Map sensor per port

[NITROPIXIE]

Would it be possible to have an individual map sensor per port, for say running a multi carb setup?? That way no vacuum robbing would happen between each port.

Obviously the voltage signal of each MAP sensor would have to be smoothed electronically, rather than using a vacuum reservoir, which could also be variable with the use of a variable capacitor/circuit. The price of MAP sensors would also increase the cost of running this setup.

I'm not sure this is a real problem anyway or if there is any point to this, but it could help someone.

Anyone else got any views on this???

[dr.occa]

Would it be possible to have an individual map sensor per port, for say running a multi carb setup?? That way no vacuum robbing would happen between each port.

Obviously the voltage signal of each MAP sensor would have to be smoothed electronically, rather than using a vacuum reservoir, which could also be variable with the use of a variable capacitor/circuit. The price of MAP sensors would also increase the cost of running this setup.

I'm not sure this is a real problem anyway or if there is any point to this, but it could help someone.

Anyone else got any views on this???

part of the design w/ some vacuum tubes between carbs was for balancing hence the existence of vacuum balance tubes. but i think i know where you're going with this.

the problem that would arise if you were to successfully setup individual map sensors per cyl is the need for multiple ignition tables to fire how ever many cyls contingent on individual load and at varying timing advance. at least that's how it would make sense to me to take advantage of specialized individual data.

BUT if vacuum is that different from cyl to cyl, then maybe the issue is the carbs not being synced. vacuum is suppose to be balanced between carbs even if it requires "robbing" an adjoining runner. it leads to an overall better running condition for the engine.

also, MJ would have to take each individual load reading, average them out for a single load to base an indicated spark time. in any case, it's going to be "balanced/averaged" out somewhere in the process.

[alexander]

i agree with dr occa, but want to add something, taking a slightly differnet tack on your question.

i think, when you mention 'vacuum robbing' that you are concerned about airflow into an individual runner, through the tube which goes to the vacuum plenum. i think that is an unfounded concern. even though the pressure pulses are transmitted from the each runner, through the tubes, and into the plenum where they are averaged out physically, that doesnt mean that any significant amount of air is flowing in these tubes. that is whole point of making them small tubes: so that the amount of air which can flow through them is very small compared with the amount of air flowing through the venturi of the individual runner.

i glued brass tubes into my intake runners, which had in ID of about 1mm. when i later had a good look at the ID of the vac tube part of the old distributor, it would have been 0.5mm at best. so even very small tubes can transmit the vacuum signal, even they involve air flow which is neglibible by any standard.

regards
alexander.

[Alexander_Monday]

I'm not sure that what you are trying to achieve here would outweigh the complexity.

If you are worried about idle or transition, maybe I am the exception, but I am running a 2110cc 4 cylinder with 44MM intake valves, 37.5MM exhaust valves, very large ports, dual Weber carbs and a very aggressive camshaft with .601 lift / 276 duration @ .050 exhaust and .578 lift / 270 duration @ .050. I would think this combination would cause vacuum problems if any would.

I used small vacuum lines, 1/8" if I remember correctly, tied them together into a 1" PVC tube 14" long. I had these hooked up for several weeks before I got the EDIS installed, and I could not ascertain any difference in idle or acceleration due to the vacuum lines.

[NITROPIXIE]

wow, some good replies there.

I guess for basic operation this sort of system is not needed, but maybe more applicable for datalogging on a race engine or similar where information on each cylinder maybe more helpful.

think i will have to have a play with this and see what sort of results i can get out of it just for an experiment. I will try to smooth the signal electronically between each cylinder instead of using a vacuum reservoir, with something like a variable capacitor circuit so as to change the amount of smoothing (kind of like changing the size of reservoir.

[Dimbit]

Then fit a servo to each idle air screw, write some software and you have self balancing carbs

[alexander]

I will try to smooth the signal electronically between each cylinder instead of using a vacuum reservoir, with something like a variable capacitor circuit so as to change the amount of smoothing (kind of like changing the size of reservoir.

in that case, have a look at the circuit diagrams for a few basic power supplies ie the type which have a transformer, a diode bridge rectifier, then some signal smoothing. the last few components smooth the pulsating DC into DC with only a slight ripple. that is exactly what you would be trying to acheive here.

alexander
sydney.

[Spockie-Tech]

I dont think that simply a big capacitor (linear power supply filtering technique) is the best way to smooth this type of signal..

A power supplys filtering goal is to keep the voltage up by filling in with stored current from the capacitor while the rectifiers output voltage is low.

This would mean that when the MAP sensors output dropped suddenly (you floored it), there would be a long delay while the capacitor discharged back down to the new average level. The more smoothing you applied, the longer the delay.

A bit more complex (but not too much) would be to sample the MAP input with an A/D input on a micro, and then run a sliding average function over just the recent samples.

So you figure out the vacuum pulsations are x milliseconds apart at your current RPM, consider your current sampling rate and average the last Y samples. you discard samples from earlier.

Y is a bit bigger than the number of samples that fit in one vacuum pulsation cycle. Bonus points for varying Y according the current RPM, but probably not necessary

You could even read the peak, 1-cycle average, and average-over-whatever-time (variable damping) all at once, using this technique

Hmm, That doesnt read back so clearly, I hope you get what I mean.. If its muddy, I'll explain better when its not bedtime..

[Dimbit]

Or just use a vacuum reservoir per sensor...

[dr.occa]

...or balance your carbs