Is this a logical result?

[richardpope50]

A few year ago having solved my cooling issues I ended up with a 75 degree thermostat. All was fine and my normal engine temp was 75 to 80 and very rarely going above that even in traffic but usually more like 75. My fuel consumption was 23 mpg.

It was suggested that a hotter engine would help improve my mpg and so I put in a new 82 degree thermostat in a few months ago. On my first test since doing this I now get 19 mpg albeit only on a 155 mile test. My engine temp is now usually 90 and rarely getting above 95 although it does seem to be at 90 + more often.

Should I go back to a 75 degree thermostst?

(My temp gauge is 100% accurate as certified by CAI.)

NB. I have a megasquirt ECU so I did get it to re-map itself automatically after swapping the stat and before this test therefore the map has changed but only to what the software thinks it should be for the higher engine temp. The re-map was approx a 30 mile trip. I haven't changed anything else (honest).

[stevieturbo]

90degC is a better place to be for a road car both for engine and oil temps.

And ecu's do not remap themselves no matter what people might say. Yes some will have closed loop corrections, and will adjust to whatever degree the tuner has allowed it to, but there is no substitute for correct mapping and after that has been done, allowing closed loop to make small corrections.

And a hotter engine will not be a cause of using more fuel, it should use less...but without solid data between the two "test scenarios"....there could be many reasons for the differences.

[DaveEFI]

MegaSquirt can auto tune. There are different ways, but the usual one (VE Analyse Live)) is to log it when running. It then compares the existing VE table to the logged one created by your wideband using your AFR table. And then shows any differences. And gives you the option to use the new VE table or stick with the old. If the new isn't an improvement, you can easily revert to the old one.

However, if the CTS compensation is good, there shouldn't be that much difference.

A rolling road is a good way of tuning for maximum power. But the more everyday tuning like low and high cruise etc best done under actual road conditions. Where VEAL comes in. Idle, though, is best done manually.
For VEAL to work well, you need a good AFR table, an accurate wideband, and any delay due to the distance the O2 sensor is from the ports set correctly.

[stevieturbo]

MegaSquirt can auto tune. There are different ways, but the usual one (VE Analyse Live)) is to log it when running. It then compares the existing VE table to the logged one created by your wideband using your AFR table. And then shows any differences. And gives you the option to use the new VE table or stick with the old. If the new isn't an improvement, you can easily revert to the old one.

However, if the CTS compensation is good, there shouldn't be that much difference.

A rolling road is a good way of tuning for maximum power. But the more everyday tuning like low and high cruise etc best done under actual road conditions. Where VEAL comes in. Idle, though, is best done manually.
For VEAL to work well, you need a good AFR table, an accurate wideband, and any delay due to the distance the O2 sensor is from the ports set correctly.

And as the above requires user input....it is not remapping itself.

And 75-90degC....there would be zero adjustments for fuel etc anyway, there wouldnt need to be.

But yes...really more data is needed.

[DaveEFI]

That is just one method with MS - you can make it do so automatically.

But I'm with you on the difference in fuelling needed in this case would be very small.

[richardpope50]

I use TunerStudio running on my laptop connected to my MegaSquirt ECU and it does just what Dave says - that's what MgaSquirt is all about. It re-maps contiinually by monitoring what is happening, records what to change based on my wideband sensor and other readings. Then you simply periodically save as you drive along. As I said, I did this for a 30 mile trip and saved regularly so the (re-)map was based on the new 82 degree thermostat. I also specifically planned a route of varying conditions to get a good re-map.

However, that was not really the point of my post. It was, can a change of thermostat from 75 to 82 realy reduce my mpg from 23 to 19? As above, I thought a hotter running engine (at 85 degrees) should do the reverse. Am I wrong?

[Ian Anderson]

I am an accountant so love numbers and stats!

I have seen 33 mpg and 21 mpg out of the Gt40 and on a single day 30 mpg and then 25 mpg based on two tanks of fuel.one drive out and fill and return and fill so same conditions etc.

I just put the change in mpg to the batches of fuel, they seem to vary a lot!

