Gee Emm

I have broken at least three throttle shafts, and now use Junk2 TB to get around that problem. That throttle stop would ring alarm bells for me, sounds like the adjustment at the pedal is not synchronised with the TB stop, and the cable is still putting pressure on the shaft at wide open. In the heat of battle it is easy to push harder on the pedal, even though you know that isn't going to make any difference (assuming you are even conscious of doing it).

[Note, AFAIK there are at least three theories on why they break - mechanical stress, aerodynamic flutter, and vibration, you might be lining up for 'mechanical'. There are a number of fixes said to stop it happening, several of which I have tried, none of which give me enough confidence to not use the Skunk2 TB. It is the easy button, even if it does require work to overcome its design/manufacturing flaws.]

Tchaps

Gee Emm - I think your comment was to Poormxdad anyway, but in my case I know I am not on the WOT stops with the pedal flat to the floor. The S90 throttle body I am running appears well made with the screws for the throttle plate ?plannished? after tightening. Maybe it isn't plannishing, we used to do it at work but I can't remember what it was called now. Rounding off the back of a fixing to form a permanent joint to avoid the fixing coming out.

Going through the logs I've noticed some more information.

The post SC air temp sensor isn't as useful as I thought it was going to be. I thought this was going to be a quick responding sensor that would show me a good comparison of WI on and off temps when giving it the beans. It doesn't appear to do that at all. It seems to measure the temperature of the chargecooler body more than anything. Temperatures very gradually climbing about 10degC from the start of a run to the end of a run. Might try a different sensor type next time the sc is off, going to a more traditonal sensor. Try and get a small 1/8npt high temp one.

The chargecooler radiator in and out temp sensors both show a trend that they are hotter while the car is sat still idling (+10degC ambient), and that they cool off on the move, settling to around +5degC ambient by the end of the run. I need to check I have them the right way round though because on the run the rad out temp is higher than the rad in temp by a few degrees. I also think these sensors would show more info on a longer track with more time on the throttle. Anyway, I will double check they are the right way round and the bias resistor values again and log them on the next track day.

One odd thing I have noticed is that on the two gear pull down the last straight before the finish I am getting what appears to be a lean spike just after the change into 4th from 3rd, for about 0.3secs. It isn't present on the log I took as a baseline without WI on, but it is in a couple of my logs with WI on (not all of them). I seems to coincide with the WI flow sensor ramping up. It is a bit odd for a number of reasons. I didn't feel a hesitation or jerkiness just after the change, and watching the few videos I have of the gear change I don't see or hear anything odd either. The injector pulse width is as it should be in the logs, so I don't think it is an issue in the way the ECU briefly swaps maps and back again during a gear change.

I am going to check I don't have any exhaust leaks, but I certainly didn't hear any on Sunday, and these logs are all from the Saturday. Not sure whether the water reactivating quickly on a gear change is causing the AFR sensor to false read for a moment until the water fully evaporates from it. I can isolate any vacuum effect of the gear change by doing some logs using flatshift and seeing if the issue shows up or changes.

Anyway, I will continue to log my drives and pay particular attention to hesitation or jerkiness during gear changes.

poormxdad

Tchaps, I apologize for thread stealing. That was not my intent.

That said, I have a burning need to say "There's an adjustment at the pedal???"

Thanks,

Tchaps

No no don't worry no issue caused - was just clarifying who the question was going to.

I think the mk1 pedals self adjust when the pedal assy frame fatigues and snaps in half, an early form of traction control. How inventive!

Tchaps

So I have been working on the WI map, namely leaning out the fuelling to between 12-12.2 and advancing timing.

I am making gains but I am finding it a little difficult to do back to back comparisons between each run with Virtual Dyno.

More often than not I get a dyno result that just doesn't match the power curve I know the engine has, i.e very flat torque curve and pretty much straight line power increase up to redline. I will get an odd hump in a weird place etc. But there are runs that don't and it's those I try to use as a comparison.

