2.0L MZR header and exhaust length testing

[digger]

Well, I’ve see Calvin pull off 1 3/8” on a low VE, non-crossflow 4 banger making about 58hp/cyl

It’s a whole engine thing. What can be done in the engine is dictated by what a talented header builder can get away with, and vice versa. Nothing is done in isolation!

if its a whole engine thing then why are you using hp/cyl as the only criteria and not considering CR, cam duration/timing of said engine and so on?

I'm not sure what "pull off" means but unless Calvin himself build 1-3/8", 1-1/2", 1-5/8" how would you know the 1-3/8" was truly optimum? as opposed to simply better than a "off the shelf" one that happened to be larger. I'm sure for prostock and nascar he did that but for his average customer are they really getting several sets of headers each $3k a pop?

At the end of the day Calvin is not going to make this guys headers so one needs to also factor the available skills at hand and other constraints at play.

[hoffman900]

Well, I’ve see Calvin pull off 1 3/8” on a low VE, non-crossflow 4 banger making about 58hp/cyl

It’s a whole engine thing. What can be done in the engine is dictated by what a talented header builder can get away with, and vice versa. Nothing is done in isolation!

if its a whole engine thing then why are you using hp/cyl as the only criteria and not considering CR, cam duration/timing of said engine and so on?

Well if you want to be an ass about it...

I am stating what is possible, not how to do it. If you break Calvin's numbers out by CSA against horsepower, it actually scales to be about 38-40hp / square inch across the board. I noticed this about 7 years ago and posted it here, but like a lot of things, it went over most people's heads.

I'm not sure what "pull off" means but unless Calvin himself build 1-3/8", 1-1/2", 1-5/8" how would you know the 1-3/8" was truly optimum? as opposed to simply better than a "off the shelf" one that happened to be larger. I'm sure for prostock and nascar he did that but for his average customer are they really getting several sets of headers each $3k a pop?

At the end of the day Calvin is not going to make this guys headers so one needs to also factor the available skills at hand and other constraints at play.

Well it was better than a 1.625 stepped Stahl header, which was mostly better than a 1.75" Stahl header (same peak, more mid). So yes, comparing against OTS, but Stahl's aren't what I would call bad headers. Overall primary length was actually the same between all three as was tailpipe.

Okay, well, maybe he should seek the knowledge to improve his header and figure out what went wrong with the Burns header. Otherwise, he should just buy a header and be done with it.

Here is the NP01 header as referenced earlier:

[stealth]

Did the EFI fueling need to be changed with the header changes?

[englertracing]

On a low overlap engine scavenging ability is not as important. The Long header is full of tight bends.
The short header looks terrible on 1 and 3. Can't believe the price of that stuff.

I think this engine may like a stepped header.

Also on that Intake
What if you created a machined adapter plate or plates that transition to from runner size to head size.

Say the plates were 1.5 or 2" thick

[BLSTIC]

If you break Calvin's numbers out by CSA against horsepower, it actually scales to be about 38-40hp / square inch across the board. I noticed this about 7 years ago and posted it here, but like a lot of things, it went over most people's heads.

I put his chart from Exhausting101 into a spreadsheet. Assuming a 'boundary layer' thickness of 0.225" that reduced effective pipe diameter. That number works out to be 75hp/sq-in. (I assumed a boundary layer as it made the hp/sq-in number more constant, literally just guessed until the numbers were even though, no boundary layer calculations were made).

I'm sure it's not completely accurate, but it allows me to extrapolate a bit further down the horsepower/cyl range.

I've attached it if anyone wants to mess around and/or correct my math (it's labelled blowdown as I was initially going to calculate required primary length to avoid blowdown interference at given RPM, but that gets hard really fast and I didn't get any math into it)

[digger]

Well, I’ve see Calvin pull off 1 3/8” on a low VE, non-crossflow 4 banger making about 58hp/cyl

It’s a whole engine thing. What can be done in the engine is dictated by what a talented header builder can get away with, and vice versa. Nothing is done in isolation!

if its a whole engine thing then why are you using hp/cyl as the only criteria and not considering CR, cam duration/timing of said engine and so on?

