Carburetor jetting?

[globi5]

I'm trying to understand jetting on a carburetor for a 125cc 2 cycle engine.
Apparently the diameter of the mainjet can be determined as:
d2=d1*(ad2/ad1)^0.25
(d=diameter and ad=air_density)

First of all, why is it ^0.25 and not ^0.5? If one doubles the diameter of the mainjet, the fuel flow should increase by the 2nd power and not by the 4th power or shouldn't it? Or how is this explainable?

Also air density (dry air) can be calculated with:
ad=p/(T*R)

Now when I use the data from the table, the pressure data calculates the correct mainjet with that formula. But when I use the temperature data, the above formula doesn't work.
Temperature change appears to have a significanlty stronger affect to the jetting diameter than pressure change does.
If the mainjet diameter solely depends on air density this is not possible. (Pressure and temperature change should have a similar affect.)
What else does the temperature affect? Or how could this be explained?

 

[thundair]

http://www.sengpielaudio.com/calculator-airpressure.htm

This is a good page

Regards

 

[globi5]

Just to make sure: I'm only talking about dry air conditions.

Ok, so if the speed of sound goes so does the air flow?

But if you increase temperature air density decreases however air flow increases. So higher air flow should actually reduce the influence of the temperature change to the mainjet diameter rather than increase it or shouldn't it?

I wonder whether this 'stronger influence of temperature' has just to do with this particular carburetor or whether this is a 'phenomena' that can generally be explained?

 

[thundair]

I understood the fx posted on the web page to be D= Pd over Rd*T

If the temp is C+ 273.15 I see that as a constant factor or am I missing something???

Also I am unsure as to the amount of velocity increase realitive to lower density air. That may be a factor that needs to be added to the fx

Regards

 

[globi5]

Yes, D=Pd/(Rd*T)

Pd is pressure and T is temperature. Both are variables and should affect the air density to the same degree. (Rd is the only constant).

However the data I have suggests that temperature has a bigger influence than pressure and I want to find out why?

Maybe the air velocity increases with higher temperature but I don't see why this would affect the jetting to that agree?

Also websites I found suggest that jetting only depends on air density. But I wanted to know whether somebody knows different? (Otherwise the data I have doesn't really make sense.)

 

[thundair]

I will await the answer myself. Maybe someone with a little more background in that area will chime in...

I have worked on an analog injector system and the input that related to this thread was a hot wire which by the very nature of it would read different by velocity and/or temperature. I only mention this as it confirms your statement about the higher airflow reducing the temperature influence..

Also with this system air pressure was not one of the sensors nor ever calculated into the system. Looking back maybe it should have been.

 

[globi5]

A hotwire should measure air density directly. The hotwire actually does measure pressure indirectly. A higher pressure will cool the hot wire more and therefore indicate a higher air density.

A perfect carburetor should do the same. But to adjust for different air densities the mainjet needs to be adapted.

I simply don't understand why the mainjet in this case doesn't appear to relate directly to the air density. I wonder whether this always like that or just in this particular case.
(I'm obviously not a carburetor expert.)

 

[patprimmer]

I think you are taking to simplistic a view of the carby.

When dealing with mechanical fuel injection, you can correct fuel flow directly to air density by changing jet cross sectional area, provided you are not approaching supersonic air or fuel flow speeds, and so long as boundary layer breakaway is not a problem.

In a carby, an increase in air density also increases the signal strength to draw fuel. Also, as some signal is used to lift the fuel from the float bowl level to the discharge nozzle, the fuel flow rate to air flow rate is not linear.

This is further complicated by "other" fuel flow circuits in the carby and air bleeds and emulsion tubes.

This will make it difficult to make an accurate mathematical model, so plotting the results of trial and error might be the quickest and most reliable method once some data is collected in practise sessions or on the dyno.

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[globi5]

Thanks for your response.

