SEAT Ibiza - rwd turbo

SEAT Ibiza - rwd turbo

Author
Discussion

anonymous-user

Original Poster:

55 months

Friday 4th October 2013
quotequote all
chuntington101 said:
Max_Torque said:
That is the plan. In fact, with the main engine cooling rad being at the bottom, there is very much a danger of convection from the hotter rad resulting in unwanted upheat into the intercooler. With the water cooling rad running at around 90degC, and the IC at under 50, this can cause significant heat soak at low speeds (where convection is dominant)

The flip side is maintaining a suitable level of ambient cooling airflow around the rest of the engine bay (a trade off between drag and cooling). I plan to use maskable ducts in the front bumper corners to duct air to specific locations as necessary
Max, can you explain the advantages in using split radiator and intercooling as opposed to running a full hight rad with the intercooler mounted low down. Especially when it restricts evac of the rad cooling air as you mentioned above.
The cooling system for this car was designed when it was effectively a rally car, and as such forced to run with a 34mm turbo compressor inlet restrictor. As such, the compressor pressure ratio is extremely high, even for what would normally be fairly moderate boost, and that means your compressor is un-avoidably inefficient. This results in high compressor outlet temperatures and the requirement to reject quite large amounts of heat from the intercooling system.

Added to which, the main water radiator is heavy (it's full of water!) so mounting it low and the much lighter I/C rad high is sensible. By steeply angling the main water rad, the overall package height is also reduced and you can get some benefits from non forced convection cooling at low vehicle speeds (typical of UK stage rallying). As a radiator only requires something like 50-60% of it's core area as intake area (because the core itself reduces the flow area, and it uses highly turbulent flow to get good heat transfer) by tilting the main rad, and packaging the IC rad above it both rads get a nice cool air flow close to ambient temp (rather than have them sandwiched and the rear rad gets hot(ter) air)

The exit area is more than sufficient for both radiators, it's just a question of subjugation to avoid upheat at low speeds!

It's really noticeable with this cooling layout, that with the vehicle not moving, the cooling rad fans hardly even cut in at idle. In effect, there is enough convection driven cooling to remove the engines idle heat flux without cooling fans.

anonymous-user

Original Poster:

55 months

Friday 4th October 2013
quotequote all
Transmission!

Probably the Achilles heal of more high power cars than any other part! Expensive and tricky to make durable and often neglected compared to the engine. Power is cheap these days, leveraging that power into useful tractive effort isn't!

So, a quaife 6spd sequential dog box, mounted at the rear of the car, receiving drive via a CTG carbon engine speed prop housed in a carbon/alluminum torque tube, and passing drive out to a BMW E46 M3 "M" differential, with the GKN speed sensitive cross axle clutch/locking system.

Required significant floor pan mods to fit (as you might immagine) but moves something like 100kg of drivetrain backwards in the car a long way to aid traction, and provide a higher polar moment of inertia for stability.

Torque tube and bell hounsings, with trial ally "propshaft" for dimensional conformation:


The real deal, CTG prop:



Assembled with ex Volvo (suprisingly nice die cast and lightweight!) bellhousing:


Concentric clutch actuation:


Transmission rear installation, squeezed between fuel tank bay and exhaust line:




Gearbox with mounts and adaptor housings:




As installed:


Driving out via a cush coupling to allow for any misalignment and provide a degree of torsional damping, into the nose of the M diff (large disc is "mech handbrake" for MOT!):


All my Gears in a line:


Manual (backup) shifter lever (strain gauged for "flatshifting" torque cut) sits on carbon centre tunnel assembly (also holds power distribution and some driver interface switch stuff). Carbon shift rod obviously heads rearwards into a modified shift mech sat on top of the gearbox:



Designed my own pneumatic paddleshift system and controller. Uses a 10bar airpump and small carbon resevoir as a pressure store, via 12v 3 port solenoids to activate a pneumatic cylinder that pushes / pulls for the appopriate gear:


Gear requests via a single carbon paddle mounted on the steering column, pull for up, push away for down. Better than steering wheel paddles for a car in which a lot of wheel twirling may occur (as the driver always knows where it is):


Uses hall effect switches and magnetic triggering for reliability:


Air pressure pump mounted low behind passenger seat:



anonymous-user

Original Poster:

55 months

Saturday 5th October 2013
quotequote all
Rollcage said:
Megaflow said:
Max, where did you get parts such as the centre console, shift lever and air tank from?

