Glycol Concentration Increasing Over Time

[twX]

I've recently read that engine coolant can increase in glycol concentration over time. The reason being that it's mostly water that gets lost due to evaporation.

My Subaru has recently hit 10 years on the original coolant, at 100k km. Subaru recommends doing the first coolant change at 11 years or 225k km. I've topped it up with 50/50 Subaru blue coolant from time to time when the overflow tank has gotten a bit low.

Checking the freeze point of the coolant with an old school antifreeze tester, it measured around -45 to -46°C, or around 57% glycol. The 50/50 mix straight from the bottle measured -37°C, or 50%. The increase in glycol percentage is consistent with a loss of around a litre of water, replaced with the same amount of 50/50 mix, which is probably about how much I've added over the years.

This engine has had knock readings greatly increase in frequency over the past few years. I've ruled out some possible causes for this, and now I'm thinking that it could be due to the cooling system performance, maybe due to worse heat transfer from the higher glycol mix. I kind of doubt that a 7% increase would make a huge difference, but it certainly wouldn't be helping any. I'm going to get the mix down closer to 50% to see if it helps.

 

[ripcord]

I doubt that will make any difference in the pinging unless the engine is on the ragged edge of overheating all the time and even then it ill not make a big change. 50% vs 60% of glycol concentration only changes the specific heat capacity of the coolant by a small amount.

What things did you rule out and how did you rule them out? My initial guess would be insufficient EGR function but this is just an off the cuff theory.

 

[TomYoung]

I would say that your theory is almost certainly wrong. But you have old coolant in there anyway that could use replacement. Why not drain and fill and look for a performance improvement. Then keep investigating, sine this is a thousand to one not the source of the issue.

 

[twX]

What things did you rule out and how did you rule them out? My initial guess would be insufficient EGR function but this is just an off the cuff theory.

I've checked the spark plugs, and they still look great.

Fuel trims don't suggest any significant vacuum leaks or MAF sensor issues. The MAF was cleaned for good measure. I've been thinking about doing a smoke test anyway just to be sure. Silicon was a bit high on the last UOA, so there might be some dirt coming in from somewhere.

I put some PEA fuel injector cleaner in the gas tank. It did reduce the knock readings a lot, but only while it was in the tank. I'm thinking that it just temporarily cleaned up some combustion chamber deposits, which then dirtied up to a normal level afterwards.

I've had a DTC for a clogged EVAP air filter recently, but I'm not sure if that could have any effect on knock. The system does introduce un-metered air and fuel vapour into the intake when it's purging, but the knock is very consistent. I'm assuming these systems should not be purging constantly in highway driving, but I'm not positive on that.

Higher octane fuel does reduce the knock a lot, which confirms that it's not just the knock sensors reacting to some other noise.

I'll start looking into the EGR system. The knock is only an issue at low/moderate engine loads where the EGR would be active.

 

[Rand]

unless its really low.. I top off with RO water. usually about 4oz a year.
Its not pristine but averages 92-96% reduction vs tap water.
Close enough for 4-5 oz.