What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Fuel Cools

DanH

Legacy Member
Mentor
In another thread we saw...

If it (ed; a fuel injection nozzle) is not fully functioning, the decrease in fuel need paradoxically causes heat, as fuel cools.

...and rather than drift the thread unmercifully, I thought it might make a good topic.

"Fuel cools" is one of those common beliefs, and although true after a fashion, it is interesting to quantify. I missed physics class, so ya'll check me. Quick 'n dirty, we have:

The BTU content of gasoline is about 114,000 per gallon. In very rough terms, about 30% of that BTU content gets wasted to the air by the cooling system, or about 34,200 BTU per gallon of consumption. That's the heat we're trying to dissipate for every gallon of gas we burn.

The latent heat of evaporation for gasoline is about 900 BTU per gallon. That's the energy absorbed in the phase change from liquid to vapor, assuming we start at about 60F. That heat comes from the local environment, in this case (to some degree) the cylinder heads. Reality is that it mostly cools the charge, but hey, lets pretend it's all CHT reduction.

So, on a per gallon basis, about 2.6% of heat dissipation (900/34200) comes from evaporation of fuel.

If we richen fuel-air ratio 20% (say from 10 GPH to 12 GPH) without changing power output (it actually goes down a tad), and assuming we evaporate all of it, we have 900 x 1.2 = 1080. 1080/34200 means evaporative cooling is now a whopping 3.2%, or 0.4% additional cooling.

I conclude that extra fuel doesn't cool..not really. For sure it lowers combustion temperature, and HP at the same time. Get real lean, and the result is the same.

 
I think the saying "fuel cools" is rooted in the second aspect you point out - that the temperature of combustion is lowered and therefor CHT is lowered. Actual vaporization and direct cooling by that route is lost in instrument error, even with very good controlled conditions.
 
Greg, please expand on your point that temperature of combustion is lowered

See Dans chart at the top of this page. The absolute peak temperature of the combustion is lowered only very slightly with a rich mixture, but more importantly the point in the downstroke of the piston where that peak heat (and pressure) occurs is moving. Heat transfer to the cylinder is a function of both temperature and pressure as well as time exposure. You get less total heat transferred to the cylinder with a rich mixture, and the richer you get the cooler you get when operating on the rich side of peak.

It's not the physical boiling of the fuel itself that results in a cooler cylinder when rich - it's the properties of the combustion cycle in the cylinder that change with a rich mixture.

As a point of pure chemistry, peak flame temperature is reached with a complete carbon burn of the hydrocarbon chain. The hydrogen atoms split off the hydrocarbon chain and burn first and fastest, but produce quite a bit less total heat energy than the slower burning carbon. If you have enough oxygen to consume all of it, and enough time to let it happen, you get complete combustion and maximum heat. In the real world a good percentage of the carbon is still combusting as it leaves the combustion chamber through the exhaust valve, which you see reflected in the EGT changes.
 
Last edited:
. . .In the real world a good percentage of the carbon is still combusting as it leaves the combustion chamber through the exhaust valve, which you see reflected in the EGT changes.
And, in the carbon build up over time as that unburned carbon clings to those internal valve seats, piston tops, etc.
 
Brad, EGT, CHT, and valve temperature are all reasonable proxies for actual combustion temperature, since we're interested in the cooling result of mixture change. The chart in the first post is a generalization, taken from (IIRC) a Lycoming guide. Compare it with this from 1942, a measurement of exhaust valve temperature vs mixture...



....an old chart, raw data from Lycoming, mixture vs EGT (with some personal notes)...



...or a detonation test, again Lycoming, with fuel flow at the bottom:



I always liked the 1942 research paper, as it shows exhaust valve temperature in lock step with EGT, just as you might expect.
 
Last edited:
I always liked the 1942 research paper, as it shows exhaust valve temperature in lock step with EGT, just as you might expect.

True, but peak valve temperature is shifted a bit to the rich side of peak EGT which follows the CHT trend more than the EGT trend. The valve is cooled by the cylinder head.

Skylor.
 
Good stuff Dan, I knew you were not going to leave the other thread alone...

It is always interesting to contemplate that the actual peak temperatures inside the combustion event are on the north side of 4000F...
 