Btw: superb build...
I'd say he's made them himself.
Correctomondo!

anonymous-user

Original Poster:

55 months

Sunday 6th October 2013
quotequote all
200bhp said:
Any estimates for power and torque?
TBH, i have deliberately steered clear of talking about power, because there is so much bull and general rubbish talked around the subject. However, the engine design work and component specification (Pmax, BMEP, Turbo sizing, IC efficency, DCR, Cam specification and timing) have been done with a target boost pressure of 3.6bar(abs) @ 7200rpm. If you apply a typical BSAC figure, you would expect to get a flywheel power output in the high 5's / low 6's.

Peak torque will be capped for driveability reasons, but again the engine mechanics should be good for 35bar BMEP, which is around 650Nm.

Enough to be getting on with i think....... ;-)

anonymous-user

Original Poster:

55 months

Monday 7th October 2013
quotequote all
chuntington101 said:
Max, can the turbo support 600+bhp? I know the standard gt30rs is struggling at that level with most opting for the gt35rs for extra capacity.

Do garrett offer other turbos in the race spec?
The turbo currently fitted is my spare, a "ringer" as it's a (well used!) TR30r CHRA with ex GrpA touring car housings and compressor wheels. It's good for ~600, but with a relatively high EBP and boost threshold and low comp eff. Once the car has run for a bit longer, been mapped, and proven itself, i will be swapping on my "real" TR30r (the pukka WRC spec one) based turbo. This too has modified compressor housing and a back trimmed turbine, because the std WRC framesize is optimised for low mass flow at massive pressure ratios due to the mandated 34mm inlet restrictor.

You might ask, "why not put on the proper turbo from the start", well as the engine hasn't been fully mapped, the possibility for a f**k up during mapping, either by a "finger fault" (mistype etc) or just pushing the envelope a bit too much is high, and a TR30r is ~£7k worth of turbo on the open market, and i'd rather not have to buy another one............



anonymous-user

Original Poster:

55 months

Monday 7th October 2013
quotequote all
Trtj said:
Thanks for the reply to my last question! So heres another biggrin

Could you explain how your antilag system works? It certainly doesn't sound like the boy racer botches that you see on youtube!

Thanks again
The so called "ALS" you often see on youtube etc is really just bouncing a turbo car off the rev limiter and backing off, where the unburnt fuel from the spark cut then goes bang in the exhaust. Sounds impressive, but doesn't actually increase turbo speed very much!

On my car there are two ALS strategies implemented, and the one in the video is the "pre-boost" strategy used to increase engine mass flow and turboshaft speed prior to a launch from stationary. It works as follows:


1)Driver selects 1st gear,with clutch pressed fully down, and pulls up hydraulic handbrake lever to hold car in place without rolling

2)Driver pressed LC button - dash lamp illuminates to show system acceptance at Pre-Launch hold

3)Driver removes foot from clutch pedal (system holds clutch line pressure high to keep clutch disengaged)

4)Driver floors throttle:
a) System initially goes into a closed loop speed control mode, using electronic throttle system to lift engine speed to setpoint (selectable via wet(2k5rpm)-dry(4k5rpm) switch)
b) Once at setpoint, EMS commanded to apply ignition retard and revlimit. Crucially, the retard does most of the work of limiting net positive flywheel torque, meaning that the spark cut rev limit isn't very active. The retard (approx 45deg ATDC) means the charge is burnt mostly in the cylinder for controlled combustion, with the rev limit cut only acting occasionally to keep the engine speed constant
c) Once boost pressure is positive, the throttle system switches into "MAP HOLD" mode, where the throttle position is closed loop controlled to maintain a fixed absolute manifold pressure target. This gives a highly consistent engine massflow for good repeatability.
d) At this point (takes approx 2sec typically from throttle input) the dash lamp indicates the system has primed and is ready at Launch Hold by rapidly flashing the indicator lamp.

