View Single Post
  #47 (permalink)  
Old 07-15-2016, 07:45 PM
olddog olddog is offline
CC Member
Visit my Photo Gallery

 
Join Date: May 2006
Location: St. Louisville, Oh
Cobra Make, Engine: A&C 67 427 cobra SB
Posts: 2,445
Not Ranked     
Default

What causes engines with a lot of valve overlap (both intake and exhaust open at the same time at TDC on the transition from exhaust stroke to intake stroke) to idle rough is a changing imbalance of gasses in the intake manifold.

The purpose of the valve overlap is to use the momentum of the high velocity mass of exhaust gasses to draw the intake charge through the combustion chamber. This replaces the exhaust gas in the dead space with a fuel air charge. The piston never moves into the dead space of the combustion chamber, and without valve overlap, the exhaust gas would remain and dilute the next intake charge, resulting in less fuel and air in the engine on the next power stroke. So the overlap gets more fuel into the cylinder to make more power.

The problem is that at idle there is little exhaust gas volume (mass) moving at low velocity, thus little momentum to do any work. There is high vacuum in the intake manifold at idle. The intake vacuum actually pulls exhaust gasses into the manifold. This is why the vacuum runs lower. The throttle plate is near closed sealing off most of the atmospheric air pressure, but the manifold is being filled with exhaust gasses typically referred to as reversion flow.

Now depending on which cylinder is at this point and what the firing order is, some cylinders actually see an exhaust pressure pulse from another cylinder opening its exhaust valve on the power stroke. The longer the duration of the cam the sooner the exhaust valves open and with more pressure in the cylinder as the piston is further from BDC. When the cylinder is on the same bank the pressure spike is more pronounced (assuming the two banks tie together). If the two banks are not tied together the pressure spike is not there at all when the cylinder is on the other bank. So the amount of reversion is different on each cylinder.

Now the reversion flow can cause air to flow backward through the carb or throttle plate if EFI. This blows an air fuel mixture up out of the carb. Then it sucks that fuel laden air back into the carb again and pulls more fuel in and causes a richer mixture. But remember the amount of reversion is different for each cylinder, so the richer mix is different for each cylinder.

Now add to this that the firing order is pulling air to different cylinders and the flow in the plenum is constantly changing directions, causing turbulence in the plenum. Each runner is pulling gasses out of the cylinder then the cylinder pulls it back the other direction. Total chaos in terms of flow in the intake.

So now the intake gasses contains pockets of exhaust and pockets of various richness in the fuel, the flow is chaotic. Thus each cylinder gets a different charge in terms of richness and in terms of how much is air/fuel and how much is exhaust gasses. Thus each cylinder makes a different amount of power, giving the rumpty rump we all love to hear.

Also when the reversion flow changes direction, velocity in the runners go to zero and rich pockets of fuel can form droplet on the walls of the runners. Then when velocity returns the droplets get pulled off the wall. As the engine idles it loads up the intake with fuel droplets and the driver will rev the engine to clear it. It will run very rich until it sucks in all the fuel.

Now a single plane verses a dual plane also has an affect as all cylinders tie into one plenum in a single plane intake, and a dual plane keeps the two banks separated. Also the single plane has a shorter path to the carb. The dual plane smooths these affects out a bit more.

Finally the 8 stack and weber. Each cylinder has its very own intake pipe that is not tied to any other cylinders. The single pipe still has reversion flow, and you can see a haze of fuel above them. They also pull the reversion air/fuel back through and richen up the mixture. However without the common plenum, much of the chaotic flows are eliminated. They definitely smooth out the idle of a big cam. If you put individual exhaust pipes on each cylinder, they really smooth out.

Now with multi port EFI and the injectors near the intake valve, you do not pull the same air past the jets and suck more fuel in like a carb. This solves the richness variation. You also do not have the issue with fuel droplets forming and loading up the engine like a carb does. However a MAP system does not work well with low vacuum at idle.

A mass air meter measures the air flowing by it. It is not smart enough to know which direction the air is flowing. So if there is reversion flow past the MAF meter, it measures it twice and reports more air than what actually flowed into the engine. Also reversion flow can coat nasty things onto the meter's heated wires. This can easily be solved by adding extra intake pipe between the MAF meter and the throttle body. However to use one MAF meter, you need a common plenum to pipe to. Same to get an accurate MAP reading, but you can get buy with just connecting tubing to all the runners. This is what makes EFI versions of the 8 stack difficult to do.

A MAF type EFI can be designed to solve all the problems. It would require one MAF meter per cylinder and one O2 sensor per cylinder and a ECU to handle them. I have read of one that is being tested. However the length on the horns required to get a MAF meter far enough away to smooth out the reversion pulse would make your 8 stack look like something other than an 8 stack. It could be designed to disconnect the piping when the hood opened. The cost of 8 MAF meters and O2 sensors and the ECU would be huge. So technically the perfect EFI system can be done, but it will not be worth the cost.
Reply With Quote