My understanding of the "cooling" effect of a rich mixture is that the speed of the flame front is slowed by the rich mixture as heat is absorbed by the greater number of fuel droplet to be vaporized. The slowing of the spread of the flame front changes where peak pressure occurs and lowers it because it happens later in the downstroke. Correct?
 
Since we've established that extra fuel doesn't do much to cool and I know many are pulling back rpm/MAP and climbing LOP to keep the CHTs down, how about exploring what else is hurting cooling/ heat transfer?

I've seen pretty heavy paint applied to new engines. I'm thinking that cannot be helping heat transfer. Someone want to build a test rig and quantify rate of heat transfer of a surface with bare metal, polished, sand blasted, thin flat black oxide type coating and heavy painted surfaces? Or maybe someone already has that data published somewhere?
 
Since we've established that extra fuel doesn't do much to cool ...SNIP

It does indeed cool, but not significantly through the heat absorbed to vaporize the fuel. It is via modification of the combustion behavior - somehow...

A full rich climb will keep cht's down, but running LOP during <75% power climbs might be a better route. Might require FI, might require EI. There are different experiences/situations out there.
 
John Deakin has a good write-up on this topic.

Those Pelican's Perch articles are really chock-full of good info...

Relevant to this thread, I copied this nugget:

Our POHs instruct us to use full-rich for takeoff. The extraordinarily rich mixture is required to assure that detonation does not occur. The conventional wisdom is that the purpose of the "excess" fuel is to cool the engine, but in fact its primary purpose is to slow the combustion rate and delay the PPP, which eliminates the risk of detonation by reducing the pressure peak. This does, in fact, result in cooler operation, but that's actually a second-order effect of the delayed PPP. (If we could just retard the ignition timing for takeoff, we wouldn't need to throw all that extra fuel at the problem.)
 
I submit that we can do the same thing usually running LOP.

Today with EIs, we can retard timing for takeoff which also allows people to safely run LOP for most of the flight. I have a couple customers running our EFI/EI who run LOP even for formation work and the climb. They don't have CHT issues and it saves a lot of fuel compared to 200+ ROP where you are safer from detonation than 50-100 ROP.

From an older thread, you can see how AFR affects flame speed: http://www.vansairforce.com/community/showthread.php?t=140596&highlight=timing+lop
 
Last edited:
Those Pelican's Perch articles are really chock-full of good info...

Relevant to this thread, I copied this nugget:

Our POHs instruct us to use full-rich for takeoff. The extraordinarily rich mixture is required to assure that detonation does not occur. The conventional wisdom is that the purpose of the "excess" fuel is to cool the engine, but in fact its primary purpose is to slow the combustion rate and delay the PPP, which eliminates the risk of detonation by reducing the pressure peak. This does, in fact, result in cooler operation, but that's actually a second-order effect of the delayed PPP. (If we could just retard the ignition timing for takeoff, we wouldn't need to throw all that extra fuel at the problem.)

All true I'm sure, but "extraordinarily rich" is (1) hyperbole, not a specific quantity, and (2) certainly not necessary for a NA Lycoming. My 390 for example (8.9 CR), is perfectly happy at about 150-170 ROP, full throttle, and obviously I have no trouble cooling it.

We need not guess. Wikileaks published the detonation charts a few years back ;)
 
Those Pelican's Perch articles are really chock-full of good info...

Relevant to this thread, I copied this nugget:

... (If we could just retard the ignition timing for takeoff, we wouldn't need to throw all that extra fuel at the problem.)

Well, we can and do retard timing for takeoff with an EI that allows a user defined curve. This is one of the emerging benefits of EI and a lot of credit goes to Ross and his SDS products for giving us the option. I've been pulling timing out at high MAP for a while now with no loss of power. Though I've been running 100LL exclusively, I intend to make the move to straight automotive fuel when the Rocket flies again. I suspect that the appropriate ignition schedule will allow detonation free ops even with the very high ambient temps we see here in the Mojave desert.
 
I don't understand the distinction between cooling and lowering the temperature. Must cooling be due to evaporation?

it lowers combustion temperature, and HP at the same time

It doesn't lower horsepower at the same time. If you look at the first chart, as you richen from peak EGT and temps go down, power increases and only drops back to the same as peak EGT at around 350 ROP.

It's actually chemistry, not physics. The key term is "enthalpy of formation", and it lets you work out how much energy you get from a reaction.