5) Driver drops handbrake, and a microswitch registers the "off" position, and commmands the system to vent the clutch pressure (controlled to a target pressure every 1ms to give a repeatable engagement) and commands the ems to remove the rev limit and ignition retard. The throttle position is then blended from the map hold value to the current driver demand value

6) Crucially, if the system isn't used, and the handbrake remains up/on, after 15sec, or by re pressing the LC request switch, the system reverts back to normal engine idle, and subsequently, selecting neutral gear will then release the clutch once the service brake pedal is pressed!

In that(pre-boost) mode, no external bypass air is required, because the engine massflow can be high without penalty.

The second ALS strategy is Overrun boost enhancement:

In this mode, the system cannot use a high internal engine massflow otherwise engine braking is lost, and the ALS system operates in a different (and more conventional mode). Once the system senses a trailing throttle after suitable entry conditions have been met (engine temp, eng rpm, peak load before entry etc), the bypass solenoid valve is opened (intake -> exhaust manifold cleanair bypass) the throttle system slams fully shut for 50ms to create a large positive pressure wave in the intake to help initiate bypass flow, the ems retards ignition, and overfuels. During this time, throttle position is maintained at a preset target verses engine speed (a trade off between maximum negative torque (engine braking) and sufficient mass flow to cleanly push the hot vapourised fuel into the exhaust manifold).
This is a much more aggressive strategy as a lot of the burn occurs in the exhaust manifold and is hence a lot less controlled! This strategy times out after 10sec to prevent overheating of exhaust line components.

Edited by anonymous-user on Monday 7th October 12:54

anonymous-user

Original Poster:

55 months

Monday 7th October 2013
quotequote all
mwstewart said:
Max_Torque said:
5) Driver drops handbrake, and a microswitch registers the "off" position, and commmands the system to vent the clutch pressure (controlled to a target pressure every 1ms to give a repeatable engagement) and commands the ems to remove the rev limit and ignition retard. The throttle position is then blended from the map hold value to the current driver demand value
How do you achieve blending of ALS throttle position (map based) vs actual TPS reading?
it's all handled within the ETC (Electronic Throttle Control) ecu ;-)


The APPS (Accelerator Pedal Position Sensor) supplies a "driver demand" throttle request at all times, the ETC arbitrates(based on mode) between this request and other requests from transmission(external) and ALS, idle speed control strategy, MAPhold strategy (all internal). Post launch blending is scheduled based on clutch position, with the driver demand given full authority with a fully engaged clutch.

(ETA: At all times the ETC reports actual throttle plate position to the EMS which helps to keep the mapping simple. This would be critical for a system using tps as a load input, however my system uses MAP (via the multirunner MAP sensor system) as it's primary load dependency)

Edited by anonymous-user on Monday 7th October 20:24

anonymous-user

Original Poster:

55 months

Monday 7th October 2013
quotequote all
ARCHES!


After fixing the new long travel wide track suspension in place, i was easy to see that some wider arches would be required, and that significant inner wing mods would be required to maintain clearance with large diameter wheels/tyres at low ride heights.

wide boy:




Innner rear arches cut out, and lifted, including turreting the rear for the coil over struts, clearance for the driveshafts, and mounts for trailing arms and antiroll bar system



Original steel rear wings cut and widened:



Bare kevlar outer wing in position:







Modified original metal wings to get the basic shape and clearance




The bare shell with the metal work (arches, tunnel, cage, mounts etc) all complete





Next, the bare shell was sent off for acid dipping and electroplating:





When back, the rear wings were flush riveted and bonded into position:






Then the joins were masked with careful application of filler and high build primer






Next job, send the shell off for a coat of "rally car white" paint....... ;-)

anonymous-user

Original Poster:

55 months

Monday 7th October 2013
quotequote all
stevesingo said:
Max_Torque said:
MAP (via the multirunner MAP sensor system)
BTW

Want one for my TBs!!!
Email me with/for details ;-)

anonymous-user

Original Poster:

55 months

Monday 7th October 2013
quotequote all
sierra1off said:
Could you please tell me what language this is so i can get it translated .

