Using Blair to spec a Cam

[GARY C]

Bob, that was friendly sarcasm. That's the problem with typed words on a forum; especially one where everyone has gone into their respective corners and are sitting on the edge of their seats just waiting to pounce. Voice inflection is gone.

You bet your ass that Jon will never and I mean never give an answer but I'm not going to blow up Stan's thread because I have zero respect for people like Jon that do little more than dance around a question but will never put their name on an answer.

Never give an answer to what?

They want you to pick out a camshaft in some cam catalog for the ST default 500hp, 383ci Chevy, with a single carburetor, AFR heads, and 10:1 compression. This will be for a street car, but we can ignore that because the dyno test won't start until 3500rpm and who cares below that point. AmIrightorwhat?

So Rev, take note. This isn’t about just yanking some camshaft out of a catalog. This is about breaking down an engine and looking at can we make more? Can other rules of thumbs (plural, there are a lot out there form multiple sources) maximize the combination? Does any given rule of thumb get you the maximum to what the combination is capable of? Is a good result good enough? This is Speedtalk, I think lots here are interested in the pursuit of more power, at least it use to be. It seems too many here are interested in some generic recipe magazine type build for some ancient architecure. That’s fine, but if that’s you, then too mucking up the threads for Thebes other guys. I know many in this thread are not interested in that. Is, for example, a 500hp 383ci pump gas engine good? Sure. Is it everything that I can be? I don’t known. We need to break it down and have a look.

Lots here come from other schools of experience or academics. If we see technical errors, we should be able to point them out. If the poster is willing to make their theories public, they should be able to take the critique and improve their product. All should be done without malice. So really, stop with the posting like you’re in the cesspool that is the political forum. You’re certainly capable of it.

Stan, I'll have some input later in the week when I get some more time to read up on this. I also want to try some other things with these two combinations (on the exhaust lengths and intake lengths - maximize overall power for both). Then we can break them down and see what's going on.

In all honesty from both personal conversation and PM from several here they would just like to see Jon for once post a # to something, especially this engine as it is in his self proclaimed wheel house but I will not hold my breath.

But hopefully others that truly understand this line of work will be willing to post some numbers as others have already.

I really wish this was an actual project and not just a theory so it could be quantified in the end.

[SchmidtMotorWorks]

In all honesty from both personal conversation and PM from several here they would just like to see Jon for once post a # to something, especially this engine as it is in his self proclaimed wheel house but I will not hold my breath.

But hopefully others that truly understand this line of work will be willing to post some numbers as others have already.

I really wish this was an actual project and not just a theory so it could be quantified in the end.

If you think it is a theory, you don't understand what we are discussing.
The discussion is about the concept of how a formula is built, not the coefficients.
Please put in the minimal effort required to understand that and your concerns about predictions will vanish.
If you need me to explain how that works, just ask and I will.

Please begin by watching this video.
https://www.youtube.com/watch?v=tl5QNhSe0Yk

[GARY C]

In all honesty from both personal conversation and PM from several here they would just like to see Jon for once post a # to something, especially this engine as it is in his self proclaimed wheel house but I will not hold my breath.

But hopefully others that truly understand this line of work will be willing to post some numbers as others have already.

I really wish this was an actual project and not just a theory so it could be quantified in the end.

If you think it is a theory, you don't understand what we are discussing.
The discussion is about the concept of how a formula is built, not the coefficients.
Please put in the minimal effort required to understand that and your concerns about predictions will vanish.
If you need me to explain how that works, just ask and I will.

Please begin by watching this video.
https://www.youtube.com/watch?v=tl5QNhSe0Yk

This is your chance to school us on your knowledge of Blair and show us how to use it to spec a cam for an engine that virtually no one on this site has even seen...Hell at this point you could just make up numbers and 95% of people on here would prays you because they have no idea.

Please don't respond with DBL talk and deflect to someone else empirical data followed be some suggestions of what I should do... If you can do the math then prove it!
If you can't than just admit it!
I can't and it doesn't bother me at all to say so but I did have the balls to put up a number...maybe way off in left field but who will ever know in a theoretical engine.

[SchmidtMotorWorks]

In all honesty from both personal conversation and PM from several here they would just like to see Jon for once post a # to something, especially this engine as it is in his self proclaimed wheel house but I will not hold my breath.

But hopefully others that truly understand this line of work will be willing to post some numbers as others have already.

I really wish this was an actual project and not just a theory so it could be quantified in the end.

