I have some new 914 2.0 repro heads from DRD. The new heads use a 12mm spark plug rather than the stock 14mm. I want to use a ring terminal type Cylinder Head Temperature Sender under the #3 spark plug just as I did on my stock heads. The problem is the outside diameter of the ring terminal of my 12mm CHT sender is larger than the bore of the spark plug hole so I can't put the sender under the #3 plug as designed. As many of you probably do, I had the 14mm ring terminal sender on my stock heads and that took a lot of manipulating of the terminal stem to get it to fit. This is way worse, the OD of the 12mm ring will not even fit into the plug bore.

I asked Darren at DRD what to do and he suggested to use the fuel injection CHT location out on a fin by the intake manifold. I don't like that idea because it seems too far away from the chambers to pick up any major temp fluctuations.

I'm considering changing to an 8mm ring terminal on the sender and double nutting it onto one of the #3 exhaust studs.

Anyone have a better location?

Without the sensor under the plug the values that you will gather is virtually worthless. I have applied as many as 16 sensors on a single engine at once in positions all over the cylinders and heads. I saw as much as a 200 degree differential between hottest and coolest reading simultaneously.
What this proved to me was the ONLY place for that sensor was under the spark plug.

The entire head doesn't achieve a single temperature, under the plug gives the most repeatable readings. With a sensor under an exhaust stud you will have insanely high temps as exhaust gas temps run 9-1200 degrees and heat soka the entire area around the exhaust port.

I'd recommend modifying the plug well to accept the proper sensor (our heads have this done as standard offerings) and you can buy various size ring terminal sensors from Westach and Dakota Digital.

Values are no good if they aren't accurate.

What is the expected range of the head temperature?

Dependent on load CHT can vary from 250-450 degrees.. Values can differ 100 degrees in just one mile of driving in certain road conditions.

I recently put in a CHT. I typically see temps around 350 with spikes near 400. The temps vary a lot during a typical drive. I installed it with the engine in the car.

My friend had taken a spark plug and cut it in half, leaving only the threads, filed it with epoxy and put a screw driver slot in the epoxy. I used his "plug" in my spark plug hole while I used a dremmel to modify the head to accept the ring terminal. No metal in the eninge.

Jake,
I have read numerous places about the inaccuracy of the VDO gauge; especially since the cold end of the thermocouple is in the hot engine bay. What do you have to say about this?

I've thought of making my own sender and running thermocouple wire all the way to the center console. I've contacted Digital Dakota and they can make a special wire with the 14mm ring terminal and the length I would need for about $20.

if you are using the cht value to run a fuel injection system, then you need to find a way to put it back on a plug till you reprogram your FI for the next location.

If you are not using an EFI system that relies on that CHT, you can put it where ever you want.

Many people run with no Cylinder Head temp gauge and the car still runs.
Do your heads have the bung drilled and tapped for the original CHT that is next to the #3 intake head bolt?

I would not put it on the exhaust stud until you verify the gauge and sender can deal with that much heat...
EGT sensors are designed for the very high temp of the exhaust, but not the typical CHT sender or gauge.


Rich

Have them make a couple extra,
Digital Dakota 'cause I sure can use one and thats a good deal

The stock CHT position for the EFI system is also inaccurate for CHT values and monitoring. That sampling position was specifically positioned to protect the switch from too much heat as well as to remain hot for a longer period of time after engine shut down.

That position is at minimum 100F lower than the under plug sampling location. I have multitudes of that supporting data.

Well, if you know it's always going to be about 100F lower in that position, you can easily design your FI logic to compensate for that and still be reasonably accurate.

zactly

So if I use the stock location for the temp sensor, would a normal gauge of up to 350F work?

Problem is the fact it isn't absolute... Thermal conductivity of different head castings creates a variable. One degree separates safety or failure.

I have never had an issue with spark plug senders; I only changed the plugs twice in my 912E engine in 8 years and 160,000 miles.. The senders under my plugs have been in service since 2002 and still are in service today. No need in changing plugs as often as most people do with these engines.

Sorry for the highjack but this seemed a good time and place to ask this:
What milivolt values should I see fromm the Dakota plug style thermocouple and it is compatible w/ the VDO gage? Also, how can I test a VDO gage if I suspect it is not working? TIA and sorry for the highjack, again.

I had the same issue with the 12mm plug and got some excellent support from "The sensor connection" guys. They worked with me sending back & forth several CHT's and finally got it right with a 12mm ring that will actually fit down in the recess on the head. Mind you, it's still a tight fit & you have to carefully bend up the tab, but their 12mm ring is a true 12mm, not 13mm as some vendors are supplying. No vested interest, just have had great service from these folks:
http://www.thesensorconnection.com/index.shtml

Just when I was ready to give up on an under plug sender.
Perfect, Problem solved!


