2 stroke thread (with occasional F1 relevance!)

[manolis]

Hello Uniflow.

You write:
“As for the rotary drum inlet (that is what we are looking at) its opening and closing is too slow as compared to a large dia rotary disc valve inlet. Rotary drums were abandoned in the 1930's, “


The OS18TZ is a 2-stroke model engine having:

16mm bore,
15mm stroke,
3.02cc displacement,
2.28bhp peak power @ 30,500rpm (755bhp/lit)
80oz-in torque from 19,000rpm to 29,000rpm
(0.58mN from 19,000rpm to 29,000rpm)

(more at http://www.pattakon.com/tempman/osmz211 ... cnitro.pdf )



The specific torque, along the 19,000rpm to 29,000rpm rev range, is above: 0.58mN/0.00302lit = 192mN/lit = 19mKp/lit

The specific torque is double than that of the good 4-stroke sport engines.
More interesting: this top specific torque is maintained along a long rev range (not tuned): from 19,000rpm to 29,000rpm; like, say, having a completely flat (and top, at the same time) torque from 5,000 to 8,000rpm in a normal size (~300cc per cylinder) 2-stroke.

The above model engine is based on a rotary drum inlet:



The diameter of the crankshaft at the drum valve is 12mm, with the diameter of the hole towards the crankcase being 7.5mm.

Note: this is for a bore of 16mm.

The above 12mm diameter is a compromise between the “fast opening –closing” of the inlet port and the need for a small diameter main crankshaft roller bearing.



From the above dyno it seems the drum inlet of the OS18TZ makes a better job (by the way, there is no reed valve capable to operate reliably at the 40,000+ rpm of the red line of this small engine).

In the PatATE the diameter of the drum valve is necessarily bigger than the bore. Instead of 12mm in the OS18TZ, it would be, say, 18mm (50% bigger) in a PatATE-OS18TZ.
This would give a substantially faster inlet opening and inlet closing, if necessary.

Unless I am wrong:
without a power valve,
and with a torque which is not only top, but it is also flat along a wide rev range,
a tuned exhaust would be more a problem than a solution.


The PatATE introduces some new "strong" parameters in the 2-stroke architecture for greener,and at the same time more powerful, 2-strokes.

Thanks
Manolis Pattakos

[Pinger]

They could shift to cross-flow to achieve a faster blow-down.
But the side effects of the cross-flow scavenging make it a worse solution.

They shifted from cross-flow. They won't be going back....

In the following post - reply to Uniflow - the BMEP of the OS18TZ is well above 10bar.
And, unless I am wrong, it is not based on a tuned exhaust.

Looks like an expansion chamber and even the text alludes to it. Ditto the 20% nitromethane fuel.

[manolis]

Hello Pinger.

You write:
"Looks like an expansion chamber and even the text alludes to it. Ditto the 20% nitromethane fuel."


Quote from http://www.pattakon.com/tempman/osmz211 ... cnitro.pdf

"DYNO TESTING

The .18 TZ was tested with 20-percent nitro fuel that contains 12-percent lubricant.

A standard Novarossi 90-degree header and a 51602 pipe is the exhaust system we use for testing small-block engines.

The argument could be made that because this engine has 50 percent more displacement, it would run better with a bigger tuned pipe, but for the sake of comparison, we chose to stay with the Novarossi pipe."

End of Quote.



The nitromethane fuel has some four times lower specific energy than the gasoline and consumes some 10 times less air than the same quantity of gasoline.

The 20% nitro in the fuel adds some 5% thermal energy and consumes some 2.5% of the air (i.e. it reduces by some 2.5% the quantity of gasoline, for a given air-fuel ratio).

In total, with the 20% nitro the thermal energy of the fuel increases by 5-2.5=2.5%.

The 12% lubricant (which enters cold in the engine and leaves the engine hot) absorbs thermal energy.

Being a glow plug engine, its ignition is anything but optimized.

Being a miniature engine (multiplying all dimensions by 4.63 it becomes a 300cc engine), it has 4.63 times larger surface to volume ratio than a normal size 300cc 2stroke (say like the KTM300EXC-TPI), which means substantially higher thermal loss on the walls around the combustion chamber.


So, the extreme specific torque cannot be justified neither by the 20% nitro, nor by the "tuned" exhaust (which is used also in several other 2-strokes engines of various displacements).


Besides, the top specific torque which is maintained constant (flat torque curve) along 10,000rpm (19 to 29 Krpm) cannot be the result of a tuned exhaust or other "tuning".