If you think it is a theory, you don't understand what we are discussing.
The discussion is about the concept of how a formula is built, not the coefficients.
Please put in the minimal effort required to understand that and your concerns about predictions will vanish.
If you need me to explain how that works, just ask and I will.

Please begin by watching this video.
https://www.youtube.com/watch?v=tl5QNhSe0Yk

This is your chance to school us on your knowledge of Blair and show us how to use it to spec a cam for an engine that virtually no one on this site has even seen...Hell at this point you could just make up numbers and 95% of people on here would prays you because they have no idea.

Please don't respond with DBL talk and deflect to someone else empirical data followed be some suggestions of what I should do... If you can do the math then prove it!
If you can't than just admit it!
I can't and it doesn't bother me at all to say so but I did have the balls to put up a number...maybe way off in left field but who will ever know in a theoretical engine.

Anyone that tries can do the math.
Stan is working on a way to present it.
If he can't do it easily in a spreadsheet, maybe one of us will show how to do it in a program, that is the most practical way to do it so that it will be useful.
If enough people here are interested in putting in the effort to learn how to program, I would be pleased to teach them how. But they have to be willing to put in some effort. I can't say strongly enough how valuable of a skill/tool, knowing how to program is, it will change how you look at solving many problems.

What engine are you talking about that "virtually no one on this site has even seen"?

Do you understand what data fitting is?
Do you understand that the 128 formula is supposed to be based on data fitting?
Do you understand that the Blair STA is also a method of data fitting that just considers more relevant data than 128?
Do you understand that Blair explains how the formula is constructed?
Do you understand that he shows the data that it is fit to?
Do you understand that any data fitting method can be improved by each new data point that it can be refit to?

If any of these things do not make sense to you, just ask and I will explain it to you.

[Stan Weiss]

Jon,

I have tried to keep this thread on track and only talk about the Blair method.

Why do you have to keep going back to David's 128?

David took a very complex set of formulas and "dummy-ed" it down to basically a one liner.

David has also created a somewhat middle of the road program which does ask what engine and then wants to if it is solid or hydraulic and also if it is solid or roller tappet .

It also take more inputs than the 128 one liner.

Finally David like everybody else has to earn a living and can only give so much away for free.

Stan

TMC.gif

[RevTheory]

Bob, you're reading what I said wrong if you think I only know how to do a 383. I chose that engine on purpose because it's probably the easiest damn engine to cam and I wanted to pin Jon down to an actual number instead of talking in condescending circles but he'll never do it.

If I remember correctly, David said that if he had to write his program from scratch, it would take him a month to enter all the equations into Excel. He was exaggerating with a little humor but the point is that he knows far more than you guys want to give him credit for. He dumbed down one part to make LSA easier on a specific platform and you guys have just gone off the damn rails over it.

It's totally cool for Jon to say that an equation from Blair can be tweaked if someone has thousands of tests and empirical data but if David does it, empirical data is meaningless.

Sorry Stan, I'll quit now

[BobbyB]

Jon, you said

"A good example of this is the Coyote engine, when people drive it without knowing what manifold is on it, the F150 manifold gets high approval, when they choose from a dyno sheet, the F150 manifold is never chosen."

What is the "take away" from this? What is it about the f150 dyno sheet and how the f150 intake feels when you drive it? What do people really want from an engine that is not what they think they want when they look at a dyno sheet?

[SchmidtMotorWorks]

Jon, you said

"A good example of this is the Coyote engine, when people drive it without knowing what manifold is on it, the F150 manifold gets high approval, when they choose from a dyno sheet, the F150 manifold is never chosen."

What is the "take away" from this? What is it about the f150 dyno sheet and how the f150 intake feels when you drive it? What do people really want from an engine that is not what they think they want when they look at a dyno sheet?

This is a comparison of the F150 vs a Ford Racing Cobra Jet manifold on the same same engine with same cam tining.
The CJ manifold looks better on a graph, but most people would enjoy a street car more with the F150 manifold.

CJvsF150.JPG

[GARY C]

Let me start by saying if you are not going to post anything that adds something technical to the thread PLEASE don't post. I know that could mean this will be the only post in this thread.

Let see how using Blair will help me select a cam.

I am going to use a 535 ci Pontiac engine for this example. bore = 4.35", stroke = 4.5", rod = 6.7", cr = ??? (Pump gas), Intake Valve = 2.22", Exhaust Valve = 1.77".

