What is the actual temp?

[k belknap]

Okay, I've got the stupidest question ever! On my '81 Vette the temp guage starts at 100 degrees and ends at 260. The number 200 is stamped at the "12 oclock" position. However, there is an exaggerated mark at the "11 oclock" position. Is the exaggerated marking really the 200 degree mark, or is straight up 12 oclock the 200 degree mark? I've asked 5 fellow employees their opinions and they all differ. While I'm here...my car runs very hot (about a mark below the red). Would a bored out block cause my car to run hot no matter how well the cooling system is functioning? My feet continue to burn :~) !

Thanks

K

[Jack H.]

Re: What is the actual temp?

Several threads in archieves on 'hot' Corvettes. Guage faces are intentionally vauge as there were a number of changes in both guage as well as temp senders over the years and system isn't really that precise once original gauge and/or sender have been changed out.

Footnotes in major parts catalogs warn of +/- 30F accuracy on current senders!!!! Give away for genuine problem(s) is constant popping of rad cap and loss of coolant and/or confirmation of high temp guage reading with external precision themometer....

[Jack H.]

Re: What is the actual temp?

Several threads in archieves on 'hot' Corvettes. Guage faces are intentionally vauge as there were a number of changes in both guage as well as temp senders over the years and system isn't really that precise once original gauge and/or sender have been changed out.

Footnotes in major parts catalogs warn of +/- 30F accuracy on current senders!!!! Give away for genuine problem(s) is constant popping of rad cap and loss of coolant and/or confirmation of high temp guage reading with external precision themometer....

[Fred Oliva]

Re: What is the actual temp?

The confusion in the graduations on temperature is actually due to the non-linear characteristics of the temperature sender. By this I mean that the resistance of the sender does not decrease at the same rate over the entire temperature range. Taking a specific example, the sender resistance may change 10 ohms when the temperature changes from 100 degrees F to 105 degrees F, but may change only 1 ohm when the temperature changes from 200 degrees to 205 degrees. I use these numbers as an illustration & not as actual performance fugures for the sender.

Now, the temp gauge is actually an ammeter measuring current thru the sender. This type of gauge is inherently linear, ie, for a given amount of current change, the same additional needle deflection will occur no matter how much current flowing thru the gauge or where the needle is pointing. Naturally, this statement applies within the rated current range of the gauge.

The problem is how do you make a non-linear sender work with a linear gauge? The solution is make the graduations on the gauge non-linear to make it compatable with the non-linear sender

That's the theory. In practice, a reasonable approximation is to assume that the graduations increase linearly from the lowest reading to the middle reading, and the same for the middle reading to the max reading. What I mean by this is assume that if the lowest reading is 100 deg & the middle reading is 180 deg, the graduation in the middle of these 2 is exactly half. The same would be true for the graduation between the middle reading and the max reading.

With this said, I would determine where the 11 o'clock reading falls between the lowest graduation & the middle graduation. If it's about 75% of the distamce, then the graduation is around 75% of the difference between the 2 graduations.

This technique is known as a piecewise linear approximation of a non-linear curve. In other words, approximate a curve with a bunch of straight lines.

With all this said, you must also consider Jack's comments. There is evidence that the factory attemped to match the sender & gauge to achieve a resonable accuracy. This was done because of the large variation in production senders. This is true of both the original senders & the ones you get today. So, unless you have the origial sender with the original gauge, it's difficult to know exactly if your gauge is reading correctly. I suggest the use of an infra-red thermometer & take readings around the sender to get an idea of the accuracy of your gauge-sender system.

As far as your car running hot, this has been the subject of much discussion & you can check out these threads in the archives.

Hope this helps.

Fred Oliva

[Fred Oliva]

Re: What is the actual temp?

The confusion in the graduations on temperature is actually due to the non-linear characteristics of the temperature sender. By this I mean that the resistance of the sender does not decrease at the same rate over the entire temperature range. Taking a specific example, the sender resistance may change 10 ohms when the temperature changes from 100 degrees F to 105 degrees F, but may change only 1 ohm when the temperature changes from 200 degrees to 205 degrees. I use these numbers as an illustration & not as actual performance fugures for the sender.