Thanks
;-)

Suggest you just watch the video i posted a few pages back! laugh

You can see the launch mode handler working in that vid actually, with an initial rev with normal sparl timing (sounds normal) then the retard coming online, and the throttle opening, the rev limiter just trimming the ocasional spark to keep the engine speed constant, and finally turbospeed (and boost) building to target. Obviously, i that video, with the garage only 3 feet infront of the car, i terminate the system after a few secs at "launch hold" ;-)

anonymous-user

Original Poster:

55 months

Monday 14th October 2013
quotequote all
And finally an easy job, send the shell off to get painted! The only bit i haven't done on the car myself really:

Nice to get back a perfectly clean and perfectly white "bare" bodyshell:












Next, the hardwork starts again with the start of systems fitout ;-)

anonymous-user

Original Poster:

55 months

Tuesday 15th October 2013
quotequote all
No, not really. As a 2wd it couldn't make use of any extra low rpm torque, and the effort would be better spent on chasing more power......... ;-)

anonymous-user

Original Poster:

55 months

Wednesday 16th October 2013
quotequote all
More revs! ;-)

I'd like to build a 10krpm engine, but to do that with my current one would be extremely difficult due to the undersquare configuration.

(Oh, and i'd like to go DI as well)

anonymous-user

Original Poster:

55 months

Wednesday 16th October 2013
quotequote all
Very tricky if i wanted to go DI, because pretty much all the modern DO engines are small bore low speed engines ;-(

Ignoring DI, then a Millington cylinder head (95mm bore) on a machined from solid block / liners with gear driven cams would probably survive the loadings, with a short stroke crank and nice long rods to keep control of piston dynamics etc

anonymous-user

Original Poster:

55 months

Thursday 17th October 2013
quotequote all
Back to the (neverending?)story ;-)


With the bare shell back from painting, the task of fitting out the cars numerous systems, mechanical, hydraulic & electrical could begin. A massive task to do neatly and properly. I've seen too many cars on a rally stage break down for want of a well places cable restraint, or a properly routed pipe etc.

I'd tried to trial fit most of the major systems before the shell was stripped/electroplates/painted so most of the holes for fixings etc were already drilled, but of course, you can never get 'em all, and constant evolution of the systems during build mean't that some drilly/hammery stuff was required (but not much!)


First work centered around the main bulkhead, as this is completely inaccessible once the engine was installed





Pipe runs for brakes, clutch, firesystem, PAS, cooling bleed, fuel, and vehicle electrical looming went in.


On the cabin side, looming and engine control units were mounted:


The centre tunnel assy carries all the power distribution and hosts the front - rear loom runs, and uses a number of multiway connectors to interface to the bulkhead and read looms to make it easy(ish) to remove


Pretty much all the systems that are not engine related are mounted low behind the navigators seat, inc water injection pumps and reservoir, PAS, Fire system, main electrical contactor etc



The hydraulic handbrake and rear break proportioning valve are also mounted on the centre tunnel cover


The pedal box and steering column were installed on the drivers (LH) side:



And the dash cover fitted over the top:



(shown here prior to being 'flocked')


Now the dummy build engine could be slotted into place to allow engine bay plumbing for the fuel, oil and cooling systems to occur:





I really didn't want to end up with the dry sump tank in the boot, as that adds masses of weight and means you need a lot of oil in the system to avoid surges etc. So that gets packaged tight at the rear RHS of the engine:




Throughout this time, numerous systems were built up as modules ready for installation:

Engine + throttle ecu's & Water injection + boost solenoid drivers on a carrier that mount them under the navigator foot rest:




Centre dash switch pannel as a "plug in" module:




Completed center tunnel unit with manual shift lever:




First stage electrical system tests were also run outside of the car (living room floor ;-)




The electrohydraulic PAS pump was not really man enough for the job and so got ditched early on for the latest TRW brushless electrohydraulic unit, being commmanded via CAN bus data:



Again, mounted low and rear wards for mass optimisation reasons.


Apparently silly little jobs, like a clutch foot rest and floor plate absorbed more hrs of time:



Luckily, being 'fly-by'wire' the throttle pedal was an easy job without the usual hassles of getting a nice smooth mech cable routing etc ;-)

As more and more systems were fitted, it became harder and harder to maintain a quality and neat install!




