Bit Ot, Anyone Good At Turbo Sizing For A Small Capacity Race Engine Project?

[Chris Wilson]

I know there are few really clued up folk on turbo sizing here. looking for suggestions for the following race engine spec:

Toyota 4-AGE 1.6 litre Formula Atlantic spec engine, currently an N/A producing about 208 BHP. Will keep the 304 degree duration cams currently in it, bit compression will be dropped from 13.5 to 1 to 8.75 to 1 with new pistons. Revs safely to 10,000 RPM. looking to achieve about 380 BHP in a 580 kilo Sports GT chassis for track only usage. Transmission is six speed sequential Hewland FTR.

Exhaust manifold will be custom made (in fact the car is in at Primary Designs on Wednesday this week, so I need to decide on a turbo by then). Can be divided or none divided turbine housing, externally or internally gated, and I have a spare new Tial 44 mm external gate to use if required. Intercooling will be by air to water and should be pretty efficient. Looking for some reasonable power starting about 5000 RPM 'ish.

Turbo Dynamics suggested a TO4R with divided turbine housing, what do you guys think? Thanks.

[The Dan]

I've supplied a GT28RS to someone with the same setup over here. They have made nearly 400hp at the wheels with power coming in strong at around 4500rpm. It holds all the way to their 10,000rpm redline. I think they would have better results with a 3076 or 30/35R. Obviously a 0.86 rear would be needed.

[discopotato03]

Engines with long cams tend to prefer twin scroll turbine housings on turbochargers to get around the reversion issues long duration wide overlap profiles give .

With the righ combination of bits I reckon you can get the power without the 10G rev ceiling and it may be worth looking into high lift profiles with a bit less duration and overlap .

My guess would be a GT3071R but with a non standard and reprofiled GT32 turbine housing which I think you can get in 0.78 A/R . The twin scrolling allows you to have the larger A/R ratio and less restriction (potential reversion issues) with no real lag penalty if everythings sized about right . You generally get to up the ignition timing because reversion/pollution/preheating is less so the detonation threshold is higher .

Depending on what octane fuel you can run you may not have to drop the static CR that low , if the engine is not expected to have torque at low revs there is less time at higher revs for the end gasses to auto ignite .

Ethanol fuels are supposed to make for lower peak combustion pressure but higher average pressure so if the rules allow ...

Just my 5 cents , cheers A .

[Dale FZ1]

Something about the GT28RS doesn't stack up there - Garrett rate them as good for ~ 320 crank hp. I couldn't see them able to pass enough mass-flow to hit on 400 wheel hp.

If I was spending the coin particularly on a track car, probably GT30 based turbine in free-floating (external gate) configuration. I'll crunch a couple of numbers and see if I can come up with anything useful.

[Chris Wilson]

Just to clarify I am looking for 380 BHP at the engine flywheel. Fuel would ideally be UK Super Unleaded (98 RON) but I would stretch to a 50 / 50 Super Unleaded / 118 octane leaded race fuel mix. Management is going to be Motec M800, engine will be on dual twin choke 48 mm throttle bodies. Pics of the thing as is, awaiting me taking it for its manifold build are here:

http://www.gatesgarth.com/4age/4age.html

Thanks for the replies, keep the ideas flowing please

Edited November 3, 2008 by Chris Wilson

[Chris Wilson]

I would quite like the simplicity of a NONE water cooled housing, but I am also keen to have a Ni-Resist turbine housing, having seen how a standard cast iron turbine housing on an IHI RX-6 (Apexi) turbo warped on my Skyline engine. I know. I ask too much....

[Adriano]

Mate 380 hp should be a walk in the park, with no need ot rev to 10000rpm to make it. If you fot a 3071, than you will only have to run about 7psi to make the power you are after, in which case you could easily leave teh comp ratio at 10:1

[bubba]

I wouldn't be using water-air intercooling for a track car. My understanding is that water is only good for short bursts but because it takes so long to get rid of it's heat means it's not that suited to circuit applications.. just my understanding anyway, I'm sure someone will pipe up if I'm completely off the mark

[Lithium]

As I said on the GTR UK forum, GT3071R I reckon is a good match for the power you are after.

Here is a dyno plot for a chap I know ofs 4AGTE MR2 (running cams) with a .63a/r GT3071R:

[rob82]

What about a Bushcur 20G - twin scroll titanium wheels - blah blah - have seen around 260-270awkw on an evo9 with all the mods. I would also be keeping the compression (10.5:1) in it as well so your average power would be better - might make it a bit nicer mid coner. I would also look at reducing the overlap as well.

[Dale FZ1]

I’d think you need to have targets for the operating rev range, and perhaps gearing needs for the change to forced aspiration and the torque curve you’ll see.

I agree with previous comments about camshaft specs and static compression ratio ie. get cams that suit turbo application, and keep the C.R. reasonably high to assist overall efficiency. Efficient scavenging via cam and turbine specs should allow an easy 9.5:1 and low pressure differential across the engine to make the target hp. Running to ~9000rpm @ 1 bar is my best guess for 400 crank hp.

