Spoon Headers???

[Honcho]

I agree with you on this, what is important is the area under the curve.

+++++1 :smile:

I just think it is interesting that of all the header manufacturers, only Spoon and Honda chose to go with a short-primary design. We'll probably never know why.

 

[Mac Attack]

+++++1 :smile:

I just think it is interesting that of all the header manufacturers, only Spoon and Honda chose to go with a short-primary design. We'll probably never know why.

It's easy.

The primary tube length is determined by the rear bank layout. Once this is set, of course, you should have the front bank dimensions equal to the rear bank.

You either need to have the primaries short enough (i.e. Honda OEM and Spoon ~12" length) to merge before you make the tight 180degree bend on the rear bank, or you have to have them long enough to merge just after the tight 180degree bend (i.e. all the other designs).

I'm not surprised these two Japanese companies picked the shorter of the two as their designs. Isn't their general philosophy more high-RPM power?

In a real race car, that's how it's preferred. Street is another story where you do want to maximize the total integrated torque over a wider RPM power band.

Like another poster said, it's all about primary length and diameter, secondary diameter, collector layout (3-to-1, 3-2-1), collector shape, and material. There's no perfect design - you have to optimize it for your intended application.

Everything else downstream of your last collector is an obstruction to the exhaust flow.

SOS has, or had, some nice information on their website about a lot of the header piping diameters for the various manufacturers that I was considering. In the end, I would just have Burns stainless make mine up with the dimensions I provide.

Dave

 

[greenberet]

Looking at rear NSX headers, some merge the primary tubes before the 180° bend, some merge in the bend, and some merge after the bend.

Generally, the shorter and thicker the primary tubes, the higher the rpm at which the engine will produce power. The longer and thinner the tubes, the lower the rpm at which power will be maximized.

For good scavenging (having the exhaust pulses help suck all the spent fumes out of the cylinder so that as much fresh fuel/air as possible can enter), equal length header primaries are very helpful. If your primaries aren’t equal length, you can start playing around with different tube diameters for the different cylinders to compensate.

 

[Mac Attack]

Looking at rear NSX headers, some merge the primary tubes before the 180° bend, some merge in the bend, and some merge after the bend.

Thanks, I forgot about the Fujitsubo headers - the most expensive ones available.

I had looked at these before, and I believe they offered the best specs for a modified engine:
Primary diameter: 45mm
Collector diameter: 60mm
(Collector Area) / (Sum of Primary Areas) = 0.59
Primary length ~22 inches

In VTEC, standard cams with a torque peak emphasis at 6500RPM's yields a calculated primary length of 22inches.

Of all the companies, Honda included, I think Fujitsubo took the best approach and didn't make any compromises when it came to performance.

Honda and Spoon on the other hand, made it cheaper for them to make. But of course, this was a big step up from the cast iron manifolds

Dave

 

[Honcho]

Wow. Posts like this are a big reason I'm a Prime member. Great insights, Mac, Ponyboy and greenberet.

If the Fujitsubo is the ideal FI header, doesn't it stand to reason that a header with slightly smaller primaries and collector would be better for NA or stock engines?

 

[Cody@LoveFab]

Hi guys, I think it's time for LoveFab to enter this discussion.

We have been building and testing headers for the NSX over the past 18 months and have shown significant gains over other aftermarket headers in High HP applications(think SC and turbo). The thread below chronicles some of the testing:

http://www.nsxprime.com/forums/showthread.php?t=131643s

Some dyno graphs from this thread:

Stock 1994 vs. our headers(and exhaust):

3.5 Liter Intercooled SC with Cantrell headers vs. LoveFab headers. The LoveFab headers not only picked up significant peak HP/TQ but they produced a HUGE amount of torque "under the curve".

A few pictures:

We build these headers in-house and can customize a few different options such as collector outlet diameter and materials(titanium, 321 stainless, Inconel).

Please contact us for anymore questions!!!

 

[Honcho]

Thanks Cody, I stand corrected.

Ahem, let me restate- if the LoveFab is the ideal FI header, doesn't it stand to reason that a header with slightly smaller primaries and collector would be better for NA or stock engines? Cody, do you think this would optimize the area under the curve for OEM power NSX engines?

