Cylinder Head Temperature

[germansupplyscott]

[Wildthings]

The length of the wires of a thermocouple do not matter, excepting a very very tiny loss do to resistance. You can make the distance between the two joints any distance you want within reason without changing the calibration of the gauge. What you can't do is splice in a section of copper (or other) wire to make the leads longer, as each connection you add is another thermocouple operating at an unknown temperature.

The thermocouple is the junction between the two wires, it would not be the ring terminal unless the ring terminal is made of two different metal plies and the wires are attached to each ply separately.

On heads that have the boss for the early TSII sensor, that boss might make a good location for attaching a smaller ring terminal with a screw. It should give close to the same readings as a ring terminal under the spark plug.

[rustbus]

now if there was a way we could get a reading FROM the TSII that would be slick, as my engine is ex-FI and still has the TS2 in it...
_________________
May of '72 Deluxe. 2.0L L-Jet CS & 091 trans conversion
my Bus thread

[fusername]

if you cut off the ring, what is really important ray is that you maintain the factory crimp, as it is the junction of the two wire ends that REALLY matters. they must be joined with no impurities, no corrosion (harder than you think) and EXCELLENT contact. this is often done in a vacuum or special atmophere, and with extreeme pressure clamps. you can cut off the ring, just don't touch that factory crimp!
I can't see how jb weld would affect the reading at all, it MAY cause the guage to read slower due to poor thermal conductivity, however I don't know that for a fact, maybe JB weld conducts heat really well. I have been trying to find a epoxy that is designed to conduct heat, but have yet to turn it up. what is important is that the wires are not too exposedto the air flow, as they will be cooled by the air flow over them, which will wick heat away from the junction, and you will be reading this cooled temp rather than the head temp itself. actually potting the couple end in JB, while still using the ring under the spark plug would probably be the most accurate technique possible, imho.
_________________
[email protected]
Need something custom bent up? shoot me an email, maybe we can make it work!

FORSALE: Thrust cut T4 and 1.9 main bearings

[busdaddy]

[raygreenwood]

Thats exactly what I said...it is NOT OK to cut the wires.

Also...unless you have checked...you have no idea whether your gauge is now only a few degrees off. This is all about conduction and radiation. The ring terminal pulls in and concentrates heat at the TC point at the ends of the wires. Without that mass...you have no idea of what it may do to the accuracy of the TC joint.
Why? because a TC is not just an open to the air joint. The two metals must be intimately connected. Factory high pressure crimping....forms almost a hermetic bond.
Now.....yes it would be much better to leave the crimp on....but you still have no idea how much losing th ring lost you.

Alot of this is seat of the pants calibration. What everyone is failing to understand...is that heat rise and heat banking are not linear across every tempereature range. So its like this.... you try it with the ring under the spark plug and drive it down the road...maybe you read 300F. So you snip off the ring and glue it down and now you read 305-310 and figure..."good enough for me". Did you bother to map this across upper range (above 300F) and middle range and mximum to see if the minor change was linear? The problem is....that in many cases it may not be...and it can be grossly off.
If you are off 25F...thats enough to fry your heads...when you think you are border line at 375 and motor on....and you are really at "seat dropping" 400F.
I could care less about a few degrees...I do care about reliability across the range.


If you are using your head temp gauge for "trending"....IMHO...you would not be buying products from dakota digital. You would be buying a VDO or similar product. DD gauges are designed for greater accuracy.

Wildthings...you are correct and incorrect at the same time. Yes...you can extend milivolt generating TC's ...if you know the wire composition. But....its dangerous to do so. Its generally not the added length of wire and any resistance properties that are at issue. Its the connections in the splice and the connectivity....every single time.....that screw you. There is no crimp system or connector systemthat can guarantee a clean splice in a thermocouple wire. Thats why whatever crimped terminals or splices there are...are calibrated with the wire as an integrated unit. Ray

[WhirledTraveller]

Hmmm, Ray, I don't really know how to say this politely but I'll try. You are putting a lot of confusing and frankly incorrect information out there.