So here is where we are so far:

Red is my current WI tune, 12-12.2AFR with an increase in spark advance.
Green is new exhaust
Blue is no WI old exhaust.

These are my non WI and WI spark maps currently. Bare in mind that WI is only active for the area in the red box on the WI map, and also uses different rpm and MAP values, so pay attention when comparing!
Currently WI fuel ratios are about 30% 4-5.5krpm leaning to 20% above that.

My first test adjusting spark was a 2deg increase from 4000rpm to redline. When I looked at the dyno traces this increased peak torque at around 5500-6000rpm to almost 250lbft, but then with far less of an increase at high rpm. So my next test was the current WI spark map, that kept the 2deg advance from 4000-6000rpm but then wound it up 6500-7500rpm up to 4-5deg advance by 7500rpm. Of course in these dyno traces the 250lbft hump at peak torque was nowhere to be seen (see comments above about dyno difficulties).

The current red trace looks 'right' though. No big humps anywhere, very little smoothing at all used, like my green trace. It shows a broad chunk of extra torque and power from 6000rpm up over the green trace, even if peak hp figure is very similar.

No sign of any knock anywhere and I have noticed that the new WI map with the extra advance has basically halved the EGT gradient during a pull so I feel like we are moving in the right direction.

I think my next steps will be to give the 4000-6000rpm area another couple of degrees and do a few pulls to see how the torque responds. If it makes a difference I will then try giving the whole WI rev range (4-7.5krpm) another couple of degrees, and again see how the torque curve responds. If I don't see any increases, particularly if I don't see any in the top end I will wind the advance down a little in those areas. No point in over advancing if it's not showing a power increase, its just increased stress on the engine if in theses areas I am already in the MBT area.

I'm still a little unsure on how the addition of water will effect the location of MBT - if we ignore any knock limitations for a moment. I.e how is the burn rate and flame propagation front speed affected by the addition of the water in the combustion chamber. With WI you can advance timing towards MBT from a non WI knock limited spark advance because the water quenches the hotspots and knock that would otherwise occur. However what I am not sure of is if having 20% water to fuel in the chamber advances the location of that MBT in the cycle in comparison to fuel and air only for the same rpm and manifold air pressure.

Just have to slowly keep testing but when I get to a point that I don't see any significant torque increase over a few dyno pulls with a small increase in advance I won't push it anymore. I don't want to have it tuned on the edge for the sake of a couple of hp and loose my track work safety net.

Tchaps

Here's an interesting read linked:

https://core.ac.uk/download/pdf/151511133.pdf

It deals with water injection from the perspective of increasing the performance of a modern 4 valve small capacity forced injection engine.

It's a big read but I think one of the most interesting parts I have read so far is the comparison of burn speeds between a performance gasoline, standard gasoline with WI and a high octane race fuel, see page 179, figure 73. It's got copyright stuff all over it so I don't think I can extract the graph and share it here but I have shared the source I got it from.

It plots the bulk burn time (amount of time in crank angle taken for 10-90% of the cylinder contents to be combusted). Basically the water injected mixture takes between 10 and 20 crank deg longer, dependant on rpm to burn the 10-90% mass fraction than both the high performance gasoline and the race fuel, which both have very similar bulk burn times.

What this means is that the spark advance with water injection needs to be considerably increased in comparison to the two other fuel options to achieve MBT, which is considered to be 50% mass fraction burned by approximately 10deg ATDC. The rough thought in my head is that you would halve the difference in burn speed angle and add that to the spark advance for a non WI, non knock limited spark timing to achieve MBT, so 5 to 10 crank degrees more advance. In more knock limited areas of a non WI map, i.e peak torque, you would add more spark advance than that because the timing is already retarded from MBT to avoid knock.

So I feel my current spark advance of between 2-5 crank degrees is likely a fair way off optimised yet, although I will still proceed with caution.