 

[Cody@LoveFab]

I think you are correct. Most of the "standard" aftermarket headers are designed for low HP(OEM) gains, and they all use relatively small tubing diameters. These are great for the OEM guys. However, throw in a SC or a turbo, and these headers will become a bottleneck.

We decided to build a header designed to fill that void, which is a high flowing header designed to evacuate and scavenge as much exhaust energy as possible.

FYI, we have tried the "short runner" headers with FI applications and there is a reason we now use long runner headers.

 

[TURBO2GO]

Cody these look longer... how does that work? Do these work with cats?

 

[Mac Attack]

Wow. Posts like this are a big reason I'm a Prime member. Great insights, Mac, Ponyboy and greenberet.

If the Fujitsubo is the ideal FI header, doesn't it stand to reason that a header with slightly smaller primaries and collector would be better for NA or stock engines?

I think the Fujitsubo header WAS the best option for those that just wanted to buy a ready-to-go off the-shelf product for their highly modified engine (increased displacement or supercharged - turbo is a different story).

However, now Lovefab is making some awesome headers (I especially like the collectors) and are willing to customize them for specific applications. The prices from the link they posted is reasonable too for such a quality product.

My $0.02.

Dave

 

[Cody@LoveFab]

We threw convention out the window for the best gains possible.

We can build adapter pipes with Cats included, or an exhaust to bolt to the headers with cat's built in. Stock cats will not work.

Thanks Mac Attack! We have worked very hard on these headers and have had nothing but great results!!!

I may be playing with a Titanium set in the coming weeks, I'm trying to decide if I want to drop the coin on materials.

 

[Ponyboy]

In the end, I would just have Burns stainless make mine up with the dimensions I provide.

Burns Stainless has a computer model design program that takes into account the engine's intended use, displacement, and rpm range and spits out the ideal:

- primary length
- primary diameter
- collector diameter
- primary entry diameter
- primary entry angle
- collector outlet diameter
- collector length

I'm sure HyTech and SMS have their own proprietary versions of this type of program as well. HyTech would, of course, add in their reversion chamber. In fact, I think Burns will custom design a set of headers and let you, or someone else, fab them up.

 

[Ponyboy]

However, now Lovefab is making some awesome headers (I especially like the collectors) and are willing to customize them for specific applications. The prices from the link they posted is reasonable too for such a quality product.

I concur. The craftsmanship evident really is remarkable.

 

[greenberet]

I think Fujitsubo took the best approach and didn't make any compromises when it came to performance.

I don’t know enough about what the best primary lengths, primary diameters, collector shapes, etc. are for various n/a engines, but those Fujitsubos don’t look like they have equal length primaries. I’ve never seen any in person so I can’t say for sure, but from the pictures, it looks like the primary for the #1 cylinder is longer than for the #3.

If there is an optimal diameter for the primaries (to achieve a certain flow speed at a certain rpm), then

• either all the primaries have that diameter and the header won’t scavenge as well as it could because the lengths are different, or
• the primaries have different diameters to compensate for the different lengths so the header will scavenge well, but it will still be suboptimal because you now have different gas speeds coming from different cylinders

Then again, since the OEM, Spoon, Fujitsubo, Comptech, B&B, and DC all look like they’re unequal length, maybe it’s not so bad.

In any case, it’s too bad I didn’t measure the diameter of the primaries of my Comptech headers when I took them off and of my Cantrell headers when I installed them. The length of the Fujitsubo primaries look like they’re between the Cantrell’s and the Comptech’s, so it would be interesting how the diameter compares.

We have been building and testing headers for the NSX over the past 18 months and have shown significant gains over other aftermarket headers in High HP applications(think SC and turbo).

Now those LoveFab headers look like they have very long and very thick primaries and of course, great welding. Since scavenging is basically irrelevant in forced induction engines as far as I understand it (no need to suck fuel/air into the cylinders – it’s getting forced in – and the valve timing should be set minimize overlap anyhow), primary length can be designed pretty much without regard to the RPM at which scavenging will be maximized. However, since it’s good to keep the primaries separate until they reach the turbo so that the exhaust pulses from each individual cylinder hit the vanes with as much energy as possible, you should move the collector as far towards the turbo as you can. Or in the case of a twin-scroll turbo, move the collector into the turbo itself. Also, given the increased volume of the exhaust gasses, larger diameter primaries are in order, too. Those headers look better than anything else out there for turbo and probably SC applications.

