Sams Polo 3.0

[sambb]

So this is the direct port water meth hardware setup I've been running on the generic APR tune that came with the car. The system is a Snow performance 'Boost cooler Stage 3 with mpg-max'. Basically unlike the cream gear eg Aquamist where they run fixed 150psi pumps and then use pulse width modulated solenoids to deliver very accurate spray, the Snow uses a variable voltage signal to the pump ie it ramps up pump pressure to match water meth to the fuel curve. With a system like this because you may be starting from low pump pressures you need to really run a small nozzle pre throttle body to get things started and then a solenoid (on/off only) then opens up a second nozzle stage for when the initial small nozzle cant meet demands.
So I basically ran a 60cc pm post IC/pre TB nozzle that would start spraying at 30% fuel injector duty and then at 55% duty the solenoid would open which would fire up the direct port nozzles seen in the pics above. The line that goes off to the left comes from the pump, it feeds into the solenoid which is mounted directly to the water meth manifold before each line goes out to each inlet runner.
Now a few things I think I discovered:
- the pre TB nozzle works super well at keeping inlet air temps down. I could start a 3rd gear pull at 26 degrees and finish on 24 degrees! If only inlet manifolds were see through but I would love to know if the water meth marketing is true and all that nozzles droplets were evapoarted by the time the air passes the iAT sensor and the sensor is truly reading cooled air. I do suspect though that part of the cooling is due to the fact that this is simply not true and what is really happening is that un evaporated droplets are wetting the iAT sensor and then as air rushes past it, of course it reads very low.
- what sort of confirms this, for me, is that well after the nozzle has switched off, iAT readings stay low and usually continue to drop (say after you've given the car a short sharp hit and then backed off). That could only happen to my mind if the sensor was actually wet. Also I think droplets make it all the way through to the runners in un evaporated form because when I was running this pre sensor only, I could use it to cancel timing pull on cylinders 1 and 4 but 2 and 3 would still suffer timing pull. If the water meth was truly in a gaseous state then you wouldnt see that.
- now the ECU in our car from what I've been told by a tuner is always trying to jump into more aggressive timing maps based in large part on favourable inlet air temps. What I saw using the pre TB nozzle was that if you were trying to fix a situation where you were seeing 6 degrees of timing pull, you could add water meth, you would see in the logs that the car was running much more timing across the board, yet you would still see 3-4 degrees of timing pull. I found this on my first water injected Polo too. I think what happens is that the water meth produces low inlet air temps, the ecu says 'sweet, we're in Sweden, i'll add loads of timing' but then the water meth you are spraying pre throttle which produces big temp drops, isnt actually sufficient to handle the knock. It might have been able to cancel the knock that you first saw, but not the knock produced by the additional timing the ecu added. Hope that makes sense but it seems to me that you see a certain amount of timing pull, but you are never really able to counter it with a pre TB nozzle because the ECU keeps moving the goalposts on you.
- so then I thought Ok the pre TB nozzle will drop the temps, the ecu will see that and step to a more aggressive timing map, so what I need is direct port water meth to handle the sum total of the timing pull that the engine was predisposed to + the additional knock that will exist due to the ecu having advanced the timing further. So the above direct port was added. The car had already come with direct port but it needed a re-jig to get it ship shape and the above is what I ended up with. What I discovered is that direct port is very very good at killing off knock. So much so that without any water meth the car had sometimes been pulling timing and spark had been coming 0.5 degrees after TDC!! With the pre TB cooling nozzle it was obviously in more advanced maps but then the addition of direct port meant that finally all timing pull was cancelled. In the same spots the car might have had 12-15 degrees BTDC. So it was hugely effective.........but bum dyno said this is not as fast as before and the logs proved it. It was down 10+ g/s of air through the maf at the top of a third gear pull.
- What I think was happening there was that the water meth was so effective at stopping knock, that it was getting into that E85 characteristic of being able to actually take so much timing without knock that it was well past MBT ie you could keep adding additional timing without knock but it wasnt actually making any more torque. So I was watering it to the point of killing off all the timing pull but at these levels it wasnt going to make any more torque and I beieve actually loosing power because by then my octane was probably at 116 due to all the meth.
- so what I then did was ditch the pre TB nozzle. My thinking was that I didnt need to fool the ecu into thinking I was on a cold winters night. The timing that the ecu wanted to run with was probaly most appropriate and if there was any timing pull happening, then the direct port water meth could deal with that over the narrow range where it was a problem. This was definitely the best solution. Timing stayed more petrol-esque rather than E85-ish. The car still had snappy boost response but you could see where timing pull would try to happen and then cancelled very quickly. Problem was, then I was getting O2 sensor heating circuit too low or open circuit errors. I dont think its possible that un evaported water meth could get though the combustion cycle at all so I dont think the sensor was getting 'wet'. It could be that either the exhaust temp was so low due to the effect of water meth that the O2 sensor wigged out, the by products of methanol were damaging the o2 sensor in some way, or more likely that last part of a third gear pull when you lift suddenly was throwing a hugely rich water meth solution through it at a time when the injectors had already turned off. Either way this aint good either so I think from here what is in order is:
- accurately measure the pump pressure and drop it from there. If that fails then choke down the feed line with a restrictor to get it so that only just enough water meth that is needed is sprayed.
-come back from 50:50 water meth to say 70:30 to cover off the possibility that meth is hurting the sensor.