My other take away point from it is that the benefit of added methanol content in a mix is possibly showing itself here. The longer burn duration of a 100% water mix, driving a larger spark advance will place a longer period of the compression stroke under a higher cylinder pressure, meaning there is a larger pressure acting for longer on the piston trying to force it back down the bore against the rotation of the crank, than if spark duration is retarded and the large increase in cylinder pressure from the start of combustion is delayed until the piston is nearer the top of the cylinder.

A faster mixture burn speed is better for power production - proven by multiple papers. This is why the AFR of 12.5 for gasoline is seen as the rich best power mixture. It is the AFR that produces the fastest burn speed for gasoline. I think this has a lot to do with being able to reduce spark advance, avoiding so much of the unhelpful cylinder pressure before TDC, while still being able to achieve the 50% mass fraction burn by 10 ATDC, i.e MBT.

It strikes me that a water methanol mix with gasoline may well behave more like a race fuel in terms of burn duration, while maintaining the anti detonation characteristics of a 100% water mix. This would mean less extreme spark advance can be used in comparison to 100% water, minimising the compression stroke pumping losses while still maintaining MBT for optimum power production.

Need to do a bit more research on burn speeds of various methanol water mixes but certainly an avenue I will be exploring.

Spaceman Spiff

Great find. Dude did a pretty good job with his thesis.

Water/meth is what I'm going with on my setup, but it'd be interesting to compare combustion rates for cheaper 30/70 (windshield wiper fluid) vs a full 50/50 mix.

Tchaps

Yes, what a great project! I'd have given my right arm to do that experiment now in a laboratory. At the time I did my degree I'd probably been to busy skateboarding and drinking though

One thing is a bit odd about the experiment though and that is that the water to fuel ratios are a bit weird, both in terms that the water content to me seems excessive, to the point I am surprised the engine didn't hydrolock. The author is talking about water to fuel ratios in the 3:1 region, and I haven't got that backwards.

If I find anything good about different mixtures vs burn speeds I'll post it up.

I pulled a tiny bit of fuel out of the midrange and gave under 6krpm another couple of degrees advance.

Red trace is yesterday Blue trace is today

Yesterday we had a bit more boost - this is weather dependant and its all over the place at the moment. When I was getting up into the 15psi area it was significantly colder and we were in a high pressure weather zone. Its a bit dreary, muggier and more humid at the moment so I think that explains the difference.

There is a little gain in torque under 6krpm with the new tune, about 5lbft. I just kind of think if I was a long way off MBT that I would see bigger torque increases than that.

Compounded with that is that in the 6 to 6750 rpm range I had made no significant changes (very slight lean out up to 6500) and there is a large difference between the two traces, probably caused by a bump. This difference is far larger than any change that I see caused by an actual fuel or spark change so it makes it very difficult to assess tune changes from.

The good thing is my headphone jack knock listening port works an absolute treat. Really is very clear. I had a little play and induced a little knock at low load and you could hear it easily so I do feel confident in that.

I don't want to seem like I'm knocking virtual dyno, I know its the quality of my runs not the software. Its just very hard to get a perfect run on the roads around me.

I think I will go out and try to get some more decent runs on the current tune, we are at about 4-5deg advanced across the board now. Then once I have built up a good average score if you like I will add maybe 3 degrees everywhere and go out with the headphones on, and get some more averages, then knock it back a couple of degrees to make sure I have a safety margin in the tune. Just going back to making sure I have it in a state of tune that is not going to cause expensive issues on track. If I'm not making significant gains then I don't really see the point in running outlandish spark advance for the sake of a couple of lbft here and there.

Maybe an answer is to just go and book a morning on a dyno somewhere and eliminate a few variables to really see what it likes in terms of fuel and spark.