Combustion to CO2 releases more than 3 times more energy (temperature) than combustion to CO. However, pressure (which translates to power) is a function of temperature AND the number of gas molecules. If you are O2 limited i.e rich you can make twice as much CO as CO2. If you're making CO instead of CO2, you get the same power at a lower temperature due to the additional molecules.

I just had my 30 year high school reunion :eek: so this is a bit rusty, but lets see if I can still do stoichiometry. Feel free to correct. I will multiply all quantities in the equation by 20 to avoid fractions.

Combustion of octane:

20 C8H18 + 250 O2 -> 160 CO2 + 180 H20
Energy released is 102352
Number of molecules produced is 340

If we add 20% more fuel, same amout of O2:

24 C8H18 + 250 O2 -> 92 CO2 + 100 CO + 216 H2O
A lot of energy goes out the exhaust in the form of CO
Energy released is 94522
Number of molecules produced is 408

So with 20% more fuel we produce only 94522/102352 or about 92% of the heat, so temperature is lower. The additional molecules of CO boost the pressure. I think combustion to CO is much faster than combustion to CO2, so the reaction is faster which also boosts power.

If we return to the chart we can see the result. Stoich mixture is 14.7:1 or 0.068. 20% more fuel is about 0.082.

Stoich is very close to peak CHT, and a little rich of peak EGT.

0.082 is close to best power, and 130? ROP

20% more fuel from stoich mixture gives us a few percent MORE power, and from calculation about 8% LESS heat. For the SAME power you are probably looking at 10-15% less heat. I guess that is why people say fuel cools.

Out of interest we can pick some AFRs typically quoted from other engines and find them in the chart.

Stoichiometric ratio 14.7:1 = 0.068
~25 ROP, peak CHT

Best Power 12.5:1 = 0.08
Chart: Max power, ~125 ROP

Best Economy 16:1 = 0.063
Very close to peak EGT

Lean cruise 17:1 = 0.059
Used by some (many?) cars.
Middle of the best economy range. About 40 LOP.

Lean Burn Limit 18:1 = 0.056
~100 LOP? Outside the best economy range.

We can also find where they ran the radials John Deakin etc. like to refer to. The technique I read was to lean until torque dropped 10% from peak. At constant RPM that is the same as 10% power drop i.e. 90% power.
That looks like about .058 on the chart, about the same as lean cruise AFR and 40-50 LOP. But the same manual required advancing the timing for improved valve life...
 
I don't understand the distinction between cooling and lowering the temperature. Must cooling be due to evaporation?



It doesn't lower horsepower at the same time. If you look at the first chart, as you richen from peak EGT and temps go down, power increases and only drops back to the same as peak EGT at around 350 ROP.

It's actually chemistry, not physics. The key term is "enthalpy of formation", and it lets you work out how much energy you get from a reaction.

Combustion to CO2 releases more than 3 times more energy (temperature) than combustion to CO. However, pressure (which translates to power) is a function of temperature AND the number of gas molecules. If you are O2 limited i.e rich you can make twice as much CO as CO2. If you're making CO instead of CO2, you get the same power at a lower temperature due to the additional molecules.

I just had my 30 year high school reunion :eek: so this is a bit rusty, but lets see if I can still do stoichiometry. Feel free to correct. I will multiply all quantities in the equation by 20 to avoid fractions.

Combustion of octane:

20 C8H18 + 250 O2 -> 160 CO2 + 180 H20
Energy released is 102352
Number of molecules produced is 340

If we add 20% more fuel, same amout of O2:

24 C8H18 + 250 O2 -> 92 CO2 + 100 CO + 216 H2O
A lot of energy goes out the exhaust in the form of CO
Energy released is 94522
Number of molecules produced is 408

So with 20% more fuel we produce only 94522/102352 or about 92% of the heat, so temperature is lower. The additional molecules of CO boost the pressure. I think combustion to CO is much faster than combustion to CO2, so the reaction is faster which also boosts power.

If we return to the chart we can see the result. Stoich mixture is 14.7:1 or 0.068. 20% more fuel is about 0.082.

Stoich is very close to peak CHT, and a little rich of peak EGT.

