Datalogging water pressure - expected results.

Hi all,

Could someone cast an eye over the the attached log. I data log water pressure as a fail safe for head gasket failure. Reviewing my data I see water pressure rise with boost and revs. No spikes. This seems to be to be expected result. Looking for some validation of the data.

Sensor is located in the top of the rad on the hot side of the block. i.e. outflow from the head.

Regards,

Richard.

I'm new here and I know this isn't an answer to your question, but more of a question I wanted to 'tack on' for follow up. I would be curious to learn the answer as well.

Wouldn't coolant pressure increase with RPM because your block has a natural resistance to flow and that causes the pressure to increase as you flow more coolant? (As RPM increases, so does speed of water pump and consequentially the water FLOW?)

Also, would your radiator cap also dictate pressure at certain temperatures? As temperature reaches the temperature setting of the cap, the coolant is going to go to the overflow or expansion and during that time it will cause pressure to drop until engine is ready to take the fluid back in?

William,

1. Yes. Pressure will increase with RPM. I agree.

2. Coolant system is sealed with expansion tank, no rad cap and overflow. This makes it trickier as the expansion tank is plastic. Hence the sensor in the rad.

That's a screen shot of a graph, not a log that can be examined.

So I can't tell -- what was the coolant pressure at the beginning of that acceleration run? A pressure rise of more than a couple of PSI for an engine already at operating temperature would be unexpected. What does the coolant pressure do when free-rev'ed to the same RPM (slowly) but not under load?

Did the water temp change during that run? It kind of looks like it went up (how much?), then down (was 81c at the cursor) -- but that could be another channel.

FYI-- My Mazda MZR-powered race car, runs 3-7psi of coolant pressure for an entire 30 minute race (73-83c coolant temp, engine speed 5000-8400 the entire time). I will admit I run a non-stock water pump impeller, which is non-optimal as we turn it in the opposite direction.

David,

I have pulled all the data out of the log at 250rpm points. The log is an Aim Datalogger, so u kind of need RaceStudio 3.

Here is the data that covers all the questions above. I have an 82deg thermo in the car so this maybe playing a part in restricting flow and having an influence on the pressure.

That's pretty much what I would expect for a block mounted pressure sensor, as you surmise, there can be significant resistance to coolant flow at high rpm.

That's why slowing the water pump by using a larger pulley, or using an electric water pump, can give significant power increases, the pump isn't working as hard against the lower pressure head.

The block pressurisation, and the higher boiling point, at higher rpm, compared to idle, is also a big factor for cars suddenly boiling over when driven hard and then idled - localised hot-spots that were below the high pressure boiling points may suddenly be above the idle pressure boiling points.

As an AiM dealer, I would not have any problem using an AiM logged data file :) If it won't upload to the forum, then just put it in a ZIP file.

So does the pressure come back down as you return to idle? What were the results of the free-rev test without load?

Good question, free rev' and under load should be basically the same pressure ranges - assuming the temperatures and thermostat openings are about the same.

David,

As a fellow aim dealer this is reassuring. :-) I have attached the log. It's a good hours driving.. with some WOT pulls. Please bear in mind the following. the engine has been compression tested and sniff tested. So I do not believe the engine has a head gasket issue. This is about understanding the shape of the data. Data without context is not information.

The cooling system is OEM sealed with an expansion tank, not a rad cap/overflow setup like on older cars. So I do expect pressure behaviour to be different.