Competition inlet valve seat prep: Fiat 8v TC head

porting, development, valve and seat work, combustion chambers, cams, head construction, etc
Guy Croft
Site Admin
Posts: 5039
Joined: June 18th, 2006, 9:31 am
Location: Bedford, UK

Competition inlet valve seat prep: Fiat 8v TC head

Post by Guy Croft » November 2nd, 2006, 11:04 am

At last a few minute spare to post something interesting. Some development tests here for interest, done 01 Nov 06 on a type 4371507 Fiat TC 8v head, 42.4mm inlet valves, the head is from a 2 liter Lancia Montecarlo head for a GC client.

First I did routine porting as promised, (more or less keeping to standard port dimensions) - smoothing the port at 80 grit throughout and around the fairly generous short side radius.

You can see how much a standard head flows, (black dotted line) and how much more flow you can get just by porting and not even modifying the standard seat (green), and how much more careful seat work adds (red).

In my experience these types of ports do not like a very narrow seat with a very deep throat grind all the way down the insert, and they seem to like a lower parallel section in the seat where it runs into the curved alloy port. The upper section needs to be 70 deg, I have not found anything better, 65 or 75 deg. Half and half 70 deg and parallel is good if you can do it, though I settle for 85 deg lower section which is as near to parallel as the Peg grinder will do for me without spending hours on it. The gain vs cost to bore the last bit out parallel is not huge and is something I'd save for StIII and full race units.

The valve/seat flow gains ultimately shown on the graph below really have come from improving the discharge through the valve and seat, not from just making the diameter of the throat bigger, it has only been enlarged about 0.3mm in fact. If the shape and thus efficiency of the valve/seat region is poor (I'm talking shape here) you cannot get gains just by enlarging the throat diameter. Not on a sidedraft port like this anyhow.

As far as inlet seat width goes I work on minimum 40 thou‚ per inch of valve minimum width, 1.7mm on a 42,4mm valve, and you can go as big as 60 thou‚ without much penalty, although, that said, wider seats seem better at lower lifts and narrower at high lifts, but I cannot be absolutely specific, everything is interlinked. I recommend that the contact face on the valve and seat match for width, anything else seems to generate a loss. I use the back grind as a tool to match the valve contact face when the seat has been finished, the lapped region on the valve shows me where to grind to.

It's interesting how much better the mid lift flow is with the seat top grind of 30 deg. I have done some where more is better but too much top grind can set the valve contact face very deep in the head and make it hard to shim up. I think what might be happening is that the air separates from the back of the valve, skating off at the edge of the 30 deg back grind (rather than flowing round the valve's 45 deg contact face as you might imagine it would) and it responds better to the 30 deg exit angle than it would a flat exit region as it enters the chamber.

Insofar as far as I am able to be specific (not far..!) radiusing the sharp edges of the throat or top grind can lead to a significant flow LOSS. Why? Air doesn't always follow a curvature, sometimes better to let it break away at the sharp edge. My goodness, there are so many unknowns; you really need CFX or some other computational software to figure out what's really best.

We all know nothing is certain in life, and what suits this head will not necessarily suit another with a different port layout. So the beauty of having a flowbench is that (if you don't cheat) it really does take the guesswork out of head prep, so here you can see how it's done. Anyone with a set of Neway cutters (20 x 45 x 70 deg angles) and a porting tool (Metabo, Dremel, Bosch) could get close to these results at home. I would always advise new guides first, esp with Neway pilots, otherwise you could have real problems with concentricity.

The flowbench says the head - even with modest port sizes and standard sized valves - has vastly more potential. There are gains beyond mere flow too; the throttle response (pickup) is vastly better because the air responds far quicker on the new seat when the valve opens, and the crossflow (scavenge) is improved too. From my dyno experience the bhp gain is around/over 18bhp alone just from this 'modest' head work, but then again, so much depends on the carburettor and exhaust layout as to whether it will 'breathe' in situ.

I am sure there are countless combinations that could be tested but I have enough dyno experience of GC units to know that this head will achieve its bhp target comfortably. You can spend weeks doing heads and making dazzlingly clever rules about ‹Å“doing this¢ž¢ and ‹Å“doing that¢ž¢, multi-angle seats etc etc the internet is full of them. But I think the truth is that you are working with an imperfect thing to begin with ¢‚¬Å“ air doesn¢ž¢t like curves ¢‚¬Å“ and however hard you try you will never know for sure if what you have is perfect.
Test 1 std seat.jpg
Original (worn) OE spec factory seat - surprisingly good with standard valve..
Test 1 std seat.jpg (121.33 KiB) Viewed 23702 times
Test 2 wide seat.jpg
The seat has now gone very wide, as I have ground it out to true valve diameter 42.4mm. The later heads had 41.8mm. This valve takes the seat very close to the edge of the insert, I would not go bigger really on a std seat insert, it may crack at the edge
Test 2 wide seat.jpg (102.26 KiB) Viewed 23698 times
Test 3 seat.jpg
Seat used on tests 3/4, appr 2mm x 45 face, 70 deg upper throat and lower section ground at 85 deg the slight groove in the port from the grinding wheel (it happens..) can be dressed out later.
Test 3 seat.jpg (102.36 KiB) Viewed 23684 times
Test head flowed.jpg
View down lightly modified ports, showing bullet nose race guide - the intrusion into the airstream of guide and valve is insignicant at this level of tune, don't shave the guide off.
Test head flowed.jpg (35.01 KiB) Viewed 23668 times
Test head on Superflow rig with inlet valve.jpg
I used 3 ports all prepped the same for the test; this is the last development port in test. When you do a final mod like 'add 20 deg top grind', you have to have a pretty good idea that it will work. If it doesn't, you could be in real trouble.
Test head on Superflow rig with inlet valve.jpg (110.25 KiB) Viewed 23666 times
Test 5 seat with 20 deg top grind just visible.jpg
The 30 deg top grind just overlaps onto the alloy, it made a big difference to the mid range flow.
Test 5 seat with 20 deg top grind just visible.jpg (119.39 KiB) Viewed 23655 times
Finished seat and short side radius at 80 grit.jpg
As they go this is quite a big short side radius, the bigger the port the smaller it will be. You have to have big ports and sacrifice the low lift flow to get high overall flow - to a great extent, if you want big bhp.
Finished seat and short side radius at 80 grit.jpg (113.19 KiB) Viewed 23631 times
Back-ground (left) and early std 42.4mm valves.jpg
The valve at left has been lapped on the finished seat and then back-ground at 30 deg (and edge left sharp) right up to the 2mm contact face. RH valve is standard.
Back-ground (left) and early std 42.4mm valves.jpg (120.97 KiB) Viewed 23619 times
Test 5 valve on full lift.jpg
Final (Test 5) valve & seat combo on full lift. Of course, the valve only reaches full lift once in the cam cycle so we cannot just max everything out for flow-at-full-lift.
Test 5 valve on full lift.jpg (115.77 KiB) Viewed 23637 times
Fiat 8v TC head 06.117 seat flow tests.JPG
Step by step flowtests with valve and seat mods, starting with dotted black line - standard head. The other tests all have the same porting, just different seat/valve combinations.
Fiat 8v TC head 06.117 seat flow tests.JPG (64.7 KiB) Viewed 23630 times


Who is online

Users browsing this forum: No registered users and 2 guests