pdf (page 6). If the 54% efficiency
isn't good enough for you then you can add a boiler to the exhaust driving a
steam turbine to reach an overall efficiency of approx 56.5% Unfortunately
the TCS is designed for engines of at least 13.8MW - not very pratical for
your average road transport!
Splin
Garrett Fulton
aircraft engine. 3,350 cu. in. Had three TC units feeding power back into
the crankcase. Didn't have a real high reliability record, however.
Garrett Fulton
"splin" <splin...@ntlworld.com> wrote in message
news:69Yf9.276$7x3....@newsfep2-win.server.ntli.net...
-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 80,000 Newsgroups - 16 Different Servers! =-----
Steve
> I seem to recall seeing somewhere that some engine manufacturers had a
> turbo whose shaft was tied through a gearbox to the engine, further
> increasing the efficiency of the engine. This way more of the energy
> heading out the tailpipe can be recycled into useful work.
>
Well.... depends on what you're asking the turbo to do.
The Wright turbo-compound R-3350 aircraft engine of the 1950s used three
exhaust-driven turbines to couple power back to the crankshaft. They
drove the crank through fluid clutches to absorb speed variations. The
engine was NOT turbocharged at all, but blown by a 2-speed
crankshaft-driven centrifugal supercharger. IIRC the PRTs (power
recovery turbines) added about 900 HP (300 each) back to the crankshaft
during high altitude/high blower gear/low propellor RPM cruise
conditions. It was 30 years before turboprops or jets could come
ANYWHERE close to the low specific fuel consumption of that engine....
but it was a maintenance nightmare and the jets were 10 times more
reliable right out of the box.
More recently, EMD has used a rather clever turbocharger on their
two-stroke locomotive diesels for years. Rather than using a roots
blower and then just daisy-chaining a turbocharger to blow into the
roots unit, as Detroit Diesel did with their turbocharged 2-strokes, EMD
elected to use a shaft-driven centrifugal blower. The blower is driven
through a sprag clutch, and shares its shaft with an exhaust-driven
turbine. At low throttle settings, it operates as an engine-driven
blower powered through the accessory gear train. As throttle/load
increases and the exhaust temeprature and volume get sufficiently high,
the turbine takes over, the turbine RPM exceeds the accessory shaft rpm,
and the sprag clutch freewheels, letting it operate as a true
turbocharger and unloading the mechanical drive shaft. Gets pretty goofy
sounding when one is running with plugged filters or another problem
that causes it to repeatedly "fall off" the turbo and back to a straight
blower near the transisition point. :-)
Steve
Search for "Wright" and "turbo-compound" and "R-3350"
They're still out there- in fact they fought a LOT of forest fires this
year in Lockheed P2 Neptunes.
dmitri
> system?
Hello, I may add, You may read
the artical is: NASA Low-Speed (huge)Centrifugal Compressor for 3D
Viscous Code Assessment and Fundamental Flow Physisc research,
estumated efficiency 0.9.(I think this is adiabatic efficiency).
Also if you read ASME paper for 74 years you can find the same maximum value
for centrifugal Impellers.
In the artical H.Krain Swirling Impeller Flow (it was printed in 1988)
Total polytropic impeller efficiency was 95, and maximum achieved total
isentropic stage efficiency was 84
percent.
I can't check this result, because I haven't any coordinats for this stages
and access in any labs.But I think It was the the best results.
That is why for impruving of maximum achieved total isentropic stage
efficiency you have to use
ideas of open systems (synergetics), for example like a heat pump.
As the term "pump" implies, a heat pump moves heat from one place to
another. In the winter, it moves heat from outside to inside to heat your
home. In the summer, it moves heat in the other direction to cool your home.
This reverses the natural flow of heat from warm places to cooler places.
Like a water pump that uses energy to pump water uphill, a heat pump uses
electrical energy to pump heat "uphill" from cool locations to warmer
locations. Heat pumps use the refrigeration cycle to accomplish this.
You take enegy from open environment( for example you may use free very hot
air for heating or use cold air
for cooling inside the centrifugal Compressor.His efficiency will
improve.But it will work for special condition
of his work.
Sundar Narayan
During WW2, Napier in England built turbocompounded diesel engines for
aircraft applications. They were very efficient, 0.33 lb/hp-hr fuel
consumption. The horsepower recovered from the exhaust gases was of the
order of 100 or more. Total engine hp = 3000. I am quoting these figures
from memory, as I do not have my copy of "Some Unusual Engines", by
L.J.K.Setright.
Sundar Narayan
Vancouver
Nano
available on the shaft.
This system is used in heavy duty Volvo (FH), Iveco (EuroStar) trucks among
others. It was 1st introduced in the 70 by Scania on trucks but had huge
reliability problems.
Advantage: the extra torque on the shaft, energy (heat) is otherwise wasted
to the atmosphere
drawbacks: complexity and reliability, small efficient operating window,
added cost
Arnaud
"carl mciver" <cmc...@mindspring.com> a écrit dans le message news:
aloumf$sgs$1...@nntp9.atl.mindspring.net...