Members Page/Marcell Gal/Power And Traction

MembersPage/MarcellGal/PowerAndTraction (2007-06-11 12:28:32)

Having seen MembersPage/NanassyPeter/CloneInstall it became apparent that it's possible to make something for appr. 8000 EUR that leaves most Ferraris, BMWs and factory Subarus behind (in normal road circumstances or even during drag-races).

TODO: cleanup and move stuff from this page to the relevant subpage: only leave overview here

[gallery] (more pics will be uploaded)

Configs and Logs

Last [configs and tables] and [logs with 3B factory injectors and factory FPR]
- injectors since upgraded to 630cc/min siemens pencils, so I will NOT be able to capture logs with factory injectors in the future; but I have some more logs on my disk available if needed
- you can check pulsewidth at cruise and power, and ignition advance (maybe conservative at power, but not extremely high fuel consumption at cruising)
- note that AAN injectors are said to be slightly higher, and FPR is also higher (4bar instead of 3.5bar means +7% alone)

Factory manuals needed

Thinking about getting the 1989 audi100-200 repair manuals. Not only engine, but the whole car (owner's manual is not enough!). I prefer digital to paper (I can print the parts that I need, maybe 10% at max.

Unfortunately not all that I need is available at this time. I'd be happy to pay 49 USD for the 3 day ebahn licence. If someone needs same set of docs, and willing to click (to print pages to pdf via pdf995 driver), possibly for hours (not sure how big chunks it's possible to print), let me know here.

Audi people suggested that I get etka, so I can look up part numbers. They said something about emule/bittorrent... I haven't investigated this at all.

I allocated gigs of storage place, and allowed anonymous uploads to http://ftp.hostmonster.com/?domain=ftp@vems.hu (into incoming folder).

I got a AAN (with K24 turbo) audi quattro running with PNP VEMS box (with motronic-55 connector) on [this config and tables]. That is not my engine, although very similar. My (old) harness is 3B (with lotsof small leafs hanging that I have no idea where to connect; and injector part had to be remade because of flaking insulation; guess it didn't like the heat for 18 years).

My 3B engine is running with PNP VEMS box and 1.1.23 firmware. While the HALL sensor was bad, it started appr every 2nd try, now fixed

good startup with appr 5/2msec cwl/cwh, using batch mode during cranking. Works slightly better than 17/11 msec in alternate mode (during cranking).
after plugging in the extra manifold nipple, air can now be controlled I tuned idle-air controller. It recovers reasonably when the big fan starts
- but if ECU turned on the big coolant fan, we could raise idle beforehand
after adding some ignition advance (and some 100..150km breakin), the engine is getting nicer and nicer
- I still have to verify with a timing light that the flywheel pin is sensed 62 BTDC cyl1 (==134 BTDC cyl5) and base timing is correct
- a leakdown test or at least compression test would also be nice to see how equal they are
- some detailed logging and loganalysis with individual cylinder power would also be nice, to compare the actual power output from individual cylinders
note that 1.1.x (at least upto 1.1.23) is not ment for distributer ignition
- but works if dwell is limited: set dwell6=0 and make sure dwell14 is short enough for upto rev_limit
- also make sure the coil has max 7.5A fuse, just in case
- a 2x2 coilpack fixes that very quickly and cheap, no more dwell limits at high boost ! I used the same: cyl2,5 and cyl3,4 of course (a given coil does not drive cyls neighbor in the 12453 sequence). Cyl1 still driven by itself (from factory power output stage + coil in my case), that makes strobing easier.