So, what we have here is a real (in metal) modern 2-stroke that, using an "old / obsolete" drum inlet valve, a fixed exhaust port timing and an "almost randomly selected" exhaust pipe, achieves top power and top torque and excellent torque curve.


In comparison to the above OS18TZ architecture, the PatATE:



introduces some new characteristics / parameters that can reduce substantially the emissions of the 2-strokes, without compromizing on the power output.

Thanks
Manolis Pattakos

[Tommy Cookers]

there's tests showing about 1.85 peak bhp at somewhat less rpm on plain fuel
(afaik the fuel is basically methanol - but you give gasoline figures ?)

a 3cc cylinder has relatively much more port area because the ratio of area:gasflow volume is much higher than eg in a 200cc cylinder
ie scaling 4x (OS '3cc' to just under 200cc) halves the port area relative to the gasflow (half rpm, normalising piston acceleration not piston speed)
so mep is halved and bhp/litre is quartered (roughly)
quarter rpm normalises piston speed and maintains relative port area but bhp/litre is also quartered
so the OS induction efficacy at 3cc proves nothing here ?

the torque curve 'flat' from 19000 to 29000 rpm - isn't this due to combustion limitations (that nitro somewhat relieves)
OS claim it performs from 4000 but don't show this

ok unlike lots of model engines the OS piston speeds and accelerations here are not trivial

[Pinger]

A standard Novarossi 90-degree header and a 51602 pipe is the exhaust system we use for testing small-block engines.

Standard, in terms of usually used with engines with significantly lower rpm potential. Had they tried a shorter pipe, perhaps the full rpm capability of the OS unit would have been exploited rather than it falling flat 75% into its rev range.

The nitromethane fuel has some four times lower specific energy than the gasoline and consumes some 10 times less air than the same quantity of gasoline.

Nitro's strength is not in supplying hydrocarbons to burn but in providing oxygen for the complementary hydrocarbons, in this case, methanol. It isn't referred to as 'chemical supercharging' without good reason.

Being a miniature engine (multiplying all dimensions by 4.63 it becomes a 300cc engine), it has 4.63 times larger surface to volume ratio than a normal size 300cc 2stroke (say like the KTM300EXC-TPI), which means substantially higher thermal loss on the walls around the combustion chamber.

And given the elevated rpm the time for heat loss is much reduced. Such are the nuances of IC engines....

So, the extreme specific torque cannot be justified neither by the 20% nitro, nor by the "tuned" exhaust (which is used also in several other 2-strokes engines of various displacements).

Chemically supercharged and deploying an expansion chamber (albeit one not necessarily best suited to the rpm range), you'd have to wonder (or explain) why they don't just ditch the nitro and run a plain parallel pipe.

Besides, the top specific torque which is maintained constant (flat torque curve) along 10,000rpm (19 to 29 Krpm) cannot be the result of a tuned exhaust or other "tuning".

Estimate the torque if the pipe were removed and the test re-run.

[manolis]

Hello Tommy Cookers.

You write:
“the fuel is basically methanol - but you give gasoline figures ?”


My wrong.

Calculations with methanol instead of gasoline:

Methanol (CH3OH) energy content: 19.7MJ/Kg
Methanol stoichiometric air-fuel ratio: 6.42:1

Nitromethane (CH3NO2) energy content: 11.3MJ/Kg
Nitromethane stoichiometric air-fuel ratio: 1.7:1

Burning all the air either with gasoline (14.7:1 air fuel ratio, 44.4MJ/Kg) or with methanol (6.42:1 air fuel ratio, 19.7MJ/Kg) it gives about the same thermal energy.


With 20% nitromethane (basic fuel: methanol), the methanol fuel reduces by 6% and the overall thermal energy of the fuel increases by 8.3%.


With plain gasoline or with plain methanol, the 192mN/lit specific torque of the OS18TZ (nitro 20%) in the 19 to 29Krpm range reduces at 177mN/lit, which is still top.




You also write:
“a 3cc cylinder has relatively much more port area because the ratio of area:gasflow volume is much higher than eg in a 200cc cylinder
ie scaling 4x (OS '3cc' to just under 200cc) halves the port area relative to the gasflow (half rpm, normalising piston acceleration not piston speed)
so mep is halved and bhp/litre is quartered (roughly)
quarter rpm normalises piston speed and maintains relative port area but bhp/litre is also quartered
so the OS induction efficacy at 3cc proves nothing here ?”


mep (mean effective pressure) is halved? No.