Thermocouples are not calibrated items in the way that, say, a micrometer is. They are two wires joined together. The accuracy and reliability of the thermocouple itself is mostly a function of the quality of thermocouple wire used, and to a lesser extent the quality of the joint. They are typically not so much calibrated as just checked, at least for general use like we are talking about. Now in my case I made my own thermocouples using high quality SLE (special limits of error) wire, but made a rather crude weld which probably contaminated the joint just bit with some of the metal from the battery post. I'm not that worried about it, it might change the reading a few degrees at most. If the joint is really bad, or the wires in a crimp get separated, you will get no reading at all or an intermittent reading. It won't "drift". Either the wires make electrical contact or they don't. You can simply twist the wires together and the thermocouple will work fine, with perfect accuracy, but probably won't last long. You can even put a different metal in between the wires. This is known as the "Law of Intermediate Metals". For example I could attach one wire to one cylinder head and another wire to a different cylinder head, and get a reading on the gauge which would be an average of the two temps. Not a bad idea, actually!

"Factory" thermocouples from DD or VDO or anybody else are *not calibrated* period. What would they adjust? They are assembled, and, if you're lucky, checked.

Now some of the stuff you have said, really, I must say is truly B.S.

The ring terminal does not "pull in and concentrate heat". That is engineering nonsense. It is made of copper and conducts heat, that's all. We are interested in the temperature of the head, so it is actually more accurate to put the thermocouple in direct thermal contact with the head, rather than through a ring-terminal intermediary.

You do not need any sort of mass attached to a thermocouple to make it accurate. Fine wire thermocouples can be practically microscopic and will respond instantly to any temperature change. Attaching a mass such as a ring terminal does not make the thermocouple more accurate, it just makes it respond slower.

As for "heat rise and heat banking" these are not scientific terms. I have no idea what you are talking about. This is a simple problem... what is the temperature of the cylinder head? A thermocouple placed in direct contact with the cylinder head will be at the same temperature as the cylinder head. It will generate a millivolt signal proportional to the difference in temperature between itself and the cold junction. That millivolt signal will travel through the wire (ANY length of wire, you CAN cut the wire), into the gauge, where a good gauge such as a DD or EMS1 will perform a conversion according to standard thermocouple reference tables, apply the cold-junction compensation if necessary, and display the temperature. This process is remarkably accurate, typically within a couple degrees. There are only a few things that can go wrong:

1st, as I said, the metal the wire is made out of is important. If the metal is chemically contaminated it may change the properties of the thermocouple. This is usually pretty minor.

2nd, since the signal is so low, the wire must be reasonably protected from electro-magnetic interference

3rd, any electrical connections need to be solid, again since the signal is so low.

4th, You want to *avoid* using non-thermocouple wire to extend the leads. However this is not an absolute prohibition. In fact the VDO sender is only 1 foot of thermocouple wire and the rest is normal wire. What this does is to move the "cold junction" to that point of splice. Now, that cold junction acts as it's own thermocouple and will throw off the reading if that junction is at a different temperature than the gauge (which is where the cold junction compensation takes place). For VDO they don't care because they never bothered with cold junction compensation to begin with! They just assume the cold junction is room temperature.
_________________
1977 Westy, Automatic. Big Valve heads, CS Cam.

[busdaddy]

[SGKent]

WT -

Q. Where is the cold junction electrically? Is it where the wires join the pins in the connector, the full length of the non thermocouple wire or at the head unit? I understand that the different electrical properties generate the signal, however would the signal from say a 50 foot long thermocouple wire twisted together be the same as a 1" long one twisted together at any given temp? I am trying to understand the process. If the connection to the pin in the connector (different metal) helps determine the signal, then is the temperature I am getting on the gauge nothing more than "indicated temp X" over that junction point? Or is it "X" degrees over the average temp along the run from the front to the back, or "X" degrees over the head unit etc? What I am saying in a way is that if the thermocouple and the cold junction were the same temp would there be no signal and where is the cold junction on the VDO?
_________________
"Most people don't know what they're doing, and a lot of them are really good at it." - George Carlin

[Wildthings]

[WhirledTraveller]

[fusername]

