2-stroke tuning, squish velocity for race fuel and methanol 1

[bob1111]

Any hard core 2-stroke tuners here? We are doing some extensive research on head designs for high performance 2-strokes. Mostly looking at MSV numbers, squish band dims, etc. We are looking to tune for specific fuels and though the MSV can be calculated, it really means nothing if you don't look at a specific fuel and I can find now good data for say target head specs for methanol. We will mostly be looking concentric squish designs with near hemi bowls.

 

[black2003cobra]

viper - So if you reduce advance with the faster burn velocity, does the engine still knock? I do recall seeing some data on methanol vs gasoline, and the burn velocity of methanol was about 45% faster. (M. Brusstar)

 

[bob1111]

That is precisely where the R&D comes in. Spark advance vs MSV. I would be interested to see any data you might hold regarding burn rates of fuels. I have tried with all fuel companies to get burn data and they all say " we don't test for that". If that is not BS, I don't know what is.

Quiet obviously, the right controlled lab testing of fuels could help a BUNCH in the R&D process but still would not exclude many hours of R&D for the right setup.

 

[black2003cobra]

It depends on equivalence ratio of course, but Brusstar showed methanol to have a peak (laminar) burn velocity of around 44-45 cm/s, and Smax = ~30-31 cm/s for gasoline. Heywood’s book has a graph, which shows the peak for methanol to be about the same as above, but ~35 cm/s for gasoline. Heywood’s data comes from Metghalchi and Keck, (Cumbust. Flame, vol. 48, p. 191.) If you have Heywood’s book, he does give some empirical expressions for burn velocity vs equivalence ratio, pressure, and temperature. There’s also a graph showing burn speed vs turbulence intensity from Groff and Matekunus, (SAE paper 800133). Don’t know if any of that is useful to you, or not.

 

[bob1111]

I appreciate those references. We do not have any of those for data. We only have some private lab tests and our own test data and calculations. One of the things that is misleading is the latent heat involved in alcohols that seems to require an increase in MSV to use the same spark advance. Otherwise, it has been proven in our shop that two identical motors, one on gas, the other on alcohol, the alky motor will require more timing for max power. Obviously, there are other fuel factors involved here like SG, volume, etc. In our testing, we have nearly proved that measures must be taken to better vaporize fuel, otherwise, a bunch of unburned fuel ends up cooling the chamber as well as going out the tail pipe.

 

[black2003cobra]

Odd. I would have expected MBT timing to go down with a faster burn speed, (i.e., less advance). That’s what the quickie simulations I did showed, anyway. Perhaps you’re simply able to get closer to MBTT with methanol due to the higher knock resistance. Or said another way, perhaps you can’t get advance as close to MBTT with gasoline because you reach the knock limit first. And yes, methanol does take more energy to vaporize it. For methanol, Lv = ~1100 kJ/kg vs ~350 kJ/kg for gasoline. Well it’s getting late. Good luck!

 

[patprimmer]

There seems to be strong support for my comment of 18 March.

It is my belief (but I hae no back to back data) that with really good squish, you can cut a degree or two from timing from maximum power, but there are many variables and when increasing squish, I would also normally add some compression and de bur the chamber and smooth off all sharp edges. This is why I have no back to back data.

I also have no back to back on petrol vs methanol flame speeds, but a great number of variables also make tests under real world conditions very difficult. That might explain the variations from text to text.

Methanol will tolerate a lot more compression than most grades of petrol based racing fuel, so engines optimised to methanol might have very high dome pistons that interfere with flame travel and in fact could cause some turbulence that divides the chamber with a squish area in the middle of the camber. I have seen a SBC that needed 60 deg advance until we installed flame slots in the dome. The slots with no other change allowed the timing to be set at 42 deg. Of course the slots slightly reduced compression by removing a few cc from the dome. This was 20 or 30 years ago, but if you want to do the sums, say 2 cc per slot, 16:1 CR with full domes, 4" bore, 3" stroke.

Racing fuel is a different composition to common petrol purchased at the bowser.

Common petrol varies quite a bit according to brand, season, region and grade, so what is the standard composition to be used for comparison in test results you might obtain, and how might that relate to your petrol.