Now, the temp gauge is actually an ammeter measuring current thru the sender. This type of gauge is inherently linear, ie, for a given amount of current change, the same additional needle deflection will occur no matter how much current flowing thru the gauge or where the needle is pointing. Naturally, this statement applies within the rated current range of the gauge.

The problem is how do you make a non-linear sender work with a linear gauge? The solution is make the graduations on the gauge non-linear to make it compatable with the non-linear sender

That's the theory. In practice, a reasonable approximation is to assume that the graduations increase linearly from the lowest reading to the middle reading, and the same for the middle reading to the max reading. What I mean by this is assume that if the lowest reading is 100 deg & the middle reading is 180 deg, the graduation in the middle of these 2 is exactly half. The same would be true for the graduation between the middle reading and the max reading.

With this said, I would determine where the 11 o'clock reading falls between the lowest graduation & the middle graduation. If it's about 75% of the distamce, then the graduation is around 75% of the difference between the 2 graduations.

This technique is known as a piecewise linear approximation of a non-linear curve. In other words, approximate a curve with a bunch of straight lines.

With all this said, you must also consider Jack's comments. There is evidence that the factory attemped to match the sender & gauge to achieve a resonable accuracy. This was done because of the large variation in production senders. This is true of both the original senders & the ones you get today. So, unless you have the origial sender with the original gauge, it's difficult to know exactly if your gauge is reading correctly. I suggest the use of an infra-red thermometer & take readings around the sender to get an idea of the accuracy of your gauge-sender system.

As far as your car running hot, this has been the subject of much discussion & you can check out these threads in the archives.

Hope this helps.

Fred Oliva

[Jack H.]

PS

Looking at GM drawings for various temp senders, we see evidence there were changes over time to the T/R curve loci used by AC for accept/reject. Exactly how the pre-64 T/R curve compared to the post-64 and post-68 curves is information that's lost. This might help to explain the +/- 30F accuracy spec mentioned in C-Central and Paragon catalogs on the AC sender parts they sell.

While there the T/R response curve consists of a loci of min/max lines to define acceptable tolerance, the fact that curves WERE changed over the years adds to total deviation based on what MY system the sender is to be installed on. Doesn't necessarily mean AC builds 'shoddy' and inaccurate temp senders today....

[Jack H.]

PS

Looking at GM drawings for various temp senders, we see evidence there were changes over time to the T/R curve loci used by AC for accept/reject. Exactly how the pre-64 T/R curve compared to the post-64 and post-68 curves is information that's lost. This might help to explain the +/- 30F accuracy spec mentioned in C-Central and Paragon catalogs on the AC sender parts they sell.

While there the T/R response curve consists of a loci of min/max lines to define acceptable tolerance, the fact that curves WERE changed over the years adds to total deviation based on what MY system the sender is to be installed on. Doesn't necessarily mean AC builds 'shoddy' and inaccurate temp senders today....

[Jack H.]

PPS

Off-line discussions indicate Fred Oliva has spent considerable R&D time addressing this accuracy issue. Apparently he's built a 'ding ding' box available for rental that can be used to fault isolate your car's temp system (sender, temp guage) and allow you to find the root cause(s) of temp reporting inaccuracy and let his firm 'custom' build/repair the affected items to put your car's guage ON THE MONEY. Might want to contact Fred individually....

[Jack H.]

PPS

Off-line discussions indicate Fred Oliva has spent considerable R&D time addressing this accuracy issue. Apparently he's built a 'ding ding' box available for rental that can be used to fault isolate your car's temp system (sender, temp guage) and allow you to find the root cause(s) of temp reporting inaccuracy and let his firm 'custom' build/repair the affected items to put your car's guage ON THE MONEY. Might want to contact Fred individually....

[Joe L.]

Re: which cam is correct ?

Bill------

The GM #3904362 camshaft was not used until 1967, at the earliest. The GM #3863143 was used in big block SHP applications from 1965-66 and, perhaps afterwards.