When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
If I were running this test, I would put thermocouples into the metal of the head itself, not into the coolant.
I agree 100%. If I had the time, money, and inclination to really answer this question, that's exactly what I'd do. One thermocouple drilled into the casting at the front and the back, in both a front-therm and rear-therm configuration.
But I don't.
Originally Posted by Engi-ninja
I think you could get away with just one at either end, since it seems likely that cylinders 1 and 4 will be the extremes of the head temp. However, I'm not sure if it's strictly correct to say that it's the delta T across the head that we care about; what we're interested in is ensuring an equal and adequate amount of coolant is getting to all 4 cylinders
Presupposing that we're talking about delta-T across the metal, then these two concepts are functionally interchangeable. If anything, I'd actually be less interested in seeing the coolant flow than in knowing the actual temperature differentials, as in the end, we care more about the temperature of the head casting than the flowrate of the water surrounding it.
On an unrelated note, I can't help but think about how much this whole thread reminds me of the sort of thing that I used to mock and despise 25 years ago when I was totally into air-cooled engines...
Presupposing that we're talking about delta-T across the metal, then these two concepts are functionally interchangeable. If anything, I'd actually be less interested in seeing the coolant flow than in knowing the actual temperature differentials, as in the end, we care more about the temperature of the head casting than the flowrate of the water surrounding it.
I agree; I guess I was ambiguous in this comment. What I should have said was that we care about the actual temps of the metal of each cylinder, and figuring out how to make those temps as low as possible, which is accomplished by getting an adequate amount of coolant to each cylinder.
What I should have said was that we care about the actual temps of the metal of each cylinder, and figuring out how to make those temps as low as possible, which is accomplished by getting an adequate amount of coolant to each cylinder.
I've been thinking about this a lot. And right now, I'm wishing I'd stayed awake in thermodynamics.
I don't think that keeping the temp in each spot along the head as low as possible is the most important goal. I think keeping the temperature across all points on the head as consistent as possible is the most important goal. It reduces the mechanical stress due to expansion (I think), and perhaps more importantly, the hottest chamber is the one that's gonna kill the engine first. Better to have three cylinders at X° and one at X+30°, or all four at X+10°?
The more I'm thinking about Steve Dallas' proposal, the more I think it has merit. Sure, the temperature of the coolant outside the combustion chamber doesn't give us an accurate reading of the temp in the metal. We can skew that relationship by adjusting the flowrate of the coolant so that it's uneven across the head.
Which is exactly what using an '01+ gasket with a rear-therm reroute does.
BUT, it is relatively safe to assume that the rate at which heat is being generated in the head is fairly consistent across all four cylinders. And 2nd Therm tells us that the rate of heat transfer between two objects is proportionate to the delta-T across them. So if we do measure the temperature of the coolant at two locations, and find that it's significantly higher at point A than point B, we can reasonably deduce that less heat is being transferred from the metal into the coolant at point A. And that means that, by definition, the metal at point A is being cooled less effectively and must, therefore, be running hotter.
I've been thinking about this a lot. And right now, I'm wishing I'd stayed awake in thermodynamics.
I don't think that keeping the temp in each spot along the head as low as possible is the most important goal. I think keeping the temperature across all points on the head as consistent as possible is the most important goal. It reduces the mechanical stress due to expansion (I think), and perhaps more importantly, the hottest chamber is the one that's gonna kill the engine first. Better to have three cylinders at X° and one at X+30°, or all four at X+10°?
The more I'm thinking about Steve Dallas' proposal, the more I think it has merit. Sure, the temperature of the coolant outside the combustion chamber doesn't give us an accurate reading of the temp in the metal. We can skew that relationship by adjusting the flowrate of the coolant so that it's uneven across the head.
Which is exactly what using an '01+ gasket with a rear-therm reroute does.