Other variables are latent heat of vaporisation, specific heat, mass of fuel used, surface energy, viscosity, boiling point, vapour pressure, effect of compression on the above, size of jets or nozzles. These all effect evaporation rate, and therefore actual burn rate in the chamber of a running engine.





Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[mattsooty]

Coming at it from a slightly different angle; in an application where the knock limit is reached before MBT an increase in squish is always going to allow you to advance the ignition further, as the knock limit is shifted, thus generating more torque.

However, if using alcohols the knock limit may well be beyond MBT so an increase in squish, in this instance, will allow you to increase the compression ratio in an attempt to regain some BSFC.

An increase in turbulence within the combustion chamber is going to cause autoignition in a positive ignition engine with a premixed charge.

In terms of comparison of flame speed for different fuels one must be careful when comparing laminar flame speeds when the majority of combustion, within a positive ignition engine, is turbulent.

The only real way to make any sense out of all of this is through thermo analysis using incylinder pressure measurements & PV/Pcrank diagrams because, as Pat has already said, there are so many variables.

On another note I must say that I am a little surprised that the OP is not familiar with John Heywoods book, perhaps not enough time spent on the 'R' part of 'R&D'???

MS

 

[bob1111]

Is John Heywoods book the only one in existence regarding fuel combustion? Seems we have compiled a decent list of data without owning the book...

I would say the "R" part of our R&D consists of plenty of dyno time working with different chamber designs. Sometimes theories only get you so far.

 

[mattsooty]

" Is John Heywoods book the only one in existence regarding fuel combustion? "

No, of course its not...but I would say that it is, arguably, the text most often used by professional engineers as reference on the subject matter.


In terms of " Sometimes theories only get you so far." I would tend to agree but I'm afraid that the majority of 'garden shed' engineers I have come into contact with do not have even the most basic understanding of the fundamentals & theories of the IC engine.

That may sound arrogant but I'm afraid that the companies that employ professional engineers, such as some of the members of this forum, spend an awful lot of money on their equipment, training & people to try to get an edge over competitors in an industry that is (was???) worth billions - its as simple as that.


By the way I have just noticed an ommision in my last post: -

" An increase in turbulence within the combustion chamber is going to cause autoignition in a positive ignition engine with a premixed charge.

should say ....NOT cause autoignition...


MS

 

[evelrod]

Hmmmmmm. That's a fairly significant change, hey?

Guess I was not paying attention. I get set in my own ideas and sometimes only 'scan' some of the other contributors posts. I'll try to do better, Matt.

Rod

 

[malbeare]

The fourstroke combustion chamber with the most squish is the may fireball design. It doesnt breath well at higher Rpm but lean AF ratios are possible and high compression without detonation.
cheers malbeare

http://www.jagweb.com/jagworld/v12-engine/page3.html

http://www.eng-tips.com/viewthread.cfm?qid=63761

http://www.me.udel.edu/meeg467/Powertrains/Ch4.ppt

 

[bob1111]

Some books have indicated that indeed, too much squish velocity will induce autoignition in a two stroke engine. TO be honest, our testing so far has been somewhat inconclusive because we have only tested the higher MSVs with their relative CRs. We have not attempted a high MSV with low CR as of yet. Obviously our search here is for the BMT to be as close to zero as possibly while still controlling the flame front.

I realize many companies spend bottomless penny banks doing these things but they are also not designing chambers for exotic fuels that I am aware of. The two stroke chamber is simple in appearance, yet complex in design because of all the variables that can be applied. Dome height, squish band area ratio, angle, MSV, spark kernel height, etc.

Main question in mind is if alky propagates faster than gasoline, why do most engines require additional timing to achieve max power?

Also, regarding squish band width, it has been taught for years that wider gives more bottom end while narrower gives better top end performance. Our testing has not definitively proved that theory either. One thing that is worth investigating is squish clearance compared to squish width. It would seem that the fuel charge trapped between the piston and squish band will not contribute to the performance of the engine, but rather burn MUCH later (well after 20*ATDC) and contribute to destructive burn patterns. Shrinking this distance to near zero seems a no brainer yet many insist that higher clearance is recommended.