Ian

[ChrisJC]

However, that was not really the point of my post. It was, can a change of thermostat from 75 to 82 realy reduce my mpg from 23 to 19? As above, I thought a hotter running engine (at 85 degrees) should do the reverse. Am I wrong?

I doubt it very much. Once upon a time I used to measure the MPG of every single tank of fuel. It varied from 38MPG to 58MPG depending on fuel, route driven, driving enthusiasm, vehicle load etc. etc. You would have to some very controlled testing to get a meaningful result in my opinion.

Chris.

[richardpope50]

My fuel is always the same - Tesco Momentum - and I measure per tank full so give or take over about 200 miles each time. I suggest my driving is generally about the same as my mpg is generall within 1 mpg each reading. I'm trying my second tankfull so will see the result next fill-up and no doubt report back. Still a bit surprised at the possible change just down to the thermostat.

[DaveEFI]

I too look at my average MPG over each tankful. Which isn't that much different from yours. But wouldn't be surprised to see a couple of MPG difference on just the one tank. More so, given the different traffic patterns over the past few months.

[richardpope50]

So today the averae mpg over the last 162 miles was 20.55mpg.

Thinking I will go back to the 75 degree thermostat if next monitor shows this sort of figure.

[SuperV8]

You mention megasquirt/tuner studio also auto-tuned (VE Analyze Live) between the stat changes.

I must say regarding VE Analyze Live - I found it amazing, on one drive, I left the garage with a rough running very rich and lumpy engine, and on that one drive it kept getting better, smoother and faster - an amazing drive.

My money would be on the change in tune rather that a hotter stat.
What were your cruise AFR's before and after the tuning?

It could have been running too lean before? or maybe your new cruise AFR could be leaner?

Tom.

[richardpope50]

Tom, yes it is VE Live I use.

During all my cooling problems (or very hot running) a few years ago the mpg then was complicated as I was also having problems with using a Racelogic traction control unit for 6 cylinders (as when I bought it they did not do an 8 cyl version) albeit correctly wired as per RLs instructions. I discovered that the RL ECU was giving me 17 to 19 mpg but when RL replaced it with an 8 cyl version I got 20 - 21 mpg. I also got 20 - 21 when I disconnected the RL ECU so proving to RL there was a problem. I also did a reasonable number of map updates, say, 1 or two a year.

Once I got the temp down to 'normal' having changed the rad and made it two way as well with the 75 degree stat and after a map or two I eventually got the mpg to 23. I kept the status quo for a coupe of years but this year I felt that I needed the 82 degree stat as my engine was running a little too cold and others have said a warmer engine would give a better mpg. So this year I switched to the 82 and was hoping for 25mpg but got 19 and 20.

I did one mapping run of 30 or so miles immediately after having changed the stat to make sure the my fuel tests were on a good map.

To answer your AFR question, no idea. Sorry but that part misses me so just assume VE Live does it all correctly. Certainly on the map run there were plenty of changes and plenty of saves as I go along. I would be surprised if the map has caused the problem as why would it not produce the best map possible (for the info it gets in the run) as that's what VE Live is supposed to do?

With lockdown I've only really done 500m this year so perhaps this or next year I'll do another map or two and then change the stat back down to 75 if no improvement. My aim is a 25mpg figure.

[SuperV8]

VE Live tries to automatically change the fuel injector duty cycles to close in on an AFR look-up table which you have pre-defined.

Have a look at your AFR look-up table and check the figures look reasonable? You may be able to lean out your cruise AFR target and do some more cruise tuning runs to reduce cruise fuel consumption?

VE live needs to spend a certain amount of time at each load/rpm point on the maps to get confidence in the AFR its measuring vs the target AFR (which you have given it) then it will make an adjustment. I found it very good at cruise tuning as its easy to spend lots of time driving at cruise so it has lots of data to make accurate changes, but found it difficult to spend enough time on the road at the high load/rpm areas of my maps simply as I would be going too fast!

Also I would have a look at your spark map - and check you have enough advance at cruise/light loads.

Tuning for Economy
http://www.megamanual.com/ms2/tune.htm

this is also interesting:
https://www.davidboettcher.com/thermostat.php

Tom.

[DaveEFI]

Quoting the MegaManual? Hope James Murray doesn't see this forum.