Multiplying all dimensions by 4, and dividing the revs by 4, you have (according the similarity) the same volumetric efficiency, the same BMEP, the same specific torque and four times lower power.

The 3cc OS18TZ running at 32,000rpm and the 4*4*4*3cc=192cc normal size 2-stroke running at 32,000/4=8,000rpm should have the same specific torque.

A 192cc 2-stroke having 177mN/lit specific torque from 5,000 to 8,000rpm on normal fuel is top.
In a similar way, and judged by its specific torque between 19 to 29Krpm, the OS18TZ induction effectiveness does appear top, too.




You also write:
“ok unlike lots of model engines the OS piston speeds and accelerations here are not trivial”


With conrod to stroke ratio 2,
while the mean piston speed at 42,500 rpm is only 21m/sec (and the maximum piston speed is only 30m/sec),
the maximum inertia acceleration of the piston of the OS18TZ at the 42,500rpm limit is 19,000g.

Thanks
Manolis Pattakos

[manolis]

Hello Pinger.

You write:
“Standard, in terms of usually used with engines with significantly lower rpm potential. Had they tried a shorter pipe, perhaps the full rpm capability of the OS unit would have been exploited rather than it falling flat 75% into its rev range.”

Without the proper exhaust, the OS18TZ still makes top specific torque along a wide rev range, proving that its simple design is correct, and its drum inlet port a good selection.




You also write:
“Nitro's strength is not in supplying hydrocarbons to burn but in providing oxygen for the complementary hydrocarbons, in this case, methanol. It isn't referred to as 'chemical supercharging' without good reason.”

No.

The nitro (nitromethane, CH3NO2) is not providing O2 to the methanol.

1Kg of nitromethane needs 1.7Kg of air to get completely burnt (while 1Kg of gasoline needs 14.7Kg of air to get completely burnt, i.e. some 9 times more).


The two oxygen atoms in the molecule of the nitromethane (CH3NO2) are consumed by the carbon atom of the same molecule, to form a molecule of CO2.
The three atoms of hydrogen need “1.5” atom of oxygen to form “1.5” molecule of water; this “1.5” atom of oxygen is removed from the air in the cylinder, leaving less oxygen for the “complementary hydrocarbons” (the methanol in this case).

On the other hand, with 20% nitro fuel, it is like increasing the thermal energy produced by the combustion of the oxygen of the air in the cylinder by 8.3% as compared to the case of plain ethanol fuel (as explained in a previous post), which can be considered as “chemical supercharging”.



You also write:
“And given the elevated rpm the time for heat loss is much reduced. Such are the nuances of IC engines....”

The heat loss happens in various ways in the cylinder during combustion; some of them are linear with the time, some others are not linear.




You also write:
“Chemically supercharged and deploying an expansion chamber (albeit one not necessarily best suited to the rpm range), you'd have to wonder (or explain) why they don't just ditch the nitro and run a plain parallel pipe.

Estimate the torque if the pipe were removed and the test re-run.”

A tuned exhaust is:
good at some rpm wherein the pressure waves help the filling of the cylinder (improved trapping efficiency),
and bad at some other rpm wherein the pressure waves spoil the filling of the cylinder.

The local peaks of the torque curve show the revs wherein the tuned exhaust helps the overfilling of the cylinder, while the holes of the torque curve show the revs wherein the exhaust spoils the trapping efficiency of the engine.

The flat torque of the OS18TZ along 10,000rpm has neither peaks, nor holes.

Thanks
Manolis Pattakos

[Pinger]

Without the proper exhaust, the OS18TZ still makes top specific torque along a wide rev range, proving that its simple design is correct, and its drum inlet port a good selection.

Yes, and in part it is due to a -ve pulse returned to the cylinder from the divergent cone in the exhaust.

No.

The nitro (nitromethane, CH3NO2) is not providing O2 to the methanol.

1Kg of nitromethane needs 1.7Kg of air to get completely burnt (while 1Kg of gasoline needs 14.7Kg of air to get completely burnt, i.e. some 9 times more).


The two oxygen atoms in the molecule of the nitromethane (CH3NO2) are consumed by the carbon atom of the same molecule, to form a molecule of CO2.
The three atoms of hydrogen need “1.5” atom of oxygen to form “1.5” molecule of water; this “1.5” atom of oxygen is removed from the air in the cylinder, leaving less oxygen for the “complementary hydrocarbons” (the methanol in this case).

On the other hand, with 20% nitro fuel, it is like increasing the thermal energy produced by the combustion of the oxygen of the air in the cylinder by 8.3% as compared to the case of plain ethanol fuel (as explained in a previous post), which can be considered as “chemical supercharging”.