Many thanx to

MembersPage/JorgenKarlsson/AudiStwo for injector pulsewidth (size, and other) hints
MembersPage/MiskaPeippo/AudiSSix for precious trigger-HW hints
thanx to Teppo for his patience
MembersPage/DaveHartnell for his help (Dave has audi docs ;-)
others for various hints

Project plan with parts and costs

New engine is running in NA mode (no cooler tubing yet)
Engine has electricity, coolant, oil and power.
Will run NA for the breakin period. The charge cooler is mounted, but no tube before or after it. The pre-cooler tube will be a pie. Post-cooler tub tube sucks... goes around the bay (it's hard to run boost without the plumbing). Maybe a 40x80 mm profile right above the cooler would work. The hood lock is in the way, (maybe [aerocatch] would help?). The other solution would bring a curved tube (above the charge cooler) out the grill (either solid black/darkblue; or showy chrome or white) just to go back at the other side. For airflow, this would be best.
injector part of the harness was worn beyond repair (the PVC insulation audi used in this 1989 luxury wagon was pealing off in flakes due to intensive heat and long time). We had to make new injector-part. It was quick hack (not by the professional harness partner which has latency of 2+ weeks), and the guy who made it did not obey lengths => the curves look a bit clumsy, but it works.
we're patching the harness to allow leaving out the distributer, by driving more than 1 coils (3 with wasted spark or 5). No idea about the distributer phase correctness
DONE: patch the harness for WBO2
DONE: ignition
- cyl1 ignited from factory logiclevel (powerstage and) coil via Stepper-D,
- cyl2..5 ignited from a power-input 2x2 coilpack via 2 IGBT outputs
- cyl2 and cyl5 ignited via ignch2
- cyl3 and cyl4 ignited via ignch3
TODO: find out how to fix/replace the HALL sensor (or the whole distributer)
- most importantly, how to rotate the HALL window into position (I have a scope)... and how to secure there
DONE: connect starter
- first we thought starter was bad. But we just had to use the hard-to-see spade connector (the other connector, that we first thought was the only one, is only some kind of feedback - currently unused)
DONE: connect alternator (seems to charge when engine is running, > 13V)
DONE: tachometer gauge (driven from VEMS genboard) works fine
TODO: hydraulic system check
- hydraulic pump is now driven by a belt (belt was missing, I had to remove the alternator belt to install the hydraulic V-belt)
- I read somewhere that one starts to pump the break after engine is stopped. If he can pump for 15..25 times, the hydraulics is OK. If only 3..5 times, than something is not very good (fluid level? leak?)
- there is power-steering and power-breaks but break pedal becomes heavy immediately when I switch off the engine => what to do ?
- what is the correct level of hydraulic fluid in the reservoir, and what fluid is in there ?
DONE: get the coolant radiator fan turn on when needed (connected a big 3pin connector at the bottom of the radiator, under the big hose)
DONE: the water manifold has the "multi-function-temp-sensor" (MFTS), the connector housing was missing in my apparently raped harness, wires were patched with blue wires (still barely reaching the sensor, so it was far from intuitive what goes where ;-), but original wire colors and nice info on [SJM page] helped, so I could connect the spades one by one with the help of a small mirror.
TODO: connect "afterrun temp sensor" some day (not applied at the moment) with 2 simple blades connector. This has nothing to do with ECU, or FAN.
DONE: the fan needs to be connected too
DONE: there is a temp-sensor with a huge 3pin connector at the coolant radiator, under the big hose at the bottom of the radiator. AFAIK this will turn my fan on if radiator temp > above a threshold.
TODO: there are a lot of wires (insulated but) flying around that I know not where to connect. Some are most likely for the (yet missing) air conditioner. Others ... who knows..
TODO: I'm a bit puzzled. S12 fuse (15A) is fine. S12 powers MFTS and transmission wheelspeed HALL sender according to http://www.sjmautotechnik.com/trouble_shooting/cooling.html
- "Fuse #12 supplies +12V to the Multi-Function Temperature sensor on the original 4 terminal sensors, so check your fuses. This fuse #12 also supplies +12V to the back up lights (==white lights at the rear, that lit up in reverse gear) and to the speedometer sending unit in the transmission."
- TODO: get cockpit lights (it is dark!)
- TODO: get speedo signal somehow. Now needle rests at 0. AFAIK the cockpit gets speedo freq from the transmission, not from the ECU.
- no wonder the speedo did not work. The 2pin connector I connected to the transmission (right side) was for the backup lights. For the speedo there is a 3pin conn on the left side of the trans, yet unconnected. I gotta find a matching piece. g60: any wire color idea?
- hopefully the ABS light will turn off
DONE: replace front right turn signal lightbulb. Easieast from outside. I almost ruined the socket when, lacking the manual and listening to bad advice, trying from the inside (engine bay)
DONE: removed springs from wastegate so car can be driven safely (as NA) until the cold-air tubing is done, without danger of overspinning the turbo
TODO: I need the partnrs for
- the audi200 1989 coolant tank
- and the 2 rear "rubber" bushings at the bottom of the rear suspension

See old photos, clutch, flywheel and enginebay [pics].

