My S6 work in progress with dyno figs and pics

[JP]

Whilst we had a nice sunny day yesterday I figured I'd have a quick play with some cams. So I chose the easiest one to do which was fit the naturally aspirated 7A intake cam. With the current best chipset.

After doing 6 very consistant runs I grabbed the average of the batch and overlaid it to the AAN intake cam.



Now unlike the dyno simulation which showed increase in top end power we had the exact opposite it was a drop as you can see.

So I proceeded to do some further testing I checked the 30-60mph check and 3.27secs was the best I got on the AAN cam whereas 3.22 was the best on the 7A.

My butt dyno was telling me it was smoother almost a little less aggressive. This, as can be seen is comfirmed on the dyno plot with an almost identical power run but a much smoother plot.

Now everything I have ever come across in real life testing has shown that an increase in intake air (i.e. cam duration, head flow etc...) = an increase in hp. Whereas an increase on the exhaust is an increase in low down torque. Balance the two correctly through headflow and camshaft configuration and you can get an engine to do anything you want with regards to power delivery/driveability. So why the strange real life results? It almost felt like the car was hitting a brick wall at 5K, much like what one of the original chipsets did in the real cold weather.

So I decided to do some back to back runs datalogging with the zeitronix unit as can be seen below.



Now this is where the interesting things starts to happen as you can see one set of graph lines overlaid onto the others shows that one took longer to complete the full rpm range than the other. This was the 7A intake cam the same one that has a smoother dyno pull and a better 30-60mph pull. I also managed to come to a stop sooner on the straight in relation to the testing from the other day of all the chipsets which indicated that it was faster despite the log.

So I started looking at the figures more carefully by exporting them to excel. This is where the interesting thing happens because looking at the info being collected of afr vs rpm vs egt relative to boost it appears that the 7A cam would have produced more top end power if it wasn't for the fact it seems to be running out of fuel? So is it possible that the hp limit of the K24 is not the limit of the K24 but the limit of the fueling system and we are hitting 99% duty cycle when fitting the 7A intake at the high end of the rpm scale? As with the right flow and cam profile you could make near enough 270hp naturally aspirated on paper so the normal boost settings would push it a lot higher than what you'd expect from a K24.

Definatelly food for thought and something I'd like to look into a little later in the thread by upping the fuel pressure a little allowing the ECU to part compensate and running the cam swap again. Before exploring the route of putting a properly ported head on.

[EspenW]

Any significant temperature difference between the two days of testing?
(probably not, as low and midrange is similar)

[S2central.net]

Interesting results... I don't think you can just whap in a different cam and expect it to produce results without some hard work. You might find that with an adjustable pulley that you could achieve better or worse results, but the real benefit cannot be known without ECU tweaks.

At WOT - the ECU delivers the amount of fuel stated in the maps - for the prevailing ambient conditions, battery voltage (and the color of wings on all butterflies within 19m range of the car - such is the extended bizarreness of Motronic). I therefor doubt that you'd be running out of fuel delivery on this setup.

My suspicion is that the mixture at WOT may not be rich enough - what is the AFR on both setups ?

Alternatively the ignition timing needs changed to maximise the potential benefit of the supposedly higher flowing cam.

Interestingly - the work I am doing on an AAN chipset has a similar profile with it flattening off too early - stock cams here - in comparison to ABY runs on identical engine. I have more work to do on the emulator to establish if its timing or mixture related. I don't have AFR data on this machine - I might have to do some judicious datalogging with VAGCOM on my reference ABY setup to see what is the difference. Only had sporadic access to car and equipment this week so a bit frustrated by it so far... Very hard to tell in the butt dyno - its only when I hooked up the GTech and saw things tailing off circa 5k rpm that I knew something was amiss.

More disruption of the neighbouring farm animals tonight or tomorrow night for me. What joy - trying to drive a laptop and a car all at the same time, keeping an eye to boost and GTech as well - in the dark with candles for lights... All in the name of science and vorsprung durch technik

[S2central.net]

A quick afterthought... WIth your whizzy Zt2 - can you log the MAF voltage output (pin 3 on the MAF) on a WOT run - then snap a screenshot as it runs the data with boost and rpm data - from that I can make a reasonable swag at the point at which the stock MAF maxes out at 5V - then a good estimate of the mass of air at that point in time would follow with some simple maths.

