The car is a S4 euro-spec NA. I am building a new engine with S5 TII housings, plates and LIM. UIM will be from the FD and rest will be from the S4 NA engine.



I am looking to get a streetcar with as much performance as possible within the constraints of being streetable. Streetable to me means : solid idle at stock rpm, no bucking, must not be too loud, must be nearly as reliable as stock.



I have ported the primaries, I have hardly changed the timing, only increased size enough so that the stock edge where bowl meets port is blended smoothly into the runner/bowl. I kept it this way because the car will get a ACT Prolite flywheel, and I have heard light flywheels can make it harder to get a good idle, so I do not want to port the primaries too much. Also because of velocity considerations.



I have also ported the exhausts. They are modelled after the RXBeetle's port-pictures found on rx7club.com. Pictures are here :

Exhaust 1

Exhaust 2

Exhaust 3

They are kept moderate in size as I don't think more will make much difference for NA, and also because I do not want to mess with removing the sleeves. I only have picture of one port next to stock port, but both ports are now ported and the second one looks just like it.



I am now considering my options for the secondary ports. I have moddeled my template from Judge Ito's wild streetport for open-timing and a bit less extreme for closing-timing(for velocity). My template can be seen here :

Template

The dotted blue lines are where the sideseal leading edge is (vertical line) and where rotorface edge is at different crank-angles (horisontal lines). The red line is my suggestion for the port. Perhaps it is difficult to see in the picture, but the sideseal edge is placed just within the port, and I will bevel the closing edge to keep the seal guided. I have verified that I get good "scissor-action" for the closing edge, almost as sharp an angle as stock. One thing I was wondering though : would it be beneficial to reverse the scissor-angle? So that the edge towards the manifold is where the scissoring starts.

Another picture of my port-template drawn on the plate :

Port



All my runners and bowls will be smoothed out, but otherwise not changed much from stock.



Any suggestions for improvements before I start to assemble the engine?

 

The TII LIM sucks, and I believe the poor transition between FD UIM and TII LIM will do you more harm than good. Especially in an N/A.

 

In what way does the TII LIM suck? I will gasketmatch as much as possible between UIM an LIM for better transition. I have heard good things about this combination, and I am hoping to utilize the tournament effect of the fd UIM.

 

Also I forgot to mention that I will be using solid cornerseals , I've read this is better for big streetports like this where only 1/4 of the cornerseal is supported.



And I am planning on beveling the rotors RX8 style without rebalancing. I figure that if I am very careful to remove exactly the same (small) amount of material at all sides, and do it both for the opening and the closing sides, I shouldn't disrupt the stock balancing anymore than carbon buildup does.

Like this :



Only with bevels at all sides.

 

DO NOT REVERSE SCISSOR the closing port if you have ported the opening side so the top sideseal drops in. This will cause the drooped sideseal tip to impact causing chipping, wear and poor sealing and very possibly broken seals. Make sure it scissors from the supported end (center of engine) all the way to the tip comming back onto the housing last.



The slower port closing of a well scissored port WILL hurt the dynamic effect a bit, but the early opening timming more than makes up for it.



You can elect not to scissor or even reverse scissor if both sideseals are supported at both ends at all times on the housing.



Rob @ Pineapple said the '89 TII LIM flows well and balanced between ports. I told him the mods I did to my '89 LIM and he said it wouldn't be a bad idea to flow it again to check balance since they are good stock.



Balancing flow is much more important in turbo application as you are flowing so many CFM and AFR is critical.



I backwelded and cut out the ACV intrusion and worked the center mounting bolt intrusion in the primary runners as well as porting, polishing and matching it to 3rd gen UIM.



There will be a spot between the 3rd gen UIM and 89 TII LIM where the primary runner volume has to be increased to match the ports- this could hurt velocity some.



This volume increase could be eliminated by using adapter plate, but this will add to runner lengths and so shift the power from dynamic effect lower in the rpm range as will adding opening or closing port timing.



