Printable Version of Topic

Click here to view this topic in its original format

914World.com _ 914World Garage _ Oil temp vs. cylinder head temp question

Posted by: Bill Shaw Apr 13 2005, 07:56 AM

Just put a new VDO cylinder head temp gauge in during the off-season (after getting scared that the oil temp gauge was off, though the car runs like it's new). After driving it around for a couple of days I have the following results (with Mobil 1, 15W50, in a '74 2 liter):

Oil temp on factory dash gauge - barely reaches the "T" after full warm up and no matter what rpm I'm at.
Cylinder head temp - around 300 driving around town (never gets beyond 325, even in traffic, such as it is in Kalamazoo); around 400 on the freeway @ 3200 in fifth at 75-80 mph

I thought the oil temp and cyl. head temp would be roughly equivalent, at least as far as I can figure from the past threads. The big difference has me stumped. Anybody have any explanation? Am I going to hear any expensive noises soon? What numbers should I be looking for?

Posted by: tat2dphreak Apr 13 2005, 08:09 AM

heads get signifigantly hotter than the oil should... and those head temps look high...

Posted by: Jake Raby Apr 13 2005, 08:36 AM

These can vary and are NOT dependant upon each other! Thats why I recommend a gauge to monitor each- nice to see you have one of each.

Here are my tolerances for temps:
Oil=
180-225 is normal
225-235 Very warm and will thermally breakdown oil in 1500 miles (dino oil)
235+ HOT- Will kill idle oil pressure and erode rod bearings and break down oil in 1K miles

Heads=
275-325 VERY good for flat running with short bursts to 350+ while pulling a hill. About the only time you'll see these temps is with a non stock, more efficient engine. Stockers with injection seldom run this cool.

325-350 good for flat running withstock D Jet with spikes to up to 375 while pulling a hill.

350-375 is warm for flat running and may indicate some tuning issues, bad engine configuration or a cooling system thats missing pieces, etc.

If your engine ever reaches 400 degrees while flat running- Its' hot and will probably go to over 425 climbing a hill. Stock valve seats drop out at about 425 so remember that. A good set of heads with modern seat material will take up to 475 sustained.

As soon as your heads go over 400F you'll need a valve adjustment, as the extra heat will more than likely stretch some valves and could move some seats.

So, your 400F head temp is WAY high and shows that you need some tuning before you kill your heads.

Remember that:
Oil temp is directly impacted by RPM. RPM = friction, friction = heat and it's the oils job to remove that heat. Oil temps are slow to resond to changes due to all the oil having to attain temp before seeing differences.

Head temp is directly related to load and tuning. It responds quickly to change and can vary by 200 degrees in just one mile of normal driving. The faster you go the more the aerodynamic wall forces the engine to work, especially in higher gears. In upper gears engine RPM is less while load is greater and this builds head temps. I often call 5th gear in a 914 "The head killer" because so many people lug the engine at a low RPM while only going 60 MPH. This increases load terriably while killing fan speed and cooling system efficiency. Thats what cracks all those pricey 2 liter heads!

With that being said I'll say that my DTM in the 914 on a 170HP engine with 10:1 CR has NEVER seen head temps over 325 degrees! With the stock cooler removed and only a full flow external oil cooler the oil temp has never hit 205 degrees yet after 3500 miles of driving.

Stock engines work hard with heavy throttle positions because they don't make much power for their displacement. Due to this the hottest temps I have ever seen have came from stock engines, especially those out of tune.

Now for the crazy part- My 3 liter TIV with 235 HP runs head temps of 300 running flat @ 80 MPH and has spiked to 350 once! Oil temp with a very small external Mesa cooler thats really considered inadequate for the huge engine has never been over 230!

Sorry for the long post, but after spending about 3 solid years testing cooling systems and etc I have been consumed by the numbers and their sources...

Posted by: Tom Perso Apr 13 2005, 08:50 AM

Bill...

I just picked myself up off the floor. I live in Kalamazoo, North East side.

Those head temps do sound high. Any blocks or obstructions in your cooling?

Maybe you need some tuning. PM me or email me and I would be happy to help you out!!!

Thanks
Tom

Posted by: phantom914 Apr 13 2005, 08:52 AM

How do you know that the gauge is accurate?

Andrew

Posted by: Joseph Mills Apr 13 2005, 12:34 PM

QUOTE (Jake Raby @ Apr 13 2005, 08:36 AM)

Sorry for the long post, but after spending about 3 solid years testing cooling systems and etc I have been consumed by the numbers and their sources...

Sorry? Get outta'here!

Thanks for the very insightful and informative post.

