MadBill wrote: ↑Thu Jun 28, 2018 10:33 pm
The first question could be "are we referencing AFR as measured by air and fuel turbines or via accurate wide band sensors?"
If they're reading differently then you've got a calibration issue.
No such thing as a accurate WB sensor, not unless it's been calibrated against something else like a mass flow sensor or 5 gas emissions box..........or both.
Ummmm, Joe. I thought we and others discussed this, quite a while ago. Measurements are not accurate without understanding what you are trying to measure.
Otherwise it is blah blah blah.
AFR, Lambda, BSFC, BSAC. All lies to the novice.
Let's all have a good convo about all of these lies.
Let us not forget about dyno vs, track lies.
Heat is energy, energy is horsepower...but you gotta control the heat.
-Carl
Delivery efficiency and output may change between different induction systems (carb vs efi ect.) But the required afr, as measured in the exhaust tract won't unless the fuel is changed.
Calculated may change, but it is tough to get a true calculated measure without some very expensive equipment that most do not have.
Have tested this back to back carb vs efi vs differing injector locations.
Even on carbureted engines you can see different power levels with O2 readings that lay on top of each other.
In one case I saw 20 HP difference from carb to carb on same engine, same day, within 2 hrs (emulsions/bleeds/boosters) while 02 readings were within normal test repeatability window.
If I'm reading the answers correctly; it seems that tuning with afr is unreliable at best, which leads to more questions.
Given the same original engine and inlet scenario:
Why would we analyze and/or adjust the tune for each cylinder using afr's?
Why would we conclude a safe/unsafe tune based on afr?
How would we know if a carb or injection inlet design is truly better than another? Or better said: How do we assure that the tune for induction system-A and the tune for induction system-B are at their maximum performance level for proper comparison?
In-Tech wrote: ↑Fri Jun 29, 2018 4:34 am
Ummmm, Joe. I thought we and others discussed this, quite a while ago. Measurements are not accurate without understanding what you are trying to measure.
Otherwise it is blah blah blah.
AFR, Lambda, BSFC, BSAC. All lies to the novice.
Let's all have a good convo about all of these lies.
Let us not forget about dyno vs, track lies.
OK lets just imagine that you're an electronic engineering guru(you're not)....and you're building a WB from scratch.
Next question......How are you going to build it so it sort of reads right if you're.........
NOT going to calibrate it against either mass air and fuel sensors on and engine dyno OR
NOT going to calibrate it against a 5 gas emissions box
OR BOTH.
Its output would be totally guesswork?
Is guesswork good enough for you?
It's NOT to me.
In-Tech wrote: ↑Fri Jun 29, 2018 4:34 am
Ummmm, Joe. I thought we and others discussed this, quite a while ago. Measurements are not accurate without understanding what you are trying to measure.
Otherwise it is blah blah blah.
AFR, Lambda, BSFC, BSAC. All lies to the novice.
Let's all have a good convo about all of these lies.
Let us not forget about dyno vs, track lies.
OK lets just imagine that you're an electronic engineering guru(you're not)....and you're building a WB from scratch.
Next question......How are you going to build it so it sort of reads right if you're.........
NOT going to calibrate it against either mass air and fuel sensors on and engine dyno OR
NOT going to calibrate it against a 5 gas emissions box
OR BOTH.
You could at least calibrate it at one point (stoichiometry) by checking against a narrow band lambda sensor, which needs no calibration and indicates stoichiometry with high precision. Of course that only calibrates offset and not the slope. I am no expert on WB sensors but having the sensor spot-on at 14.7, the slope would have to be wildly inaccurate for readings at say 12.5 to be unreliable.
EDIT. A bit of research shows that the WB sensor incorporates a narrow band sensor (Nernst cell) and measures deviations from stoichiometry by measuring how much "pumping" of oxygen ions is required to balance the Nernst cell (balance = stoichiometry). It would be reasonable to assume therefore that the WB sensor (when functioning correctly) is accurate at stoichiometry. The slope can be checked using a "free air" calibration technique. http://blog.innovatemotorsports.com/o2- ... important/
They're naturally calibrated at a lambda of 1.
The user does a user calibration at the lean end (free air) to give 20 to 1 or whatever it maxes out at.
There is never any calibration at the rich end which means it's floating......or totally guesswork.
If you've got nothing to compare it with then you'll never know.
When these instruments were first developed, they were for use in lean burn engines which means that you're only using them between the lean end of the scale and about lambda 1.
As far as any "slope" goes, it's not.
At lambda 1 the pump cell changes polarity so there's no straight line to try to extrapolate.
Ok here goes probably with more ? than anwers. using a carb with good calibration and optimal droplet size one would think that the intake charge cooling from the latent heat used to vaporize the fuel would be a big advantage.But how about the fact the fuel has to displace some air thus removing oxygen from the equation?This is not even allowing for fuel separation which can harm burn characteristics.Simple answer to me is it boils down to the particular carb and intake setup vs. another particular injection and intake setup I don't think there will ever be a clear overall winner.
Years ago some SpeedTalker (think it was David Redszus) referred to spreadsheets he had done for several fuels, merging the evaporative cooling and charge displacement effects of various percentages of vaporization to determine the combined effect on mass air flow. AIR, the displacement effect generally (always?) outweighed the cooling.
FWIW, alcohol WoO cars all seem to used 'down nozzles' at the intake valves, suggesting they are more concerned about minimizing displacement.
AFR required for peak power with each carb or EFI will vary because of the difference in distribution to the cylinders and how well the fuel is homogenized with the incoming airstream. The more efficient the engine and induction is the less it is affected, engines with less than ideal to poor induction will require more fuel to get the lean cylinders rich enough, making others too rich, combustion quality suffers. I had a customers car pick up a tenth in the 1/8 with primarily a booster change, minimal calibration change, the new boosters atomized the fuel better and distributed more evenly and in a more homogenous mix. Alternatively my SB2.2 doesn't care much, I've had from a 2.200 throttle blade to a 2.800, and minimal difference between them. Smaller runs a little slower, pumping losses come in to play.
Mark Whitener
www.racingfuelsystems.com
____
Good work isn't cheap and cheap work can't be good.
cab0154 wrote: ↑Fri Jun 29, 2018 1:52 am
Not sure why anyone would target a given AFR. Its just a number. If you went from a junk carb to a carb with great emulsion and atomization you could theoretically get the same AF from both with the great carb using less fuel volume but making more power; due to far greater combustable surface area of the mixture.
OK, been running my dyno since 1975 and lets dispel some myths. Making no changes but changing carbs and going from one droplets size to another and setting up for AFR, you may find that large droplets dont always completely burn and a fully atomized system will burn more fully.
reed
cab0154 wrote: ↑Fri Jun 29, 2018 1:52 am
Not sure why anyone would target a given AFR. Its just a number. If you went from a junk carb to a carb with great emulsion and atomization you could theoretically get the same AF from both with the great carb using less fuel volume but making more power; due to far greater combustable surface area of the mixture.
OK, been running my dyno since 1975 and lets dispel some myths. Making no changes but changing carbs and going from one droplets size to another and setting up for AFR, you may find that large droplets dont always completely burn and a fully atomized system will burn more fully.
reed
Were the carbs the same size (area)?
Did the large droplets produce different afr's than the fully atomized system initially? If so; did you adjust the tune to get back to the same afr's? If you did get back to the same afr's - was there an apparent change in power?