Calling Max_Torque - manifolds and turbos question

[Luther Blisset]

Sounds like a great project, you've got enough room to make the runners as straight as possible, and avoid a sharp 90* bend straight out of the port or sharp angles at the merge collector. 38mm seems to be sufficient for 700hp on a 4-cylinder. As said length is not a big deal until you're into extreme territory, just avoid having the exhaust flows fighting each other, just like a good NA design. DP in the Netherlands make turbo manifolds and they focus too much on equalised lenghts and they end up looking like a ball of spaghetti.
Avoid this! Keep things as straight and simple as you can.

[sl0wlane]

Interestingly I gave all the engine, cam, desired usage, power band and turbo specs to Burns stainless (http://www.burnsstainless.com) who offer a header design service.

And they came back with a suggestion that was not a million miles away from what I was going to go with: 1.7/8 diameter primaries of a 23" length (I had worked out 24")... so I guess we will start here for the first iteration!

[sl0wlane]

Luther Blisset said:

DP in the Netherlands make turbo manifolds and they focus too much on equalised lenghts and they end up looking like a ball of spaghetti.

Wow! They certainly do! I'll take my manifold with some built in interheating lol :

[jontysafe]

This is my custom twin scroll manifold for a YB I'm building. Severely space restricted so very short runner.

Don't over concentrate on one aspect of build, give the job to someone that knows what they are doing. I over concentrated on intercooling and it probably added 6months to build.

[stevieturbo]

sl0wlane said:

Interestingly I gave all the engine, cam, desired usage, power band and turbo specs to Burns stainless (http://www.burnsstainless.com) who offer a header design service.

And they came back with a suggestion that was not a million miles away from what I was going to go with: 1.7/8 diameter primaries of a 23" length (I had worked out 24")... so I guess we will start here for the first iteration!

Even at OD, unless it's very thickwall tubing ( ie to reduce the ID ), that's seriously large diameter pipe for only 500hp

I've made over 800hp on larger 4cyls with tubing smaller than that

Or is that a n/a exhaust setup they're calculating for ?

[Luther Blisset]

I agree Stevie, OTTOMH that is 48mm which is way out of the ballpark of what people use for 500hp on a 4cyl.
Lets be honest it's an easily achievable power target so surely the point of this exercise will be getting it spooling early and having a wide operating band.
Not a task a 48mm runner is conducive to IMO.

[sl0wlane]

stevieturbo said:

Even at OD, unless it's very thickwall tubing ( ie to reduce the ID ), that's seriously large diameter pipe for only 500hp

I've made over 800hp on larger 4cyls with tubing smaller than that

Or is that a n/a exhaust setup they're calculating for ?

Burns designed for the turbo I have, even down to the TiAL v-band flange.

Most N/A Honda manifolds are 1.7/8" (at least the ones Ian @ clockwise specifies)... however we have actually gone for 1.3/4" with a flare to the head for the anti-reversion step with a gentle taper.... and maybe even allowed for some slip joints so we can alter the length and experiment later on.

[stevieturbo]

sl0wlane said:

Burns designed for the turbo I have, even down to the TiAL v-band flange.

Most N/A Honda manifolds are 1.7/8" (at least the ones Ian @ clockwise specifies)... however we have actually gone for 1.3/4" with a flare to the head for the anti-reversion step with a gentle taper.... and maybe even allowed for some slip joints so we can alter the length and experiment later on.

Is it a divided V-band housing or open scroll ?

[sl0wlane]

It's single / open "M-Spec" version of this: http://www.owendevelopments.co.uk/product/389/Owen...

[sl0wlane]

jontysafe said:

This is my custom twin scroll manifold for a YB I'm building. Severely space restricted so very short runner.

Don't over concentrate on one aspect of build, give the job to someone that knows what they are doing. I over concentrated on intercooling and it probably added 6months to build.

Nice fab work!

Don't get me started on intercoolers (already controversial in Lotus land, most opt for water-air)... I'm all hammer formed end tanks and aerospace cores already...

[stevieturbo]

Why go to all the concern about manifold design, spool, power, whatever. Then not even invest in a proper twin scroll design ?

You're already losing spool and power from the outset running the open housing.

[jontysafe]

stevieturbo said:

Why go to all the concern about manifold design, spool, power, whatever. Then not even invest in a proper twin scroll design ?

You're already losing spool and power from the outset running the open housing.

Absolutely! Didn't realise he was
going open scroll. Time to sit down and look at where you're going I think.

I should squeak 500bhp on a twin scroll Borg Warner EFR7163 turbo. You should absolutely look at either 7064 or 7670 because you're going to be revving higher. I'm not interested in headline numbers though.

Let me guess you bought the GT35 from OD?

[stevieturbo]

Here is one extreme....ish example from those who might follow Street Outlaws.

Murder Nova unveils his new car, from around 12:30 when he takes the front end off the car one turbo and manifold can clearly be seen. Now obviously spool etc isnt a huge factor for these guys given the usage. But it's very clear to see those manifold tubes are no larger than 2" OD, might even be smaller. And you can be sure that engine will be making in the 1500-2000hp range. Perhaps even more at the upper end.

https://www.facebook.com/187Nova/videos/1173548382...