The clutch looked burnt (and weak anyway).

a strengthened SACHS clutch was ordered (and delivered) 700 EUR (was solved with a different type of flywheel). Pressing the clutch required impressive force, sg. like 120kg * g (I didn't measure, just by feeling)
- Teppo suggests: I can't see why you needed a new flywheel (see [3B - AAN differences] : lists the flywheel clutch difference, though not many know why it was neseccary). Sachs Race engineering pressure plate from Speed for about 300 eur. New Sachs standard cclutch disk about 150 eur. Those would hold 600 Nm without a problem
- Anyway, the workshop that delivered the clutch handled the clutch subproject 100%, and they are good partners (they are our customers at the same time), also they provided professional services and ordered factory parts (gaskets and other) during the install which they charged none for.

New, Bosch044 fuelpump was installed (without the factory-designated accessories that would have cost the same as the pump)
- stronger hardness=12.9 flywheel bolts were used
- oil cooler nipple was broken beyond repair. New (perhaps bigger) oil-cooler would be needed. There is a fall-back (standard size, used, for 45 EUR) oil-cooler if we don't find a better one.
- all necessary gaskets and flywheel bolts are getting ready as well
- most importantly, we hopefully get all gaskets, screws/bolts, the small timing wheel and other small items in time.

Since I have no mechanical installation experience (especially with this 5cyl engine), I made a spreadsheet to track the parts necessary for install. It would be nice to prevent major stalls due to the lack of small gaskets or bolts or other accessories.

See [project plan audi-costs.xls] and add your hints below.

totals > 6500 EUR ;-) for engine + install: without any air-conditioner related items
Hungarian and English name for parts
contains past and future (estimated) costs (4 EUR units)
current location of parts
- all should move to location 'X' in next few days where most of the assembly takes place
any further (charge-cooler to throttle) cold-air tubing might continue in 'K' later, if necessary

The car should preferrably come out from the assemly shop on it's own power. Even if in NA form (with turbo+oil installed, but wastegate open and perhaps without charge-cooler to throttle tube, with improvised air-filter)

Add your hints here (anything missing from the spreadsheet, or errors, bad concepts, etc...):

air filter !!
fuel ;-)
flywheel screws
flywheel front and rear simmering
intake manifold gasket
exhaust manifold gasket
turbo gaskets are available (came with the turbo)
distributer needed for cam-HALL signal (even if 5 coils are used)
the (factory) oil-cooler tube is worn, must be replaced. Might be a simple hydraulic-task, or it might suck ? Does this connector have a name ?
more gaskets ?

1.1.x firmware for InputTrigger/AudiTrigger will have some advantages over the old 1.0.x (especially above 7000 RPM). But 1.1.x had problems with auditrigger, all these problems must evaporate this weekend.

www.vems.hu/download/v3/firmware/ for 1.1.19 experimental firmware, already has some auditrigger (135 tooth setup) related changes. I would have no concern trying it on my engine (first at low and mid-RPM, low load).
- But more testing is needed before release. Continuing with the bench-tests

[pictures about the engine pieces]

note that [almost_factory_k26_26] folder contains pictures about and installed almost factory K26-26 setup. That is NOT mine (the main coolant radiator is in front, not at the left side)
- on pics 36..39: NOTE THAT THE INTAKE MANIFOLD THROTTLE PLACEMENT ON THAT ENGINE IS RATHER SUITABLE FOR A FRONT-mount charge cooler. On the other hand on my 3B intake manifold throttle placement seems very bad for a turbo install. One end of a frontmount charge-cooler connects to compressor outlet fine; but the left-side of the charge-cooler is VERY FAR from the 3B throttle, must go around the engine bay as bad as it can get. Should I try to find an intake like that on pic 36..39 ?
look at [head]
- looks nice, no cracks at the sparkplugs
- from pic0012 and pic0013, can you confirm this is the turbo head ? The holes for the extra-coolant manifold under the intake valves confirm this, right ?
- for now, the normal camshaft will stay (not changing to 7A)
do you notice anything noteworthy ?

