3/5/7 angle valve jobs?
Discussion
Looking at the photo, and with the caveat that it's pretty hard to interpret those, the seats already on there look very nice in terms of geometry and I suspect there really isn't going to be any great scope for finding more flow. The 45 is a decent width as opposed to the ridiculous 1mm seats some people think they need to cut and it appears there are two bottom cut angles, a 60 degree and a 75 or thereabouts so perhaps already a bit better than a straight 3 angle seat although only a flowtest would tell.
What is clear is the seats are pretty badly burned where the inlet and exhaust are adjacent and I suspect some distortion has been taking place in the head, probably because the inlet and exhaust valves are near as dammit touching with very little bridge thickness between them. With valves so big this is going to be a fact of life in severe usage and is going to require the occasional refurb.
So in summary I'd just get your chap to copy the geometry that's already there if he has the cutters to do so. The American CNC head producers spend a bunch of time on the flowbench optimising what they do in what's a fiercely competitive world in the USA and the chances of beating them by much are slim. Any bhp gains you are going to see are only because the freshly cut seats will seal properly again.
A better test would be to have conducted a before and after flowtest somewhere independent but chances are this would only have shown that what was already there was pretty much optimal.
BTW, I didn't really comment originally but it's obvious to most what a 3 angle seat is - a 45, a single angle top cut and a single angle bottom cut. Presumably a 5 angle cut is two bottom cut angles and two top cut angles although I'd like to see flowtests if anyone is claiming any great improvement. However WTF a 7 angle cut could be, and indeed the purpose of such, other than to relieve the punter of more money eludes me.
What is clear is the seats are pretty badly burned where the inlet and exhaust are adjacent and I suspect some distortion has been taking place in the head, probably because the inlet and exhaust valves are near as dammit touching with very little bridge thickness between them. With valves so big this is going to be a fact of life in severe usage and is going to require the occasional refurb.
So in summary I'd just get your chap to copy the geometry that's already there if he has the cutters to do so. The American CNC head producers spend a bunch of time on the flowbench optimising what they do in what's a fiercely competitive world in the USA and the chances of beating them by much are slim. Any bhp gains you are going to see are only because the freshly cut seats will seal properly again.
A better test would be to have conducted a before and after flowtest somewhere independent but chances are this would only have shown that what was already there was pretty much optimal.
BTW, I didn't really comment originally but it's obvious to most what a 3 angle seat is - a 45, a single angle top cut and a single angle bottom cut. Presumably a 5 angle cut is two bottom cut angles and two top cut angles although I'd like to see flowtests if anyone is claiming any great improvement. However WTF a 7 angle cut could be, and indeed the purpose of such, other than to relieve the punter of more money eludes me.
Valve seats - Part Deux
It's been said many times, and bears repeating, that the biggest gains you can get from modifying most heads where the ports are at least about as big as necessary come from the area half an inch below the valve seat to half an inch above it. That means the whole region from the valve throat through to the chamber walls adjacent to the seat need to be treated as a single entity where mistakes in one place will have knock on effects in others.
One of the most important factors in cylinder filling is gas speed in the valve throat just as the inlet valve is closing. At this point the piston is rising rapidly up the bore and trying to force the last of the incoming gas to turn round and go back down the port. If the inlet valve closes too late this is indeed what will happen and the engine will run poorly until the rpm is high enough to let the cam duration work properly.
If we can maintain a high gas speed just below and around the valve seat the kinetic energy of the incoming gas will resist this piston pressure and fill the cylinder better than might otherwise be the case. The conclusion this immediately leads us to is the the valve throat must not be too large or the gas speed will drop.
I've already said that one of the biggest mistakes many tuners make is to cut the valve seats too narrow. This hurts heat dissipation and seat life but also means the valve throat is bigger than it would otherwise be. It's not uncommon to measure heads with narrow seats and short bottom cuts where the valve throat is well over 90% of the diameter of the inlet valve itself. This is pretty much curtains for optimum cylinder filling in a competitive race series where the last few bhp are vital.
In most of the seats I cut the geometry works out that the inlet valve throat ends up at about 86% of the diameter of the inlet valve. (On the exhaust side I go bigger than this but that's another story.) Only on very high specific output engines, well over 100 bhp per litre which means mainly 4v engines do I go much bigger than 86% and even then rarely as high as 90%. In most cases 86% to at most 88% in higher output engines will get the job done very nicely.
