Testing Effect of PEA Fuel Cleaner on Knock

twX

Joined
Dec 5, 2008
Messages
959
Location
Canada
Over the past couple of years, I've noticed more knock readings while datalogging on my 2015 Subaru WRX, mostly in highway cruising, but I'd semi-regularly get a knock reading at WOT as well. I'm no longer able to tune out the knock since the EPA nerfed my tuning software, so I decided to try some fuel system cleaner. The engine has around 55k miles, and still has the original spark plugs, which are due to be changed at 60k. It's tuned for 87 octane E10.

I used Gumout Regane High Mileage (Canadian blend, 26% PEA), at 70% of the recommended dose, which works out to 940 ppm PEA.

Here are the results for knock during highway cruising, in knock events per hour, before treatment, on the first tank, and on the subsequent two tanks with diluted PEA content. Average engine speed, load, temperature, and manifold air temperature were all similar between runs.

Knock Events Per Hour:
0 ppm PEA: 96
940 ppm PEA: 3
210 ppm PEA: 299
15 ppm PEA: 86

After the first treatment, the knock was virtually eliminated. After diluting the PEA to 210 ppm on the next tank, the knock events increased 100-fold. This is consistent with some studies that show that PEA has a "hump effect", where combustion chamber deposits are reduced at high doses, but they actually get worse at low doses (~130 ppm), which increases the likelihood of knock. With almost no PEA left in the tank, knock was similar to before the treatment.

Previously I'd get a knock reading on at least half of all WOT pulls. After PEA treatment, I haven't had a single knock event at WOT, but I've only logged 5 WOT pulls. I'll need to do more pulls to be sure, but the PEA seems to have helped.

My next project is to replace the spark plugs. I might just replace the two plugs on the more knock prone cylinder bank and use the other bank as a control to test the effect of new plugs on knock frequency.
 
how much PEA is present in top tier fuel?
I believe it's a lot less than the 100-200 ppm required for the hump effect. Also, while the hump effect increases combustion chamber deposits temporarily, it doesn't necessarily affect the cleaning of intake valves or injectors.
 
Is 87 Octane the factory tune? If not, what is it?
how is an 87 octane e10 tune different from 87 octane tune?
The recommended fuel is 91 octane. The choices for fuel around here are 91 E0 or 87 E10. Since GDI engines benefit so much from ethanol, the effective difference in octane between them is pretty small for this engine.

The engine makes a bit more power than stock due to a bit more boost and richer fuelling. Knock is less frequent as well.
 
Wait wait wait, E10 in Sweden did not get higher Octane when it went up from E5. Still the same 95 RON.
Its still marketed as 87 (ron+mon) why would you expect it to have any benifits with E10.
My guess is they just start with lower quality fuel.
 
Wait wait wait, E10 in Sweden did not get higher Octane when it went up from E5. Still the same 95 RON.
Its still marketed as 87 (ron+mon) why would you expect it to have any benifits with E10.

The knock reduction doesn't come from higher octane, it comes from injecting and vaporising a bit more fuel which lowers the temperatures in cylinder. Like you get less knock when it's colder outside aswell.
 
The recommended fuel is 91 octane. The choices for fuel around here are 91 E0 or 87 E10. Since GDI engines benefit so much from ethanol, the effective difference in octane between them is pretty small for this engine.

The engine makes a bit more power than stock due to a bit more boost and richer fuelling. Knock is less frequent as well.
Run premium on stock tune and then let us know about how much knock you're getting. Rich fueling is done to reduce emissions by reducing combustion temps from running too lean. GDI doesn't "benefit so much from ethanol".

Off the cuff I'm thinking your tune is causing the problem.
 
Run premium on stock tune and then let us know about how much knock you're getting. Rich fueling is done to reduce emissions by reducing combustion temps from running too lean. GDI doesn't "benefit so much from ethanol".

Off the cuff I'm thinking your tune is causing the problem.
I just checked a random log of a 3rd gear WOT pull with the stock tune on 91 octane. It had up to 3.2 degrees of ignition retard caused by 4 real-time knock events in a single pull. This is despite the fact that the DAM (coarse/global knock multiplier) was down from the default 1.00 to 0.68, which would have resulted in at least an additional 2 degrees of initial timing retard throughout the pull.

My tune has never had more than a single knock event at WOT, which pulls 1.4 degrees of timing, which is the smallest possible knock correction. The tune is not the issue. As for low-load knock, I don't have logs to compare, but I don't remember the stock tune being any better.
 
I just checked a random log of a 3rd gear WOT pull with the stock tune on 91 octane. It had up to 3.2 degrees of ignition retard caused by 4 real-time knock events in a single pull. This is despite the fact that the DAM (coarse/global knock multiplier) was down from the default 1.00 to 0.68, which would have resulted in at least an additional 2 degrees of initial timing retard throughout the pull.

My tune has never had more than a single knock event at WOT, which pulls 1.4 degrees of timing, which is the smallest possible knock correction. The tune is not the issue. As for low-load knock, I don't have logs to compare, but I don't remember the stock tune being any better.
So you've run a full tank of 91? BTW..all engines continually adjust for minor knock. From logs I've seen it will never be zero. I think you're chasing ghosts.
 
So you've run a full tank of 91? BTW..all engines continually adjust for minor knock. From logs I've seen it will never be zero. I think you're chasing ghosts.
I ran 91 and datalogged with the either stock or generic aftermarket tunes for a couple of years before starting with the custom tuning.

The amount of knock it has now is still minor and I'm not concerned about it, but since it's more frequent than it used to be and mostly limited to one cylinder, I thought it would be interesting to test whether it was due to a fouled injector or something.
 
The recommended fuel is 91 octane. The choices for fuel around here are 91 E0 or 87 E10. Since GDI engines benefit so much from ethanol, the effective difference in octane between them is pretty small for this engine.

The engine makes a bit more power than stock due to a bit more boost and richer fuelling. Knock is less frequent as well.
Where in Canada are you getting E0. I believe it can be had in the maritime provinces but besides that ?
 
Where in Canada are you getting E0. I believe it can be had in the maritime provinces but besides that ?
I'm getting it at a vp racing station 20 minutes north of the city I work out of. I tested it and it was in fact free of ethanol. There is also a shell station that shows up on pure gas.org and is labelled as no ethanol (this is for 91 octane only at both stations) but shell could change at any time. The vp racing is unlikely to.

I came back to this thread because I found this moto master cleaner on sale at Canadian Tire and under "first aid treatment" it says it contains Polyether amine. Guess they didn't realize that was something to brag about instead of hiding it under first aid treatment.
Maybe this is rebadged Regane? It was $11 which was a few $ cheaper than regane.

20240731_114426.webp

20240731_114456.webp

20240731_114502.webp
 
Aside from the dramatic inconsistency in fuels. Isn't the PEA bottle primarily filled with naphtha and other petroleum products? Your taking unknown chemistry and combining with more unknowns. All while logging something that can't tell you the content of the fuel.
 
Back
Top