0.082 is close to best power, and 130? ROP

20% more fuel from stoich mixture gives us a few percent MORE power, and from calculation about 8% LESS heat. For the SAME power you are probably looking at 10-15% less heat. I guess that is why people say fuel cools.

Out of interest we can pick some AFRs typically quoted from other engines and find them in the chart.

Stoichiometric ratio 14.7:1 = 0.068
~25 ROP, peak CHT

Best Power 12.5:1 = 0.08
Chart: Max power, ~125 ROP

Best Economy 16:1 = 0.063
Very close to peak EGT

Lean cruise 17:1 = 0.059
Used by some (many?) cars.
Middle of the best economy range. About 40 LOP.

Lean Burn Limit 18:1 = 0.056
~100 LOP? Outside the best economy range.

We can also find where they ran the radials John Deakin etc. like to refer to. The technique I read was to lean until torque dropped 10% from peak. At constant RPM that is the same as 10% power drop i.e. 90% power.
That looks like about .058 on the chart, about the same as lean cruise AFR and 40-50 LOP. But the same manual required advancing the timing for improved valve life...
Welcome to VAF!

Hang around here.....we would be glad to have you.

My in-flight recorded data agrees with you. My engine can always run cooler ROP than LOP making the same power. Power verified by establishing the same air speed and rate of climb for comparison.
 
I don't understand the distinction between cooling and lowering the temperature.

Andrew, welcome to VAF.

Semantics enter here, as "cooling" can indeed mean different things to different people. Some consider cooling to be the removal of thermal energy from a system, while considering adjustment of combustion temperature to be a method of governing how much thermal energy is added to the system.

It's admittedly my view, as separating the two very different processes helps understanding. For example, I may elect to open my cowl exit door, and run at peak EGT in cruise. The first removes thermal energy (by increasing mass flow), the second adds thermal energy (by increasing combustion temperature). The result is the same CHT.

Must cooling be due to evaporation?

Obviously not. That's the point of the first post...in the big picture, evaporation cannot do much of the required energy removal.

It doesn't lower horsepower at the same time. If you look at the first chart, as you richen from peak EGT and temps go down, power increases and only drops back to the same as peak EGT at around 350 ROP.

You'll have to trust me, but yeah, I know that ;)

Seriously, I'll concede that the statement lacked an important detail, to wit, a reference point. I was thinking best power, and you're suggesting peak EGT.

Starting from peak EGT, yes indeed, increasing fuel increases power until it doesn't. However, a carb/FI shop sets up a fuel delivery device by measuring flows; no RV'er is near peak EGT at full throttle unless the technician made a serious error. In the real world, when folks add fuel to lower CHT ("Gotta drill that carb jet!"), they're fighting a full throttle climb problem, not leaned cruise...and they're starting somewhere around best power.

Of course starting at peak EGT, we can better reduce CHT by leaning. Enriching from peak initially increases CHT.

The key term is "enthalpy of formation", and it lets you work out how much energy you get from a reaction.

Combustion to CO2 releases more than 3 times more energy (temperature) than combustion to CO. However, pressure (which translates to power) is a function of temperature AND the number of gas molecules. If you are O2 limited i.e rich you can make twice as much CO as CO2. If you're making CO instead of CO2, you get the same power at a lower temperature due to the additional molecules.

I just had my 30 year high school reunion :eek: so this is a bit rusty, but lets see if I can still do stoichiometry. Feel free to correct. I will multiply all quantities in the equation by 20 to avoid fractions.

Combustion of octane:

20 C8H18 + 250 O2 -> 160 CO2 + 180 H20
Energy released is 102352
Number of molecules produced is 340

If we add 20% more fuel, same amout of O2:

24 C8H18 + 250 O2 -> 92 CO2 + 100 CO + 216 H2O
A lot of energy goes out the exhaust in the form of CO
Energy released is 94522
Number of molecules produced is 408

So with 20% more fuel we produce only 94522/102352 or about 92% of the heat, so temperature is lower. The additional molecules of CO boost the pressure.

Ahhh, interesting stuff, thanks!

I think combustion to CO is much faster than combustion to CO2, so the reaction is faster which also boosts power.

Merely making the reaction faster doesn't translate directly to power. A big factor in torque (and by extension, power) is the point of maximum cylinder pressure, a matter of crankshaft mechanics. Faster combustion makes it earlier, which is not necessarily beneficial.