In fact if you bore the inlet throat to about 86% of the inlet valve diameter, cut a seat that's 4.5% of the inlet valve diameter wide you'll end up with a 70 degree bottom cut about 2.5 times as long as the valve seat. So for example with a 45mm diameter inlet valve this will end up as a seat 2mm wide and a bottom cut 5mm long. It's a very distinctive appearance to the seat that's unlike what you'll see in most heads with narrow seats and short bottom cuts that are only about as long as the seat itself.
If you use a bottom cut with two angles (the most I'd ever recommend on the inlet side), let's say a 60 degree and a 75 degree, then each of these will end up about the same length as the 45 degree seat itself if the throat is the correct size.
Even for those who know little about cylinder head modifications it's not much trouble to measure the valve throat, compare it to the inlet valve diameter and see if it's in the ballpark 86% to 87% target area. If it is then almost certainly the valve seat will also be a reasonable width although I suppose you could still find a narrow seat with an unduly long bottom cut that also gave an 86% diameter throat.
So for what's very complex subject where only the flowbench and dyno will find the optimum geometry in every specific case these simple rules condense 20 years of testing into something anyone can measure, or specify to a machinist, to see if their cylinder head is good or bad in the critical seat area. The ports are of course another story and not relevant to this particular thread.
If enough people want I'll add a part 3 to explain why exhaust ports are fundamentally different to inlet ones in how they need to be modified and why most tuners fail to realise this and treat them like inlet ones.
It's been said many times, and bears repeating, that the biggest gains you can get from modifying most heads where the ports are at least about as big as necessary come from the area half an inch below the valve seat to half an inch above it. That means the whole region from the valve throat through to the chamber walls adjacent to the seat need to be treated as a single entity where mistakes in one place will have knock on effects in others.
One of the most important factors in cylinder filling is gas speed in the valve throat just as the inlet valve is closing. At this point the piston is rising rapidly up the bore and trying to force the last of the incoming gas to turn round and go back down the port. If the inlet valve closes too late this is indeed what will happen and the engine will run poorly until the rpm is high enough to let the cam duration work properly.
If we can maintain a high gas speed just below and around the valve seat the kinetic energy of the incoming gas will resist this piston pressure and fill the cylinder better than might otherwise be the case. The conclusion this immediately leads us to is the the valve throat must not be too large or the gas speed will drop.
I've already said that one of the biggest mistakes many tuners make is to cut the valve seats too narrow. This hurts heat dissipation and seat life but also means the valve throat is bigger than it would otherwise be. It's not uncommon to measure heads with narrow seats and short bottom cuts where the valve throat is well over 90% of the diameter of the inlet valve itself. This is pretty much curtains for optimum cylinder filling in a competitive race series where the last few bhp are vital.
In most of the seats I cut the geometry works out that the inlet valve throat ends up at about 86% of the diameter of the inlet valve. (On the exhaust side I go bigger than this but that's another story.) Only on very high specific output engines, well over 100 bhp per litre which means mainly 4v engines do I go much bigger than 86% and even then rarely as high as 90%. In most cases 86% to at most 88% in higher output engines will get the job done very nicely.
In fact if you bore the inlet throat to about 86% of the inlet valve diameter, cut a seat that's 4.5% of the inlet valve diameter wide you'll end up with a 70 degree bottom cut about 2.5 times as long as the valve seat. So for example with a 45mm diameter inlet valve this will end up as a seat 2mm wide and a bottom cut 5mm long. It's a very distinctive appearance to the seat that's unlike what you'll see in most heads with narrow seats and short bottom cuts that are only about as long as the seat itself.
If you use a bottom cut with two angles (the most I'd ever recommend on the inlet side), let's say a 60 degree and a 75 degree, then each of these will end up about the same length as the 45 degree seat itself if the throat is the correct size.
Even for those who know little about cylinder head modifications it's not much trouble to measure the valve throat, compare it to the inlet valve diameter and see if it's in the ballpark 86% to 87% target area. If it is then almost certainly the valve seat will also be a reasonable width although I suppose you could still find a narrow seat with an unduly long bottom cut that also gave an 86% diameter throat.
