Valve seat development.

[4v6]

Good morning Guy, I hope youre well.

My question concerns the development of valve seats.

I've now been modifying ports on Toyota V6 heads with some success and I have surprised myself with an ability to repeat results, however I feel I would be able to improve on these results even further by careful attention to valve seats and valve shapes.
The issue of cutting the seats accurately and to various angles and with repeatability is now no longer a problem as I've purchased a Peg10 (pre-owned) with which to carry out such work, I intend to get it serviced and checked as soon as possible although it appears to work very well indeed, what a machine!

I'm wondering whether or not I should start such development work on an unmodified port first to see what effect the seat may have or should I start on a modified port first? I'm swayed to starting modifications of the seat profiles using the unmodified port.
I also propose to machine several seat "inserts" from aluminium, lightly pressed into the seating counterbore so that I can rapidly cut a combination and then remove it to cut another and compare results before transferring the finished profile for real.
Do you think this is a coherent and reliable method?

Any light you may be able shed on the above as ever, will be gratefully appreciated.

Best wishes, Tony.

[Guy Croft]

Tony, hi

do you propose to use a flowbench to prove out your results?

I assume so, and if that is the case be aware that the seat/valve combo response depends very much on the short-side radius. You often can't perfect the SSR unless the insert insert size and coresponding throat (parallel region - or lower diameter of seat insert) has been done. So I would tend to do the throats only and do the porting but leave the seat angle till last. As for a removable insert, probably a good idea - but I would just do the dev work on a scrap head.

If you are able to provide a photo of valve and ports and do a decent 'cutaway' sketch showing the port shape and valve head/stem inclination angles I can then have a guess at what your optimum angles will be and we can compare my guess with your dev work.

G

[4v6]

Hello Guy, yes i intend to utilise my homebuilt flowtester to keep me pointing in the right direction hopefully, its certainly proven useful so far.

I think i may be able to do better than a sketch Guy, ive taken some measurements and photos for you if they will help.

Valve dimensions.
Stem diameter= 6mm.
Valve head diameter =34mm. X 2 ( multivalve head).
Seat is 45 degrees X 3.0mm total width from margin to the back cut angle.
Contact width of seat =1.50mm.
Back cut angle=28 degrees X 3.5mm wide.
Stem to head radius =0.375 inch ( 9.525mm).

I have observed that the actual contact area of the seat is of less width than the 45 degree ground angle and im tending to think that it may be beneficial to grind back from the contact point so removing material that would otherwise project into the flow at low lift.

Valve head detail. Radius at stem to head of 9.525mm. Seat contact area less than total 45 degree width. backcut of 28degrees.

S8007893 (816 x 612).jpg (37 KiB) Viewed 13537 times

I sectioned one of these heads some time ago although the port profile isnt exactly as it would appear in standard form due to alteration of it.
Intake port is on the right, please ignore the apparent cross sectional change in area at the entry its due to the way i had to cut the head- a hacksaw isnt the easiest method.

Sectioned head.

DSCF0030 (800 x 600).jpg (113.3 KiB) Viewed 13537 times

Head dimensions.
Throat directly under insert= 29.00mm
Seat angle= 45 degree X1.50mm width.
60 degree second cut X 1.70mm width.
75 degree 3rd cut X 1.10mm width.
Remainder of insert at 90 degree X 1.0mm width.

Valve inclination is 11 degrees from vertical.
Port angle is 35 to 40 degrees as close as i can measure it measured from combustion face.

Seat details. Multi-angle profile.

S8007887 (816 x 612).jpg (92.83 KiB) Viewed 13537 times

I also took some port moulds in silicone of an unmodified port and one i modified.
The short side radius is modestly altered based on your advice previously Guy and it makes a huge difference on these ports.

Standard port shape, very poorly finished with ridges and misalignments, rough surface also.

S8002540 (816 x 612).jpg (79.19 KiB) Viewed 13537 times

Comparison photo. Modest changes to shape and size of pink mould but flows much better than standard.

S8002535 (816 x 612).jpg (77.73 KiB) Viewed 13537 times

Back to back comparison.

S8002522 (816 x 612).jpg (75.58 KiB) Viewed 13537 times

Please let me know if you need any more information Guy and ill do my best to supply it if i can.

Thank you for you help, Best wishes, Tony.

[Guy Croft]

MODEL POST!

Excellent presentation Tony, well done!

I assume the pink mold was developed on a scrap port and you haven't done porting yet on the 'real thing'?

At this juncture and before suggesting any more work I'm going to ask if you can present your flowtest results?

1. BPF inlet and ex
2. Inlet valve in
3. BPF inlet with inlet man
4. As 3 with valve in


In view of the inlet port shape I'm particularly interested in the effect of adding the inlet manifold. Gain? Loss?

GC

[4v6]

Ive actually fettled a number of these heads for others now Guy, with good results; The pink mould was taken from a scrap head i used to investigate the inlet ports with and flowed the best of any of the others i modified - 153.6cfm @10" but the guides were shaved in that particular one, i usually see no more than 145 @10" with them at full length and the same profile.

I can certainly fill in some of the blanks, although i recently had a hard drive crash that eliminated some of my figures, so i will have to re-run them in the future when i get another manifold.

The inlet manifold on these is a split item with an upper and lower section.
The upper section is a dual plane design with an "acis" Toyota-speak for " acoustically controlled induction system" with an internal throttle plate that opens at certain rpms to change the tuning of the intake system.

Here are the figures you required.

