Xmt - Sarty's Xsara GTI6 Turbo

[petert 586]

From your picture, the XU7 looks identical to the XU10, with two oil pressure sensors tapping directly into the oil gallery.

[DrSarty 90 1 Cars]

Thought so. The RH (green connector) I'm sure is a level sender/low level switch, so I agree wouldn't be pressured; but it could be a good turbo oil drain.

 

The LH (grey) one is the factory OP warning light switch, which I'm going to take to a new funky dash gauge per Andy's suggestion - including scaled reading and low pressure warning light - and then I'll still need a pressured feed tapped in for the turbo, ideally at the rear or end above the flywheel.

[petert 586]

Both are pressure. One is oil pressure light, the other is oil pressure gauge. Neither can be drain. Either can be feed.

Edited October 30, 2017 by petert

[petert 586]

Is there an oil level sensor (or bung) on the rear of the block which could be used as a drain?

[welshpug 1,657]

there's only one pressurised port on the xu7 block which is where the pressure switch is fitted iirc.

[Andy 105 1 Cars]

The green connector is an oil level sensor so drops to the sump. Might make a good turbo oil drain if it is a large enough diameter. In JIC hose speak, I reckon you need -10 hose at least for the drain , so do check that the hole vacated by said green sensor is large enough. If not, hole in front face of sump above the oil level should do the trick.

Andy

[Anthony 1,003]

there's only one pressurised port on the xu7 block which is where the pressure switch is fitted iirc.

I've got to admit that's what I thought too from the XU7JP4 block I had a few years back. I replied saying that this morning, then deleted it as I doubted myself given the elapsed time and didn't want to give bum info.

 

Servicebox shows an M10x1 threaded bung in the front of the block behind the starter and the level sensor probe on the rear, which only further confused me as that doesn't look like an M10 threaded part in the photo! Wondering if that's another SB error, perhaps using the XU7JP 8v block (which does have an M10x1 thread into the oil gallery same as the XU5/XU9 block).

 

[petert 586]

I just assumed the green sensor was a regular M16x1.5 gizmo. Unscrew it Sarty.

Edited October 30, 2017 by petert

[wicked 103 3 Cars]

 

 

 

 

Did you check the spec's of this turbo?

 

This is what I've found:

 

IHI RHF5

Min CFM @ 2.0 PR (15 psi) = 50

Rated CFM @ 2.0 PR = 382

Practical Max CFM = 363

Choke Flow CFM = 415?

 

360cfm won't give you 300bhp...

[DrSarty 90 1 Cars]

@Wicked.

The turbo specs you've quoted may be right, but are you sure they're correct for the IHI VF38 Legacy GT twin-scroll I have?

 

EDIT: I gather you zoomed in on the photo and got the RHF5 bit? Does the RHF5H make a difference, i.e. later/larger capacity model? From my investigations I saw people getting 300 plus off that turbo, assuming what I've bought is the same thing of course.

 

EDIT 2: This is what I read. Peak numbers not that key for me; driveability is. If I'm 'limited' to 260ish with buckets of get up and go I'll be happy:

"VF38

Twin scroll turbo with titanium turbine and shaft. Smaller compressor housing than VF36/VF37 provides tremendous spool up capabilities but less top end than VF36/37. The spool capabilities of this turbo are demonstrated on the JDM Legacy GT, which reaches peak torque at 2400RPM."

It's quoted as 276bhp and 285torque, admittedly from factory on a 2.5L lump.

EDIT 3: Now noted this turbo is for the 2.0L JDM (Japanese market) Legacy GT.

 

If (and probably) yes, then what's the power potential on 1bar max boost pls? It may just be a good starting point and I'll have to change the turbo later. At least I'll have proof of concept and the plumbing will be there.

Edited October 30, 2017 by DrSarty

[welshpug 1,657]

I believe that turbo is on the 05 on auto EJ207 engines running 13.5psi peak for 265-280 bhp, they do have AVCS, aka variable valve timing, its a 1994cc engine running 9.5-1

[DrSarty 90 1 Cars]

Cheers Mei. So it seems a decent start, at least for driveability if not peak.

 

Notwithstanding me not having 'VVT', perhaps my 9:1 on different pistons and a damn fine (RS) head, plus a bespoke exhaust will help this turbo shine.

[welshpug 1,657]

it seems like it yes, you may want a larger unit to break the 300 mark but at the expense of driveability, I think Dans 98 Legacy has a TD05 of some kind running about 17 psi peak and about 320 bhp.