I agree with Lithium on the 3071, but perhaps the 0.82 A/R might be preferable. You wouldn’t know until you tried both, but I’d say the 0.63 would give a bigger torque hit in the mid range (chassis stability + traction problems?) and stifle the torque production at higher revs due to the turbine mass-flow efficiency difference. The larger A/R size may retain higher rev ceiling (cam spec notwithstanding) and not compromise response if you don’t fall below the boost threshold rpm. And if packaging / budget allows, definitely go with an external gate.

The above package might not be exotic, but should work.

[Chris Wilson]

Asked this question on a couple of other forums, too. The vast majority of people who replied suggested a GT3071R with a .64 A/R, and Turbo Dynamics and Owen Developments concurred, so I have ordered one of these to do initial trials with. the T28 footprint helps a lot, as the installation is very tight (see IrfanView HTML-Thumbnails if interested). I took the car down to Thame this morning and Primary Designs are making and fitting an Inconel manifold, support bracketry for the turbo, and a silenced exhaust system, all thermally coated in Zircotec. I have used primary designs for an F3 system, and for some prototype work before, and can't speak highly enough of this company. They now take on some general motorsport fabrication work, too, so there's less toing a froing than there used to be, as they can fabricate ancillary stuff in house. I justify their high prices by telling myself 60% of the F1 grid can't be wrong in using them They were making a new Arrows BMW Megatron turbo manifold, and it was truly a work of art, more deserving of the Tate Modern than any exhibit I have yet seen there They should have a gallery on their web site, some of things they make are very deserving of photographic record. http://www.primarydesigns.co.uk is their place.

Thanks for all the fast advice, committed to this turbo now, so we'll see how it goes.

I have decided to struggle and remake the rear bodywork to accomodate and air to air I/C. I may well run some cams with less duration when I have the engine out for the low compression pistons and refresh.

Thanks again

[Dale FZ1]

The vast majority of people who replied suggested a GT3071R with a .64 A/R, and Turbo Dynamics and Owen Developments concurred, so I have ordered one of these to do initial trials with. the T28 footprint helps a lot, as the installation is very tight (see IrfanView HTML-Thumbnails if interested).

Thanks for all the fast advice, committed to this turbo now, so we'll see how it goes.

I have decided to struggle and remake the rear bodywork to accomodate and air to air I/C. I may well run some cams with less duration when I have the engine out for the low compression pistons and refresh.

If running something with a T25/28 flange, do you realise it's in all probability a housing designed to run with a smaller spec T25/28 rotor - and not likely to offer great flow efficiency? Especially with a 0.64 A/R, that combination would tend to limit what happens as rpm and mass flow increases - torque would likely fall away savagely at a particular point as the turbine hits its max capacity and effectively chokes the engine from the exhaust side. Think of it as mechanical constipation.

Garrett do offer a slightly larger spec 60mm rotor version GT30 (there'slots of discussion on this forum about it), and the advantage it offers is a much more mass-flow efficient turbine combination. See their catalogue here: http://www.turbobygarrett.com/turbobygarre...T30/GT3071R.htm

With regards the cams, I could only offer the opinion that they have to be considered part of the whole engineering package ie. static compression ratio, cams, turbo + manifold. You don't want specs that are not going to complement each other. A set of 4AG cams to suit your application would be only a fraction of the price of your manifold so you'd be mad to not change them.

Intercooling is a must, and I see packaging that setup as probably your biggest challenge. Water to air is more complex and probably has a weight penalty, but it can be very efficient at scrubbing temps and offers short-run pipe work between turbo and throttles.

Good luck and keep us informed.

[bubba]

Water to air is more complex and probably has a weight penalty, but it can be very efficient at scrubbing temps and offers short-run pipe work between turbo and throttles.

Only in the short term, it's not really suited to sustained rpm circuit racing because while water can absorb a lot of temperature, it takes quite some time to lose that temperature too.. my understanding is it's more suited for street applications where you're not on boost practically ALL of the time.

[Chris Wilson]

Having all been so helpful I thought you guys might like to see where I am now with this. Inconel manifold, supported turbo, split gate dump, and silencer now dry built, awaiting final detailing before ceramic coating, along with turbo turbine housing:

Thumbnails at http://www.gatesgarth.com/4age/primary/primary.html

[Dale FZ1]

No issues with the workmanship on the pipework there - looks very neat.

Was there any sort of modelling done to arrive at the length and diameter of your runners?

The ID tag on the cartridge indicates a GT3282 plain bearing unit. I'm presuming it is a split pulse housing you've used. Any more information on it?

Edited November 17, 2008 by Dale FZ1

[Chris Wilson]

Th cartridge is just a dummy the turbo people sent to make the manifold from. The runners sze and length were based on RPM expected, as well as getting the runners within 10 m of equal length. Availability of Inconel at a sensible price had some influence, too. Kept the ID fairly small to keep gas speed up.

Edited November 17, 2008 by Chris Wilson