So for what's very complex subject where only the flowbench and dyno will find the optimum geometry in every specific case these simple rules condense 20 years of testing into something anyone can measure, or specify to a machinist, to see if their cylinder head is good or bad in the critical seat area. The ports are of course another story and not relevant to this particular thread.
If enough people want I'll add a part 3 to explain why exhaust ports are fundamentally different to inlet ones in how they need to be modified and why most tuners fail to realise this and treat them like inlet ones.
I have a spreadsheet that I've used for the last 15 yrs or so that calculates seat width, valve throat dimensions, inner seat diameter, bowl size and port size for any given valve size. It's pretty consistent with the previously suggested parameters. If anyone would like a copy, let me know.
Dave
Dave
Pumaracing said:
If enough people want I'll add a part 3 to explain why exhaust ports are fundamentally different to inlet ones in how they need to be modified and why most tuners fail to realise this and treat them like inlet ones.
Yes please, im learning a great deal reading your posts 
Valve Seats - Part C
It might seem like a statement of the blindingly obvious that exhaust ports flow in the opposite direction to inlets but the significance of that simple fact is still lost on many engine tuners. In large part what determines the flow efficiency of inlet ports is how well the port guides the air to fully utilise the entire circumference of the valve seat. In nearly all cases this cannot be achieved because air travelling at high speed can't get round the short side bend in the port and just skips across the back of the valve head and tries to exit through the long side of the valve seat. Only on the most steeply downdraft and straight ports is there such a minimal bend on the short side of the port that the air can come close to using the full circumference of the valve seat at high valve lifts.
If we take a round straight tube of the same I/D as the valve head as representing 100% flow efficiency then the average stock inlet port only flows at about 50% of this at full lift. Well modified road heads, especially the more downdraft ones, can reach 60% but to get any more requires custom designed ports that are as close to straight as possible.
The exhaust port suffers from none of these problems. The air already uses the full circumference of the valve seat by virtue of the fact it's going into the seat from all round the combustion chamber anyway. Provided the port itself is big enough then almost any design of exhaust port will flow at 65% efficiency or even a tad more because the valve seat is being used so effectively. Even the shape of the short side bend is not that critical because it's not having to guide the air into a seat - all it's doing is guiding the air into the straight section of the port.
Nevertheless most tuners will port the exhaust in exactly the same way as they port the inlets i.e try and maximise the short side radius as if the air needed to get round this in the opposite direction. In fact it's more important to try and prevent reverse flow in the exhaust port which can help the engine run with longer duration cams and come "on the cam" at lower rpm. To this end it can actually be helpful if the short side bend is more square which actively prevents any reverse flow from utilising the valve seat effectively. Anyway that's drifting into the complexities of porting so back to valve seats.
Because exhaust ports flow so efficiently they need proportionally bigger valve throats (and ports) than on the inlet side. A seat width of about 5.5% of the valve diameter is ideal for heat dissipation and good flow. In most cases this will result in the same seat width as was used on the inlet side. On very highly developed race engines with large inlets and small exhaust valves it can even result in the exhaust seat width being less than the inlet seat width which might raise eyebrows but is nonetheless the best way to do things.
Under this a bottom cut of 70 degrees of about the same length as the seat or perhaps a tad longer will result in a valve throat of 88% to 89% of the valve diameter. Only on very high specific output engines (120 bhp per litre plus) might it be necessary to alter these guidelines. Fully radiused bottom cuts actually work well on the exhaust side but this requires either special cutters or very delicate hand work. For the average engine it's not worth the money. BTW you don't want radiused bottom cuts on the inlet side because sharp angles between the cuts helps keep the fuel droplets in suspension.
As for the top cut this can usually be a shallower angle than on the inlet side because it's guiding air into the seat rather than out of it and the problem of the airflow breaking off the seat and becoming turbulent is not an issue. In practice most heads are inlet flow limited rather than exhaust flow limited so worrying too much about the tiniest details of exhaust seats and port shapes is fruitless and won't add to power. Top cuts of anywhere between 20 and 30 degrees will do the job.
So there we have it. Guidelines for seat widths, angles and throat size on both inlet and exhaust sides that will in all cases be infinitely better than either a single angle 45 seat or the very old hat 30/45/60 seats that really should have gone out with the ark.
