[widetyres]

This is a super interesting discussion and perhaps highlights the source of some of my issues with the BM3 and knock.

I had a lot of knocking on Stage 1 93 and PTF suggested it was to do with fuel (which was UK Shell V-Power 99 RON - supposedly the best pump fuel available here). I then dropped to the Stage 1 91 map and the severe knocks disappeared.

On reading these posts, I have found the datasheet for UK Vpower (dated 2013) here:
https://events.imeche.org/docs/defau...3.pdf?sfvrsn=2

This details that the MON is 86, which from the above conversation seems to be on the lower side). This would actually give a result of 92.5 AKI.

So I am below OTS 93 spec to start with, and then the occasional "off" tank could easily see this drop further.

I will have to investigate the other brands considering your comments that MON is more important to consider on the M2 engine.

[Nezil]

widetyres, I was worried that our discussion had gone off topic and you’d be annoyed by that so I’m relieved that you’re not!

The PDF link you posted is great. I have been looking for similar documents for pump gas in the US and worldwide. Your document shows the MON and RON for Shell V-Power and for the two standards used in the UK for Regular and Super Unleaded fuel:

EN228 (UK Regular Unleaded): min RON 95, min MON 85 = min AKI 90
BS7800 (UK Super Unleaded): min RON 97, min MON 86 = min AKI 91.5
Shell V-Power: min RON 99, min MON 86 = min AKI 92.5

So it seems that I was partially correct and partially incorrect... UK Regular is a little better AKI than US mid-grade (90 vs 89 AKI), and UK Super is marginally better than US premium in some states where you can only get 91 AKI, and marginally lower than other states where you can get 93 AKI.

One other important point in that document is that the max Ethanol content is 5%, whereas in many US states it’s 10% or even higher!

Unfortunately the document was from 2009, and may not longer be accurate. I’ll see if I can do some more digging on the standards above to see if they’ve been updated or changed at all.

[Nezil]

I did some more digging and found that the latest EN (British Standard) specs, although changed in the last 10 years, haven’t changed with respect to minimum RON & MON, and maximum Ethanol.

I also found this article, which is a 2016 study of the effects of RON and MON on performance: https://www.sciencedirect.com/scienc...52146516302629. Interestingly, this article concludes that RON is more important than MON for modern engines, but I’m suspicious of the results. Both of the engines used for the test were naturally aspirated (I believe one was a 2.5l BMW IL6), rather than high performance turbocharged.

I’m going to look at some of the other documents cited in that one, and perhaps send an email to the author asking some questions about the topic.

So based on this new evidence from the British standards, my theory about the balance of RON and MON being different between US and UK fuel, particularly ACN US fuel is still possible. I really want to get to the bottom of this!

[Nezil]

I've looked into one of the documents cited in the 2016 paper. One of them from 2012 is particularly interesting because it's a) the paper that proposes the K constant for calculating an 'Octane Index', and b) the tests they conducted were using a turbocharged engine, and are therefore closer to our cars:

https://www.researchgate.net/publica...gnition_Engine

In this paper it again suggests that RON is more important than MON, and in fact that as load, pressure and temperature increase, fuels with a higher 'Sensitivity' (RON - MON) perform better.

The paper suggests that at the time that the MON test was developed, it was the most appropriate octane measurement method for engines at the time, but as time has passed, things changed, first toward an equal combination of MON and RON, hence the US AKI equation of (MON+RON)/2, and now to the point where RON is most important, and actually lowering MON, or increasing the difference between RON and MON helps with performance!

A new equation for Octane Index is proposed, which includes a variable 'K' for temperature and pressure of the unburnt gas mixture. You could think of this as being very very loosely related to the age of engine technology because inter-cooled high pressure forced induction is now common place.

Octane Index (OI) = RON - K * ( RON - MON )

• If K = 1, OI equals MON
• If K = 0.5, OI equals the same as the AKI equation
• If K = 0, OI equals RON

Modern engines are said to be 'Beyond RON', because their operating conditions are beyond those of the RON test in terms of pressure and temperature. For this reason, negative K values should be used.

In the paper, a K value of -0.93 is observed at 4,000 rpm when the intake pressure is 2.6 bar absolute, which corresponds to a boost pressure of 23.2 PSI. Our engines usually run slightly lower boost than this at around 18 PSI (2.24 bar abs.). In the paper, a 10:1 compression ratio engine results in a K value of -0.33 @ 1,500 rpm, and -0.38 @ 4,000 rpm. Based on this paper then, for our engine, we should be assuming a K value of between -0.33 and -0.63; perhaps even up to -1 if tuned aggressively.

I'll go back to my Excel sheet and calculate these values again, but looking at the British standards to start with, they have the following OI:

When K = -0.33

• UKRegular Unleaded: - 95 - -0.33 * (95 - 85) = 98
• UK Super Unleaded: 97 - -0.33 * (97 - 86) = 100.63
• 2013 Shell V Power: 99 - -0.33 * (99 - 86) = 103.29

When K = -0.63

• UKRegular Unleaded: - 95 - -0.63 * (95 - 85) = 101.3
• UK Super Unleaded: 97 - -0.63 * (97 - 86) = 103.93
• 2013 Shell V Power: 99 - -0.33 * (99 - 86) = 107.19

When K = -1

• UKRegular Unleaded: - 95 - -1 * (95 - 85) = 105
• UK Super Unleaded: 97 - -1 * (97 - 86) = 108
• 2013 Shell V Power: 99 - -1 * (99 - 86) = 112

The paper concludes that standards should be changed so that MON can be lowered, or RON increased without raising MON to provide better fuel for newer engines allowing them to be down-sized for efficiency.

So I might have got things completely upside down, and MON is negative towards our engine performance. Perhaps in fact ACN fuel has low 'Sensitivity' (not much difference between RON and MON), and that's what makes it worse for our engine than rest of US and European fuel?

[Nezil]

So now I have some more data... Please feel free to correct me if anyone knows better than I do about the subject.

California passed the Low Emissions Vehicle (LEV) III standard in 2012, and this includes requirements for pump gasoline. There is detail in this document that covers the requirements:

https://ww3.arb.ca.gov/msprog/levpro...i/attacha9.pdf

But in summary, they are as follows:

• 10% Ethanol, no other oxygenates
• 91 AKI minimum for 'Premium'
• 87 ~ 88.4 AKI for 'Regular'
• Octane 'Sensitivity' minimum 7.5
• Sulphur 10ppm max

So if we assume the worst, we get the following:

CA Regular:
- AKI of 87
- RON of 90.75
- MON of 83.25
- Sensitivity of 7.5
- OI @ -0.33 of 93.225
- OI @ -0.63 of 95.475
- OI @ -1 of 98.25

CA Premium:
- AKI of 91
- RON of 94.75
- MON of 87.25
- Sensitivity of 7.5
- OI @ -0.33 of 97.225
- OI @ -0.63 of 99.475
- OI @ -1 of 102.25

If the Octane Index theory is to be believed, this fuel is terrible! The sensitivity is key here though, because CA Premium with a sensitivity of 10 (the UK spec for 'Regular') would result in 96 RON / 86 MON, and a sensitivity of 11 (the UK spec for 'Premium') would result in 96.5 RON / 85.5 MON, which is pretty close.

I imagine though that blending fuel to achieve a higher sensitivity is more expensive, in part because it requires RON to rise as much as MON decreases (to achieve the same AKI). CA fuel may therefore be stuck at these low numbers...