Yes on the ends matter. if you have a good guage hooked to the end of a thermocouple, and have the sensing end at 100 deg, the guage at 0, and the middle of the wire at 200, it should read 100 deg. only the difference in temp between the endpoints matters. A good guage knows its own temp and based on the voltage figures out the temp of the other end by adding or subrtacting it from its own temp. Any autmotive guage and most guages used anywhere else use voltage to detect the temp difference, as it is constant for one wire type. A guage that sense by current can vary due to wire length and must be calibrated for every isntallation, and takes much finer equpitment. We would never see such a thing in a dash guage.
The voltage generated by the heating of the far end does not suffer from voltage drop over the lines, becasue the guage looks to the wire as an open connection, as its resistence is rediculously high, to the point of being an effective infinity. However the cheaper the guage, the lower the resistance, most likely. Now you have unpredictable voltage drop which simply turns up as error. Another issue You can have with a cheap guage is the onboard electronics can start generating heat. They generate enough heat to mess with realative readings, but not enough to heat up everything inside the guage, meaning one half of the electronics can be many degrees over that of its neighbors due to poor thermal conductivity, so even tho the guage has cold junction compensation, its not compensating for the temperature of the end of the thermocouple, but of one part of the board. This here is splitting hairs, but intersting to those who plan to build thier own guages from scrap, or who just like knowing a bit too much!

Now the DD guage uses a two piece thermocouple wire as well, using a special "thermocouple exentions wire" I don't really know what this means. It is some special wire, I am not sure if its cheap thermocouple wire, or just high grade copper, or what, as it would move the junction, just like was said before, but DD says all thier guages are CJC, so this some how must not effect it, however they may simply be assuming the enging compartment, or wherever the junction i, approx 8 inches from the ring, is going to be at the same temp as the guage, a terrible assumption. This is somethign I need to do a bit more looking into, I have yet to figure that one out.
_________________
[email protected]
Need something custom bent up? shoot me an email, maybe we can make it work!

FORSALE: Thrust cut T4 and 1.9 main bearings

[babysnakes]

Wow, this was a great read. Although I was planning to install a CHT before I installed the motor anyway,understanding how they work and what to and not to do will ensure the best install. Thanks for starting this thread Skidoo

[skidoo]

Really i'm really happy to learn !!! It's interresting to exchange about eveybody experience !!!

I like to thanks everybody who help me about my question !!! i'm new on the Bus world and it's great to see that the AIR COOLED people are so helpfull and give time to other one for helping !!!

See you somewhere on the road !!!!

Jeff

[skidoo]

Just a last question:

What is the normal temperature range for the cylinder head, at what temperature should i become worry about having trouble ???

[Randy in Maine]

Stock valve seats likel to drop out about 450º F.

I would not let it get much above 400º F.

"Improved" heads can take a bit more, but don't bet the farm on it.

[WhirledTraveller]

[rcnotes]

I went for a drive today and couldn't get head temps above 380.

That was at about 80 mph and 35 degrees F.

Man, talk about wind noise!

[GusC2it]

If it will help anyone, I cut my ring off, not the crimp, and wedged it in between the fins, at their base, making sure of good thermal contact, and using JB weld only to hold it in place. It works fine, but here's the thing.
I chose the exhaust side of the spark plug, not thinking that the plug base, where the normal ring goes, is in between the cool intake valve and the hot exhaust valve. My sensing point is much hotter, therefore, I can't compare my readings with those who use the plug ring.
So when I get a chance, I will relocate my sensor closer to the plug base.

I agree with BusDaddy, that any head temp gauge is really a trend indicator, because temperatures vary greatly at different points on the head. (so its difficult to compare with other engines)
IMHO, this is the only positive reason for using a plug ring.
The gauge needs to be calibrated in our heads, the head on our shoulders, , by observing under different driving conditions.
_________________
75 type 2 http://gusc1.tripod.com/1975vwtype2
Old vices have now been upgraded to bad habits.
http://www.facebook.com/pages/VW-Wild-Weekend-2011-St-Pete-Fl/197064760336111

[jackstar]

Ok, after reading some interesting posts I decided I needed to go to the "horse's mouth" so to speak. I called JB Weld and asked them if their product would conduct electricity. I explained the application and they said that JB Weld would NOT conduct electricity. However, that is not to say you could not take the ring terminal and cut it in a half moon shape, place the metal from the terminal in the pre-cut groove and have the half moon shape fit around the plug hole and have it contact the head and JB Weld over that. It would leave the metal from the terminal in contact with the head and the JB Weld over that. In theory, it should work.

Then I called DD and talked to tech assist. They did say that the end of the lead is special. How special he could not tell me. He directed me to the engineers. They didnt have time to talk. The replacement lead from the 2 philips head screws to the terminal is 7.50 plus shipping. $16 total to NW Florida. I ordered one.

I dont know if this helps or not. I would like to get an answer from the engineers. I will try later. Just my .03 worth.