That all sounds a bit depressing but I have to keep insight that I am consistently getting virtual dyno runs in the high 270s to low 280 whp area - a good 25whp up on where I was - and the car feels like it makes that power for the whole track session with all temperatures visible to me under control. That is from a 53ci supercharger that the is only meant to make 280crank hp from the manufacturer info I have - not too shabby I think!

I will certainly have a play with methanol content at some point though, I think that would be very interesting. Also end of the season this year the engine is coming apart for maintenance, re ring etc and I am looking to revisit the head, specifically improving the short side radius, maybe going +1 intake valves and maybe swapping out the cams for something with a bit more duration to help top end breathing.

Tchaps

So having had a conversation with a couple of people on a water injection group the plan is to reduce the water to fuel ratio down to 10-15%, from the current 20 - 30% to see if it makes a difference.

The thought is that I am over-quenching combustion with the reasonably high water ratio I am currently running, and this is likely to be causing the unresponsiveness to spark advance changes.

To bring the required water quantity range into the valve response range I will need to reduce the size of the restrictor before the valve.

With my current 0.6mm restrictor I have good linearity between 300cc and 400cc min water flow - this represents the flow range between 20% and 80% duty where the valve is responsive.

I need to move the 20% - 80% valve duty to provide flows between 150 and 300ccmin to give me about 15% water to fuel over the 4000-7500rpm range. I will try a 0.4mm restrictor and I can always drill it out to 0.5mm if required.

I will go back to having a map that only pulls fuel, then build up the spark advance as before to see if I can see a better torque response than with the higher flow rates.

Tchaps

So I tried to make a 0.4mm restrictor, but it ended up as a 0.5mm because I broke my 0.4mm drill bit and had to drill it out with the 0.5mm bit.

I tried out the 0.5mm restrictor anyway and it certainly moved the responsive flow range in the right direction, down to 250-330ccmin from 300-400ccmin.

I went for a drive and the car certainly felt like it pulled more eagerly, despite turning off the spark advance under WI, just leaning out the fuel to 12.0-12.2AFR.

I got a log and virtual dyno and figures were basically the same as more water and more advance.

I've ordered some more 0.4mm drill bits because I think it will be the optimum orifice size for my target 10-15% water fuel.

Should be able to have that up and running for testing some point this week, hopefully we can make some ground with the reduced water ratio.

BarbyCar

iTrader: (1)

I'm late to the party and sorry if you already covered this earlier. 0.4mm is a standard 3D printer nozzle and available cheaply in brass or stainless. Would it not be better to do that and make it an easily replaceable consumable item?

Tchaps

Oooo ok thankyou for that. I will investigate.

I don't mind doing my short term experimentation stage with my home made JBweld restrictors, they seem to do the job, atleast with water only but when I finalise the setup I would rather do something a little less... bodgy

Tchaps

So water injection tuning was slightly delayed by my son arriving a little earlier than expected, but stealth commissioning has resumed under the guise of taking my daughter to the play park in a nearby village. I will see how long I can continue this without being rumbled.

I installed the 0.4mm restrictor with not entirely positive results. It has lowered the flow rates for a given valve duty, but also narrowed width of flow response, i.e with the 0.5mm res I could target flows from about 250ccmin to 330ccmin across the range of 20-85% valve duty. The 0.4mm res narrows this to 190ccmin to 230ccmin. This gives me 15-12.5% water to fuel, just that I am only able to supply up to about 12.5% by redline. I sort of wanted to run right on 15% at high rpm because that is where the sc is getting hottest, just with the ability to not be running 30% water in the midrange. It seems quite fiddly targeting low water to fuel ratios, even with the pwm valve on a setup that in the grand scheme of things is only making around 300hp. Richard at Aquamist said that these systems are really targeted at the 500hp+ user. But we shall battle on.

The system felt laggy, with a noticeable time gap between getting on the throttle and the map switching over. It is about 1sec with the 0.4mm res, with the 0.5res it is 0.5sec.