So this was mostly an experimentation with water meth on a tune that had some pretty manic timing pull and allowed me to get my head around its workings in case I choose to run it down the track. The car is in the process of getting remote tuned right now so with a sweet dialled in 98 ron tune, the water meth I have here will probably be null and void but it'll be good to have in the back ground in a much less aggressive state for those insanely hot summer hillclimbs when I'm sure timing pull will creep in.

all testing was done with 50/50 water meth and yes I did pull the primary fuel back a bit allowing for the additional fuelling that the meth would add to the mix. So where I had 0.82's with petrol only, I was still on 0.82's with the 50/50 water meth throughout.

[Sydneykid]

You already know the answer Sam, you need to move to E85, be it 100% E85 or by adding some E85 to the Pump 98. Having 2 tunes and being able to quickly swap between them would be good. Removes all the water injection hardware and most of all the metal gymnastic working around the shortcomings.

Cheers
Gary

[sambb]

Indeed. Yep having seen how effective alcohol is at killing knock in a pretty knock limited engine, I'm pretty well sold on the idea of E85. As soon as the car gets a Link G4+ or an Elite 2500 plug n play it'll be tuned for E85.

In terms of doing it sooner, I do have two ECU's now. The one in the car has the immobiliser on it so will always be the petrol/road ECU but my spare ECU is immob defeated and would be perfect to switch in for track duties. What concerns me though is:
- can I run with existing pump/underbody fuel lines with E85 provided the E85 is only ever in the car for a weekend maybe 8 times a year? The adjustable FPR is E85 tolerant and so too are the EV14 injectors. The APS fuel pump isnt classed as E85 safe and I'm not sure how long those hard nylon click fit fuel lines will tolerate things. The rubber fuel lines in the engine bay are easily changed though.
- If I was to tune on a factory ecu there's no way to integrate a flex fuel sensor. If I'm flying a little blind RE specific actual ethanol content then the tune would have to be a conservative one and to add to that it'd be a remote tune because flash tuning on a dyno is not something I have the dyno time/dollars for - so I'm unsure if THAT sort of E85 tune would still be that much better than a decent 98 tune.

Yeah water meth injection is great but once I got it working effectively it astonished me how much water it was going through. Its fine to have a little 2L container on the street but on the track I dont think you could even do 5 laps at full send without it going empty and then your engine goes bang. I dont like the ide of having to run with a 10L container of extra fluid on board.
Also while I was testing, the Snow nozzles use 1/4 inch fittings but I only had 6mm tube on hand. The fitting felt secure but in service a 6mm hose blew out of one of the nozzles and sprayed 50/50 meth all over the engine bay. If I'd been running richer than 50/50 on the mix I may have had a fire, the fluid emptied the tank almost immediately and if that happened on one cylinder like that on the track you may not feel it at all and thats the cylinder that would then rattle itself to bits. So for the amount of hardware you have to have in place, the risk to the engine if any one part of the system fails and the added hassle of carrying around a second fuel that you always have to keep an eye on....... I think I'd rather just go E85.

[Sydneykid]

I have an E85 tester (tube) which, if I remember rightly, also has markings for E10 and E50, could easily add more. The United E85 is pretty consistent, never seen any below 80% in 12 years.

Something like this CHECK YOUR E85 CONTENT!!! EASY HOW TO WITH SIMPLE TOOLS - YouTube

Quite a lot of guys mix their own E20 (103 ron), being ~5 litres of E85 mixed with ~15 litres of Pump 98 in a 20 litre drum. That seems to overcome the detonation issues and enable decent ignition timing. The common quick tuning is to add 5% more fuel and then tune the ignition timing to suite. It's not enough ethanol to cause fuel systems issues (which I think are overrated anyway) and 5% extra flow from the fuel pump is not normally too much of an ask.