I have an accurate oil temp gage - now I've got to decide - do I get a head temp gage first, or the external oil cooler next (which I know I need).

..

Posted by: BMartin914 Apr 13 2005, 01:41 PM

As far as a cyl. head temp SENDER, what would any of you recommend: lead that goes under the spark plug, or a lead under a head bolt (both on #3)?

Posted by: type47 Apr 13 2005, 01:50 PM

cylinder head temp ring goes under the #3 spark plug

Posted by: 7391420 Apr 13 2005, 01:51 PM

Good to hear that there are a few 914's in Kalamazoo, I went to school at K college, graduated in 98, and come back annually to see friends!

-I have what sounds like the same head gauge in my 73 2 liter, your temp's do sould high, my engine is basically stock, but it's carbed, and I average around 350 on the highway. Around town, even in the summer I stay in the 325 range, the highest I've seen would have been on a 300 mile drive last summer, all highway, and I peaked at 370 or so.

-What do you have for a sending unit-is it matched to the gauge? if not you're likely getting off readings.

-Adam

Posted by: brer Apr 13 2005, 01:52 PM

I've seen several posts about head temps and no one has mentioned that these guages, or should i say the VDO ones I have always used, read a temp. number that is relative to the outside temp.

Your heads may be at the same temp everyday, but depending on the outside temp your guage will read differently because of the type of sender it uses.

Posted by: Jake Raby Apr 13 2005, 01:56 PM

VDO gauges are calibrated at 76 degrees for every degree of outside temp hotter than 76 the gauge will read 1 degree cold. For every degree below 76 the gauge will read one degree hotter..

ONLY install the sender under the sparkplug, NO WHERE else is accurate and the absolute worst place for it is the position of the stock CHT switch- It was not designed to read a given temp, but more to say "hot or not"

Dyno proven.

Posted by: Joseph Mills Apr 13 2005, 02:17 PM

So if it's 96 degrees and your temp gage reads 200, the true temp is 220?

Posted by: Jake Raby Apr 13 2005, 02:19 PM

Yep, thats the way it was explained to me by a tech at VDO. Because of this I run westach instruments and their thermocouples.

Posted by: Bill Shaw Apr 13 2005, 03:52 PM

What a great fountain of information! Thanks, guys!!

Tonight I'm gonna take the sender off the plug (yep, I did put it on no. 3) and immerse it in boiling water to see if the gauge reading reading is somewhere near accurate. It is the sender that came with the gauge, so I think it should match.

I know I have one piece of tin on the bottom of the engine that is missing, but it's on the side opposite the sender, so I assumed it wouldn't affect the reading I am getting. I'm certainly no colling expert, though, so I'll start looking for a replacement and hope that'll help.

Just got back from a 100 mile round trip on the freeway, with air temps in the high 50s. On the flat and going up long, but gentle hills, the gauge topped out at 400. Downhill and in construction traffic it dropped to about 300-325.

By the way, the engine is, at least as far as I know, a bone stock 2 liter with about 100,000 miles on it. It still pulls strong to 5700, though.

Looks like I'll be in the market for one of Jake's DTM setups when he gets the ones for 14ers ready to go. Hope it's before the Michigan summers hit!

Posted by: lapuwali Apr 13 2005, 04:05 PM

QUOTE (Jake Raby @ Apr 13 2005, 12:19 PM)
Yep, thats the way it was explained to me by a tech at VDO. Because of this I run westach instruments and their thermocouples.

Are Westach gauges temperature compensated? The VDO "problem" is a function of how thermocouples work. Unless there's some active compensation for ambient temperature and how it varies from the calibration standard, any thermocouple setup will have the same problem. Note that if the temperature is BELOW the calibrated value, the gauge will read HOT. So, if it's 20dF outside, a cylinder at 300dF will show 350 on the gauge.

Thermocouples work by showing the difference in temp between two ends of the thermocouple pair. One spot under the plug, and the other spot at the other end of the thermocouple pair. On the VDO senders, the thermocouple pair is only 12-18" long, so the "cold" end of the pair is still in the engine bay. Since the VDO gauge is calibrated under the assumption that the cold end of the pair is at 70-76dF (I've heard 70, Jake has heard 76), the gauge will always read cold in a 914, as the ambient temp at the cold end of the pair will usually be around 120-130dF.

There are some that contend that it's the temp of the gauge, not the temp of the cold end of the pair, that determines the "cold" calibration temp. I don't know for certain either way.

If the Westach gauges have a second temp sensor to correct for ambient temp errors, that's a wonderful thing, given that they're no more expensive than the VDO units.