Custom manifold is another great option if you like to fabricate- which you must



My set-up ran 11# flywheel, 720cc primaries, the ported/polished manifolds, Pineapple early opeing streetport. Idle was strong at 700rpm. It by no means purred like a house cat; more like a lion, but with the Haltech it was possible to get a good strong fairly steady idle



The primaries can actually ported to close a bit LATER than 2ndary ports to take advantage of their higher velocity as mine were- or leave them conservative for a more tame idle as you have in mind.



Though, sidecutting the rotors for early opening will make it idle much more like a small half bridge port from what I have heard- as does porting for early opening.

 

Yeah, I see what you mean - either support the sideseal top and reverse scissor, or port intake timing some more and normal scissor. I would think that for the same size port (with supported sideseal) reverse scissoring would be better for flow though, do you know of any comparisons?



My suggested port design, as you see in the pictures I have verified to have only marginally sharper scissorangle than stock. This should fit my constraint of only allowing marginally worse reliability than stock. My stock engine ran for 20 years, my goal for the rebuild is 5+ years.



This is good news, as I beleive a balanced flow can also help with the tournament effect which I think is vital to good NA performance. I am planning to run 2 EGT probes so I can experiment with adjustments in case my porting work ends up unbalanced, but starting with balanced flow makes it much more probable that it will end balanced too.



I do not have access to TIG welding equiptment, so wont be backwelding, but other than that I will try my best to do as you did. I still haven't made up my mind whether I'll make an adapterplate, I'd rather not, like you say my added port timing moves the powerband up, so I want to keep it a bit in check - especially since I am using S4 rotors and can't rev as high as with S5 or Renesis rotors. I'm thinking that perhaps a 3 or 5 mm adapterplate wont change it much though.



Custom manifold is definately not for this rebuild, but might be in the future if I am not satisfied with the performance.



I don't need a house cat idle, and will be using the stock 460cc for primaries, so this is encouraging. My primaries have about 1mm later closing than stock, with beveled rotors bot open and close timing will change a bit more, but since my bevels wont be big, I am thinking the resulting timing for the primaries will be close to a mild streetport, which should still allow me a strong idle.



How big were those rotorcuts and for what ports? I have stock primary intake timing and only marginally later than stock exhaust closing. With some small (2-3mm) bevels on the rotors I hope to end up with timing like a mild streetport, but with better flow due to the small ports and guided portflow from the bevels.



Thanks for your comments, they are very appreciated as your UIM/LIM setup is what I am modelling mine after. I was wondering where you sourced connectors for the UIM TPS and air temperature connections?

 

I would think it would be good for flow verses keeping velocity up as you want on the primary ports since the smaller volume from not having to extend the top of the port and the roof of the runner leading up to it will aid velocity and the straight line from runner top to port top flows nice.



So, what some people do for the early opening primaries is port it reverse scissored and add a large beveled area out to where when the sideseal actually hits the housing it is scissored. I have seen RETed port like this.



I felt this type of port had advantages but was causing the flow at the top of the port to fan out back toward the sidehousing (diffusing the flow) and I wanted to direct it to the back of the intake chamber.



So, I made a port the shape I wanted it for flow direction and small volume and then cut a ~2mm deep recess above the unscissored section of the port (a step) that has a bevel on its scissored edge. I felt the direction of the flow leaving the port is influenced by the release angle just as a thrown ball is.



I haven't run this engine yet- Probably next week.



Don't know of comparisons, I think porters find a port that works well for them or has for others and slowly tweak it for reliability or performance.









It is all a compromise. My professionally ported early opening streetport motor was not scissored all the way to the outer edge of the port (the last 1/4" unscissored) and had no bevel at the top of the port. It left marks of the sidehousings and chipped/wore 1/32" off all the corners on the end of the sideseals after only 12,000miles. The engine builder said that is fine and he has seen the sideseals wear to half their thickness this way at the tips and run fine. Obviously it would affect compression though! He said that it was really the low rpms that cause this wear as that is when the sideseal has time to "fall" into the port. I drove in the 1,500-2,000rpm range a lot as the turbo boosted so low.