This is particularly pronounced if you go from a low rpm, say 3k. The flow sensor isn't up to speed until 5krpm with the 0.4res, whereas with the 0.5 res the flow sensor is spun up by 3.8krpm. Both of these runs started mid 3krpms, so a fair comparison. The valve doesn't seem to invoke a response on the flow sensor until about 50% duty with the 0.4res.

Power wise I think I need to give the 0.4mm restrictor another shot, as the dyno trace I have showed it down about 10-15hp over my 20-30% +4deg advance and 0.5mm restrictor (25-17.5%) runs. I can't remember which bit of road this logged pull is on though and the route was quite hilly, so probably not a fair test so far.

The 0.5mm dyno run I did about a week back now did show promise, picking up a nice chunk of torque through out the rev band over the 20-30% water ratio, despite having 4-5deg less timing in it. However it also had God's boost on it's side (it was freezing!) which meant almost a pound more boost, and when I used an air temperature correction factor the traces where almost identical.

My gut feeling at the moment though is I think the 0.4res is really pushing it in terms of the minimum amount of water the system can deliver, just from the fact that it doesn't seem to do anything until about 50% valve duty and it reacts far slower. The 0.5 and 0.6 restrictors do not show the lack of sensitivity below 50% and have the same fast response time. I think I need to maybe open out the 0.4 to 0.45mm and thin the depth of the orifice to see if this improves response in terms of range and speed. If that works then I will start to add some spark advance and see if that shows any increases in torque anywhere.

Tchaps

Had a play with the 0.4mm restrictor, and also a changed the PWM valve mapping slightly and I am now in between 15% to 20% water to fuel from 5000 rpm onward.

I was trying to target this ratio from 4000rpm but it looks like the PWM duty cycle needed (10%) doesn't really get the water or flow sensor moving.

Dyno results look positive so far. Blue trace below is the current adjusted 0.4mm restrictor (more like 0.475 now) and new PWM valve map, still only pulling fuel on map switch, no spark advance change. Its not a full pull unfortunately. I don't have quite the same time freedom at the moment and the roads were a little busier than ideal, but I'll get one soon.

The red trace is the old 0.6mm restrictor running about 20-30% water to fuel pulling fuel and about +4degrees of advance in it over the base timing.

Green trace is WI off, new exhaust and higher rev limit. It was on the 'winter' tyres - slightly smaller diameter.



Both pulls done in decent warm dry weather, if anything today's blue run was hotter, so I don't really have any idea on the 0.5psi boost difference. Both runs done on the same bit of flat road, blue one slightly further down it and in 4th gear, but it is still nice and flat there so as fair of a comparison as I can achieve road tuning.

You can see a fair chunk of extra torque through 5500rpm to the where I let off at 6700 odd - despite the lack of additional spark advance, just the reduced WI ratio.

Particularly over the green non WI trace it is a convincing gain, really only being achieved with a bit of water and a mild lean out of fuel. Green trace is likely to have been done in colder weather too, it was a month or so ago now before spring really got going.

This looks very promising, the engine feels fresh and pulls well. It's a decent chunk of torque to win over the old water to fuel ratio and with reduced timing. If it now responds to spark advance better then gains could increase further.

As it stands if the blue torque continued in its normal pattern up to redline then we could be looking at 290+whp up at 7400rpm. I don't think that's an unreasonable leap based on the data around me, and it's good to see torque increases through the whole upper rev band where I drive it on track.

Hopefully I am still spraying enough to retain the excellent engine temperature control I saw on my trackday, but I can always turn it up on the day if required.

Next steps will be to use the same valve duty at 4000rpm as 5000rpm, just to get the system up and running earlier during a pull from lower rpm, even if it is at the expense of a slightly higher water ratio than I want very briefly at those lower rpms. Then I will put 2 or 3 degrees of spark advance into it and see if I see a response, either on the det cans or the dyno traces.