So you're basically saying that every carburetor with a given engine behaves differently, right? Does that mean that all race car engineers need(ed) to make up tables by trial and error for any given carburetor and engine?

Also, is it possible that the temperature increase somehow increases fuel flow in a carburetor? Since at equal air density but higher temperature and lower pressure a leaner mainjet is required according to the table I have.

I do have a table for one engine (which must have been empirically determined), but I wanted to know whether I could use that data to make up a simple mathematical model and then generate a table for another engine that uses the same carburetor (but generally runs richer)?

 

[patprimmer]

No

Many carbies react in a similar manner as many designs are similar. I am saying that there are a lot of non linear variables that makes accurate mathematical modelling difficult.

For example:-

You are looking only at the main fuel discharge nozzle. All other air bleeds and fuel circuits are disabled.

You start with no airflow then increase it until fuel starts to flow. From that point on, the fuel flow will conform to a formula, and you can modify the formula to accommodate the fuel flow starting point. You then add the idle circuit. It draws fuel through the main jet, robbing it from the main discharge circuit. The fuel flow there starts virtually instantly at closed throttle, but falls off with throttle opening. The transition circuits act somewhere between the two.

You then add air bleeds and emulsion tubes. They respond at a different rate to the main circuit, but influence it.

The accelerator pump discharge nozzles might supply some fuel at steady running, but it will have a very different supply curve and starting point to the main discharge nozzle. It may have a power valve which will most likely have an on off type action, or may in some cases be a needle valve. What is the taper on the needle, and is it uniform?.

Does the fuel level in the float bowl fall with speed and how do you model that?

Now we come to pulsations in the airflow and fuel stand off. That is about all I can think about for now.

Once you have collected actual data, you can see a pattern evolve and you can guess accurately at points between the dots. This data will quickly adapt to other reasonably similar engines with reasonably similar carbies, but will not adapt to very different situations.

By similar I don't mean similar brand names or V8 vs V8, but similar number of cylinders drawing from a carby, similar tuned lengths, similar valve timing, similar cylinder size, similar compression ratio etc.

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[globi5]

I thought about the issue again and found a very simple answer, why at least these relatively simple carburetors flow disproportionally more fuel at higher temperatures. With higher temperatures vapor pressure of gasoline goes up, so more fuel will automatically evaporate at higher temperatures even if the air density and therefore the massflow (airflow) is reduced.

 

[patprimmer]

Higher vapour pressure and higher evaporation rates have no effect on fuel flow rates in a carby, however fuel boiling in the float bowl can cause flooding of the carby.

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[globi5]

Could you please elaborate, why higher evaporation rates should not have an effect on fuel flow rates?
These carburetors flow more fuel at higher temperatures and this appeared to be a reasonable explanation. I don't think that the fuel boils in the float bowl, since the carburetor is exposed to the environment and the curve (fuel flow vs. air flow) is linear (but of course I don't know for sure).

All the data I have is at WOT, so the needle shape, the idle or 'accelerator circuit' should not affect that data and even if they do, they're constants anyway.
The curves (fuel flow vs. air flow) from that data are pretty much linear. It's just that the fuel flow vs. temperature curve is way steeper than what it (according to the actual air flow based on air density) should be.

 

[patprimmer]

I would have thought it was extremely obvious that under normal circumstances that the fuel does not start to evaporate until after it leaves the discharge nozzle.

If it evaporates before the main jet, it will lean the mixture unless it boils then it will percolate out the float bowl air vents.

Thinking about it, if it evaporates where it can enter the main jet as vapour, it is boiling in the float bowl.

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[globi5]

In this case the fuel is mixed with air (atomizer) before it exists the discharge nozzle in the venturi. The nozzle (not main jet) in the venturi draws an air/fuel mixture which then gets diluted with more air.
If part of the fuel in the atomizer evaporates more rapidly (because of higher temperatures) the air/fuel mixture might end up being richer at the same airflow.