BUT, it is relatively safe to assume that the rate at which heat is being generated in the head is fairly consistent across all four cylinders. And 2nd Therm tells us that the rate of heat transfer between two objects is proportionate to the delta-T across them. So if we do measure the temperature of the coolant at two locations, and find that it's significantly higher at point A than point B, we can reasonably deduce that less heat is being transferred from the metal into the coolant at point A. And that means that, by definition, the metal at point A is being cooled less effectively and must, therefore, be running hotter.
I really want to see that data...
It's definitely better to have all 4 cylinders as close in temperature as possible, which is why I originally said "equal and adequate" amount of coolant. Even if the hottest cylinder is not knocking, it still leads to inconsistent combustion characteristics across cylinders. I'm actually working on a project now in which we're trying to minimize the coefficient of variation across 12 cylinders...even assuming we can design the intake to get the same amount of air into each cylinder, which is very difficult, casting and assembly variations are enough to cause unacceptable amounts of variation. Anyway, getting consistent combustion in each cylinder is very important.
Your application of the 2nd law of thermodynamics is correct, however, I'm not sure I understand your conclusion. If the coolant is flowing from point A to point B, the coolant temperature at point A is always going to be lower than the temperature at point B, because it's going to be picking up heat as it flows along the head. If you mean that, because of this, point B is always going to less efficiently cooled than point A, then that is correct. However, this doesn't actually tell us anything interesting, since we can easily deduce that just from looking at the direction of flow.
Your application of the 2nd law of thermodynamics is correct, however, I'm not sure I understand your conclusion. If the coolant is flowing from point A to point B, the coolant temperature at point A is always going to be lower than the temperature at point B, because it's going to be picking up heat as it flows along the head. If you mean that, because of this, point B is always going to less efficiently cooled than point A, then that is correct. However, this doesn't actually tell us anything interesting, since we can easily deduce that just from looking at the direction of flow.
I may be misunderstanding your point though.
Put another way:
Let us presuppose that:
1: The amount of thermal energy being generated inside the combustion chamber, and subsequently transferred from the valves and chamber wall into the "meat" of the head, per unit time, is equal across all four cylinders.
2: At the back of the head, there is a steady flow of coolant, and we observe its temperature to be X°.
3: At the front of the head, there is coolant which is relatively stagnant (not flowing freely as compared to the back), and at this point we observe the temperature of the coolant to be X+ 30°.
We don't need to know the actual temperatures of the metal of the head in order to deduce that since each part of the head is being heated equally, and the front of the head is less able to reject heat into the surrounding coolant, that the metal at the front of the head is going to run a lot hotter.
I think we're saying the same thing here, and I'm not trying to convince you of anything. Just kinda typing it for my own benefit to make sure that it makes sense, and also to clarify my line of reasoning for anyone else who might not have followed the same train of thought.
Yup, we're on the same page :-) So, the application of this, then, is to take coolant temp measurements in both front and back of the head in the following configurations:
(A) Front Therm / Pre '01 gasket
(B) Rear therm / Pre '01 gasket
(C) Front Therm / '01+ gasket
(D) Rear Therm / '01+ gasket
Based on our discussion, it seems likely that the results will show the following:
(A) #4 much hotter than #1
(B) 1 and 4 pretty close together, and overall cooler
(C) not as good as B, but close
(D) #1 much hotter than #4
Now...someone with no children and an ungodly amount of time and number of Miata's on their hands, get to work proving (or disproving) this hypothesis.
.. but I *think* that this test should give a reasonable indication of the delta-T across the metal of the head casting as well, which is what we *really* care about.
The GT350 has many temp gauges, all in your choice of digital or scaled analog. One of which is cylinder head temp, like virtually all piston engined aircraft. What it does not have, is a scaled coolant temp gauge. Just a needle arc on an unscaled display, separate from the center dash where all the real temp gauges are.
My takeaway is that Ford is more concerned with actual CHT than CLT and thinks I should be too. They know more than I do about engines I think. So the Supermiata S1 (EFR6258) race car we're putting together will get not one, but two CHT sensors. You guessed it, #1 and #4.
