Dyno testing on different fuels?
 

Recently I have been using Shell's V-Power 98 RON, mainly because the station is on the way to work, and my wife gives me the coupons on the bottom of her Coles supermarket dockets. I'm not sure I can tell much difference over the previous 95 RON, but in correspondence with Shell's technical dept. here in Oz, they say that you should get more power from 98 vs. 95 RON. They also say they haven't tested V-Power on a Renesis. Soooo.... has anybody done any side-by-side proper dyno testing on 95 vs. 98 vs. 100 RON on an RX-8? Note I don't mean the much vaunted but probably highly subjective "butt dyno" but the real deal.

I used the Shell v power racing fuel at the track this Wednesday and people where commenting on the flames shooting from both rear exhausts. I used to get some flames before but not as much as what I had this time.:Eyecrazy:

Although I was experimenting with blipping the throttle during down shifting this time for better shifting, less wear in the clutch and more revs out of corners.

I cannot comment if I felt more power however I did post my best ever time on the sprint circuit.

skc

its probably from the fact that higher octane fuel is slower burning, so there may be more unburnt fuel exiting the engine and igniting on the way out the exhaust.

If you are not losing power from the previous fuel, you will not gain any by going up in octane.
Seeing as how our knock sensor doesn't actually work, if the fuel octane is too low you will get ping. Otherwise, there is nothing to gain.

Now, I'm paraphrasing the Shell Oz website who say that the effect of the knock sensor is to effectively de-tune the motor (reduce power) until the pinging ceases. In correspondence to me they said that if the car is optimally tuned for the new fuel, all things being equal, an increase in octane will give you more power. Hence my question: has this been proved/disproved by dyno testing under controlled conditions?

For info to Americans

100 Octane V Power racing = 98 Octane V Power + 5% ethanol.

I don't see the difference on the track. I am happy with both.

I did see a film within the last 12 months that tested various levels of octane. Low, medium and hi rating fuels. They dyno tested power output. The validity of the test was increased by using 3 different cars. Cheep daily, medium family and performance - I can't remember what they were.

The results showed that if you own a cheep small car the octane level made no difference. Buy the cheepest and save the money.

The medium family showed a power increase in moving away from the lover level but not any great difference between the top two levels.

The performance of course showed that the engine was capible of using the higher ratings to advantage. For this class of car it was worth the extra fuel.

I normally run 98 but put 95 in yesterday (for some reason I can't explain) and ther is a lost edge to the response.

I think I read somewhere that ethanol didn't add much to the performance even though the figure was higher.

I'll see if I can find a link to the test I described and we can see if my memory serves me well.

Would that be Fifth gear? British?

Yes, Taka-san. I just found it on YouTube. '5th Gear - High Octane, better?' Basic but a good view, if you look ast the Korean subtitles.... It might help.

I did some acceleration tests using scanalyser to see If I could see a difference between 91 95 & 98 . On stock tune no difference . With optimised fuel & timing I found that I got better results with higher octane .
I also found that with 91 and the optimised tune - if the motor/intake was heat soaked , I had really bad detonation under WOT .

Brett hit on what I was going to post. The only time you are really going to see much difference is how the ECU is tuned. Aus spec RX-8's are tuned to our fairly poor quality fuel (annoying really as we export all our crude which is some of the best in the world). So we are going to see very little benefit from VPower Racing as opposed to VPower.

Interesting thing is I was at a dyno day a few weekends ago and there were a fair number of largely stock Corolla Sportivo's. Fairly consistantly those running VPower were getting 7 or 8kw lower readings than those running Ultimate.

Just wondering about the upcoming Knightsport ecu dyno day. I will most likely be attending but as I am based in Albury, I have to plan well in advance.
Anyway, what fuel octane are those attending KS ecu tune session looking at tuning with?
My concern with running 98 octane for the tune session is that the dyno may not expose the possibility of getting some detonation as the weather warms up later on ( seeing that the KS tune will occur in cold melbourne weather) and if so , we really dont have anywhere to go other than adding octane booster regularly during the hot summer months and during tracking.
Therefore personally, if I attend , I will be using my normal 95 octane shell brew, knowing that if any detonation issues arrise as temperatures soar in summer, I can counter this by running 98 octane shell V power.
Any thoughts?

