puff of blue smoke at higher rpm decel

[Belgian1979]

I have exhaust backfiring during decel at high rpm quite frequently. On the logs i can occasionally see a map spike which indicates a spit back but it equates to 4-5 kPa higher. An explosion of a fuel mixture would probably cause a spike all the way up to 105 kPa maximum reading. I can’t remember seeing that in the logs.

For the prevention of flow back into the chamber a check valve would a possibilty. Although I’m worried it may break up and parts my get sucked into the intake.
Maybe when there is a reserve of vacuum , during intake valve closure the pressure returns to atmospheric and the chamber pulls in a mixture from the runner.

This definately is something I need to sort.

[mk e]

Careful with reading too much into the log data. The sample rate is relatively low and the sensor response time is lower. When the intake opens and the exhaust is in tune there should be about a 50kpa shift in the runner pressure......but you don't see that in the logs either right? Nor any of the tuning pulses that must be there. You need much faster sensors and logging rates to see stuff like this.

An explosion is the only viable explanation for the force it would take to balloon the plate like that.

[Belgian1979]

The problem here is that it can happen again. David Redzus once commented that there would be a certain amount of cross talk between cylinders when they are connected to a vacuum chamber. If this 'cross talk' caused mixture to be sucked into the chamber, it can and will happen again, no matter if the connection is made above the manifold or under it.
If it was indeed and explosion the effects will always be damaging. The pressure/force has to go somewhere.

[Newold1]

I think the explosion issue is really a matter of size and not necessarily some cross talk or air/fuel mixture in the vacuum vessel. Your vacuum plenum is really quite large in volume in terms of what is needed for a MAP signal or even for vacuum assisted brake master cylinder, etc. I like the idea of taking that vacuum from points in the runner closer to the head manifold runner junctions and routing it via small 1/4" lines to a much smaller vessel located up top and centered to allow those line lengths to stay as equal as possible. Just abandon that area under the intake. I think your results will be better and you won't have to deal with the sealing and oil burn issues you are trying to correct now. JMO

[mk e]

The problem here is that it can happen again.

Plus thinking about it, plain old simple gravity will want to put fuel in that chamber and then it will pop.....its a terrible design

[mk e]

The problem here is that it can happen again.

Plus thinking about it, plain old simple gravity will want to put fuel in that chamber and then it will pop.....its a terrible design

Plus with all that volume every time you open the throttle the pressure goes up in the manifold and air/fuel flows into the chamber, some of that fuel will bead up and build up so even if by luck fuel isn't into running in, and it probably is, you'll still get fuel in there....then it will pop if you get ever a small pop back in the runner, which on my engine is every time I crank it.

[MadBill]

..For the prevention of flow back into the chamber a check valve would a possibilty. Although I’m worried it may break up and parts my get sucked into the intake.

Check valves wouldn't work, because they would only allow (pulsating) flow out of the chamber, leaving it fully evacuated at all times and thus unsuitable for a M.A.P. reference, etc.

[Belgian1979]

Agreed.

I've mocked up my setup partially to evaluate fitting positions on top and the placement of a reservoir. There is a (small) possibility that I can use. The main question is now 1/4' or 3/8' outside diameter alu tubes. The 1/4" is 0.18" inside. I guess the 3/8' is 0.314' inside. Obviously the larger the better but in view of space constraints I may not have this option as the 3/8 tubes (4 on each side) take up a lot more space and may prove difficult to properly route such size tubes.
I will plug the holes on the vacuum chamber side with 1/8 plugs and remount the plate to prevent oil from splashing/heating up the underside of the runners

[mk e]

Perfect.

Don't be afriad of rubber hoses, they can be much easy to run/fit but do have more wall thickness to deal with. so aren't always easier, just usually easier.

[Belgian1979]

Perfect.

Don't be afriad of rubber hoses, they can be much easy to run/fit but do have more wall thickness to deal with. so aren't always easier, just usually easier.

Yes, the rubber hoses that fit the 1/4 fittings are doable, the ones for the 3/8 line are about 15 mm on the outside, which is a totally different story when having to deal with 8 of these.
I will have to use rubber hoses to make the connection when going from motor to firewall, so this is an important issue to deal with.