Next time, engine support systems (fuel, exhaust, cooling, heat shielding etc) ;-)

anonymous-user

Original Poster:

55 months

Sunday 20th October 2013
quotequote all
With the engine low and far back in the chassis, and the turbocharger mounted as low as possible, the exhaust line was fairly straight forward, but i wanted the exhaust system to be tucked up nice and tight into the floor pan and transmission tunnel so i could run the car low without danger of knocking low hanging parts off it!

Space constraints also mean't that quite a lot of "hot" bits would be crammed into a small and not necessarily well ventilated space.

To limit thermal issues a significant amount of thermal protection was needed between the hot exhaust line components and the rest of the car:


Particularly around the RHS engine mount & dry sump tank:


Underfloor heat shielding:





Post turbine exhaust and wastegate exhaust integrate nicely togther:




Exhaust is pretty conventional 3" stainless system:


Slip joints and springs allow for thermal expansion to occur without excessive strain on the fairly solid mountign system.

Fan style 'tailpipe' sneaks out under the back of the off side rear sill:


That saves having to route the large diameter exhaust through the rear suspension and diffuser

And straight onto the.........

anonymous-user

Original Poster:

55 months

Sunday 20th October 2013
quotequote all
.... fuel system!


The small (35l) "sprint sized" ATL tank sits in steel housing low and rearwards:


Dual (in rearmost LSH & rear RHS corners of the tank)internal low pressure lift pumps sit in the tank using a "trapdoor" system and foam to prevent excessive slosh:


Fuel filler is routed upwards to a panel let into the nearside rear window aperture. The 2" fuel filler hose is fitted internally into a larger diameter alluminum pipe to protect it, and isolate any leaks, that are vented out the bottom of the tank well rather than into the cockpit:






The filler cap and fill vent (via quick fit connector) mount into the window panel:

(as do the GPS & RF Burst Telemetry antennas)

The tank internal LP pumps feed into a swirl pot mounted in front of the tank (but under a sealed cover from the cockpit). The outlet of this swirl pot runs to a dual series set of BOSCH 044 gradeA high pressure pumps.



The -6 high pressure fuel line is routed forwards along the nearside sill (in cabin) and up through the main bulkhead:


At that point the fuel pressure and temperature is measured, before the fuel is passed along to the two engine mounted rails (primary & secondary rails). The primary rail is mounted in the normal position, close to the intake valves, and the secondary rail with the much bigger injectors is mounted in a "staged" location up in the plenum, firing directly down the intake trumpets:



The secondary rail also features a pressure damper to help remove dynamic pressure fluctuations:



The system is returnless, and uses a pressure controller to supply fuel to a target pressure (actually an injector delta pressure as it takes into account the plenum pressure). The pump controller sits above the pump housing:


It also controls the lift pumps, carrying out both key-on priming and de-aeration strategies and also sequences the lift pumps depending on lateral g (biases towards the lift pump where the fuel is sitting due to cornering g). The high pressure control has various modes, including an overpressure start assist (8bar) to help prevent vapour locking during hot starting. Normal injector delta pressure is maintained at 6 bar during engine run, and at 3 bar during idle(lengthened minimum pulse width) and engine off( reduced noise/power consumption)



Next up: Engine cooling ;-)


anonymous-user

Original Poster:

55 months

Monday 21st October 2013
quotequote all
The exhaust manifold is conventional 4-1 stainless affair really, as it was designed for when the car was going to have to run with a compressor inlet restrictor (and hence have limited mass flow). As such it is aimed at maximising pulse recovery from low speed rather than being absolutely low restriction/high flow. It's one of the things i would change if i get the time tbh:







The tank has been re-certed don't panic! (because it isn't a bag tank, the re-cert just consists of an internal / external inspection for damage etc)

anonymous-user

Original Poster:

55 months

Tuesday 22nd October 2013
quotequote all
When i made it, WRC turbines were single scroll only and with properly calibrated ALS there is little benefit to a twin scroll housing really.

anonymous-user

Original Poster:

55 months

Friday 25th October 2013
quotequote all
AER said:
Max_Torque said:
[/URL]
You really did build your engine in the lounge room. Well done!
Yup, and at the time, it was the most expensive thing in the house........ ;-)