[JP]

MAF maxes out at 4750rpm like mentioned in an earlier post in this thread

[S2central.net]

OK - sorry I missed that

[S2central.net]

I see you had approx 17psi boost at that point where MAF maxxes out...

So thats ~1.2bar and assume 85% VE - we get to approx 200-220 grams per second of air (or approx 750kg per hr if you prefer) - which makes sense if you look at scales of K24 or RS2 compressor map... It makes sense for that application on factory boost levels. In chipped mode there is a way avoid fuel cuts when MAF voltage exceeds a certain level, but as posted in my other MAF threads - this is a problem as shortly after 4500rpm on full throttle the MAF is essentially adding no valuable info due to its limited dynamic range... and we rely purely on the rpm axis on timing and fuel maps at full load - far from ideal but not the end of the world.

I'm glad you shared this info as its given me more insight on some next steps - and it makes it easier to select an alternative MAF with suitable scale... My gut feel for the HFM5 1000kg is looking real good. If only I knew how to rescale the MAF calibration maps Ahhh one day I'll get there.

[JP]

Interesting results... I don't think you can just whap in a different cam and expect it to produce results without some hard work. You might find that with an adjustable pulley that you could achieve better or worse results, but the real benefit cannot be known without ECU tweaks.

Yes I would agree that a cam comes into it's full potential with a map to suit however unless fueling is on it's limit I would expect to see a visible change based on an ECU's ability to compensate within given variables. We see it time and time again when fitting an aftermarket cam to a standard motor and seeing hp and tq increases. Heck aftermarket manufacturers advertise power gain claims for mild cam profiles. Not until the more radical profiles are sold are they advising the need for custom fueling.

Likewise the same was shown at the beginning of this thread when I did a swap on the 7A exhaust and a visible hp difference was seen along with spool up. Obviously once I've tested the final two chipsets I'd like to redo that cam swap and see if gains are made again or we are going in reverse like what we have just seen in the 7A intake swap. As this will then confirm whether we are on the fueling limit or not.

If power is lost at the top end with the exh. swap then I would like to test again using more fuel pressure. If gains are made then it is clear that without a vernier the 7A int. is no good for high rpm power despite the simulation.

[S2central.net]

My view is that if a new cam provides more airflow then the ECU will of course compensate for that by way of higher readings from the MAF - but only if the MAF has a dynamic range such that you can measure the difference.... Lets say for sake of argument the 7A inlet cam allows the MAF to flow 250g/s of air at 5000rpm - the MAF is already maxed out at 200-220 and so the ECU has now way to tell that more air is being ingested.

I don't think you are on the fuelling *limit* of this cam - I just think the ECU cannot automatically cater for it cos of the MAF issue at high revs - so you need to manually tweak the timing and fuelling in the upper rev ranges at max load to find where improvements can be made. Maybe you are running lean - a variable FPR would be a quick and easy way to cater for this without getting into binary code. Only then would you know if the ECU can adapt in cruise conditions for mixture control. That might work automatically of course - or need a tweak in the fuelling maps.

I hope to have an AAN chipset ready for you soon...

[JP]

Cool I should have another set for testing arriving sometime next week aswell.

I am having some concerns as how to carry out the 1/4mile tests though with regards as to what rpm to launch at as I just did the first hard launch in the car from 5k rpm and there was a distinct wiff of clutch which I have not had before under normal loading conditions of the Gtech runs and roadgoing tests. Nothing major but it was there. But we'll get to that as I get to Gti Festival at the end of the month.

[JP]

OK due to unforeseen circumstances I wont be completing the test in it's fullest extend as I currently have a car that has a horrible burning smell coming of the clutch when she is not coughing and splurting from a failed headgasket. One of the chipsets I tried produced a tad more boost than the rest, although desireable to drive it has taken it's toll on the engine and clutch.

If there is enough interest I will write up a report in the next couple of weeks when I got a moment of the comparisons that I have done so far. As the fuel economy has now gone down hill rapidly the last bunch of chipsets can't be commented on in the fullest. Likewise what where smooth power producing chips before now are suffering from flatspots. Due to the occasional misfire.