Posted by: Jake Raby Apr 13 2005, 05:01 PM

QUOTE
Tonight I'm gonna take the sender off the plug (yep, I did put it on no. 3) and immerse it in boiling water to see if the gauge reading reading is somewhere near accurate.

Remember:
Water only bolis at exactly 212F at sea level.

As for the Westach thermocouples. I have used both gauges side by side and I would say that the westach is not as picky on the ambient temp as VDO.
The tale will be told when I receive my datalogger with fool proof (and super expensive) SPA digital instruments and test them all against the logger results on the same engine at the same time....

Posted by: guywan914 Apr 13 2005, 07:34 PM

Jake, we definately all appreciate your input and knowledge. Can,t wait to hear the further results of your testing.

Posted by: Bill Shaw Apr 14 2005, 01:58 PM

Tested the sender and gauge last night with boiling water and it looks like the gauge is pretty accurate. Guess that means I've got some tuning and valve adjustments to do over the weekend - and the yard cleaning will just have to wait.

Glad to hear there are some folks who know where Kalamazoo is; those of us living in the middle coast can feel a little left out sometimes except, of course, for the Midwest Classic weekend!

Posted by: tat2dphreak Apr 14 2005, 02:18 PM

QUOTE (Bill Shaw @ Apr 14 2005, 02:58 PM)
Tested the sender and gauge last night with boiling water and it looks like the gauge is pretty accurate. Guess that means I've got some tuning and valve adjustments to do over the weekend - and the yard cleaning will just have to wait.

Glad to hear there are some folks who know where Kalamazoo is; those of us living in the middle coast can feel a little left out sometimes except, of course, for the Midwest Classic weekend!

I know where Kalamazoo is... my 2056 came from there

or very close... right Tom?

Posted by: Tom Perso Apr 14 2005, 02:35 PM

Yup, that 2056 came right here from Kalamazoo...

Got that thing "DRIVING" yet? I know it run!!!

Later,
Tom

Posted by: phantom914 Apr 14 2005, 02:59 PM

QUOTE (lapuwali @ Apr 13 2005, 02:05 PM)

QUOTE (Jake Raby @ Apr 13 2005, 12:19 PM)
Yep, thats the way it was explained to me by a tech at VDO. Because of this I run westach instruments and their thermocouples.

Are Westach gauges temperature compensated? The VDO "problem" is a function of how thermocouples work. Unless there's some active compensation for ambient temperature and how it varies from the calibration standard, any thermocouple setup will have the same problem. Note that if the temperature is BELOW the calibrated value, the gauge will read HOT. So, if it's 20dF outside, a cylinder at 300dF will show 350 on the gauge.

Thermocouples work by showing the difference in temp between two ends of the thermocouple pair. One spot under the plug, and the other spot at the other end of the thermocouple pair. On the VDO senders, the thermocouple pair is only 12-18" long, so the "cold" end of the pair is still in the engine bay. Since the VDO gauge is calibrated under the assumption that the cold end of the pair is at 70-76dF (I've heard 70, Jake has heard 76), the gauge will always read cold in a 914, as the ambient temp at the cold end of the pair will usually be around 120-130dF.

There are some that contend that it's the temp of the gauge, not the temp of the cold end of the pair, that determines the "cold" calibration temp. I don't know for certain either way.

If the Westach gauges have a second temp sensor to correct for ambient temp errors, that's a wonderful thing, given that they're no more expensive than the VDO units.

The temperature of the "cold" end of the thermocouple junction is what is important as far as the compensation for ambient temperature. But, depending on design, the gauge itself can also introduce error if it is not compensated for temperature well.

So in other words, the ambient temperature of both the gauge and the gauge end of the thermocouple wire are important. The effect of the ambient temperature on the gauge end of the thermocouple wire (junction) is predictable and can be compensated for, but the effect of ambient temperature on the gauge itself would be different for each design.

In my work with thermocouples, an IC temperature sensor was in the measurement device near the junction to measure its temperature in order to compensate in software. Any meter you buy will need to do this to be accurate and the gauge electronics itself will also have to be consistent under varying temperatures.

Andrew

Posted by: tat2dphreak Apr 14 2005, 03:00 PM

QUOTE (Tom Perso @ Apr 14 2005, 03:35 PM)
Yup, that 2056 came right here from Kalamazoo...

Got that thing "DRIVING" yet? I know it run!!!

Later,
Tom

I need to put the seats in and it will drive... I am going to paint the floor this weekend I think... I got sidetracked with rust

other than seats, all I need is the normal legal hurdles.. insurance,inspection, and registration...