[Gunt]

Hi

i'm kinda in with the other guys on this , you've asked for all the math / data to build the best , then purchased the turbo presetting a lot of your manifold , not that single scroll will make too much of a difference on a Honda but every bit counts , when going single why not the tial stainless exhaust housing , it nearly matches the twin scroll , and i know doing the twin right bring twin waste gates added expense , how ever another large aspect in making and playing with the manifold is stepped headers , a good link below . on your chosen route , i understand all the math point in favor of the 23+ length but that's biased on optimal out put at rpm for your engine , what it dose not tell you is the driving experience especially where you want to be , not flat shifting on antilag up the 1/4 mile , you need throttle response partial throttle load points , this will lead you to something far shorter

http://speedtalk.com/forum/viewtopic.php?f=15&...

but in truth i do envy your build , build two first with mild steel then work from there , just to put a cat among the pigeons , id use a different intake manifold too

[sl0wlane]

Thanks, full reply when m not drinking

If you are subscriber, great webinar here, interestingly covers the points and logging requirements Max_Torque covered and a great practical example... Andre obviously loves thetwin scroll borgs too... but I'm too invested to change now (turbo was actually purchased around 18 months ago - LONG project!)

https://www.hpacademy.com/previous-webinars/107-an...

If you are going to suggest the skunk ultra manifold... sorry but I do actually prefer the one I have, bar a custom made one it's the best available off the shelf in my mind... the proof will be in the (eventual) pudding.

Thanks for the link though, kinda confirms that "as short as possible" is not the answer to Turbo exhaust manifold design doesn't it?!

Edited by sl0wlane on Friday 16th December 19:01

Edited by sl0wlane on Friday 16th December 19:37

[sl0wlane]

Thanks for the link to the speedtalk forum...

Lots of threads covering pulse tuning with turbo applications... it appears the general consensus is that it by and large works much like an NA application and can be tuned as such, threads like this : http://speedtalk.com/forum/viewtopic.php?f=15&...

[stevieturbo]

sl0wlane said:

Thanks for the link to the speedtalk forum...

Lots of threads covering pulse tuning with turbo applications... it appears the general consensus is that it by and large works much like an NA application and can be tuned as such, threads like this : http://speedtalk.com/forum/viewtopic.php?f=15&...

Although a lot of any pulse tuning can be lost when you use an open scroll turbo setup, and would also be very camshaft dependent, of which turbo cams tend to have far less overlap etc than a racey n/a cam

Your best chance of improving performance would be with a proper twin scroll setup. Throw away the open scroll and do the job properly if you're going to all this trouble and worry about the manifold.

[227bhp]

Max_Torque said:

1) Exhaust port shape and area at the exhaust manifold flange - you want a SIGNIFICANT anti-reversal flange, which means your exh manifold flange needs to be at least 3mm bigger all round than the port exit. This will drive you towards a larger overall manifold tube diameter. However, depending on the CSA you end up with, you may want to cone down to a smaller main tube dia to maintain reduce manifold volume. Typically, you'd come back down, using a 3deg taper to the same dia as the exh port at head exit.

One thing you don't do is mess with different exhaust diameters, you choose the diameter based on the port CSA, valve sizes & events, RPM range you want it to work at and leave it at that. Any change in diameters slows the gasses and that is not something you want. You need it to flow out fast, drag out any residuals and start to pull the intake charge in.
There is very little chance of reversion on this engine, it's VTEC. Reversion happens at low engine speed with long duration/tight LSA cams, they won't be so at low rpm.

It's a shame the OP can't divulge cam specs, I'm highly suspicious of 'Turbo cams' as they are quite often Snake oil.
You do wonder what cockeyed theories someone has come up with that make their cams better than the ones Honda invested millions of Yen and hundreds of hours in perfecting. The addition of a free flowing turbo and manifold shouldn't affect things much in that department.

[anonymous-user]

227bhp said:

There is very little chance of reversion on this engine, it's VTEC.

What has "VTEC" got to do with anything? You have, for a typical road sized turbine, EBP at around 2 to 3 times IMP, so "reversion" is un-avoidable.

[227bhp]

Max_Torque said:

227bhp said:

There is very little chance of reversion on this engine, it's VTEC.

What has "VTEC" got to do with anything? You have, for a typical road sized turbine, EBP at around 2 to 3 times IMP, so "reversion" is un-avoidable.

I explained it there:

227bhp said:

There is very little chance of reversion on this engine, it's VTEC. Reversion happens at low engine speed with long duration/tight LSA cams, they won't be so at low rpm.

Without using obscure terminology to put you off the scent.

To explain further:
Reversion happens with long overlap, long duration cams at low rpm. The Vtec system adjusts the cams duration, overlap and lift so you don't get that effect at low rpm, low rpm being where it occurs, not high rpm.
High rpm is where the high speed lobes come in (longer duration, higher lift, more overlap) and reversion isn't a problem.

Is that easier for you to understand?