[DrSarty 90 1 Cars]

And before I forget, is a 2.5" stainless suitable for up to 300bhp?

 

I can make the tip on the back box as big as some of the idiot Scooby or Evo ones I see if so inclined (but I won't), but for the main system, is 3" necessary? I hear people often relate big exhausts with big power, when engines don't need as large a diameter zorst - perhaps disappointingly - as they think.

 

Googling indicates that 3" is needed for 340bhp, whereas 2.75" peaks at 280. Estimates of course, but 3" looks a smarter choice. Shame my current supplier can bend 2.5" max!

Edited October 30, 2017 by DrSarty

[DrSarty 90 1 Cars]

Found this an interesting read. N/A vs turbo exhaust science from a guy at Garrett:

http://www.tercelreference.com/tercel_info/turbo_exhaust_theory/turbo_exhaust_theory.html

[wicked 103 3 Cars]

Regarding the turbo; the VF38 is a variant of the RHF5, whereas the VF37 is a variant of the slightly bigger RHF55.

So the flow number might be a bit off, but it will still be in the same ball park. Given that it already spools at 2400rpm indicates that it is a small turbo.

 

The power (energy/time) you get from an engine is more or less correlated with the amount of fuel you burn (efficiently). To burn it efficiently, you need an correlated amount of air (afr 1:12).

So there is a rule of thumb that indicates the amount of air that you need to push trough your engine. It's roughly 1.45 times the power. So for 300bhp you need about 435cfm. (again: rough number)

The rated flow numbers for turbo's are in general at pressure ratio 2.0, which is more or less 1 bar boost and where the efficiency drops below 65(?)%.

 

So the VF38, if numbers are correct, will be able to push 360cfm at ~1 bar boost; ~250bhp. But that is the turbo.

Next thing is your engine; will your engine flow 360cfm if you put ~1 bar boost at it? An 2.0 8v turbo won't. That will need more pressure to push 360cfm through an 8v head; but the turbo won't do 360cfm at 1.3 bar. So you won't make the 250bhp with a (stock) 8v. Your 16v probably will. And a 2.5 16v might not even need 1 bar boost to make 250bhp. Maybe it only needs 0.8 bar. At 0.8 bar the turbo could maybe flow 400cfm and make the quoted 280bhp.

If you want to know the details, you should start studying turbo flow maps.

 

So is this turbo suitable for your engine? It will work and spool fast, but I would not have chosen this one. Given that you have 16v and probably the 16v gearbox, you don't need boost at 2400rpm. That is way too fast for this engine imho. That is not the rpm range you use on a NA 16v as well. Don't let people scare you with loss of drive-ability. You can have bigger turbo's that will work fine on a 16v. A gt28rs (500+ cfm?) will work fine on an 16v. If you enter the big GT30xx arena, driveability will be a point to consider. If your turbo spools between 3000 and 3500rpm, you'll still be fine. If you really want to have the boost at 2400rpm, you could argue if you need a 16v at all, because an 8v will be more efficient there.

On my 8v turbo with 1.9td gearbox, I've Td04hl-15t (430cfm) and that spools at 2700rpm which is easy to drive, but on a 16v I would definitely trade some spool for more top end.

 

Of course you can upgrade the turbo later on, but think ahead on what you will upgrade; complete turbo or just compressor wheel? If turbo; you probably need to modify your manifold by then. If compressor wheel; you can upgrade it, but you can't go nuts, because your turbine will remain small and get bloody hot if you try to push too hard. You can't make it flow 500cfm.

These were my 2 cents... Probably you'll stick to using this turbo (which is fine; it will work and it's your project), but don't expect it to make 300bhp.

Edited October 31, 2017 by wicked

[petert 586]

A friend owns and runs this place:

 

http://www.gtpumps.com.au/

 

Some great info there and lots of ideas.

[wicked 103 3 Cars]

Some turbo sizes and flows: http://www.stealth316.com/2-turboguide.htm

[dcc 855]

We went from a laggy td06 hybrid (full boost from 5000rpm) with cfm approx 680, to a td05 18g, with cfm 550 (boost from 4400 approx.) we wanted more mid range and less need to rev out (to change gear and not drop out of boost every time).

 

The td05 18g or VF34 i would suggest for your needs (circa 300bhp on 2ltr 16v)

 

We will be going back to TD06 hybrid is we can get the mapping done a bit better off boost

[DrSarty 90 1 Cars]

Hi.