Thanks for the link

I have the factory CHTS in the right head. Should I be running the plug style instead? Do they wire in the same?

If you are using the cht for the fuel injection, you must use the stock location, stock sensor etc.
If you are using that reading for either an aftermarket gauge or aftermarket fuel injection, you can use the plug sensor.

In your cold weather location, you may want to contact Brad and get his spacer for the sensor to give your car a bit more time on warm up....at 914ltd in Illinois

Rich

Aaah.. I thought this was some new fangled thing I needed. I'm cool.

nope, you just need to start the damn thing already...
What is missing?

Rich

Sorry but I am reviving this old thread. I want to install the CHT under the spark plug like all the cools kids have. So is this the one to get or have there been better solutions? I am preparing myself for the idea of dropping the engine to perform the modification on the head. That's not the only reason to drop the engine. I need to install the crank sensor for my Microsquirt conversion too.

Thanks All!

Is this sensor for the gauge, or is it for your microsquirt?

If for the MS, the stock thermistor-type sender in the stock sender's location should be more than good enough. Since the ring-under-the-spark-plug sender is a thermocouple, and generates voltage instead of changing resistance, you'll have to program the MS differently to use that for warmup enrichment.

If it's for a gauge, then the spark plug sender is the way to go.

--DD

The issue is that it can be difficult to properly thread a spark plug into the head, and it is even harder with a copper ring around the plug threads. Tradition has used such an instrument to measure head temps, so using number three spark plug has become a kind of standard. Is it etched in stone? Heck no! The factory made a special drilling for its own cylinder head sensor, but it is something that can be related to the experiences of others. Hmmm, my cylinder heads melted away at xyz degrees. George can correct me, but my recollection is that when I called AA he said that a CHT gauge would “just scare me”. And it has on many occasions, but the motor still runs…
Moral of the story is that the instrument will allow you to start making measurements. These can come in useful. It certainly tells when you stress the motor, and when you let it rest. It is summer while we have it. Good luck!

Yes it is for my Microsquirt conversion.
Ok that is good to know I can use the stock CHT. I also acquired this CHT from the dub shop. So between the 2 hopefully I can get my MS to work. I think I was overthinking the CHT part.
Thanks DD!

Click to view attachment

I don't know what kind of sender that one is from the Dub Shop. It might very well be a thermocouple, like the ring sender that goes under the spark plug.

Check on what the MS system expects to see for warm-up enrichment. Most cars that I know of use a thermistor in the coolant circuit.

Again: Thermistors are things that change resistance with changes in temperature. Thermocouples are things that generate voltages based on the temperature difference between their "cold" side and their "hot" side. Something that is set up to look at resistance will work poorly if at all when fed a voltage, and something that expects a voltage will not work when fed a resistance.

Make sure the sensor is compatible with the expected signal.

The stock CHT sensor will likely require some extra calibration to be done in the MS to get the enrichment right. The resistance range might be different from what the default setup expects, and it is likely that using HCT instead of coolant temp will throw things off to some extent. Check through the setups that people have used for MS on aircooled motors to be sure.

--DD

DubShop makes that thermocouple/thermister-thing for tuning warm-up on the Micro/Megasquirt. It's intended to be inserted under one of the engine tins screws (I use the one behind the #3 intake manifold). It's not intended as an accurate measurement of CHTs in the context of how we discuss them and the allowable temperatures.

One major advantage to that sensor is that it already has the pull-up built in, so all you have to do is wire it up to the MS (5V, ground, sense). It's the easy button for the purpose of managing engine warm-up and enrichment. Works great.

I also installed a standard K-Type thermocouple under the #3 plug to be used for tuning purposes, to ensure I'm not running too lean or hot. I ported that into the Microsquirt's ADC7 input via an Adafruit amplifier. It will not provide feedback to the ECU for corrections but I can log the data and later compare it to trends to see if I'm getting too hot (good so far). It can also provide warning limits if I'm using the ShadowTuner dash on an Android tablet.

So Ted, I suggest moving forward with the DubShop sensor for warm up enirchment and then work later toward adding the CHT for info.

I just noticed the Dub Shop has a Type 4 version of the sensor that I didn't notice a few months ago. Looks like the difference is just that it has a larger hole do it can be mounted in the stock sensor location. Here is the description:

If I am not mistaken it goes on the tin screw and not the actual location where the factory CHT goes, but maybe I am wrong?

Click to view attachment

Contrary to what the experts are saying, I will be installing a new CHT gauge sensor in the factory location. Autometer setup. I will be able to monitor a "normal" reading vs an "elevated" reading.

Well, that is great and means that all of the cautions I mentioned above are meaningless! Sounds like it's pretty much plug in and go!!