I added more scissor and a nice bevel as reliability and low rpm power (compression) was more important to me than maximizing dynamic effect.





I just put dual EGT gauges in my manifold so I can check it out- though builder said by exhaust port carbon coloration it was probably pretty balanced. He had balanced the ports on a flow bench already, so the manifolds are probably OK. We will see how my ports balance...





I should have said I HAD it welded. I prepped all my aluminum and had it welded in a batch- still pricey. Now I have a friend that TIGs for beer- Thanks!









The early opeing will have more effect on idle than late closing. A 6 port idles OK with ports wired open as long as the 2ndary throttle plates are still closed to prevent compression stroke reversion from affecting primary ports. With exhaust overlap from early opening the throttle plates do not have an effect as it is exhaust gas reversion into the intake cycle that is killing the idle quality and that happens in the engine not the intake manifold.



So since that ealy opening 2ndary port can not be closed off to help idle, you might want to consider making the primary port early opening as well. I feel keeping the primary later opening at this point will only help idle marginally from the smaller port area keeping the velocity up to fight exhaust reversion, but exhaust pressure will probably be higher at idle than intake pressure so it is a losing battle. Remember, that engine I was running had primaries and 2ndaries opening at the same time and primaries closing LATER and it idled fine.



If you want to keep the primary port stock opeing so you don't have to worry about scissoring; that is another story.



Rotor cuts add more overlap to affect idle from exhaust dillution. From what I have read Chrisspeed wrote a fully beveled rotor on a large streetport will idle much like a small half bridge.



What is your engine control? Having lots of adjustability for ignition timing and throttle pump settings becomes important for driveability the more you port and work the manifolds.



I think this set up has good potential. When I was running my car NA it was exhausting through a restrictive stock S4 turbo and manifold (just no turbo in there). I was running just the 720c primary injectors as the 1600cc 2ndaries would not stage well. I was running 100% duty cycle on the 720cc injectors at ~ 4,500 -5,000rpm as lean as 13:1 AF from what I remember. That should translate to some good power. I know in turbo trim the power really took off right AFTER 5,000rpm though full boost was at 3,500rpm- so the manifold length might work well enough.



I have read another forum member that is into the Japanese scene say NA streetports with 3rd gen engines/manifolds are making big power there (claimed 250RWHP??).

 

Do you have pictures of this port? Is this like the "skateboard ramp"-style runner?



Yeah I totally forgot to take that into consideration. Makes me reconsider my strategy, as I really want a good idle.



I think what I will do instead of my previous plan, is to open primaries and secondaries at the same time, and then be more conservative about the timing. This lets me port for supported sideseal tip and do more of skateboard-ramp like the ports by kahren in another thread. Then I can make up for the lost timing by the bevels on the rotors. I can always port larger later, but the other way around is not so easy.



This is where I have an unusual plan. I plan to run the stock S4 NA ECU. This is because I am currently reverse engineering the ECU and hopefully can make my own custom chips for it some time in the future. I have experience with embedded systems so it should be doable. Until I am ready to modify the stock chip I will use a SAFC for fuelcontrol. I was originally planning to use Megasquirt, but I like the challenge of modifying the stock ECU. Also the japanese car with supposedly 250rwhp(which is the car that inspired me to try this NA/TII mutant) was using stock TII ecu, so that is indicating that it is possible.

 

I am really looking forward to drive your car, Alex

Let me know when you plan on assembling the engine...

 

Me too! I wanted to assemble it last weekend, but since that was a bit optimistic, hopefully it will be this weekend.. That is the shortblock only though, I still need a new oil pump chain before I can fully assemble it, and need to figure out the length of cable needed to control the OMP by the fd throttle body, but other than that I have received all my parts (including the flywheel I was worried about).