In non WI related news I have ordered up a spare bonnet that I am going to put some louvres in for hot weather trackdays. I have noticed particularly as the weather has heated up that the new headers really do put a lot more heat into the bay than the old more compact setup, which was also wrapped. I don't want to wrap these headers (it will make fitment even tighter than it already is) so I plan to louvre right over the top of them to evacuate the heat.

Playing around with the dyno trace of that really cold day opened my eyes quite a bit that a few degrees of air temp into the back of the charger really can make a decent difference, and my intake is hot to the touch after a drive at the moment, where it wasn't before the headers.

Also as a slow burner I found another cheap BP4W head casting to mess about with. Plan is to build a diy flow bench over the summer and start messing about with some ideas, test and validate them then incorporate what I find works into the rebuild of my cylinder head either this winter or next spring. I'm not going to just hog the ports out to absolute caverns and loose all the low lift flow, I plan on keeping cross sections pretty close to stock, just paying attention to the bowl, short side radius and valve shrouding then seeing what a +1 intake valve and a 3 angle seat does. Nothing too mad just hopefully seeing what the numbers say and going further than I did on the current cylinder head.

I also intend to go with bigger cams, again nothing too wild, but something akin to a 'fast road' cam. Something with a little more overlap, i.e some overlap, to let me purge the last bit of the chamber better now I (assume, still need to check actually!) have low exhaust backpressure, and also something with a mild bump in duration and or lift on the intake to aid high rpm cylinder fill. I have the shim under bucket kit with the titanium retainers, so I am looking to exploit the lower valve train mass with a cam that has high valve acceleration and a large flow area without having to go too far on duration.

My current cams are really quite small. Quite a lot smaller duration than the stock BP5A / BP4W cams, just with an extra 1mm lift over them. I think high rpm breathing can be significantly improved, bringing down boost pressure and therefore both AITs and supercharger drive power, while increasing charge density and cylinder fill. Always got to have a project, I can never leave it alone :-)

Tchaps

I added a bit of water low down in the rev range to get the system active quicker and also added in 3deg advance across the board.

The system comes online a lot quicker now which is good, but the torque / hp traces are pretty much identical, the non advance run actually has it slightly from 6krpm up but its by about 2lbft. The sort of difference I would find run to run anyway.

Peak torque is coming out as 222lbft at about 5500rpm, peak whp occurring right at the top of the trace, 270whp@7200rpm. Oddly I am being short changed here a little by virtual dyno because torque at 7200rpm is 209lbft - which should equal 286whp. Not sure why this is happening as I am using a smoothing factor of 1 so there should be very little correction taking place. Also I am actually revving out to 7400rpm and my runs are consistently clipped a bit short of that rpm, but I assume its something in the way the software calculates the rpm gradient and so power / torque that means it does this. Anyway, the main point is that I haven't seen a difference with the spark advance ramped up a bit.

It's not letting me attached the dyno sheet at the moment but as soon as it does I will post it up.

I feel like I want to bookend my water injection findings on my setup so far, because there's been a lot of tuning, learning, theories etc spoken about in previous posts, and I want to condense that down into a concise this is my setup, this is how it is tuned and this is the effect it had with a back to back same day WI and non WI dyno run, because I think I have reached the development limit currently on my setup with 100% water.

I think it would be better done when I can post pictures again so I will do it in the next few days.

Tchaps

This is my same day back to back non WI vs WI dyno pull. Blue is with WI on, red is WI off.




100% water sprayed at a 17.5-20% water to fuel ratio. Spark is advanced by 3degrees and fuel is leaned out to around 12AFR.

Manifold pressure is increased by about 0.5psi.

Manifold air temps climbed from 27degC to 38degC without WI, delta 11degC. With WI temps were 33degC to 42degC, delta 9degC.