Looking at one test engine we use, rod stretch is calculated to only .005" yet we hear this engine should have .040-.050" clearance. Puzzling at times. Our test engine has run MUCH tighter tolerance but with a totally different chamber design so results there are questimates at best.

 

[mattsooty]

Viper,


" I realize many companies spend bottomless penny banks doing these things " Yes, right up until the recession !!!

Not sure what you mean by 'BMT' but if you mean that you are looking for complete heat release to occur at TDC then I think you are on the right track!


" Main question in mind is if alky propagates faster than gasoline, why do most engines require additional timing to achieve max power? "

I would suggest that a contributory factor is that the stoic AFR of alcohols is relatively low & the latent heat of evaporation relatively high such that the charge has a much reduced enthalpy, during compression, compare to an equivalent gasoline/air charge.

" It would seem that the fuel charge trapped between the piston and squish band will not contribute to the performance of the engine "

A classic example of quench and is of detriment to any chamber design and an obvious HC emissions raiser.

MS

 

[bob1111]

Quite obviously, we are working with performance based designs here so the increase in HCs is of little concern.


"I would suggest that a contributory factor is that the stoic AFR of alcohols is relatively low & the latent heat of evaporation relatively high such that the charge has a much reduced enthalpy, during compression, compare to an equivalent gasoline/air charge."

I agree with that! It just makes me wonder what we can do to improve the performance. WE have always run into mechanic limits to obtain ridiculous CRs but often wonder about trying difference chamber designs with CRs as high as 18:1. We currently have only ran 16:1 here with promising results.

 

[j79guy]

Viper: What application is your 2-stroke being used for, and do you need a wide power band, i.e. Motorcycle, with a gear type transmission, or CVT clutches? The snowmobile guys with CVT's have been pushing 17.5:1 compression for a few years now, but obviously, they tune for one RPM. (Clutch shift-out RPM.) Regarding squish, it may bear for you a bit more investigation into the parallelinity between the piston and head within squish band. No one has said that the squish band must be perfectly parallel, and indeed if slightly tighter at the cylinder wall, than at the pocket, may help to avoid the phenomena of autoignition in an otherwise "tight" squish band. Slight mishaps while machining the head's squish bands can and often do happen and end up with the band slightly tighter at the pocket, than the cylinder wall, and as you can imagine, the piston does not last for long.

Robin Sipe.

 

[HKOZ]

I have a retired racing two stoke that ran on Methanol.Squish band is same Pump gas as its about MSV to creat turbulance ,the squish area is about 50%. not sure off compression ration as I havnt cc the head but visualy its about half volume to pump gas head. 28 m/sec is concidered Idea, this is usaly about 30thou clearance. for those that think less is better below this the pumping inefficency robs the engine of power.
you need the "gap" to expell the mixture across to the combution chamber. contry to populare belief retarding the timming increases the power for two reasons ,it alows the power stroke to be still active with greater angle on the crank ,and it delivers hoter gas to the pipe giving faster sound wave reflection allowing higher effective revs(more power)
Methanols main function is its cold burn to protect the engine under high loads .needs captive CR of +16 to 1 above the exhaust port or it wont burn properly.
Hope this helps

 

[tw05tr0k35]

so, given same state of tune, CR and engine dimensions and assuming that none of the incoming air is replaced by the copious amounts of fuel required with alcohol to reach optimum a/f ratio; if no other changes are made except fuel and jetting then we can assume:
1 the compressed alcohol mixture will have a higher density

2 the higher heat of vaporization will cool the compressed gas, piston crown and cylinder head more, thus creating more density

3 the mixture will be harder to light off

therefore we can assume that;

1 higher CR will be needed to heat the gas more to create a comparable ignition point between the gasoline charge and alcohol charge

2 MSV should not change because as i understand it msv's are tailored to operating range (this may be wrong, as i have never seen an actual equation that determines MSV, and the parameters may be changed slightly by the flow of different gas densities, seems the equation is a closely guarded secret and those with the secret have been less than forthcoming with their information sources, any help pointing me in the right direction would be greatly appreciated)

back to my "R"

Tom