Do 8 Stack EFI or Weber Carbs Run Good On The Street?

[FUNFER2]

Originally Posted by Dimis

Great to know! Its always perplexed me how 8 trumpets theoretically can't supply enough air for each cylinder when compared to a single carb

If you don't mind kindly sharing please:
What are your cam specs?
Hydraulic or solid rollers?
and Rpm red line?

Thx


My understanding is a wide lsa is required to limit reversion, and the more vacuum the better the end result. I'm no expert. I defer to greater minds for info on this.

"A wide lsa is required to limit reversion"
How much lobe seperation, was my question.

[G-Pete]

Originally Posted by Gaz64

Anthony,

That's because a single 4 barrel carb supplying 8 cylinders via a single plane intake, has only 3 inlet valves open at any time supplied by a carb of example 750 cfm.

------

In an IR carb application (48 IDA), each cylinder can only draw from the volume within it's own runner, through the carb to the trumpet.

An IDA flows about 330 cfm per barrel at best.

??? 330 cfm/barrel

[CHANMADD]

Just by the nature of design........the shorter the intake runner the better for higher rpm use and the motor produces less torque...........the longer the intake runner ...the engine will produce more torque and hence operate better at lower rpms..And obviously an intake manifold with a plenum also aids in leveling the intake charge and help to produce more torque...

[moore_rb]

Originally Posted by FUNFER2

"A wide lsa is required to limit reversion"
How much lobe seperation, was my question.

there is no magic LSA number to make things all warm and fuzzy...

valve overlap is the real factor, and overlap is a function of LSA and duration taken together.

but like I posted earlier, 112 degrees LSA, or smaller, is typically where most aggressive street/strip cams begin messing with your MAP and your manifold vacuum

114 could be considered a safer, but again- it also depends on how much duration at .050 you are running- a super long duration cam (240+ at .050) is probably still going to introduce overlap instability, regardless of how wide the LSA is...

But when it comes to the question of how well your cam might (or might not) co-exist with EFI, just let your eyes and ears be your guide-

If your cam makes your idle sound like a Harley Davidson, and if your engine shakes the whole car while it's idling, then it's a fair bet that tuning a speed density EFI system to work smoothly with that cam is going to take time, effort, and money.

It will be a little easier if you have a MAFS , but again- no current 8-stack system uses MAF

[twobjshelbys]

Originally Posted by moore_rb

If your cam makes your idle sound like a Harley Davidson, and if your engine shakes the whole car while it's idling, then it's a fair bet that tuning an EFI system to work smoothly with that cam is going to take time, effort, and money.

I can vouch for this! The combination of the lopey cam and DFI 8-stack in the Roush 427IR are a really tough combination to get working at other than sea level. I hear they changed their controller from Accel to something else, but I struggled for almost a year to get a balanced tune that properly accounted for altitude.

[fordracing65]

Try it with a self learning system. Even harder to get right.

[olddog]

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.

[FUNFER2]

Originally Posted by twobjshelbys

I can vouch for this! The combination of the lopey cam and DFI 8-stack in the Roush 427IR are a really tough combination to get working at other than sea level. I hear they changed their controller from Accel to something else, but I struggled for almost a year to get a balanced tune that properly accounted for altitude.

What are you cam specs and what changes did you make so it runs better ?

Maybe I'll drop the computer and just run it old school,.......

[twobjshelbys]

The cam document says it was used in a 1985+ Mercury 302. It lists a part number of 449571 but that doesn't look like a Ford part number.

First, you've got to remember that the Accel DFI controller is a complete controller. All firing timing and fuel firing PWM are in tables. I had to do two things. They were done by a tuner that had received Accel training.

1. Tweeks to the basic configuration to account for the unique characteristics of the 8-stack throttle bodies (note 1).

2. Properly creating an altitude compensation table.

Note 1: The position of the injector with relation to the butterfly valve in the trumpet creates some unique air flow issues. At lower throttle positions the air flow at the time of the injection fire gives adequate flow and turbulence to properly mix the fuel. This is about 0 (idle) to about 60* attack. As the throttle opens and the blade becomes more vertical the air simply flows straight down the neck. The effect is that there isn't enough air flow against the side of the trumpets to deflect and cause the fuel to mix. The result is that the fuel is show almost directly through and doesn't have a good air mix. So, counterintuitive as it is, as you get from that threshold to WOT you actually need to cut the fuel back otherwise it runs rich and will backfire.

I got to know much more about DFI and the dynamics of the 8-stack than I wanted to

This is a copy of the cam spec sent to me by Roush

[FUNFER2]

Tony, that's actually a little larger cam than I expected. Is the Accel programing better suited for the trumpet style EFI than the FAST system, like the EZ-EFI Multiport Retrofit Kit ?
Can it also be modified or is it not as friendly to custom tuning ?