Burns Stainless has a computer model design program that takes into account the engine's intended use, displacement, and rpm range and spits out the ideal:

- primary length
- primary diameter
- collector diameter
- primary entry diameter
- primary entry angle
- collector outlet diameter
- collector length

I spoke with Jon Martin (of 3.8 liter stroker fame) a few weeks ago and he felt none of the NSX headers available on the market were really ideal. It would be interesting what that Burns model spits out as ideal for a 3.0 L n/a NSX looking to maximize peak horsepower, a 3.6 L n/a NSX looking to maximize the area under the curve, etc.

 

[StmpoRaceProducts]

Burns Stainless has a computer model design program that takes into account the engine's intended use, displacement, and rpm range and spits out the ideal:

- primary length
- primary diameter
- collector diameter
- primary entry diameter
- primary entry angle
- collector outlet diameter
- collector length

I'm sure HyTech and SMS have their own proprietary versions of this type of program as well. HyTech would, of course, add in their reversion chamber. In fact, I think Burns will custom design a set of headers and let you, or someone else, fab them up.

LOL... could be related to Burns Stainless has an engine dyno.... That company has there sh*t together for sure

We threw convention out the window for the best gains possible.

We can build adapter pipes with Cats included, or an exhaust to bolt to the headers with cat's built in. Stock cats will not work.

Thanks Mac Attack! We have worked very hard on these headers and have had nothing but great results!!!

I may be playing with a Titanium set in the coming weeks, I'm trying to decide if I want to drop the coin on materials.

We have a set of Love Fab headers here and they are by far the highest quality header I have seen. I have yet to get my hands on the spoon and the Fujitsubo.. but the spoon would be the only one that I really want.

Cody is actually right about his design and FI applications... the torque gain and also the lack of volitility of the torque level below 5k rpm is impressive. (seen on the second dyno graph posted)

Cody... I would be interested in seeing a NA header with a colector style similar to the spoon...if we didnt have so much going on we would go after it... since your knee deep in header fab right now... I hope you consider it as a possible product entry.

Regards

 

[Mac Attack]

Burns Stainless has a computer model design program that takes into account the engine's intended use, displacement, and rpm range and spits out the ideal:

- primary length
- primary diameter
- collector diameter
- primary entry diameter
- primary entry angle
- collector outlet diameter
- collector length

I'm sure HyTech and SMS have their own proprietary versions of this type of program as well. HyTech would, of course, add in their reversion chamber. In fact, I think Burns will custom design a set of headers and let you, or someone else, fab them up.

Besides those inputs, you have to know the camshaft specs too. For folks with OEM cams (like me who was just interested in playing around with the numbers), you can find that info in Honda's NSX propaganda :wink:

Also, you don't need a proprietary computer program - the formulas are simple and available for free on the net if you google it.

Dave

 

[Ponyboy]

LOL... could be related to Burns Stainless has an engine dyno.... That company has there sh*t together for sure.

Though Burns may indeed have an engine dyno, the design is all done via the modeling program after a jillion parameters are given by the customer. Here's the spec sheet for example. That said, I'm sure there'll be some type of fine tuning needed.

 

[Ponyboy]

Besides those inputs, you have to know the camshaft specs too.

I know! Check out the extensive parameters listed on the Burns spec sheet. There's a lot of info they need.

Also, you don't need a proprietary computer program - the formulas are simple and available for free on the net if you google it.

Certainly - with the available tools, materials, and basic information, someone could design their own headers according to their own intended use. But make no mistake, the proprietary software that an exhaust manifold company uses versuses what one finds for free on the interweb is not an apple-to-apple comparision.

EDIT: this is a great thread, BTW.

 

[Mac Attack]

But make no mistake, the proprietary software that an exhaust manifold company uses versuses what one finds for free on the interweb is not an apple-to-apple comparision.

Actually, they are unless the header designer is using a tried-and-tested stepped expansion sections for a narrow range of applications, or are using (in our case) a 3-2-1 header instead of 3-1 header that all of the aftermarket headers for us are. Those "tricks" may be built into their software.