Wikipedia says:
''The oxygen content of nitromethane enables it to burn with much less atmospheric oxygen.

4 CH3NO2 + 3 O2 → 4 CO2 + 6 H2O + 2 N2
The amount of air required to burn 1 kg (2.2 lb) of gasoline is 14.7 kg (32 lb), but only 1.7 kg (3.7 lb) of air is required for 1 kg of nitromethane. Since an engine's cylinder can only contain a limited amount of air on each stroke, 8.7 times more nitromethane than gasoline can be burned in one stroke. Nitromethane, however, has a lower specific energy: gasoline provides about 42–44 MJ/kg, whereas nitromethane provides only 11.3 MJ/kg. This analysis indicates that nitromethane generates about 2.3 times the power of gasoline when combined with a given amount of oxygen.''

which pretty much corresponds with how nitro has been used historically. That is, to provide the cylinder with oxygen relieving the induction system of the task.

The heat loss happens in various ways in the cylinder during combustion; some of them are linear with the time, some others are not linear.

All heat transfer is time related. Hold your hand over a naked flame for proof.

A tuned exhaust is:
good at some rpm wherein the pressure waves help the filling of the cylinder (improved trapping efficiency),
and bad at some other rpm wherein the pressure waves spoil the filling of the cylinder.

The local peaks of the torque curve show the revs wherein the tuned exhaust helps the overfilling of the cylinder, while the holes of the torque curve show the revs wherein the exhaust spoils the trapping efficiency of the engine.

The flat torque of the OS18TZ along 10,000rpm has neither peaks, nor holes.

It is flat until 19000 rpm then flat again at 29000 rpm - some 13000 rpm short of its design speed. So, a 13000 rpm hole either side of the torque curve.
That, is a tuned pipe (just not well enough attuned with that particular engine). And without it I doubt the torque figure would be much more than half what it is. Why, would that engine behave any differently from any other utilising a tuned pipe?

[Tommy Cookers]

@ Manolis
you have not responded to what I wrote ......
(if the 4x scaled engine is run at half the OS rpm it may be strangled because it wants to pass 32x the gasflow through 16x the port area)

we agree - because this is what I said originally .......
(the 4x scaled engine run at a quarter the OS rpm it is not strangled because it wants to pass 16x the gasflow through 16x the port area)
ok the so-called specific torque is unchanged
but (as I implied originally) what's outstanding about the 4x scaled engine having a quarter of the OS bhp/litre ?

the OS tested on straight? fuel has far less power than the 20% nitro test (OS link below says 1.8 PS 1.83 hp at 29000)
http://www.os-engines.co.jp/english/lin ... l/18tz.pdf (back page)

isn't the action of nitro in a mix eg with methanol etc that the methanol's carbon will strip oxygen from the nitro ? - so burning more methanol
this effect is probably greater with nitro/gasoline than nitro/methanol
ie carbon reactions dominate hydrogen reactions - the reason why rich mixtures in aviation give methane in the exhaust
and the link below seems consistent with this .....
alcohols etc burn faster (less particulate) because carbon combustion 'steals' the alcohols oxygen, restitution is from oxygen induced in combustion air
of course carbon combustion yields less heat from a given amount of oxygen than does hydrogen combustion
but people should remember that this is all rich mixtures and one or other or both fuel ingredients are going partially uncombusted

anyway a good read ?
http://digitalcommons.mtu.edu/cgi/viewc ... ntext=etds

about nitromethane and nitrobenzene etc
http://www.rcuniverse.com/forum/rc-fuel ... print.html
more useful
http://www.turbofast.com.au/racefuel15.html
this latter says nitrobenzene is used as combustion aid not fuel (W196 used 0.5% - presumably prewar Mercs the same)
agrees nitrobenzene as a fuel gives less power than gasoline or alcohol
and nitromethane needs lower CR

[manolis]

Hello Pinger.

During the combustion, the nitromethene is not providing oxygen, but it is absorbing oxygen from the air inside the cylinder.
The difference from the other fuels is that the nitromethane is using substantially less oxygen (per MJ of thermal energy provided) in order to get burnt, so the quantity of the other fuel is only slightly reduced.



You write:
“All heat transfer is time related. Hold your hand over a naked flame for proof.”


As everything else, similarly the specific heat loss is time related, but not linearly.


The OS18TZ is a miniature 2-stroke engine.