For completeness I should just cover the valves themselves briefly. The O/D of the seat in the head should always be exactly the same as the O/D of the valve head. Similarly the seat widths on both valve and seat should be identical. Behind the 45 degree seat on the valve there should be a backcut of 30 degrees leading into the back of the valve head. This backcut alone on the inlet valve can be worth 3% of the engines potential horsepower compared to a standard valve with a much longer 45 degree seat on the valve than in the head itself.
If I was competing against the American drag race engine builders who spend thousands of man hours a year on flow and dyno testing I would no doubt have to vary these rules of thumb slightly for specific engine designs. In fact state of the art engines don't even use 45 degree seats anymore. 50 and 55 degrees work better at very high valve lifts. However compared to what most engine reconditioners, or even race engine builders will inflict on your cylinder head just going straight to what I've suggested will be so close to optimum in nearly every case you'd have to spend a small fortune trying to better it.
Summary
Throat diameter - Inlet 86%, Exhaust 88.5% (of valve head diameter)
Seat width - Inlet 4.5%, exhaust 5.5% (of valve head diameter)
Top cut - Inlet variable depending on chamber roof, Exhaust 20-30 degrees.
Bottom cut - 70 degrees in both cases. On the inlet side an intermediate bottom cut splitting the 45 and 70 degree cuts (i.e. 57.5 degrees) would be my idea of perfection if cost warranted the extra work. On the exhaust side a fully radiused bottom cut could be used with the same proviso. We're only gilding the lily here though.
It might seem like a statement of the blindingly obvious that exhaust ports flow in the opposite direction to inlets but the significance of that simple fact is still lost on many engine tuners. In large part what determines the flow efficiency of inlet ports is how well the port guides the air to fully utilise the entire circumference of the valve seat. In nearly all cases this cannot be achieved because air travelling at high speed can't get round the short side bend in the port and just skips across the back of the valve head and tries to exit through the long side of the valve seat. Only on the most steeply downdraft and straight ports is there such a minimal bend on the short side of the port that the air can come close to using the full circumference of the valve seat at high valve lifts.
If we take a round straight tube of the same I/D as the valve head as representing 100% flow efficiency then the average stock inlet port only flows at about 50% of this at full lift. Well modified road heads, especially the more downdraft ones, can reach 60% but to get any more requires custom designed ports that are as close to straight as possible.
The exhaust port suffers from none of these problems. The air already uses the full circumference of the valve seat by virtue of the fact it's going into the seat from all round the combustion chamber anyway. Provided the port itself is big enough then almost any design of exhaust port will flow at 65% efficiency or even a tad more because the valve seat is being used so effectively. Even the shape of the short side bend is not that critical because it's not having to guide the air into a seat - all it's doing is guiding the air into the straight section of the port.
Nevertheless most tuners will port the exhaust in exactly the same way as they port the inlets i.e try and maximise the short side radius as if the air needed to get round this in the opposite direction. In fact it's more important to try and prevent reverse flow in the exhaust port which can help the engine run with longer duration cams and come "on the cam" at lower rpm. To this end it can actually be helpful if the short side bend is more square which actively prevents any reverse flow from utilising the valve seat effectively. Anyway that's drifting into the complexities of porting so back to valve seats.
Because exhaust ports flow so efficiently they need proportionally bigger valve throats (and ports) than on the inlet side. A seat width of about 5.5% of the valve diameter is ideal for heat dissipation and good flow. In most cases this will result in the same seat width as was used on the inlet side. On very highly developed race engines with large inlets and small exhaust valves it can even result in the exhaust seat width being less than the inlet seat width which might raise eyebrows but is nonetheless the best way to do things.
Under this a bottom cut of 70 degrees of about the same length as the seat or perhaps a tad longer will result in a valve throat of 88% to 89% of the valve diameter. Only on very high specific output engines (120 bhp per litre plus) might it be necessary to alter these guidelines. Fully radiused bottom cuts actually work well on the exhaust side but this requires either special cutters or very delicate hand work. For the average engine it's not worth the money. BTW you don't want radiused bottom cuts on the inlet side because sharp angles between the cuts helps keep the fuel droplets in suspension.
As for the top cut this can usually be a shallower angle than on the inlet side because it's guiding air into the seat rather than out of it and the problem of the airflow breaking off the seat and becoming turbulent is not an issue. In practice most heads are inlet flow limited rather than exhaust flow limited so worrying too much about the tiniest details of exhaust seats and port shapes is fruitless and won't add to power. Top cuts of anywhere between 20 and 30 degrees will do the job.