On the other hand, when tapped, the fittings can be taken out without disassembly of the manifold and I could screw in a 1/4 to 3/4 type fitting that would allow more flow. For this reason I tend towards 1/4 to -4 to alu tubing, then rubber to a barb fitting.

[Belgian1979]

I made a small calculation with an online pipe diameter vs airflow calculator. I used an airflow choke point of .5 of the speed of sound. For a cylinder that has 43.5" capacity and making the airflow requirement calculation from 600 cold rpm to 1200 hot rpm, I would roughly need to supply an additional 5,9 cfm to each cylinder.
The airflow of the 1/4 line is only about 4.5 cfm. So that would be too small imo. A 3/8 has a flow of 5.2 cfm and is closer but still too small. All calculated on the basis of internal diameter of the tubing.
Imo a 3/8 line might just work but 2*1/4 per runner would be better. The 1/4 tubes bend a lot better but placing 2 fittings in each runner will prove difficult if not impossible. Have to look at things again.

[mk e]

When I set up IAC, I start out with it plugged and set the warm ideal at the TB(s). Once that is done and everything happy, then I stat messing with the ICA to to high and cold idle. The result is you a warm idle that pretty much has to be stable (I've had a couple ecus that would just surge no matter what else I tried) and more importantly you need a lot less air. The bad news I also had a setup I did this way then had to port the GM IAC valve to get cold start working.

My gut tells me 1/4 tubing will get you warm idle no problem and probably get you cold idle but you'll probably need to manually open the throttle a bit for cold start. 1/4" hose has a bigger ID than 1/4" tubing...ID vs OD measurements. 3/8" tubing is probably where you need to be to get everything working just right.....but cracking the throttle to start isn't the end of the world.

[Belgian1979]

The rubber lines I have here are 8 mm outside 4 mm inside. The throttles can be set more open, but that basically also reduces the effectiveness of the iac to control idle speed.
Another issue I'm dealing with is equal air distribution to those vacuum ports. With the chamber as is, I had the problem the rear cyls were running leaner than the front ones. Any vacuum log or canister where the ports are on a line versus the entrance will have this issue to some extent. So the best solution would be round can with the ports around the perimeter.

[Newold1]

I don't think I have ever seen an EFI engine where all the air being used for idle air and MAP was contributed only by the manifold vacuum provision of the intake system for both cold or hot idle. I know a lot of the air for the idle is coming past the throttle plates even with a nominal crack open. The throttle plates would have to much vacuum pulling on them to allow a smooth opening off idle throttle application if they were totally shut at idle.. I would therefore in my opinion probably guess only half the CFM in your calculations would be adequate along with the CFM coming past the throttle blades and hence I think 8-1/4" lines feeding a much smaller vacuum chamber would be adequate for both idle and MAP input.

[Belgian1979]

Thanks. To explain the iac system a bit : if throttles are opened further, the effectiveness of the IAC to control the idle rpm is reduced because more air flows through the throttle. So you do not want to set the throttles at idle so that when hot, you reach your rpm target without iac. In that way there is no possibility to compensate for things like AC,... In my particular case the idle rpm as dictated by the throttle position is further changed by the changing lengths of the linkage. To compound the problem, when the car sits during an amount of time when hot, the linkage tends to expand and in my case pull the throttle open more. So, ideally my idle rpm as dictated by the throttleblades is ca. 200 rpm lower than what would be actually needed.
On the other end of the spectrum, when cold this position of the throttles creates a very low rpm at idle which would have to be compensated by the iac as well. Hence my calc between 600 and 1200 rm.
If the flow through the vacuum lines is too low, the engine will not reach the set 1200 target and remain at say 1000 with the iac fully extended.

My gut also says that a 3/8 is better, but has proven impossible to fit. I spent an entire day yesterday trying out several things. So far the best option is rubber lines and a converted Moroso oil catch can as reservoir.

I'm thinking of make a sort of crossover tube between 2 1/4' lines and from those 2 tees go to short 3/8 lines that connect to the runner. That way a cyl would be able to pull from 2 lines at the same time and capacity is effectively doubled.

However, I've not found a suitable tee with 1/4*1/4*3/8.