At the other end there are some giant 2-stroke marine engines, like the modern Warstila X-92 (920mm bore, 3,468mm stroke), wherein the thermal loss is minimized (they run at more than 50% BTE) despite the fact that they run slow, really slow, providing plenty of time for heat loss.

Relative to the OS18TZ (which provides its peak power at 32,000rpm), at 70rpm wherein the Wartsila X-92 provides its peak power we have:
30,500rpm / 70rpm = 435 times more time, per combustion, for heat loss.

It is a good practice to examine the "extremities" of a kind: a giant engine versus a miniature engine.

Think of the combustion of a molecule of fuel inside the OS18TZ when the piston is at its TDC.
When a fuel molecule is reacting with the oxygen of the air, the maximum distance of the combustion flame from the surrounding walls is 0.75mm (with 15mm piston stroke and 11:1 compression ratio, the combustion chamber (piston at TDC) is a cylinder having 15mm diameter and 1.5mm height; 1.5mm/2=0.75mm).

The flame (anywhere in the combustion chamber) cannot help being in touch with the “cold” metal walls.

Think now the combustion of a molecule of fuel in a giant Wartsila X92, wherein, when the piston is at the TDC, the combustion chamber can be considered as a cylinder having 920mm diameter and ~300mm height (compression ratio ~12:1).
The maximum distance of the burning molecules (i.e. of the flame) from the metal walls is ~300/2=150mm (compare it with the 0.75mm in the OS18TZ).



By the way, the size of the combustion chamber and the lean burn allow a thick cushion (say 30mm?) of air between the metal walls and the fuel, which isolates the flame from the metal walls and absorbs most of the thermal energy emitted by the glowing “core” of the combustion chamber.

I.e. instead of heating the metal walls (as inevitably happens in the miniature engine due to their small dimensions), in the giant engine the thermal energy emitted during the combustion heats the surrounding air or air cushion, which during the following expansion provides the desired mechanical energy.

The BTE (Brake thermal efficiency) of the slow revving giant engine is above 50%, while the BTE of the miniature engine is well below 25%.
The main cause is the heat loss.

So, the less time per reciprocation of the piston is one of several factors that affect the overall heat loss.


From another viewpoint:

The “cylinder capacity” ratio is 760,000 (the giant marine runs lean; so the ratio of fuel quantity per combustion is, say, ~600,000).

The ratio of the “wall surfaces” (piston at the BDC) of the above two engines (the giant marine and the miniature model engine) is ~10,000.

At peak power, for one piston stroke the giant engine needs 435 times the time the miniature engine needs (time ratio: 435).

In a simple approach, the specific heat loss (i.e. per Kg of fuel) is linearly proportional to the wall surface, linearly proportional to the time, and inversely proportional to the quantity of fuel burnt.

On such reasoning the ratio of the heat loss would be:

(10,000/600,000)*435 = 7.25

i.e. the giant engine should have more than 7 time higher specific thermal loss than the miniature engine.
In practice, the giant engine has substantially lower specific thermal loss.


Thanks
Manolis Pattakos

[manolis]

Hello Tommy Cookers.

You write:
“you have not responded to what I wrote ......
(if the 4x scaled engine is run at half the OS rpm it may be strangled because it wants to pass 32x the gasflow through 16x the port area)”


Doesn’t similarity mean same piston speed? (same Reynolds number).

4x scaled running at half speed ?
Doesn’t this mean double piston speed?.

On the same reasoning, why not to double the rpm of the OS18TZ ? (from 42,500rpm to 85,000rpm).



You also write:
“the OS tested on straight? fuel has far less power than the 20% nitro test (OS link below says 1.8 PS 1.83 hp at 29000)
http://www.os-engines.co.jp/english/lin ... l/18tz.pdf (back page)”

The substantially higher specific heat loss – due to the small dimensions of the OS18TZ, as explained in the previous post to Pinger – is justified only partly by the increased thermal energy of the 20% nitro fuel.

So, the 192mN specific torque (20% nitro fuel) is an impressive number indicating good breathing and trapping efficiency.

This was the idea behind the presentation of the OS18TZ.



You also write:
“isn't the action of nitro in a mix eg with methanol etc that the methanol's carbon will strip oxygen from the nitro ? - so burning more methanol
this effect is probably greater with nitro/gasoline than nitro/methanol
ie carbon reactions dominate hydrogen reactions - the reason why rich mixtures in aviation give methane in the exhaust”

No.
The nitro uses its own oxygen, which is not adequate, so it uses a part of the air oxygen, too.