 

Is the XU7JP4 functional head height the same as the RS head pls? (Question relates to head bolts and cam belt to be used)

 

Does the XU7 block have the 'spacer' like the alloy Mi? Will the Mi spacer fit if I want it (as part of an increased oil capacity solution)?

 

Any comments on use of a rising rate FPR that I have?

Edited November 5, 2017 by DrSarty

[welshpug 1,657]

head is the same height yes (137.00mm +/- 0.05) you use XU7 bolts as they are M11 as with all alloy XU blocks, not M12 like the Iron blocks.

 

in respect of the belt use the XU7 kit.

XU7 doesn't use the stiffening spacer, but its just two drilled and taped holes away from being bolted on.

 

 

 

not sure on the FPR, do you mean the standard boost/vacuum referenced unit or some fancy aftermarket thing? surely if its being mapped you wont need anything like it?

[Andy 105 1 Cars]

For the FFR. It depends on how your mapper will want to map the engine. On a turbo engine, because the manifold pressure can rise well above atmospheric , it is usual to use a 1:1 rising rate FPR to keep the effective fuel pressure at the injector constant . I imagine that most mappers will wish to do this. It was certainly my experience with a Magneti Marelli ecu on a turbo engine. In fact, we had to be a bit careful not to set the base fuel pressure too high (55psi) as on full boost with 1.6 bar ( 23psi) the FPR was close to its limit . However, you are not using quite that level of boost, so any decent quality 1:1 rising rate regulator with vacuum takeof will do the job, set at around 45psi, much as Welshpug has indicated . New , one will cost you around £100 though.

[DrSarty 90 1 Cars]

Just in case anyone's interested, things are moving, and here's a 'situation dump':

 

Comment at will.

 

- Xsara gone off to Mike Edwards (nr Peterborough, as recommended by Colin Satchell) for the custom turbo manifold and downpipe fabrication

- Above is going to cost circa £1,200! I was 100% out, estimating £600-800!

- I have the new head, crank, pistons, rods etc etc all ready to become a kit for the engine builder using the original XU7JP4 block

- Wiring loom 80% sorted, mapping the OE loom - and cutting out what I don't need - into a new Emerald K6-ready loom a la 'plug and play'

- It seems a knock sensor is a practical impossibility, unless I spend 3x the normal amount on mapping. The K6 doesn't accommodate it as standard, so I'm presuming they map in the real-world, which makes me think odd things about knock sensors; perhaps only the realm of manufacturers with millions to spend, and/or an 'air bag' (last resort)

- I've decided to run staged (inner and outer) injectors. This exploits the advantages of this set up, even using a single manifold/plenum rather than ITBs, and means I can achieve the fueling rate needed via 2 sets of Bosch 'red devils'. Anticipated demand for 300bhp FI (@80% cycle) is 520cc/min, which I can deliver using twin sets with the advantage of injector resolution and staging at 320cc/min x 2 per cylinder. I could be talking bollocks, but I'm going to get bosses welded in and try it. I have all 8 injectors.

- A swirl pot is in the bag but I'll need a secondary pump - on the hunt for that

- Fuel rail solution is in progress, as I'd like both rails to be 'dumb', i.e. control the FP elsewhere, i.e. use dummy FPRs in the OE rails

- K6 will be purchased shortly

- Engine will go away next year as a kit to have the block and liners cleaned and sorted. Someone else will then build the bottom and top end (with my verniers) to give my XU7<>XU9J4Z<>10JP4RS hybrid lump (XU7JP4 block & liners / XU9J4Z {DFW} pistons+rods / XU10J4RS crank+head) a chance of living

- Once Mike is done and the engine is built, it'll just need an exhaust system from downpipe back and.....it'll start

 

Then I'll ship it to John, Dave and Karl at Emerald and get the tw@ mapped and then give it death.

 

Rich

Edited December 7, 2017 by DrSarty

[wicked 103 3 Cars]

Good progress!

600-800 quid for a custom was maybe a bit low estimate; compared with the dp manifolds that are made in small series. Given the challenge of routing a twin scroll manifold, I think 1200 quid is not very expensive.

[Andy 105 1 Cars]

Great update . I will keep at Steve until he delivers the extra parts  . 

Expensive manifold you have planned, but I guess if it is bespoke, there is no short cut . Yours will be a very interesting  build to watch unfold. New territory indeed. 

Andy