Big Turbo Problem

Now the last thing I want you to think is that it was lack of planning that caused this problem - bigger is not always better (even for sports car drivers!) and it wasn't sheer greed for "more blow" here... or I could have just sought out the "mother of all turbos" as in the photo here! (now that IS a big turbo!!!)

I actually spent many hours studying these parameters of the latest and greatest ball bearing turbos, to find the perfect match for the engine:
matching graphs (pressure Vs air flow) - it's not just all about maximum boost, you need the turbo to pump as much air as the engine requires for a given power output too), trim (compressor/turbine rotor diameter ratio) - affects turbo size, spool time and boost/flow), A/R ratios (housing geometry ratio for both turbine ane compressor) - again affects turbo size, spool time and boost/flow) etc.

I also wanted the turbos to bolt straight in place of the old T25s if at all possible, so size was always a factor and also using a turbo with the same T25 exhaust mounting flange - the RIGHT turbos for the engine tune level was the ultimate concern though - I wasn't going to all the trouble of modifying the engine to extent that I have, only to limit the final output by the selection of turbos

The common use of "reflowed" T25s with 360 degree bearings, or "hybrid" T28s, are really comprimises to try and achieve more flow in the existing T25 form factor - they are never going to supply the boost and flow plus rapid spool up that I want

So Pick the best turbos - if they fit, all well and good - if they don't, they will be fitted somehow, whatever it takes!

My fnal choice was the GT28RS - also known as the "Disco Potato"!

Actually this name even appears on the Garrett product catalogue and was coined by them from the shape of the turbo map

The Garrett GT range all have a dual ball bearing assembly instead of floating oil bearings of older designs like the T25 - they spool much faster and also have greatly improved turbine/compressor aerodynamic properties

It's always difficult to get a "free lunch" though and the more common .86 A/R turbine vesrion gives very high power but would be some what laggy on capacities under 2.0 litres (the 3.5 litre V8 equates to 1.75 litres per bank)

More research (and cutting though sales talk!) revealed that there is a lesser known .64 A/R GT28RS - with this more efficient aero and ball bearings, this should provide spool as quick as the old T25s but hugely increased power capability

Before buying, I also emailed a few suppliers for their views - Click here to view the trail

On first inspection the GT28RS look impressive and definitely "man enough for the job"

The Exhaust turbine housing, although having a very different outlet (5 bolt exit, for a V-band adapter with larger bore exhaust) didn't seem much larger than that of the T25s

The inlet air compressor is much larger though - even the inlet is 3 inch, a full inch larger than the T25s, as the photo on the right shows (GT28RS on left T25 on right)

The alignment of the CHRA (Centre Housing Rotating Assembly) wasn't correct for the oil banjos and coolant lines to line up, but this can be loosened and moved without affecting the balancing

The exhaust flange was the right pattern, but the drilled holes were smaller than that of the T25 - this isn't too hard to correct, so I started trying the turbos in position and, as the photos below show, they looked magnificent.....

Take a closer look at the photos though - when I offered the lovely new GT28RSs up to the exhaust manifolds - although I couldn't put them on the studs because of the hole sizes, it was clear that even if I could, the compressor housing comes very close to the block on the near side and the large 5 bolt exhaust exit would foul the bell housing on the off side... OH DEAR!

Here's a photo borrowed from a recording of 5th Gear (the full video is on the 5th Gear section)

My contact, GW, is holding up a GT28RS next to the T25 in situ - it's obvious how close the T25 small exhuast exit comes to the bell housing and that the GT28RS won't fit in the same position! 

Sitting there with the engine and turbos, a thought occurred to me (probably at the same time as realising that the foolproof, but expensive option, was repositioning the turbos completely and fabricating completely new, one off, manifolds) that more clearance could be afforded by spacing the exhaust manifolds out at their natural 45 degree angle (moving the turbos downwards and outwards)

It was time to share these findings with the PM71 project team -  Click here to read the email trail

I made up the annotated pictures below to explain the situation and how a manifold spacer could possibly be used


 

In a positive vane, I also took these photos with the turbos in the positions they would adopt with manifold spacers - they just might just be propped up on mallets, but they do give a good impression of how it would look

They certainly look a significant upgrade to T25s and clearly what's needed to match the rest of the modofications to the induction... I know I'm doing the right thing insisting on getting them fitted!

Now, whilst I was surfing the Net, I happned across a post about an Esprit V8 being prepared by Garrett themselves for a racing series in the USA

It was clear from the photos that, as you might expect of Garrett, they were changing the turbo installation - the shot on the right shows the engine in the workshop, with modified plumbing for connection to their chargecooler (photo in "Engine" section of "Design" section)

This shot shows the car with an engine on stand at an exhibition - but click for the enlargement and look closely at both turbos... the one on the right has the old 3 bolt exhaust exit (probablt a T25), but the one on the left has a new 5 bolt exit!

I emailed the message board, got the Garrett and race teams contact info and emailed them to find out what was going on - Click here to read the email trail

So the answer was that they were using GT25R turbos (that have the 5 bolt exit but are not as large as the GT28RS) - the photo on the right shows how the team have cut a slot in the bell housing to fit the right hand GT25R. 

The dimensions of my GT28RSs look like they make even drastic measures like this, less than what is needed - they need to be spaced away from the engine/transmission

However, this means that they encroach on the chassis tube that runs very close to the outer side of the turbo

The annotated picture from beneath the car illustrates the problem here - without the engine being fitted to the chassis, it is very difficult to gauge how much room I have to play with here

I may, of course, have to fit the engine in order to find the optimum position for the turbos and the exhaust manifold arrangement 

This isn't a step I planned on doing  though - I'd rather find a way to keep to my plan of developing the complete engine/induction, map the ECU at the dyno/engine cell and then start the assembly   

You'll notice from the underside view that the exhasts have to take a fairly sharp vertical "dog-leg" over the drive shafts too. So moving the turbos rearwards would exacerbate this problem.

If  I do completely reposition, I'll probably use the position shown on the right here - this will allow straight exhaust exits to the rear of the car, won't present oil drain issues and provides an even shorter inlet to the chargecooler

The diagram shows a cheaper alternative of using an adapter pipe from the standard manifolds - but this position could also be used with fabricated stainless steel manifolds

Either way I'm concerned about the length of the pipework to the turbines - exhaust manifold design is quite complex because of the way the pressure pulses from each cylinder combine - this is important on normally aspirated engines to provide an "extractor" effect but it is key on turbocharged engines to ensure spooling of the turbine

The odd and even cylinders are combined in the primaries and the two bores are kept seperate right up to the turbo inlet on the standard manifolds) - the thing to avoid is what's called pulse reversion - where the exhaust gas travles in the wrong direction! It is possible to merge the primaries at  a "collector" point and then run a secondary pipe to the turbo, but distances/dimensions are important....

Well that's the Big Turbo Problem defined - what the solution is, I'm not absolutely sure yet!

Oct 2011:

I am definitely having stainless steel manifolds fabricated now - after a chat over lunch during a track day at Brands Hatch, the ideal company has come to light...

Here's one the GT28RS turbos propped up on the clutch (which will probably also need an upgrade for high power/torque) showing the approximate position where the new manifold should meet it