I want 900 HP @ 7500 RPM

Torque = (Horse Power * 5252) / RPM
► (900 * 5252) / 7500 = 630.24

BMEP = (Torque * 150.8 ) / Cubic Inches
► (630.24 * 150.8 ) / 535 = 177.645

Since Blair uses metric 177.645 = 12.2482 Bars

► BMEP_BARS = 12.2482
► (5.02 * BMEP_BARS + 57.78 ) / 10000 = 0.0119265964
► (1.7775 * BMEP_BARS + 74.822) / 10000 = 0.009659317549999999
► (4.1185 * BMEP_BARS - 17.985) / 10000 = 0.00324592117
► (3.0296 * BMEP_BARS - 11.363) / 10000 = 0.002574414672
► (1.6329 * BMEP_BARS - 7.1871) / 10000 = 0.001281298578
► (2.4022 * BMEP_BARS - 14.57) / 10000 = 0.001485262604

OK, so what is my next step?

Stan

Blair.gif

So do you feel it's possible to turn this into an entire cam shaft or is it pretty much for a single lobe?

[mk e]

This is a comparison of the F150 vs a Ford Racing Cobra Jet manifold on the same same engine with same cam tining.
The CJ manifold looks better on a graph, but most people would enjoy a street car more with the F150 manifold.

Which is better apples or oranges?

I have a buddy who does H-D stuff and the standard customer is "this cam makes 5 more hp, I want it!". Buddy points out "buts its 20hp less where you actually ride so your friends will destroy you in the roll-on game you love to play with the cam I gave you, I'll do it but you won't be happy". .....and they almost always have it changed back.

It's really hard to design for what's "best" when best isn't defined....and trying to define best is pretty hard once you leave the race track and don't have the luxury of knowing what was best on the day.

[SchmidtMotorWorks]

Which is better apples or oranges?

I have a buddy who does H-D stuff and the standard customer is "this cam makes 5 more hp, I want it!". Buddy points out "buts its 20hp less where you actually ride so your friends will destroy you in the roll-on game you love to play with the cam I gave you, I'll do it but you won't be happy". .....and they almost always have it changed back.

It's really hard to design for what's "best" when best isn't defined....and trying to define best is pretty hard once you leave the race track and don't have the luxury of knowing what was best on the day.

Exactly, ^^^^^^
If you do what is most reasonable, people will complain that you left power on the table.
Then you optimize for power and they complain about drivability.
Then you make it back to reasonable and they like it.

The first two moves are a disappointment for most people even when you do what is most reasonable.
That is why calling out best specs is a fools errand

[vannik]

When creating an equation from data there is a certain process one follows. What Blair did with STA is to start with the required horse power and split the equation into its components. He then manipulated the equation to end with the one that states that BMEP is proportional to Delivery Ratio.

Back_to_Basics-P38.JPG

Next he relates DR to STA:

Back_to_Basics-P39.JPG

So now we have BMEP proportional to STA.

The key word here is "proportional", it is used where two sides of an equation starts off as being equal but by removing some of the less important terms they are no longer equal but proportional. This implies that if the one increases by 10% so will the other one, but they are not equal. A way to make them equal is to use a constant value to multiply the one side. This constant is usually determined from measured data. This is not the same as fitting a trend line to measured data and using the so derived formula.

Both have their uses but one should never confuse the one with the other, one is based on physics and data and the other on data only.

The person in the tuning world with the most data to get really good STA constants is probably DV but as Gary C said, that does not pay.

[Stan Weiss]

If I was looking to do this for myself or someone was paying me for the answer. I would use some routines I already have. I would get some engine specs (RPM, intake and Exhaust valve sizes, ...). I would then input cam specs (max lobe lift, seat-to-seat duration, 0.050" duration, 0.100" duration, ...) , generate a lift profile every 1 degree, and then calculate the TA's. I would have the LSA and ICL (also if used rocker arm ratio) as separate specs so that I could adjust these while still using the same lift file. This would be a trail and error method of looking at this output against Blair's numbers and than going back and generating a new lift profile until they just about matched. I believe once I had done a few cams it would not take that long to get the needed TA's

Stan

Let me start by saying if you are not going to post anything that adds something technical to the thread PLEASE don't post. I know that could mean this will be the only post in this thread.

Let see how using Blair will help me select a cam.

I am going to use a 535 ci Pontiac engine for this example. bore = 4.35", stroke = 4.5", rod = 6.7", cr = ??? (Pump gas), Intake Valve = 2.22", Exhaust Valve = 1.77".