--DD

They are selling two versions. One that goes on the #3 tin screw but another special for the Type 4 that has a larger hole to bolt to the stock CHT sensor location with a 10mm bolt. Same sensor, just modified with different size holes. Doesn't seem like it would matter which you use, but I like having the option of using the stock location. Feels cleaner somehow.

I went with the one on the engine tin screw because:

- It's outside the engine tin, so very easy to access - like, right there; and
- It doesn't matter where it is, since you're going to be managing/monitoring/editing your warm-up sequence based on what the sensor shows over time.

I do not, not, not suggest you use this sensor at the stock D-Jet true cylinder heat temperature. It should ONLY be used for ECU warm up sequence.

All of the years of collective "true" CHT evaluation and limits (such as Raby's recommendations) are based on a specific thermocouple (Type K) installed in a specific location (under the #3 spark plug).

If you try to use the D-Jet location in lieu of all of that collective knowledge, then you are setting yourself up for potential problems (that location is likely to be simultaneously neither as not, nor as responsive, as under the spark plug).

Of course, if you want to install both, and develop a stastical program to demonstrate correlation between the two, then everyone that follows you would be your biggest fans...

I run a thermocouple under spark plug #3 and a BMW CHT sensor in the stock CHT sensor location with MS.
I'm using Lens heads with the ceramic inside and the black outsidecoating.

Both Sensors report the same temperature with a little advantage in reaction time on the under the spark plug sender.
However temps are always within 2% of eachothers reading, which is neglectable.

Will take the car for a spin today and can post a pic later...

Would love to see that correlation! It would thus be an excellent alternative (and easier to install) location for monitoring engine safety. I've got the correlation data for the under-the-tin-screw versus spark plug and it is decisively not a good spot for The Raby Recommendations.

But if your correlation between CHT boss and spark plug proves strong, then let me offer a secondary reason why this is not a good place for using that boss for Mega/Microsquirt (or any EFI) warm-up: because it's historically shown to not be a good place for measuring D-Jet warm-up.

D-Jet has a problem with warm ups. The prevailing thought is that the CHT sensor warms up quickly and causes the the D-Jet to lean out too fast, well before the engine itself is ready for it. As a result, we have this "hole" where the engine fires up nice and rich, but soon leans out because the CHT is telling it to; idle and driveability suffer. VW/Porsche tried to fix it with a resistor and a retune, and even today we battle it by putting in steel spacers to reduce the rate of thermal rise.

The DubShop sensor in an engine tin screw near the intake manifold solves this. This location warms up much more gradually and slowly, and offers a much wider band of time to fine-tune the engine warm-up. I can start the engine foot-off with it dead cold with ambient temps in the 30s (I don't use drive the car colder than that) and it'll immediately settle into a solid idle, and core that idle all the way to full engine warmup (I use ignition timing to manage it).

My "full warm" point is 160F, and it takes about 3-5 minutes of light local driving to get there, depending on ambient. That's plenty of time to get the long block warmed up, the oil flowing nicely, and about the time I get to the interstate on-ramp. And given this wide range of time/temps to get there, I have a lot of flexibility in tuning to improve driveability.

For reference, on a typical highway drive the engine tin screw "CLT" (how it's logged in Microsquirt) is about 75-100F lower than the spark plug CHT. Whereas the CHT at cruise is ~330, the CLT is ~240.

I've read where some people use oil temps for warm-up enrichment management, but I suspect that this rise is a little TOO slow. I do want some compromise where I can reasonably inferentially measure the combustion chamber temps and still have a wider band of time than the spark plug. Plus, where would you measure it? If you kept your stock oil temp gauge you'd need to tap in somewhere to get oil sump temps.

So for all these reasons, I prefer the engine tin screw for EFI input, and the spark plug (or maybe the CHT boss?) for true CHT monitoring.

GA

The MS doesn't care if you use Mario's sender or the one I'm using or if under engine tin or in CHT location. You just need to adapt the WUE settings and it will run fine.

Here before engine start:
Click to view attachment

And here warmed up:

Click to view attachment

You need to look for coolant temp.

This correlates up the highest temps I have ever seen 183°C
Just if you push it really hard like WOT uphil the under the spark plug sender will show a max of up to 4°C more for some time and then the other one will catch up. You never miss more like this 4°C in spikes...

It's working absolutely flawless.

But again, I'm not using the OEM heads!

I totally agree, I use the Micro 1000 CHT, besides the quality it looks a lot like a VDO gauge. For CHT I won't use anywhere but under the plug. The Djet sender, like my SDS uses, is just a rough ballpark for warm up.

I have the under the spark plug thermocouple at cylinder #3 anyway, just saying it is reading exactly the same values as the other sensor...

I am probably going to MS my engine over the winter and this thread has been super educational. Thanks everyone!