Noise filtering in Automotive environment

[spamiam]

You all make some excellnt suggestions! I will be putting them to the test sometime soon.

As far as measuring throttle position, it is not so simple.

The carburetors I want to test are SU type. Stop reading here unless you really want to know the details!

SU carburetors work on the "constant depression" principle to make fuel metering more simple. What essentially happens is that the carb choke size is variable so that there is a constant depression (vacuum), and therefore constant air velocity across the venturi. With constant velocity fuel metering is quite constant. Under ideal conditions it is perfectly constant, but real life is not perfect. So, there is a metering needle being used as well.

The constant depression is achieved by having a spring-loaded piston close off the throat of the carbutetor. The piston lifts to a certain level depending on the pressure drop across the piston. More engine vacuum will cause the piston to lift higher. There is a dashpot (dampener) to slow the motion of the piston so that it is not flipping up and down at high frequency, and to provide more vacuum across the venturi for sudden throttle openings which will give the effect of an accelerator pump.

The metering needle is also attached to the moving piston. The needle is tapered so that with more piston lift, there is a narrower needle crossection in the venturi, allowing more fuel flow. This compensates for some of the unresolved non-linearities in the air-fuel metering.

There are 2 issues.

1) Do the pistons hit full lift at maximum engine power. Or do they never hit full open, or hit it too soon. Ideally, the piston will reach full open at full throttle/peak RPM.

2) What is the air-fuel mixture like at each piston position?

The factory had (in the 1960's and 70's come up with a combination of spring strength (holding the piston down) and shape of the metering needle to give good air-fuel ratios at all throttle positions.

But, if you have fitted non-stock intake or exhaust systems, then you may well need some other settings for the springs and needles.

If you have the car on a dyno, then you MIGHT be able to measure the piston lift if you are brave enough to lean over the engine while it is running.

But if you are not brqave enough, or can not afford dyno type, then you have to do a driving dynoamometer with an Oxygen probe up the tail pipe, and a sensor to measure the piston lift as you drive.

So, I have designed a little doodad that fits where the dashpot piston usually goes and this is connected to a linkage. On one of the pivots of the linkage, I have a pot which acts as a voltage divider..

I can calibrate the pot with various spacers under the piston, and then I am able to read piston lift dynamically. Probably one second response time is adequate. I suppose that this would correspond to a time constant of about a quarter or a third of a second to get a reasonable ability to read the piston lift dynamically.

On the desk this device works well. I will soon try it in the car!

-Tony

[pdubinsky]

[spamiam]