Slightly less temp increase under WI but nothing particularly mind blowing. Equally my pre sc nozzle, the only nozzle before the AIT sensor, is small so I don't drown the chargecooler,

What is clear though is a good increase in torque with the WI spraying, between 10 and 15lbft, about a 5% increase. This might sound small for all the time and investment, but keep in mind I think one thing that is I think is becoming clear is that my setup wasn't massively knock limited in the first place. The WI has increased the torque but also given me a large window of knock protection for track work when MATs will be higher, aswell as keeping engine temperatures right under control.

I have had difficulties spraying little enough water reliably; 17.5%-20% seems to be the limit for my setup and amount of nozzles. When I did back the midrange out from 30% I saw a gain in torque in those areas. It is quite possible that if I could reduce the water % further that torque would increase again. I'm interested to try a 70% water 30% methanol, which should effectively drop the water down to between 10 and 15% fuel, but keep the volume of liquid being sprayed high enough to keep my system in its operating window. ​​​​​​​

Tchaps

I have been playing with the cam timing a bit. We had a play last year on the dyno but we didn't have a lot of time so I wanted to have another go.

Up until the last few days I have been running both cams set with max lift at 114 A/BTDC intake / exhaust. The starting recommendation for both intake and exhaust cams was max lift at 110deg.

So I had my intake set 4deg retarded and my exhaust 4deg advanced. This was to reduce overlap and reduce dynamic compression ratio.

As I don't appear to have any issue with over compression and knock I wanted to try advancing the intake cam to increase dynamic compression ratio to see what the effect was. I advanced so max lift was at 106 ATDC, so a decent swing. I also retarded the exhaust to 110BTDC, so as to try and keep the expansion ratio relatively similar to before.

The changes caused where quite dramatic, three main effects.

> Fuel ratios at WOT were leaned - in the order of 0.5AFR. I had to add fuel into both the WI and non WI maps in these areas.

> Idle quality noticeably dropped, with idle occurring at a higher MAP. I had to add more air with the idle valve. Idle was still very choppy, I think due to the large increase in overlap. It sounded ace! Proper cammy.

> Unfortunately, despite the leaning of AFRs at WOT, suggesting an increase in air flow, power was reduced reasonably significantly, to around non WI levels.

It did show however that my cams are not so mild that I can't induce a change in engine behaviour.

I surmised that by advancing the intake cam you effectively make the cylinder capacity larger, so an increase in air flow is not unexpected, but that also the increased overlap may be being unhelpful in power generation with my setup.

So I backed the intake advance off by 2deg to max lift, 108ATDC and I retarded the exhaust cam back to 114BTDC, reducing the overlap by 6deg, still more than before I started playing.

> The requirement for more fuel remained, so air flow is still increased.

> Power jumped back to normal WI levels before the cam experimentation.

> Idle still a bit cammy, but less pronounced.

See traces below:


Green trace is cams set at 114 / 114 - pre experiment, low overlap.
Orange is intake advanced to 106ATDC, exhaust retarded 110BTDC, high overlap - note the drop in power and also the leaned AFR.
Blue is intake at 108ATDC, exhaust back at 114BTDC, medium overlap, fuelling has been corrected in the VE tables.

All runs are with the WI on, same bit of road. Green run was a cold day (extra boost) so I'm interested if on a cold day with the intake advanced I see a gain over previous results.

So from this I am fairly confident that increasing overlap does not increase power on my setup (glad I did this before just buying new cams!) and that there is the possibility that if I reduce the overlap further while maintaining the extra airflow the advanced intake cam seems to provide I may make some more power.

Tchaps

So I advanced both intake and exhaust together 4degrees, to see if the advance with the same overlap gave any power difference.

Green is before I started messing with the cams
Blue is the less advanced 108 / 114 timing
Red is the latest 104/118 timing


Latest most advanced timing hasn't provided any increases in power, in fact a drop in top end. EGTs are also slightly higher with this cam setting by about 20degC across the board.