Which mixture is faster or slower? This from APS, which suggests that combustion is fastest around stoich. They measure actual in-cylinder pressure on their dyno.

 
Last edited:
welcome to VAF.

Thanks, I have been a long time lurker but have resisted the urge to post until now. I don't have a RV currently, but have ambitions to build a RV-9 (or maybe 7).

Semantics enter here, as "cooling" can indeed mean different things to different people.

I agree, I was primarily going by the opening quote, which referred to a fuel restriction causing hotter temperatures. I am happy with the definition that cooling is the removal of heat rather than preventing the addition of heat.

It means that essentially the only cooling in these engines is via air and oil cooling. But it also means that you are not "cooling" the engine by running LOP :)

I would say net cooling from evaporation of fuel is in fact zero, because all the energy still ends up in the combustion chamber.

Seriously, I'll concede that the statement lacked an important detail, to wit, a reference point. I was thinking best power, and you're suggesting peak EGT.

I believe adding fuel from peak EGT reduces temperature all the way to the point where the engine will no longer run. But yes, after peak power you are reducing both power and temperature. Leaning from peak EGT you are reducing power along with temperature from the beginning.

Merely making the reaction faster doesn't translate directly to power.

I think you're right. If you have variable timing you can change the point of peak pressure so the speed of reaction doesn't matter but the best power mixture is still the same.

I guess it must be where the peak of energy times gas molecules occurs. I wasn't sure about this, I don't know how to calculate the change of temperature from the energy released - only the relative energy numbers.

I'm not an expert on this stuff, I just started with the John Deakin articles when they were on Avweb and have been researching from there. (Plus high school physics & chemistry.)

Which mixture is faster or slower? This from APS, which suggests that combustion is fastest around stoich. They measure actual in-cylinder pressure on their dyno.

Hmmm.... no-one else seems to say fastest combustion is at stoich. Everyone else is in the range around 1.1-1.2 (0.075 - 0.082).
 
As Andrew stated, The APS data does not agree with what's been published for decades in various texts on combustion with regards to flame speed vs. AFR. Who's correct here?

The methodology of using pressure transducers to measure flame speed needs to be sound. In the "old days" this data was obtained mainly with high speed photography through a quartz window although they had pressure transducers as well.

flamespeed33_zpsetkdixdl.jpg


This data from a well known text shows highest flame speed at around 11 AFR- very far from stoich.

I suspect that people assume the charge is homogeneous which may very well not be the case in all instances. With stratified pockets of rich and lean, local flame propagation will vary.
 
Last edited:
What is that text?

I'll poke around in Taylor.

Liston. He quotes work from Taylor, Hersey, Wood, Purdue labs, Weigard, Eberhardt, Hottel.

I'm running around in my head validity of the pressure method to determine flame speed. Can we assume that at a fixed rpm and fixed ignition timing, if we achieve PCP earlier in the cycle, does that conclusively prove that flame speed was fastest at that AFR? I am thinking that more/ less overall energy will be released with changes in AFR but that the rate of release changes the pressure vs. piston position curve. An interesting thing to ponder.
 
Last edited:
Can we assume that at a fixed rpm and fixed ignition timing, if we achieve PCP earlier in the cycle, does that conclusively prove that flame speed was fastest at that AFR?

Interesting question. I don't know the answer.

There's probably also a difference between flame speed and burn time, i.e. when the mixture starts to burn and finishes burning. Airguy in post #5 was describing the reaction. I suspect first H goes to H2O, then C to CO and finally the CO to CO2. When rich, the reaction stops when you run out of oxygen. So the first reaction to CO probably goes at similar speeds. It is possible excess oxygen makes it faster but I am really just speculating.

Delaying peak pressure using mixture may be unwise, as slower burning is associated with higher risk of detonation. Adjusting the point of peak pressure is normally done with spark timing. Engine manufacturers work hard to maximize the combustion speed to allow higher compression etc. without detonation.

An interesting article covering that here:
http://www.contactmagazine.com/Issue54/EngineBasics.html
 
Interesting question. I don't know the answer.

There's probably also a difference between flame speed and burn time, i.e. when the mixture starts to burn and finishes burning. Airguy in post #5 was describing the reaction. I suspect first H goes to H2O, then C to CO and finally the CO to CO2. When rich, the reaction stops when you run out of oxygen. So the first reaction to CO probably goes at similar speeds. It is possible excess oxygen makes it faster but I am really just speculating.