With 20% nitro you will burn some 6% less methanol (which means, the 6% of the oxygen of the air is used for the complete combustion of the nitro).
However the overall thermal content of the fuel (nitro+methanol) increases by 8.3% (as calculated in a previous post).

If it is still confusing, let me know to write an example with number.

Thanks
Manolis Pattakos

[Pinger]

From another viewpoint:

The “cylinder capacity” ratio is 760,000 (the giant marine runs lean; so the ratio of fuel quantity per combustion is, say, ~600,000).

The ratio of the “wall surfaces” (piston at the BDC) of the above two engines (the giant marine and the miniature model engine) is ~10,000.

At peak power, for one piston stroke the giant engine needs 435 times the time the miniature engine needs (time ratio: 435).

In a simple approach, the specific heat loss (i.e. per Kg of fuel) is linearly proportional to the wall surface, linearly proportional to the time, and inversely proportional to the quantity of fuel burnt.

On such reasoning the ratio of the heat loss would be:

(10,000/600,000)*435 = 7.25

i.e. the giant engine should have more than 7 time higher specific thermal loss than the miniature engine.
In practice, the giant engine has substantially lower specific thermal loss.


Thanks
Manolis Pattakos

I will recheck my arithmetic, but currently it shows the OS engine to have a surface to volume ratio some 168 times greater than the W's. Dividing that by 435 (rpm factor) gives 0.385. That is the OS's heat loss is 0.385 of the W's. Factor in fuel equivalence and the figure will move toward the W's advantage but it would have to be pretty lean to match it.

Does someone else want to run the numbers? (mine were rushed).
Volumes are obvious, areas I calculated as twice piston area plus exposed cylinder wall at TDC (using M's 1.5 and 300mm figures) despite neither engine having disc chambers and the W departing furthest (I'm assuming bowl-in-piston).

[manolis]

Hello Pinger.

Here are the calculations:



Thanks
Manolis Pattakos

[manolis]

Hello Tommy Cookers

You write:
"a good read ?
http://digitalcommons.mtu.edu/cgi/viewc ... ntext=etds"


Thanks for the link; at a first look, the plot at the page 102 shows a peak pressure of only 22 bar inside the cylinder.
And at the page 88, the IMEP (indicated mean effective pressure) is only 3.3bar (i.e. BMEP under 3bar).

In comparison, the OS18TZ has a BMEP over 10bar along a wide rev range.


Talking for powerful model engines running at extreme revs (30,000++ rpm):

In case of 4-stroke operation,
while the poppet valves are out of the question (inertia loads etc),
the PatRoVa 4-stroke rotary valve:



has no rev limit, at all.

In addition, without poppet valves and pockets on the piston, the combustion chambet of the PatRoVa can be optimized to enable a good part of the fuel to get burnt away from the walls.


Question:
When a molecule of fuel is burnt at the center of the combustion chamber heating only the surrounding air (i.e. the working medium),
and when a molecule of the fuel is burnt in touch with the walls (heating mainly the metal and not the working medium),
what is the difference in thermal efficiency?

Heating the working medium means delivery of mechanical energy during the following expansion.
Heating the wall means only need for more cooling and no mechanical energy during the expansion.

Thanks
Manolis Pattakos

[Tommy Cookers]

regarding the power benefits (stated by Manolis as 8,3%) to the OS of '20% nitro' fuel over 'straight' methanol fuel ......

with 100% nitromethane fuel 1 kg of air induced into the engine requires for combustion 588 gm of nitro - and increases power by 130%
so, very crudely, 20% (111 gm) of that 558 gm of fuel might appear to increase power by 20% of that 130% ie by 26%

but 1 kg of induced air in the engine running on 'straight' ie 100% methanol fuel requires 156 gm of methanol
if we take 80% of that and add nitro to make a 20/80 nitro/methanol blend we add far less nitro than 20% (111 gm) of our 100% nitro
we actually about 6% (of the nitro in our 100% nitro) to make '20%' nitro/methanol fuel - crudely this appears to increase power by 8%

this likely misappreciation of the meaning of '20%' accounts for the benefit of '20%' nitro appearing disproportionately less than the benefit of 100%
it also accounts for the corresponding notoriously disproportionate cost of running on 100% nitro

btw 100% nitrobenzene would give less power than 100% methanol, but is valued at 0.5% as a combustion improver for the 99.5% constituents
so there must be combustion improvement due to the nitromethane in '20%' nitro fuel
OS originally stated the power as 1.83 hp - but 2.28 hp on 20% nitro