Well if you want to be an ass about it...

I am stating what is possible, not how to do it. If you break Calvin's numbers out by CSA against horsepower, it actually scales to be about 38-40hp / square inch across the board. I noticed this about 7 years ago and posted it here, but like a lot of things, it went over most people's heads.

I'm not sure what "pull off" means but unless Calvin himself build 1-3/8", 1-1/2", 1-5/8" how would you know the 1-3/8" was truly optimum? as opposed to simply better than a "off the shelf" one that happened to be larger. I'm sure for prostock and nascar he did that but for his average customer are they really getting several sets of headers each $3k a pop?

At the end of the day Calvin is not going to make this guys headers so one needs to also factor the available skills at hand and other constraints at play.

Well it was better than a 1.625 stepped Stahl header, which was mostly better than a 1.75" Stahl header (same peak, more mid). So yes, comparing against OTS, but Stahl's aren't what I would call bad headers. Overall primary length was actually the same between all three as was tailpipe.

Okay, well, maybe he should seek the knowledge to improve his header and figure out what went wrong with the Burns header. Otherwise, he should just buy a header and be done with it.

Here is the NP01 header as referenced earlier:

I'm not trying to be an ass but lets look at this way. Almost every thread with header sizes you always post and imply that because Calvin got some teeny tiny pipe to work anyone not using those sizes based exclusively on hp/cyl is either buying crap, building crap or doesn't know what they are doing when Larry who wrote pipemax and has probably tested more headers than probably anyone else (and those that are accessible to 99% of people) posts the following:

that does not work at all for all Engine combinations
relatively speaking example : a NHRA Record setting low VE% and low CompRatio Pontiac Stocker will use a 2.125" OD Tube , sometimes 2.250" OD
making "half" the HP of the 950bhp 410ci Sprint Car engines use 1 7/8" off the head. That's 119hp/cyl.

This is not directed at Larry but as an aside
the specs i posted earlier with 1.5" off the head for a different engine (higher CR, more cam duration i.e. time and not a 4V head) were for more hp/cyl than the OP's engine and that's what both Burns and PM both recommended

[pcnsd]

If you break Calvin's numbers out by CSA against horsepower, it actually scales to be about 38-40hp / square inch across the board. I noticed this about 7 years ago and posted it here, but like a lot of things, it went over most people's heads.

I put his chart from Exhausting101 into a spreadsheet. Assuming a 'boundary layer' thickness of 0.225" that reduced effective pipe diameter. That number works out to be 75hp/sq-in. (I assumed a boundary layer as it made the hp/sq-in number more constant, literally just guessed until the numbers were even though, no boundary layer calculations were made).

I'm sure it's not completely accurate, but it allows me to extrapolate a bit further down the horsepower/cyl range.

I've attached it if anyone wants to mess around and/or correct my math (it's labelled blowdown as I was initially going to calculate required primary length to avoid blowdown interference at given RPM, but that gets hard really fast and I didn't get any math into it)

Blstic,
Thank you for sharing your thoughts and work. Because I work primarily on motorcycles, I also was looking for a way to expand Calvin's numbers beyond his provided data and also allow application to thinner wall pipes as seen on most motorcycles. (His numbers assume a .063" wall).
I found what I'm calling a constant. It was arrived at by dividing HP by Area by Circumference. It seems to follow his numbers closely for both Broad and Narrow power bands and allows adaption to other wall and pipe sizes. I have attached my work sheet for you and others to review.
Of course to use the constant you multiply the constant x area x circumference.