Delaying peak pressure using mixture may be unwise, as slower burning is associated with higher risk of detonation. Adjusting the point of peak pressure is normally done with spark timing. Engine manufacturers work hard to maximize the combustion speed to allow higher compression etc. without detonation.

An interesting article covering that here:
http://www.contactmagazine.com/Issue54/EngineBasics.html
I am just a slug out here who built an airplane for my recreation and education and then once it was completed now fly it for personal pleasure. Consequently I am by no means an expert on this topic. In fact I would consider this still part of that education part of building my own airplane. Never the less, this statement is really not making sense to me. Andrew, you state:
Delaying peak pressure using mixture may be unwise, as slower burning is associated with higher risk of detonation.
This is what is confusing me since it was my understanding the lead added to our fuel is put in for the purpose of slowing the flame front down to prevent detonation. So, Andrew, Dan, anyone else, please comment on this concept so that I can get the cobwebs out of my brain.
 
Delaying peak pressure using mixture may be unwise, as slower burning is associated with higher risk of detonation.

Not with our engines, given standard timing. No speculation. The detonation tests are based on a mixture sweep, and they don't detonate at very rich or very lean mixtures, both slow burning.
 
Last edited:
Not with our engines, given standard timing. No speculation. The detonation tests are based on a mixture sweep, and they don't detonate at very rich or very lean mixtures, both slow burning.
Thanks for swinging that straw broom above your head and clearing the cobwebs for me! :D
 
Slow burning mixture promotes detonation?

Ignoring a specific engine type and what made the mixture burn slowly, the slower the flame front the longer the end gasses are exposed to the radiant heat and pressure of combustion. So this may be why slow burning mixtures, in general, increase the risk of detonation.
Cheers
Nige
 
Ignoring a specific engine type and what made the mixture burn slowly, the slower the flame front the longer the end gasses are exposed to the radiant heat and pressure of combustion. So this may be why slow burning mixtures, in general, increase the risk of detonation.
Cheers
Nige

Pressure is the key. Given short 20~25 BTDC timing, for example, a slow burn mixture just delays peak pressure until further ATDC, which means pressure is then a lot lower, so no detonation. For us, the risky thing is a faster burn, pushing peak pressure closer to TDC while at high power. But you know that part.

You're not wrong. Let's assume Joe Redneck hogs out the ports in his stump jumper (lowering velocity), adds some high dome pistons (raising compression and blocking chamber swirl), and throws lots of fuel at it because after all, "Ya gotta have fuel to make power!". Given little charge motion and a slow burn rate, he then finds he also needs lots of advance to make power, i.e. move peak pressure back where it belongs. Given the same RPM as our Lycoming, 40 BTDC means twice as long for radiant heating, so when he stabs the throttle (high MAP), "the valves rattle".

Returning to the question of "What mixture burns fastest?", here's a good paper from 1940. Rather than me posting images, everybody just download the whole paper here:

http://naca.central.cranfield.ac.uk/reports/1940/naca-tn-772.pdf

The Liston curve Ross posted and NACA 772 generally agree. Neither matches the APS illustration, which shows the fast mixture as 40 ROP. That's about 0.07 AF on the dyno reports, or 14.3:1. A stoichiometric mixture for gasoline is generally assumed to be about 14.7:1 air-fuel, or 0.068 FA, just a little leaner. If I recall correctly, 100LL is closer to 15:1.

The APS illustration does appear to be correct if we ignore the 40 ROP note. Mixture leaned to (somewhere on) the best power side of peak is the fastest, full rich and LOP are slower.

The Liston illustration places the fastest burn run rate around 0.092 FA, or roughly 11:1 as Ross stated. NACA 772 doesn't actually attempt to find the absolute fastest burn rate, but rather plots six data points, with 0.082 FA, or 12.2:1, being the fastest measured.

It's important to note that the test cylinder was from a Wright Cyclone (think B-17 bomber), so the researchers were interested in power vs fuel economy. The data points focused on the lean side, with a very large spread between 0.082 and the richest data point, 0.118 FA. Thus I would not take 0.082 FA as the absolute fastest, just the fastest of the six data points. It's entirely possible that the addition of 11:1 to the test would have moved the curves a bit.