Cheers
Gary

[sambb]

hmm you've got me thinking on the E20 method there Gary. I'm really wondering why not. Thinking I may not even have to adjust the FPR as the injectors probably have enough duty cycle head room in them on the current small turbo.
If on E20 the engine was still knock limited before MBT (like with petrol) then I'm thinking it'd be totally within the scope of the remote tuner to sort an effective E20 tune. Yeah It'd be no big deal leading up to an event/track day to run the tank right down, go to the servo and test the E85 is legit and if so mix my own 20L containers of E20, plug in the 'track' ecu and then head off. Once all the covers are off changing an ECU takes 5 seconds. Just like pre-mixing an outboard (or a rotary ha ha).

[sambb]

I'll have a completed 98 ron tune very soon. Before going down the path of a dedicated E20 tune, I'm pretty sure I can test E20 out using some rudimentary tuning myself. The old Lemmiwinks software lets me globally adjust my main fuel term and have control over low load fuel trims and allows global timing adjustment too.
Assuming I'll have duty cycle head room in the injectors I'm more than prepared to have a fiddle with E20 with two eyes on the logs of course. How it'll respond to timing changes is usually pretty clear but I'm not sure RE fuelling. Is E20 so different that the ideal lambda will need to deviate much from the 98 ron position? If say my best safe 98 ron fuelling is at 0.81 lambda or the like, will I likely need to move a couple of points richer or leaner from there once E20 is in the tank or will it most likely just be a case of nudging up the timing and seeing where that heads? If the results are tangible then for sure I'll get the other ECU tuned specifically for it.

edit: actually the VAG logs put fuelling into lambda, not afr and lambda is lambda not matter what your fuel is so if I go to E20 then I should just adjust for fuelling for whatever lambda the 98 was roughly on I think.

[Sydneykid]

You do a fair few k's so "2 tunes" maybe worthwhile, most of the E20 guys just fill up every time with 5/10 litres of E85 and 15/30 litres of Pump 98. For track days etc they take a 20 litre drum of E85 to the track and then balance up to E20 with Pump 98 from the local servo. Particularly for Wakefield where the nearest E85 is Canberra. You have United/s close to your work so you could pretty much run E20 all the time.

Next time you are over at the workshop ask Justin for some Nulon E85 stabiliser, that stops the ethanol from separating from the petrol if you don't use it for a long time. I also race tape up the can lid to make sure no moisture sneaks in, ethanol is hygroscopic like brake fluid.

Cheers
Gary

[sambb]

I ran at SMSP Gardiner/GP circuit on Monday as part of a Renault day organised by Renotech. Ive never seen so many Renaults (or hatch backs) in one spot. There were 120 Clios, Meganes etc with an Alpine A110 and even one of those V6 rear engined Clios done by Williams F1.
It was my first experience of the proper SMSP layout (that the Supercars do), having only done South Circuit a couple of times. Apart from Bathurst and Wakefield where I have only ever fleetingly gotten into 4th gear it was quite the experience to be hitting 190 on the straight, having to actually make a 4th -5th gear change at full noise, turning in near 170kph at turn 1 etc. I'd be lying if I said I wasnt a little apprehensive going into the day. The chassis is all mine on this car now but the engine, gearbox/clutch condition are a complete unknown so there was that to consider and also my last hillclimb experience on these 225/45/16 medium tyres wasnt great RE the car trying to swap ends on me. It was a different sort of apprehension to hillclimbs. There because you are competing and know you'll be going hard without ever really being dialled into what the car/setup/tyres/track is doing I get run to the toilets type nervous sometimes. This was different because no one cares what you are doing on track and you know you have all the time in the world to get used to things and work up to a good speed, but yeah I knew turn 1 was going to be a sphincter clencher as its several steps above anything I'd done before.
Well as I had suspected after Bathurst, the handling problems I'd had there were due to running wide, hard to heat mediums on a setup I'd previously always used with narrower easy to heat softs. I had no such issues on the circuit and once I got my eye in I could only describe the handling as the best I've ever had. The only handling change I made for the track was to swap in my rose jointed rear droplinks and drop the rear another 5mm on the collars. I started the rear damping on the same softened off settings I'd left it at from Bathurst, as my main consideration was to have the rear safe for turn 1. From there though I found myself going 4 clicks harder throughout the day and was still safe at turn 1 in the last session so I could have gone up another 2 I'd say which would have put me squarely back where I normally would have run at Bathurst or south circuit. Given that, I think that if I wanted to just follow the 'keep it simple' principle in the future, I could more or less start on the same settings everywhere provided I'm using rear softs at hillclimbs.
Using Racechrono Pro using the iphones internal GPS I went for a high windscreen mount to make sure the satellites could see me and had reliable lap by lap timing. I didnt get any missed finishes giving 4 minute laps or anything like that and the only mishap where it missed the whole session was my fault I think. My best lap was a 1:55.76 on the last lap of the last session which I was pretty happy with considering that I couldnt top 190kph down the straight. That lap was also my theoretical best meaning I'd nailed all the sectors in that session too which is a double plus. So for the first time there, where I was still well and truly building up to a solid entry speed into turn 1, still had no idea what I was doing at turn 2 and the engine tune isn't finalised (had some funny boost surge moments here and there), I was pretty stoked with that.
My take aways were:

- the TT brakes despite still lacking much feel and still having too much pedal throw, were fade free and really hauled it down going into turn 2. Prior to this day I was close to going back to the stock smaller solid disc rears that are stock on the Gti (because they are lighter) but I think they may have struggled to be honest. I'd just flushed with Motul RPF600 and the pads are the DS3000's that came with the car. Once they are worn down I'll change for something from the Hawk catalogue (as they actually make something for the these TT brakes) and then I think I'll be satisfied with the brakes for a while.........until Simon guinea pigs a bigger master cylinder!

- the car needs the extra toe out that the old car had. It uses the whiteline toe correction shims that give no more camber than stock and 1mm total toe out. I could feel that the rear wasnt as lively as the last car that had 3mm toe out when negotiating the back of the circuit. 3mm total toe out on the rear might be too much now that I'll probably get back to SMSP much more in future (I'm all ears on that!) but it'll still need a re-shim anyway to get a another half a degree of neg camber into its rear even if I dont add any more toe out. It could have been the wider rears but it was just missing that rear steer that could have had me on the throttle earlier around the south circuit parts of the track.

- even though the tune isnt finished, its at the diminishing returns point. So there's not much more to be had from it and I can honestly say that a K04-001 is virtually identical to the last cars K03s that'd had wastegate porting and ran a fresher wastegate actuator. It may produce the bulk of its torque 500-800rpm shifted higher in the rev range than the tickled K03s but I dont think their peak power would vary at all. The K04-001 falls over just as early as the stock modded turbo did. So coming down the straight I was still changing at 5800rpm which in the context of that straight felt SLOW. Its only advantage is that the K04-001 will be comparable to the modded K03s with say 2psi less boost so probably runs minisculy lower iAT's. But I wouldnt be removing a K03s to put a K04-001 on expecting a justifiable jump in performance. I'd be K03 hybriding (littco L280) or K04-023'ing.

- tyres. I want to find the japanese guy who made A050's happen and hug him tightly. The grip is just insane. I'm still curious to see if a tyre width down on the rear (225->205) is possible without loosing out. At the level I drive at I suspect that I'm over tyre'd at 225 on the back and maybe 205/45/16 will cut it, and may even work in hillclimbs that way too, which would be good given that 16 inch softs are like hens teeth.

All up it was one of the best days on the track I've had. I had my own little wins and the guys that were there with me - Metalhead with his Vette and my Clio mate - had really good days too. We all got the fear of the unknown Turn 1 monkey off our backs and survived with everything intact which made for a top day.

[Sydneykid]

There will be some 205's available shortly, maybe after the next meeting.

The rear wing is it's most useful for SMSP Turn 1, without it we have to compromise on the set up for the rest of the circuit. Even then a bit gentler on the rear toe out than at Wakefield, ~4 mm total instead of ~6. Camber wise the reverse of Wakefield, 0.75 degree more neg on the RHS front and same camber (no stagger) on the rear. You get a bit more roll in Turn 1 due to the higher lat G, so the rear camber helps contain the tail a bit more. Less front toe out as well, only 2 mm.

For a FWD 1.42 it's ~220 kph in 6th, then ~180 kph down to 5th on turn in but accelerating to 190 kph at the apex, if you are accelerating it shouldn't want to swap ends, it's a long run to turn 2 and we almost get to the same top speed. Turn 5 is important as you carry the speed up the hill, a small sacrifice in turn 4 can be of benefit, so it doesn't ruin your entry into 5. Need to be on the power as early as possible out of the hairpin (turn 9) as it's a decent run out of there too. They are currently resurfacing turns 2, 3 & 5 which should help remove some of the bumps, that can be unsettling.

Cheers
Gary

[Sydneykid]

And, I don't like DS3000 (or DS2500 either) they always feel "vague" to me.

Cheers
Gary