So I think I have proven increasing the advance further won't realise any power gains, so next I am going to roll back intake to say 108 leaving the exhaust cam at 118 to see if a reduction in overlap drives any gains.

If it does I will probably put the intake at 110 and the exhaust at 116 to lower the EGT slightly and see what that does.

Tchaps

Well I thought I was making reasonable progress on the cam timing, in my mind having isolated overlap as a bad thing and that additional intake advance beyond 106 not showing any midrange gains and an apparent drop in top end.

I took another look through my high overlap log this morning though and there is a 3rd gear pull I hadn't looked at previously that is on flat ground, just in 3rd gear, that does not show the performance drop I saw in my 4th gear pull. There is also another 3rd gear pull that I had previously discounted because it had some funny business going on in the top end that shows the same midrange (no performance drop) where the trace is what you would expect to see. So I don't think overlap is out of the game yet as it turns out, although this new higher overlap trace is still not quite as good as either the original 114 / 114 low overlap or the 104/118 medium overlap I tried yesterday, just less of a gap than I previously found.


Green is 114/114 Red is high overlap 108/110 Orange is high advance med overlap 104/118


So I think the current status on cam timing is:

> Advancing intake cam has leaned out fuel mixture, suggesting an increase in airflow

> My advanced intake and exhaust (104/118) medium overlap is as good as 114/114 through 5k to 6k rpm, but between 6k to 6.7k shows a decent hike in power, before all traces seem to catch up at 7k, and the 114/114 just takes it right at the top end. Need more runs on this though because I'm not 100% convinced, differences could easily be run to run variation. Needs more testing.

> The 108/110 medium intake advance high overlap doesn't really take it anywhere, slightly lags through peak torque but then is very close to 104/118 and above 114/114 through 6 to 6.7k. Possible the overlap coming into play in about the area my manifold was attempted to be tuned for? Maybe.

Realworld quickest performance currently would be 104/118 - it is on average ahead of the other traces from 5krpm upwards.

Need to get some more pulls and see what is what.

Tchaps

Had a bit of a break from trying to work out best cam timing and went back to chopping bits of metal and bashing things with hammers.

I think previously I mentioned making a vented bonnet. I like the look of louvres, I think they suit my car better than some more modern looking vents. I'd tried making a louvred bonnet a couple of years ago by sticking on a louvred vent panel but it didn't look very good and I couldn't really be bothered with the amount of work it would take to get it looking alright. Getting somewhere to press the louvres in also seems a bit tricky, because of all the support bracing on the underside of the bonnet, and I don't want to cut it all away and end up with a really weak panel that will flap about.

Anyway last year one of my friends put some louvres in his bonnet and I thought they looked great. He cut slits in the actual skin and then just prys them up with a screwdriver. Sounds barbaric but I swear he does them so well you wouldn't know. They almost look pressed in.

Thought I would give it a go myself. I had a go first on the old bonnet I'd written off a couple of years back and couldn't really get good results just prying the slits open to make the forms, so I ended up making a little forming tool. What you do it drill a two small holes just over the width of the tool apart, and cut between them with a thin slitting disk. You then pry up the material a bit just so you can get the tapered edge of the tool in the slit, and hammer the tool into the slit so the panel rides up the taper forming the louvre. So you form the louvre all from the top side of the panel.

I marked out where I wanted the louvres on a new bonnet I got and started drilling, cutting and hammering. I think they have come out pretty well. Need some filing and a couple of corrections to the forms here and there but generally quite pleased. Protecting the paint with masking tape resulted in minimal paint damage which is great. It will still need a some finishing work but I don't think I will need to spray the whole bonnet.

I've done the louvres just behind the radiator, in a nice low pressure area to optimise radiator efficiency and I will most likely put some above my headers, just to try and get some of that hot air vented away from warming up my intake pipe.


Practice bonnet

The high tech forming tool, see the taper I ground into the leading edge and the chamfered corners to form the panel

The bonnet proper

Seems to have worked ok