Do you suggest any paticular Accel part number that may work for Ford and a stacked injection system ? (Ford probably doesn't matter)

[Gaz64]

Originally Posted by twobjshelbys

The cam document says it was used in a 1985+ Mercury 302. It lists a part number of 449571 but that doesn't look like a Ford part number.

First, you've got to remember that the Accel DFI controller is a complete controller. All firing timing and fuel firing PWM are in tables. I had to do two things. They were done by a tuner that had received Accel training.

1. Tweeks to the basic configuration to account for the unique characteristics of the 8-stack throttle bodies (note 1).

2. Properly creating an altitude compensation table.

Note 1: The position of the injector with relation to the butterfly valve in the trumpet creates some unique air flow issues. At lower throttle positions the air flow at the time of the injection fire gives adequate flow and turbulence to properly mix the fuel. This is about 0 (idle) to about 60* attack. As the throttle opens and the blade becomes more vertical the air simply flows straight down the neck. The effect is that there isn't enough air flow against the side of the trumpets to deflect and cause the fuel to mix. The result is that the fuel is show almost directly through and doesn't have a good air mix. So, counterintuitive as it is, as you get from that threshold to WOT you actually need to cut the fuel back otherwise it runs rich and will backfire.

I got to know much more about DFI and the dynamics of the 8-stack than I wanted to

This is a copy of the cam spec sent to me by Roush
Attachment 28495

That cam is a hydraulic roller from Crane.

Interesting to note their quote of RPM, not minimum and maximum, but best operating range. Note that it will valve float at only another 200 rpm.

I would have thought it would float a little higher than that.

A good camshaft though.

[FUNFER2]

Yes, interesting.

[twobjshelbys]

Originally Posted by FUNFER2

Tony, that's actually a little larger cam than I expected. Is the Accel programing better suited for the trumpet style EFI than the FAST system, like the EZ-EFI Multiport Retrofit Kit ?
Can it also be modified or is it not as friendly to custom tuning ?

Do you suggest any paticular Accel part number that may work for Ford and a stacked injection system ? (Ford probably doesn't matter)

Unless you have an excellent tuner with a dyno and a lot of time I would personally avoid the Accel system. It has way too many knobs, and while they have starting point tunes, they are way off for an engine with anything other than a standard engine. You would want to find someone in your area that has been Accel factory trained.

The thing started and ran with the canned tune I got from Roush but didn't operate "well". I specifically wanted the full closed loop system so I could drive in the mountains - I had altitude ranges from 5000' to 12000'. Roush's suggestion before I "divorced" them was to run (I think it is called) Alpha-N mode which basically runs from the throttle position sensor, and the TPS of the Roush top side doesn't have a very wide range or sufficient granularity.

My tuner looked at the factory ones and compared them to the Roush one, and found that the Roush had some cliffs. The whole system is based on table lookups and they hadn't properly blended some adjacent cells. They figured that the engine wouldn't ever "get there" but didn't allow for altitude. As I said, it ran perfectly at sea level (actually, the 1000' or so in Detroit).

[twobjshelbys]

Originally Posted by Gaz64

That cam is a hydraulic roller from Crane.

Interesting to note their quote of RPM, not minimum and maximum, but best operating range. Note that it will valve float at only another 200 rpm.

I would have thought it would float a little higher than that.

A good camshaft though.

This is the one that Roush uses (or at least used, I don't know if they still do) in their Windsor/SBF based 427R and 427IR. I don't know if they might use it elsewhere. I did love the sound at idle though.

[FUNFER2]

My friend has a auto shop and a chassis dyno. He's a whiz,......at tuning anything and a hell of a engine builder. He can do anything & everything, so I'll have him read this thread and see what he thinks.

We live in Nebraska, so no mountains creating elevation extreme changes. Although I'd love to drive the Cobra in the Colorado winding mountains ! Doubt it will ever happen.

Do you know of a system that works best for the stacked injection ?

[FUNFER2]

By the way Tony, no photo's of your CSX4005LA ? !!!

[fordracing65]

The Kirkhams drive their carbed cars all over Utah with 10,000 elevation changes. no problem whatsoever.

[FUNFER2]

Yes, but that's a carb. Not cool stacked injection.

[fordracing65]

Originally Posted by FUNFER2

Yes, but that's a carb. Not cool stacked injection.

Any fi engine will run well on the street just tune it. Self learning efi another story. Especially on a big cam car.

[FUNFER2]

I just do not hear enough good things about the self tune's. The programing isn't quite there yet.