Also, for those who will attend the KS tune day, who will be running the latest Mazda recall flash that has only very recently been released and who will be running the previous flash( which I currently am running)?

Regards
Rexi

I have heard that Shell is gradually withdrawing 95RON from sale (certainly the Shell I frequent now offers only V-Power 98 and 100-Racing).

I have the latest recall flash done at my last service. I have noted that starting is now very fast and smooth, even when our garage is at a freezing 10C.

Is it possible to have the ECU tuned for a particular octane rating? In my total naivety, I thought the thing was so smart it could adjust tuning based on sensor data it receives. Perhaps I'm givig the ECU too much credit.

With higher octane comes slower, more predictable burn of the fuel. As a result you can have more precise control of when the fuel will ignite and advance the timing a lot closer to TDC. The lower the Octane the car is tuned for the more margin for error required and so they retard the timing past TDC. But in doing this you lose some of the power of the fuel burn as it has less time to fully combust.

The ecu does have a knock sensor which when "activated" retards the timing even further.

You mean earlier before TDC. Typical ignition timing is 12° to 30° BTDC.

You mean closer to TDC - less advance of TDC (i.e. 5° instead of 12° BTDC).

In theory, though it turns out it is only good for detecting bad motor mounts and the like.
Ping on any level in the Renesis does not trigger the knock sensor because the sensor is improperly tuned.
If you look at a datalog of an engine during a pinging fit, the ignition timing remains nominal.
However, hitting the block with a mallet seems to do the trick.

Yes thank you. I was just re-reading my post (based on limited knowledge) and double checking it (probably should have done that before hand :|).

So just to confirm, ignition timing is based on degrees of eccentric shaft position during the compression stroke? Goal is to ignite the fuel as close to TDC as possible?

Do you normally tune just leading and then lock trailing a fixed number of degrees after that or is trailing as variable as leading?

The idea is to light off the charge so that it is at maximum pressure 15° ATDC.
At light load, this usually means firing the leading around 30° BTDC and the trailing at the same time or even before.
At heavy load, because the flame front moves more quickly through a dense charge, the leading is fired anywhere from 18° BTDC to 5° ATDC with the trailing firing 0° to 15° after that.
At the torque peak, you have the fastest moving charge, so the ignition event is the latest with the largest split. On either side of the torque peak, the advance picks up and the split decreases.

Sorry timbo for going off topic.

MM, again, awesome info. If I may a few more questions and feel free to ignore or just some hints on what to search on.

Whats the significance of 15° ATDC?

Before? wouldnt you just fire the leading earlier and then the trailing? Or do you mean fire the trailing so early that it does nothing?

why is the charge denser at heavier load?

:Eyecrazy: I need to read that a few more times i think :)

Cheers

Andrew

Heh. Yeah, there is a lot to learn.

15° ATDC is the point where the the rotor is at maximum acceleration (not maximum speed). Having the biggest boom at that point will mean it has the greatest effect.
Any earlier and you will have the charge expanding faster than the combustion area which is detonation. Any later and you are wasting energy as the chamber is expanding faster than the burning charge.

It has been found that firing the trailing ahead of the leading in light load yields better burn characteristics and, therefore, better power and economy. This can only be done at light load and idle.

At heavy load, we have much more air (either because the throttle is open more or the turbo is pushing, creating a greater effective VE) and more fuel, which means more stuff being compressed into the same small space as when we were under light load. This yields a more dense charge, which burns faster.

The last paragraph was just pointing out that, at the torque peak, you need the lease ignition advance. In the RPM band before and after the torque peak, you need more. That is the crux of it, anyway.

try this article I haven't read it but it might have some info

More about this pdf file.

http://www.motec.com.au/pdf/fueltest.pdf