There's a heap of other good stuff in this old paper, so enjoy.
 
Last edited:
Interesting data Dan and your second paragraph had me chuckling.

The big question is why the APS data is so far away from other data points?
 
And now we know who "Joe Redneck" was! I bet there was a time when our buddy Dan didn't fancy himself quite so smart and just HAD to experiment with hoggin' out them ports to see if he could make the valves rattle. Giddy with the failure mode I bet he was.
 
Returning to the question of "What mixture burns fastest?", here's a good paper from 1940.

Very interesting, thanks for posting it.

The big question is why the APS data is so far away from other data points?

I think the second chart in figure 3 is the equivalent to the APS chart.

Looking at the APS charts I think they were adjusting the manifold pressure to achieve constant power, whereas the NACA chart they used a constant throttle setting.

All the other NACA pressure curves were below the best power curve, whereas the APS curves were above the best power curve in the later part of the cycle - which they must be to achieve the same power.

So they were measuring different things, which might explain the different result.

it was my understanding the lead added to our fuel is put in for the purpose of slowing the flame front down to prevent detonation.

I did a bit more research on the lead reaction, this is my understanding:

As heat and pressure act on the unburnt mixture the fuel molecules start breaking down. The new compounds formed are unstable and can spontaneously combust, which is detonation.

The lead combines with the unstable molecules and prevents them from combusting until the main, hotter flame front reaches them.

Faster burning is anti-detonation because it reduces the time unburnt fuel is exposed to the heat and pressure, and therefore reduces the opportunity for the unstable compounds to form.

Slower combustion is part of the equation - heat is the other. Slow and hot is the problem.

I suspect that a large part of the octane rating of fuel is due to how easily the specific hydrocarbons break down.

Not with our engines, given standard timing. No speculation. The detonation tests are based on a mixture sweep, and they don't detonate at very rich or very lean mixtures, both slow burning.

A detonation map would be multi dimensional, covering manifold pressure, RPM, mixture and air temperature at least (assuming spark timing is fixed).

I would wager there are conditions where the engines will detonate, probably achievable on an engine with a constant speed prop. I doubt that 2700 RPM WOT is the worst case. Worst case is more likely to be lower RPM.

I would expect Lycoming have the data, but they probably don't share it outside what they put in the operators manual.
 
Andrew,
Your discussion is very interesting. However, for the purpose of sweeping out the cobwebs in my brain, I am afraid your comments show me you are not the man with the broom. You are the spider! :p
 
Looking at the APS charts I think they were adjusting the manifold pressure to achieve constant power, whereas the NACA chart they used a constant throttle setting.

Adjusting for constant power would be consistent with APS teaching, which has the pilot adding manifold pressure to restore lost power after setting a LOP mixture.

So they were measuring different things, which might explain the different result.

Probably not.

Increasing manifold pressure (or compression ratio) does increase flame front velocity. However, when quantified, a few inches Hg just isn't going to shift peak pressure very much. Here's a relink to a previous post from Ross. A move from (for example) 20" to 22" would only bump flame speed about 1 ft per second.

http://i1105.photobucket.com/albums/h341/rv6ejguy/mapflame_zpsy4iawhpj.jpg

Not that it matters, because it doesn't apply here. The mysteriously fast mixture in the APS illustration is, well, fast. If APS was adjusting MAP to bring power up, they would be doing it for the slow burning LOP curve, not the one already generating the highest pressure.

I would wager there are conditions where the engines will detonate, probably achievable on an engine with a constant speed prop. I doubt that 2700 RPM WOT is the worst case. Worst case is more likely to be lower RPM.

I would expect Lycoming have the data, but they probably don't share it outside what they put in the operators manual.

Correct on 2700/WOT.

Re sharing, a university study got leaked some time back. It's since been pulled from the net. There are some charts in this post, as well as a comparison to the "limit line" on a standard power chart:

http://www.vansairforce.com/community/showpost.php?p=566697&postcount=81

The FAA's Swift fuel study is another good source.
 
I knew there was something about Dan I liked !

Here is my own weapon of choice....

Trease.JPG




Naa, no stump jumpers for me. I liked bikes. Faster, and the chicks had better teeth.

[/QUOTE]
 
Back
Top