So there we have it. Guidelines for seat widths, angles and throat size on both inlet and exhaust sides that will in all cases be infinitely better than either a single angle 45 seat or the very old hat 30/45/60 seats that really should have gone out with the ark.
For completeness I should just cover the valves themselves briefly. The O/D of the seat in the head should always be exactly the same as the O/D of the valve head. Similarly the seat widths on both valve and seat should be identical. Behind the 45 degree seat on the valve there should be a backcut of 30 degrees leading into the back of the valve head. This backcut alone on the inlet valve can be worth 3% of the engines potential horsepower compared to a standard valve with a much longer 45 degree seat on the valve than in the head itself.
If I was competing against the American drag race engine builders who spend thousands of man hours a year on flow and dyno testing I would no doubt have to vary these rules of thumb slightly for specific engine designs. In fact state of the art engines don't even use 45 degree seats anymore. 50 and 55 degrees work better at very high valve lifts. However compared to what most engine reconditioners, or even race engine builders will inflict on your cylinder head just going straight to what I've suggested will be so close to optimum in nearly every case you'd have to spend a small fortune trying to better it.
Summary
Throat diameter - Inlet 86%, Exhaust 88.5% (of valve head diameter)
Seat width - Inlet 4.5%, exhaust 5.5% (of valve head diameter)
Top cut - Inlet variable depending on chamber roof, Exhaust 20-30 degrees.
Bottom cut - 70 degrees in both cases. On the inlet side an intermediate bottom cut splitting the 45 and 70 degree cuts (i.e. 57.5 degrees) would be my idea of perfection if cost warranted the extra work. On the exhaust side a fully radiused bottom cut could be used with the same proviso. We're only gilding the lily here though.
Thanks for all that Mr Puma,I've toyed with stuff for years [on my own engines ] mainly by guess work ,some good some bad ,some no change but never really had the principals explained to me your posts have been realy great and will help me to see probs . I've been playing for 40 years plus but keep finding guys like you who can teach me a lot thank you big style !!
Pumaracing said:
If we can maintain a high gas speed just below and around the valve seat the kinetic energy of the incoming gas will resist this piston pressure and fill the cylinder better than might otherwise be the case. The conclusion this immediately leads us to is the the valve throat must not be too large or the gas speed will drop.
How does this rule alter when you introduce forced induction? We can keep the valve open a little longer, i'd assume, and use a bigger throat as the gas speed is artificially boosted? Would you alter the seat specifications from what has been talked about?
DaveL485 said:
Pumaracing said:
If we can maintain a high gas speed just below and around the valve seat the kinetic energy of the incoming gas will resist this piston pressure and fill the cylinder better than might otherwise be the case. The conclusion this immediately leads us to is the the valve throat must not be too large or the gas speed will drop.
How does this rule alter when you introduce forced induction? We can keep the valve open a little longer, i'd assume, and use a bigger throat as the gas speed is artificially boosted?The port gas speed doesn't alter a jot when an engine has forced induction. If the manifold pressure doubles then so does the gas density and therefore the mass flow and power. The speed stays the same.
DaveL485 said:
Would you alter the seat specifications from what has been talked about?
I don't change anything with forced induction engines, even very high boost ones, with the possible exception of my underwear.Brummmie said:
Puma..
Thinning those guide bosses out and matching the throttle bodies must help out too?
Not having been present to see what was done and what they were like to start with I couldn't possibly surmise. I very much doubt it though as guide bosses tend not to be anything like as critical as people tend to think and the manifold end of ports are usually too big rather than too small.Thinning those guide bosses out and matching the throttle bodies must help out too?
Pumaracing said:
Not having been present to see what was done and what they were like to start with I couldn't possibly surmise. I very much doubt it though as guide bosses tend not to be anything like as critical as people tend to think and the manifold end of ports are usually too big rather than too small.
And reducing guides could have a negative impact on guide life and it's ability to remove heat from the valve.Unless there is absolute proof of very good gains, leave a nice full guide in there.
Especially given the high lifts used ( ie 16-17mm range ) which can see high side loadings on the valve despite roller rockers.
Gassing Station | Engines & Drivetrain | Top of Page | What's New | My Stuff