Don Terrill’s Speed-Talk

First I would like to say I started surfing this forum and it has inspired me to do what I am doing to this cylinder head. The knowledge on this forum is beyond anything I have known before and its such a great group of guys that are willing to share. I unfortunately don't have data on my old head I did.

I have been playing with the 4g63 for a while. Mind you I am 25 years old but it has been pretty much my only platform that I have pursued since I was 18. Years back I ported a head not knowing what I was really doing, just having the thought in my head that I would like to see more of the valve opening looking down the port entrance. What I essentially did was destroy the SSR and lifted the entrance to the gasket. This was a "2g head" that has a raised port and a rectangular port entrance vs the "1g head" that has a much larger port.

That engine I built was a compound turbocharged 2.3L stroker with a 1.5 RR/ 9.5:1 cr and a 272 advertised 226 @ 1mm with 11mm of lift. I ran a stock converter and was able to stall it in about 5 secs if the weather was decent and build pretty much what ever psi I wanted. The engine was very responsive and trapped 140 in the 1/4 at 37ish PSI. I was told that the head did not follow good methods and I should sell it to do a 1g head. I sold that head to a friend and he said once he put it on it was like night and day. More responsive and more powerful but again.... No data

The next year was hell for me and I ended up selling it all and moved onto a newer engine design which is the evolution head. In between the compound car and this car I worked for a shop that had a very experienced porter. We worked on just doing a simple short side clean up and a slight change in the port design in the throat area and some clean up on the exhaust. When I dyno'd the car back to back we saw very little gains. Almost an identical power curve but picked up about 8hp or so at 400whp.

This really got me thinking about the tumble port theory vs the short side theory. I have an EXTRMEMELY hard time finding back to back cylinder head tests on the web but it sure is easy to find some CFM numbers. My old head seemed to perform so well even though it followed "no one I can find on the webs" port design.

This leads to today! I came across a head that needed some work. Broke some valves off and had a dance party in the combustion chamber but sounds like a perfect opportunity to me. I started dropping out the floor of the port essentially deleting the short side that was there. Mindset is to make the air move over the top of the valve and not worry so much about the little short side that I have left. I don't have access to a flow bench but maybe in the future I could purchase one my self. I bought some silicon to mold the intake port to see whats going on and this is what I have.

This cylinder head is going on my evo for dyno testing. It is going to run a 70mm turbo putting out a max of 110 lbs a min( roughly 1500cfm) with a 290 adv 235@ .050 and 11.5mm lift.

Questions I have are the roof and the throat.. It seems like it sandwiches right before the port split and gets rather large around the bowl area. I can either raise the roof to try to get a better volume change or I can leave that and fill in the roof of the port with an epoxy.

Throat measurement- 1.275 diameter(pre valve job) considering maybe a 1mm over valve to compliment the tumble effect
Guide measurement- 1.185x.965
splitter measurment- 1.185x.990
Pre splitter- 2.050x .900

The valves undercut diameter is .225 and half way up the tulip is .360. I wonder how you take that into account when trying to do volume calculations. When the valve is closed the ports volume is less and as the valve moves the volume increases.

Thanks for looking and hope to share the data I learn as i learn it

Like me, ... you need a simple flow bench
dig around and find plans for David Vizards home bake one
once you've made one, keep the size valves your going to use, hog out a port to the throat size, then start filling up ports with plasticine
Erland Cox has done a lot of work on 4 val heads, try him

Hi , Yes you will find a vast amount of great information on speedtalk forum and your young have many years ahead of you to learn.

Get pipemax software from Larry @ http://www.maxracesoftware.com/ join his forum as well. You will save a lot of time on guess work but rather with factual maths that works. Will be the best dollars you ever spend.

If you don't have a flowbench ? Get in touch with Bruce @ PTS buy a set of his bench plans and join his forum also http://www.flowbenchtech.com/forum/index.php

Happy reading

Cheers

I'm in the works of getting a chassis dyno in my shop. I hope that sometime this summer I can get the time to build a bench and be able to relate flow numbers, air speed, and port design to actual chassis dyno hp increases. This head will go on un-flowed and will compared with a dyno sheet. Then when the time comes I will flow it.

jwhalen07 wrote:....In between the compound car and this car I worked for a shop that had a very experienced porter. We worked on just doing a simple short side clean up and a slight change in the port design in the throat area and some clean up on the exhaust. When I dyno'd the car back to back we saw very little gains. Almost an identical power curve but picked up about 8hp or so at 400whp....

IMHO - when dealing with OHC heads the first thing to do is change the valve job as most don't have any bowl angle cuts and putting a radius style can pick up a good bit of flow on them without touching the rest of the head. Then play with adding a back angle to the intake valve, then flow the head both with and without the manifold attached to see if the head port is really the restriction at that point. You typically want an intake port to neck down slightly to the valve bowl, and if your intake manifold is the limiter then removing that neck down area in the head might not gain anything. Most times with modern OHC stuff you will be quickly humbled by the intake manifold limits once the area 1/2" before and 1/2" after the valve seat has been addressed.

Also most important - when you flow the head only - flow it like in real life which means you have to fabricate a fuel injector mount. I would guess 99% of flow numbers on the web are ASSuming the fuel injector and the boss it mounts in flows air, and being at the roof and so large it has a major affect on flow numbers. Modern FI OHC heads cannot be flowed like a carbed pushrod head but most seem to do it that way and the numbers are usually grossly inflated as a result. Simply mounting the intake manifold after the valve job changes will show what potential can be gained.

Mitsubishi ports from that era were all ridiculously large, which was evident in the NA engine - high fuel consumption and no mid range power - They realised their mistake and in the next turbo engine and made the ports a lot smaller - same head smaller port -.

In performance turbo engines large volumes are safe and easy to do, probably why your large port was more effective. You probably didn't loose flow hogging out the port in the wrong place because the port is probably being restricted somewhere else, like the seat. In turbo racing large is safe, but the correct size is usually a touch faster

When you start flow testing 4 valve heads you will quickly realise it hard to get more flow out of the actual port, because the design is modern, and that there is usually more to find just in the seat.

There is a lot to gain in the seat alone because Mitsubishi make the std seat too narrow and position it very inward.

My first thought seeing those heads was the port is way too big to make good torque NA. It probably is.
That does not mean your results are wrong!

At that many PSI boost you are injecting a LOT of fuel into the port, where does it all go?? Lets try to duplicate that on a flowbench LOL

Jwhalen, where are you located. You might find somebody close by with a bench.

Tom

If you have a search around here you should find some good info. Try this link: viewtopic.php?f=15&t=38864&hilit=Tumble+port
Also on page two of that link is another speedtalk link from twincam about half way down the page with some more info.

Naive question. If one has a four valve turbo engine that runs on pump gas, does it even make sense to run the head on a flow bench that doesn’t have a tumble wheel? The heads usually flow so much anyway that it’s mostly about the charge motion and burn times. That’s my theory, but just a theory, so this was more of a question than a statement.

I do think that flowing the intake manifold and the head together will be informative even without a tumble wheel.

Building a workable flow bench with electronics for about $1000 bucks these days is easy, but constructing something to open both valves at the same rate on a 4 valve head and within .003" accuracy when testing at lifts under .250" is a project, a real project!