NB2 head with 94-00 gasket though so a worthless data point.
Perhaps, amongst our many overlapping and overdue projects, we can stick two CHT's on a stock NB2 with a QMax and do some collectin'.
Your mention of aircraft jogged my memory, no drilling is necessary. You get the temps of the source of most pre-ignition by simply getting the spark plug temp.
Your mention of aircraft jogged my memory, no drilling is necessary. You get the temps of the source of most pre-ignition by simply getting the spark plug temp.
Interesting. An aluminum ring would have minimal total effect on the thermal conductivity of the plug to the head. Compression ratio theoretically affected but I think it would be nominal. Smart thing would be to put sensor rings under all plugs, even if you were only logging 1 and 4.
It's how most piston aircraft are retrofitted with individual cyl. Temp sensors if they didn't come from the factory that way. Have a few friends with planes and one that builds experimental aircraft with auto engines. These thermocouples are what I noticed on all of them. They can catch a cylinder going cold before they totally loose fire. They literally treat an engine like each cylinder is it's own sub system, so if you loose one the rest can get you home. I'm talking individual ignition and injection controllers on every cylinder.
I searched and was unable to find a CHT ring that would fit into the miata head without either hogging a channel into the aluminum or cutting a channel in the boot. The common rings outer diameter even have to be ground down just to fit down the bores.
Possbly an interesting data point: as a result of my CHT plug ring sensor attempt fail, I mounted the sensors adjacent to each intake port on a 1.6 w/reroute. I did not log the data, but used an aircraft grade gauge from Westach. Never observed more than a few degrees difference between any of the sensors. They all measured within 5-10 degrees of the CLT reported by the MS.
Edit: the sensors I used were from the above linked vendor.
Last edited by Ted75zcar; 08-08-2017 at 12:33 AM.
Reason: Add Sensor Connection vendor note
I can grab an IR gun for my next track day. I recall doing that a few years ago after doing my begi racer reroute and I did see a difference in temps. Will not be accurate data but can give an idea of the delta perhaps?
IR guns are not really reliable in measuring absolute values. However, if measuring under the same conditions, they can give a decent qualitative measure (what is hotter)
Your mention of aircraft jogged my memory, no drilling is necessary. You get the temps of the source of most pre-ignition by simply getting the spark plug temp.
I don't have any surplus BKR5s lying around... Is it possible to remove the crush washer without destroying it? This is as close as we're likely to come to getting some real, hard data.
Not sure if this will fit on a Miata spark plug "well". However, there are a few similar thermocouples in various dimensions, which could be bolted at various locations outside the engine block as an alternative; in fact we use them to measure battery cells temperature in our electric racecar, and they work well.
I had written these rules:
-----------
The following is true for any block 94-05. (I don't know about the MSM)...
I had a question about how all this pertains to the MSM, since that's my (semi)-dedicated track car. I looked through the thread until I found this post, the executive summary of which reads:
The 2004-05 MSM has a 2001+ head gasket with a 99-head.
The MSM head casting is indeed different from the '99-'01 casting. It incorporates cooling passages which differ somewhat.
I was considering adding the Qmax re-route but, given the lack of solid information (plus reasonable conjecture/supposition about possible harm), I'm going to leave well-enough alone.
Even with a stock cooling system, the car appears to run fairly cool on the track (jibing with its reputation for having the bestest OEM cooling in the NA/NB world). Of course, the idiot gauge doesn't give me any real temps (and I haven't datalogged the Hydra yet) and I do feel the timing being pulled towards the end of a 20 minute session. I may do the rad in the future, possibly Spal fans/shroud and, definitely, some ducting).
For another data point here is a picture of the gasket that came with my MSM top end kit, and a '99 gasket that I installed because I went with a DIY reroute on my MSM. There is a picture of a MSM head in my build thread if you want to check out the coolant passages (I dont want to clog up this thread with too many pictures)