We accept that the max power of this engine will be limited by the following parts:

K26/27 turbo, especially with the #6 turbine housing
- though K26/8 housing might be tried later
the factory exhaust manifold
- though nicer manifold is not ruled out
the factory charge cooler
- this will be changed later
maybe pistons as well
- Pistons are fine for 500++ hp but rods are maxed out after 500 Nm, they can take even more beating if bearings are ok and there is no detonation
- and JE pistons and Eagle-H-profile rods are clearly overengineered: much stronger than needed for K26-27 turbo. But the package cost was reasonable (high-specs engine with head under 3000 EUR - without turbo and exhaust manifold)

Effectively, max power (on the crank) that can ideally be achieved with this setup is estimated max 380 .. 400 HP.

We don't want to stress it further because:

many parts (at least the ones in above list) should be upgraded to get even just +50 HP
- this would delay the project and draw attention from our main focus: to get the electronic, firmware, config and docs part completed
380 HP is already dangerous enough
we want a streetable car. This means we want boost at around 3000 .. 3500 RPM, and we don't want to sacrifice reliability for further performance

Engine specs

JE pistons
- CR ???
eagle H-rods
Gotz sputzen bearings
metal head-gasket
factory intake manifold
exhaust manifold for K26/6: under investigation (using the factory manifold if a new proper manifold does not arrive)

As you see, this is inhomogeneus. Sounds stupid to use those pistons and rods (good for 600..900 HP) when the turbo limits power to appr 400 HP. However, we had a short timeframe and limited supply. So we chose a built (by experienced parties) bargain engine for appr 2500 EUR instead of a used "junkyard-engine" for 1600 EUR (though that had more spare parts). This way we can reduce the risk of wasting (more!) time because of mechanical problems. And possibly have some room for more HP for future upgrades.

The 20VT (5 cyl, 20 valve turbo) engine is quite amazing

intake and block are very good (intake allows >400 HP if other parts do support the air), with good supply of aftermarket pistons and rods if one wants more
factory turbo is restrictive unless one settles for 250..270 HP
factory (casted) exhaust manifold is a joke. Noone knows the real performance, probably it gets restrictive above 350 HP, and makes airflow differ in the cylinders. Most tuners will install a new exhaust manifold (with a bigger turbo of course)
- though some say even the factory exhaust manifold works well up to 400 Hp. It's hard to tell the flow difference between cylinders.
the drivetrain (gearbox, differentials and other pieces) of the audis that came with this engine is very durable, this makes it a low-budget option to make a high performance car
- though they understeer, which can be helped somewhat with right suspension, but some understeer will remain.
and last, but not least, leaves enough room in the engine bay for extras, and maintenance

We purchased a

street-legal Audi 200 quattro turbo (paper says 1400 kg without driver and fuel: a bit heavy, but not insanely)
with (known broken, see pics below) 3B (20V 5 cyl turbo) engine (and harness)
- though 3B harness is different than AAN harness, all the main points are the same. The small wiring differences, including 1 or 5 ignition outputs are very easy to apply

We understand the potential in these engines and want to support tuners around the world who chose this engine.