It's simple physics, found in every exhaust theory textbook. While the textbooks aren't free, the formulaes found in them are and have been posted on the net. Don't believe everything you read of course, but I've checked and the ones I've seen have been correct. The only tricky part is calculating the crank angle - which is why you need the camshaft specs.

Don't get me wrong - I personally have the Comp Cams software that lets me quickly select a cam based on what I need it for on another vehicle. There's nothing wrong with purchasing any of these softwares.

My main point is to try and dispel some myth that exhaust theory and any proprietary codes are special black magic boxes.

Dave

 

[Honcho]

Imagine if Cody hooked up with Burns's data, you would get a precisely tuned header with LoveFab quality welds, in other words - the ULTIMATE NA HEADER! :biggrin::biggrin:

It costs $75 to have Burns run a spec sheet...

 

[greenberet]

I'm sure there'll be some type of fine tuning needed.

I think you’re right. You can design a good header with formulas but to really optimize it, you’re realistically going to need to test it.

The diameter of the primaries should be set to achieve a certain flow velocity at the rpm where you want to maximize torque. The flow velocity is determined by the volume of the exhaust gases, the time the cylinder has to evacuate them, and the diameter of the primary tubes. The steepness of the camshaft flanks impacts what volume of gases can leave the cylinder when. However, that’s not an input in the Burns sheet. Also, what rpm your intake manifold resonates at impacts the amount of air entering the cylinder and therefore exhausting through the header and that variable is also not considered.

Entering the lift of the camshafts at every degree of rotation, the dimensions of each intake runner, the volume of the intake plenum, etc. is probably more than 99.9% of customers would ever be able to answer. That said, I’m sure you can get a good header design even if you limit yourself to the questions that Burns asks.

It's simple physics, found in every exhaust theory textbook.

Someone in the know! Getting back to the Spoon headers – do you know approximately how different the diameters of the primaries should be given how different their lengths look in the pictures?

 

[Ponyboy]

Actually, they are...

Meh, I don't know about that. I think we may have to agree to disagree here with all due respect. Unless you can comprehensively provide the modeling design software equations from the free interweb software vs. the version that Burns uses or HyTech or SMS, I don't see how that can be written w/ any validity. IMHO, that's a much too simplistic statement to make.

I've asked Burns to show me how their software differs, in general w/o disclosing exclusive design data, from what is available for free online and will definitely report back on what is communicated.

My main point is to try and dispel some myth that exhaust theory and any proprietary codes are special black magic boxes.

Certainly the basics are well covered and I don't think anyone is arguing basic exhaust theory, but there's some distance b/t theory and practical application.

 

[Ponyboy]

The diameter of the primaries should be set to achieve a certain flow velocity at the rpm where you want to maximize torque. The flow velocity is determined by the volume of the exhaust gases, the time the cylinder has to evacuate them, and the diameter of the primary tubes. The steepness of the camshaft flanks impacts what volume of gases can leave the cylinder when. However, that’s not an input in the Burns sheet. Also, what rpm your intake manifold resonates at impacts the amount of air entering the cylinder and therefore exhausting through the header and that variable is also not considered.

I think you're describing some very fine aspects of the entire equation, that .1% that you mention. However, I know race teams in ALMS and NHRA and NASCAR would definitely ask that and many of those teams use companies like Burns to design their headers. B/c it's not listed on the spec sheet doesn't mean that that data couldn't be added.

And Helmholtz Resonances! I haven't heard those discussed since our intake manifold discussions! But you're correct, tuning those Helmholtz Resonances (the second resonance is as far as you want to go, IIRC) on the intake side can ensure better scavenging of exhaust pulses when the intake and exhaust valves are simultaneously open. However, timing those are largely a function of primary length and collector dimensions. Those two factors are, in my understanding, the most indispensible.

 

[TrustJDM]

LoveFab headers look very meaty compared to the Cantrell header. As for the Spoon headers, 10hrs till the auction ends on Yahoo Japan Auction! Someone get it and show us some #'s!

 

[StmpoRaceProducts]

LoveFab headers look very meaty compared to the Cantrell header. As for the Spoon headers, 10hrs till the auction ends on Yahoo Japan Auction! Someone get it and show us some #'s!

If I knew what it said and knew how to buy them I would... under 1000.00 before shipping....

LMK if anyone can do that

Regards