Test3.xlsx

[Grant]

PipeMax uses only Peak HP RPM input value ... then calculates a 2500 RPM Range with "a best inner strong 1400 to 1500 wide RPM band"
to create best specs for your Engine .... Example : your Engine's Peak HP RPM point = 7000 RPM
PipeMax then calculates the best specs for between 5000 RPM -to- 7500 RPM
with a strong inclination to calculate best inner TQ and HP Curve between 5500 RPM -to- 7000 RPM

Thanks! Our files are more or less identical. I wasn't expecting VE % to be so high. Especially as the help text recommends starting at 100%.

By the way, EAP also says 1-7/8" is the best diameter, at 20" of length. Exactly the same as PipeMax's recommendations.

[Grant]

Did the EFI fueling need to be changed with the header changes?

No, it's a MAF car so the AFR did not change much.

[Grant]

On a low overlap engine scavenging ability is not as important. The Long header is full of tight bends.
The short header looks terrible on 1 and 3. Can't believe the price of that stuff.

I think this engine may like a stepped header.

The NP01 header was originally designed for stock cams, though the new NP01-EVO uses aftermarket cams and as far as I can tell the same header. Why do you say it might like a stepped header?

EAP thinks the car will make a bit more power with the exhaust cam retarded 5 degrees, which would be allowed by our classing. Doing this does not seem to change the optimal header size.

Also on that Intake
What if you created a machined adapter plate or plates that transition to from runner size to head size.

Say the plates were 1.5 or 2" thick

Thanks that's a good idea. We might be able to fit that if they angled the intake away from the firewall slightly. I could 3D print some for dyno testing.

After watching a zillion hours of intake tests on YouTube (mostly Richard Holderner's stuff), and doing a lot of EAP runs, it seems pretty clear that intake design is a lot more important than header design. At least on a motor with limited overlap.

[Grant]

The best plastic porting tool I found was a sharp aluminum (coarse) carbide bit. Before & after:

nc_intake.jpg


I opened it up to the start of the Y, which leaves a really big throat, which I hear is bad for harmonics. Total length to the start / end of the throat is 12.4 / 13.5"

Update:

We tested this change a while ago, opened Y vs. unported manifold with the butterflies removed. The dyno I have access to (Dynojet 248) isn't super repeatable, but the opened Y intake might've made a bit more mid-range. Maybe.

So we said screw it, lets invest a ton more hours into shortening the runners to hit PipeMax's 2nd harmonic:

nc_intake_ruined.jpg

Yeah one of the runners ended up a bit shorter than we wanted, but it didn't matter because the experiment was a total failure. It gained a few ft-lbs from 2800 to 4000, then lost lots of power after 4000. I'm done messing with this intake. The factory short runners at right at PipeMax's 3rd harmonic, and the motor seems happy with that.

(porting is open in our series, but we're penalized for an aftermarket intake)

[Grant]

Also, on the subject of headers, this pathetic little 2.0L wants 2" primaries! Barely.

1.75_v_1.875_v_2_headers_nc_miata.jpg

Anything < 4500 seems to be heavily influenced by post-collector harmonics (these tests were open header), and is below the RPM we race at anyway. Higher RPMs don't seem affected by what's on the other side of the collector.

[maxracesoftware]

So we said screw it, lets invest a ton more hours into shortening the runners to hit PipeMax's 2nd harmonic:

are you sure about this ?

to go from 3rd Harmonic Length -to- 2nd Harmonic Length, you should have "lengthened" it , and not made it shorter !
sounds more like you went from 3rd Harmonic Length -to- 4th Harmonic Length or 5th ??

PipeMax's help text :
"the 3rd Harmonic Length typically creates the best overall combination of Peak Torque and Peak HP"

The factory short runners at right at PipeMax's 3rd harmonic, and the motor seems happy with that.

[Grant]

So we said screw it, lets invest a ton more hours into shortening the runners to hit PipeMax's 2nd harmonic:

are you sure about this ?

to go from 3rd Harmonic Length -to- 2nd Harmonic Length, you should have "lengthened" it

I mis-typed, we meant to go from the 2nd to 3rd. 16" to 11" total length. Obviously a lot more changed, like taper and radius, so it's no surprise it didn't work.