The most important target of this install is to get the InputTrigger/AudiTrigger install procedure as simple, painless and reproducible as possible

doc improvements
v3.x assembled with 55pin motronic for PNP install (reusing the MAF and NBO2 wires for the WBO2), possibly min 2..3 versions:
- 1 coil harness
- 5 coil harness
- TODO: evaluate the other 55pin pinouts out there
- not sure if the K-jet harness is easy to support at all
perhaps with firmware and logging mods / suggestions, to help finding and fixing trigger issues

Near-future plans with this install

20VT - 5 cyl 4 valve, turbo
4WD - leave as is ;-)
leave factory intake for now
- DROPPED idea of highend intake: after the throttle, the ends of 5 pipes arranged in a circle around a cone, in the firing order of course (tube from charge cooler to throttle might need to go around the radiator, not a big problem)
- CHOSEN: factory intake is good enough for us
get a working bottom-end (the block in the car we purchased was broken). This should be hammered before 2007-03-15
engine management VEMS

K26-6/27 turbo, the K27 compressor looks like (cm scale):

Exhaust and turbo - advice accepted, I have no experience here

We missed a nice looking 500 EUR exhaust manifold (auctioned 2007-03-29 on ebay.de)

A friend purchased [187 USD RS2 exhaust manifold] from [ebay]

it's actually 187 + 100 USD with shipping to Eastern EU
as you might expect, the weldings look good outside, not so perfect inside. But not very bad either
friend will try to stick (bigger than K26/27) turbo on his manifold, without extra strengthening (for street use). I think that is a bad idea, the exhaust manifold will crack (sooner than later) due to heat and vibration.
- I would definitely strengthen: connecting to near engine-mountpoint (seems best: turbo is very close to engine mountpoint in the audi, 8..10cm: only watch some clearance for securing screws). Connecting to head is another option: but that is further away, seems suboptimal.

Turbo (compressor side) upgraded

to K26/27 (appr 6xx EUR). Good for max appr 400HP. Minor compromise is that it is now an oil-only turbo (=> no watercooling: if I understand right, keeping watercooling would require some overpriced factory parts that just aint worth it).

The K26 turbo was upgraded by http://www.ahturbo.hu/ with instructions:

to upgrade compressor side to K27 with rotor sizes 53mm/76mm
- turbine wheel dia=56mm, compressor dia=53mm (would be useless with turbine smaller than compressor).
- turbine outlet is 56mm diameter (the 6 in KKK numbering scheme). It's the clone of a proven setup (that worked well in an engine built by them earlier). Perfect for the street.
turbine renewed (but housing kept)
oil-only accepted (no water-cooling)

K26-6 turbine housing (cm scale):

Kalske says: Remember that RS2 turbo is 48/54 which is good for about 360, you are getting 53/55 which will do 400++
- Nanassy said exactly the same !
Kalske and jhwh say: the same turbine wheel goes into 3 different A/R housing, marked (in KKK numbering) 6,8 and 10. Higher nr gives more power (appr 45 HP each step), but boosts later (3000, 3500 and 4000 RPM, respectively).
- so turbine housing="8" is the best choice for this turbo. Stay away from 10. 6 is not optimal, but works. (and 6 might match the restricted exhaust manifold and factory charge-cooler and street-operation better than the other turbine housings)
- you should look for a cheap #8 house. On [ http://forums.rennlist.com/rennforums/forumdisplay.php?s=&forumid=72 rennlist] you can probably get a 8 housing next to free as many people upgrade
- if you come over a good deal on one it should only be a minor mod, remove turbo, remove housing, bolt back on
AFAIK KKK uses cm2 rating on turbines instead of A/R (on Garretts, and so on...). I guesstimate the 8cm2 turbine will provide a quick spool-up (maybe in the 2200-2800 rpm range on a 2.2l engine), which means big fun on street, and some restriction in the top-end power. -GergelyLezsak-

Later

bigger injectors will be necessary to utilize the bigger turbo
- maybe higher pressure FPR (or FPR springs?)
possibly water injection

You can find some pictures under [files], including

old engine with cyl1 wall knocked out by rod; cyl1 rod shrinked 10mm !!

Links

[Audi 200 specs]
- this engine does not seem to have cruisecontrol capability (huge "idle-valve" or throttle actuator)
[Repair manual] - awesome resource
http://en.wikipedia.org/wiki/Audi_100
[ self-documentation wiki] original selbst-doku.de is German, and the English is just a babelfish auto-translation (not recommended to edit the English)