I may take a gamble and do one dyno run just to see how the gtech results compare to that on a dyno as it will then give comparable information to verify the results on the other gtech runs.

At the moment the headgasket failure is not mixing water and oil but I am loosing oil from the side of the head at number 1 and burning water at a very slow rate. Also if I feed in the clutch on part throttle it won't slip. So I'll be easing of and running a standard chipset until I have the time to get a new clutch, a set of rods and a properly ported head ready to put on the engine. Actually will also need to do the brakes as they are a tad on the warped side now

After which I'm going to go a different route and see how much can be had out of the little K24 until I'm ready to invest in something a bit larger if I decide it's not enough.

[merlin's S4]

Sorry to hear about your troubles! Do you know why your headgasket has failed? I thought you'd had it all done a few months ago!

[JP]

Headgasket was done at the same time as the head obviously back in May last year. As the option was availoable and I was intending to change heads at a later date I opted for ARP headstuds.

These are torqued to spec and are then advised to be checked and retorqued where necessary 'x' amount of miles down the line to accomodate any settling through thermal expansion and contraction of the head as they are not stretch bolts but fixed studs. Through my own personal laziness these we're never checked until today.

I noticed slight smoke on startup around 2 weeks ago after I had done some testing getting progressivelly worse until I finally put an hour aside today to check. They where all down on the required tq spec so even now tighten up it is too late as the gasket has blown and is not sealing properly. I can say for definite that the oil is still leaking, It hasn't started smoking on startup yet and I haven't put enough miles on it since topping up the water to see if it is still loosing some. I may find that all I need to do is slide a new gasket in to resolve the oil leak as the car did feel smoother again after the re torque this morning.

as for the troubles I'm not really too worried as I always knew the clutch was a temporary measure until I decided what the final aim of the engine was going to be. Intending to change it when I was ready to put the rods in and the new head on. It just means I can't fullfill what I set out to do on the testing front. Still lesson learned for being lazy when it comes to maintaining your own cars. Come to think of it the wifes MK4's service light has been beeping for the last 10K

[JP]

A little slack on the updates due to work commitments but in the interim had a proper chance to actually work everything out that has happened to the motor and what needs doing to it.

It looks like my personal laziness was not necessarily the cause but unfortunately can only assume this from the following findings.

First of I put the smoke on start up in the morning down to the burning of coolant and as temps where slightly up from 90 degrees assuming headgasket failure causing the burning of coolant (due to tinted windows I couldn't tell the colour).

After having re-torqued the head to the spec on the ARP literature the smoking on start up stopped. Only a slight oil leak near the front where the high pressure oil seal on the headgasket has obviously let go. However no real oil consumption (approx 1/2 ltr every 1500 miles or so down the side of the block).

I did however still have a coolant loss and raised temperature so I carried on investigating and discovered a small hole in the radiator core which appears to have been made by a small stone of sort. (Possibly something hit the radiator when at 6000rpm in 6th?)

So I figured as a quick fix I'll use some radiator stop leak to get me out of trouble. However every time I came to a stop it would squirt out the hole again at high pressure unless I manually shorted across the afterun pump thermo switch. So I've replaced this and the afterun system is now working.

So I am now wondering whether the sequence of events was radiator leak, failed after run system and the engine cooking itself on shutdown? As I've just had to re torque the head again and it was 10ft/lbs of target yet no misfiring hesitation or smoking on start up.

So I've bought a new radiator from Audi, got a Sach's clutch on the way, awaiting a set of uprated con rods to arrive, new big end shells and rings, also awaiting completion of a fully ported cylinder head to the same spec as G60Rally's one. A second 7A exh cam so I can run a dual setup, a modified MAF housing to bring the MAF voltage below 5v and just incase I've got a set of injectors earmarked aswell if we run out of fuel. After which we can see what the poor little K24 is capable of. It's already over spooling at 1.4bar big time so probably not a lot more but hey when the turbo goes it's a good excuse for an upgrade .

In the interim some time in about a fortnight I'll take the car up to the rolling road and get one last power reading on the best chipset just to see how it compares to the gtech results. After which I'll write up the report. As once the new head goes on it will definitely have to be a custom map.