I want 900 HP @ 7500 RPM

Torque = (Horse Power * 5252) / RPM
► (900 * 5252) / 7500 = 630.24

BMEP = (Torque * 150.8 ) / Cubic Inches
► (630.24 * 150.8 ) / 535 = 177.645

Since Blair uses metric 177.645 = 12.2482 Bars

► BMEP_BARS = 12.2482
► (5.02 * BMEP_BARS + 57.78 ) / 10000 = 0.0119265964
► (1.7775 * BMEP_BARS + 74.822) / 10000 = 0.009659317549999999
► (4.1185 * BMEP_BARS - 17.985) / 10000 = 0.00324592117
► (3.0296 * BMEP_BARS - 11.363) / 10000 = 0.002574414672
► (1.6329 * BMEP_BARS - 7.1871) / 10000 = 0.001281298578
► (2.4022 * BMEP_BARS - 14.57) / 10000 = 0.001485262604

OK, so what is my next step?

Stan

Blair.gif

So do you feel it's possible to turn this into an entire cam shaft or is it pretty much for a single lobe?

Gary,

The purpose of this thread was to convert Blair's STA's into a camshaft and then see how accurate that cam was.

Using the above quoted STA's and method (No I am going to write that program) I would get the following:

Intake and exhaust lift files that needed to be smoothed and check that mechanical they would work. I would also be able to tell you seat_to-seat, 0.050", 0.100", 0.200" 0.300" duration's of each lobe a long with the needed LSA and ICL.

Stan

[GARY C]

If I was looking to do this for myself or someone was paying me for the answer. I would use some routines I already have. I would get some engine specs (RPM, intake and Exhaust valve sizes, ...). I would then input cam specs (max lobe lift, seat-to-seat duration, 0.050" duration, 0.100" duration, ...) , generate a lift profile every 1 degree, and then calculate the TA's. I would have the LSA and ICL (also if used rocker arm ratio) as separate specs so that I could adjust these while still using the same lift file. This would be a trail and error method of looking at this output against Blair's numbers and than going back and generating a new lift profile until they just about matched. I believe once I had done a few cams it would not take that long to get the needed TA's

Stan

Let me start by saying if you are not going to post anything that adds something technical to the thread PLEASE don't post. I know that could mean this will be the only post in this thread.

Let see how using Blair will help me select a cam.

I am going to use a 535 ci Pontiac engine for this example. bore = 4.35", stroke = 4.5", rod = 6.7", cr = ??? (Pump gas), Intake Valve = 2.22", Exhaust Valve = 1.77".

I want 900 HP @ 7500 RPM

Torque = (Horse Power * 5252) / RPM
► (900 * 5252) / 7500 = 630.24

BMEP = (Torque * 150.8 ) / Cubic Inches
► (630.24 * 150.8 ) / 535 = 177.645

Since Blair uses metric 177.645 = 12.2482 Bars

► BMEP_BARS = 12.2482
► (5.02 * BMEP_BARS + 57.78 ) / 10000 = 0.0119265964
► (1.7775 * BMEP_BARS + 74.822) / 10000 = 0.009659317549999999
► (4.1185 * BMEP_BARS - 17.985) / 10000 = 0.00324592117
► (3.0296 * BMEP_BARS - 11.363) / 10000 = 0.002574414672
► (1.6329 * BMEP_BARS - 7.1871) / 10000 = 0.001281298578
► (2.4022 * BMEP_BARS - 14.57) / 10000 = 0.001485262604

OK, so what is my next step?

Stan

Blair.gif

So do you feel it's possible to turn this into an entire cam shaft or is it pretty much for a single lobe?

Gary,

The purpose of this thread was to convert Blair's STA's into a camshaft and then see how accurate that cam was.

Using the above quoted STA's and method (No I am going to write that program) I would get the following:

Intake and exhaust lift files that needed to be smoothed and check that mechanical they would work. I would also be able to tell you seat_to-seat, 0.050", 0.100", 0.200" 0.300" duration's of each lobe a long with the needed LSA and ICL.

Stan

Will you be able to adjust for valve angle and different seat angles? I found if I rerun my numbers based on a steeper seat I am close to Larry's 108 LSA cam.

[pastry_chef]

Will you be able to adjust for valve angle and different seat angles?

I think it was mentioned those STA values don't require flow CFM.
Also noted if you had a set of values perfected for a certain port design you would need to re-evaluate for a deviation. Port, valve angle, seats etc.