1) BPF in and ex-
In= 204cfm @ 28" (121.9 @ 10").
Ex=154cfm @ 28" (92.0 @ 10")

2) Inlet valve fitted- 204cfm @ 28" (121.9 @ 10").
3) BPF inlet port with manifold fitted- 190.1cfm @ 28" (113.6 @ 10") - edited to 10" by GC
4) Inlet valve fitted with manifold- this info was lost in the crash but im almost certain it was per number 3 above.
190.1cfm @ 28" (113.6 @ 10").

The manifold always gave a rather big loss Guy, on the modified port it lost 26cfm@10".
Its a peculiar item in that the front bank of ports with the manifold fitted would have a swan neck shape to the induction path.
The rear set have aproximately a 120 degree redirection and it is these that seem to lose the most.
the lower intake casting has port sizes that differ front to rear and have different angles also- a bit of a nightmare really.
I also took some photos of one of the runners on the lower casting that i sectioned for you to look at.

lower intake- runners not all the same shape/angle.

S8002006 (816 x 612).jpg (132.01 KiB) Viewed 13447 times

[Guy Croft]

This one will be rather awkward to perfect - a bit like the Fiat Coupe 20v I did a while back and worth re-reading the latter phases of inlet valve-in testing, because whilst the outer port region has a reasonable downdraft angle and progressive curvature it's got one of those 'deep drop' short side radius profiles. That's the region betw the blue lines. I think you might find two rather distinct flow regimes there - at low lift the air does tend to follow the SSR (RH black line) and at higher lift, say around 5-6mm it just skates of toward the valve stem.
You've done the right thing smoothing but attempts to increase the radius of that short-side region (ie flatten it) are unlikely to be successful and with heads like that where the port floor is a long way from the valve contact face you might be stuck with it.
Your excellent cutaways tend to suggest that there will be an adverse bias of flow towards the port roof which may go some way to explaining the bolt-on loss. Of course there will likely be quite a high viscous loss too if tested in rough (as-cast) state. Can't do anything about the former.

I'll come back to valve seat/angles but I at first glance I don't like the look of those inlet multi-angles, never got them to work with little valves like this (or even up to 46mm come to that!) might work on massive USA V8s and and sure, even Japanese manufacturers get them wrong...

Do you have a graph of valve-in flow?

G

[4v6]

Guy, many thanks for your efforts thus far, its very very much appreciated, i realise you are a very busy man indeed!

Please see the graph i generated from another of these heads.

Top yellow line- Modified inlet port.
Second line- standard inlet port.
Third line in blue- modified exhaust port.
Fourth line- standard exhaust port.

Once again, thank you, Tony.

[Guy Croft]

Bit of a shift in gradient at 7mm (you agree?) but better than I expected.

More tomorrow!

G

BTW - what is the cylinder swept volume?

[4v6]

Apologies for not responding sooner Guy, i failed to spot your swept volume request.

Bore= 87.5mm
Stroke= 81.5mm

I calculate that as being 490.1 cc per cylinder.

Tony.

[Guy Croft]

Thanks!

On a 490cc cylinder if your figures above are the standard inlet BPF they are not atypical at 204cfm @ 28" (121.9 @ 10"). Some heads with same valves and downdraft with contemporary (pretty straight) ports eg: Fiat/Lancia 16v manage more in std form - nearer 130cfm @10". I would want to try and achieve 150+ @10" - though in saying that I don't know if or how it's possible on this unit. If they are modified figures they are very low.

If the figs are modified the next 2 questions may not be that relevant:

1. You won't get more by valve and seat work (well, not much anyway) - can u see any room for improvement anywhere else?

2. Can you do porting on one dev port before any valve & seat work; always better to do seat dev on a fully (or near fully) developed port. The only exception is that I would keep off the SRR within 15mm from the lower region of the valve insert.
The trick will be to locate the regions of loss and the controlling sections that are holding back the flow - and it's not the valve size that's for sure. A 34mm valve pair can easily permit 150cfm plus - on the right head.

GC

[4v6]

Thank you Guy, thats extremely useful information.

The BPF in its unmodified state is at that 121 cfm @10", however, after porting to the pink mould shape in the photos, it allows the port to go as high as 153cfm@10", but that is with guides that have been shaved down, probably not good for longevity i would think.
More typically is see 143-145cfm @10" with the guides in, so not a million miles away from your suggested 150cfm figure, but maybe as you say not achievable.

I do have a pair of these heads currently sitting patiently under the bench so i can certainly do as you suggest and spend some time more closely investigating one of these ports with an eye to developing it further.

I feel the splitter region, specifically the diameters of the separate barrels/runners may yet yield some extra flow, from memory theyre 25.0mm in diameter, although enlarging them on the scrap heads didnt appear to give anything, but that was rather a long time ago before i was quite as savvy with the flowtester so it could well be myself in error there.

One thing i have noticed Guy; Whilst testing a basic stock port, (i use a ball of plasticine on a wire and see what happens when i move it to various points in the port)-at certain points nothing happens, other times i get a loss.
But, when the splitter "beak" is slimmed down and the pea sized plasticine ball positioned at the base of the splitter, the flow will go up markedly from that basic 121cfm@10" figure to around 140cfm@10".
I presume the air is being diverted advantageously.

The effect does not occur to any degree that i can detect either before or after the rest of the port is modified, only when the splitter is altered on an unmodified port.
Its fairly peculiar.

Maybe i can market the idea? no probably not.

Thank you once again for your time on this Guy, i will do a more in depth study of the ports and report my findings.

Tony.

[Guy Croft]

I'd just remark briefly at this moment that it is by no means unusual to generate quite remarkable gains in BPF with deflectors - I have great fun with that from time to time. However - with valve fitted - none has worked - so far!

GC