PDA

View Full Version : Running Hot


97COB
05-03-2006, 10:22 PM
I Have A 97 Cobra 4.64v. Runs Hotter Than Id Like (at The A In Normal). I Have Installed New T Stat, And Sending Unit. Any Ideas?

SNAKEYE
05-04-2006, 09:45 PM
I Have A 97 Cobra 4.64v. Runs Hotter Than Id Like (at The A In Normal). I Have Installed New T Stat, And Sending Unit. Any Ideas?
Welcome to the world of hot Cobras! Actually they are supposed to run like that. The stock t-stat is 195F and the sending unit controls the radiator fan through the the ECU and the CCRM to turn the fan on at 221F and off at around 200F. And (believe it or else) the fan is turned off at speeds above 45 MPH, except when the A/C is on. And nobody will defend the stock temperature gauge as being an accurate instrument.
SNAKEYE is a second car for me, a toy if you will, that I have never had on a drag strip, road course, or even a dyno. Just a fun driver. I have owned SNAKEYE for 5 years. I bought it from my son (with 20K on the clock) who bought it new and equipped it with a Vortech S/C (non-intercooled) right off of the car hauler. He warned me about it running hot. What an understate-ment! A couple of times the gauge almost got to the red "H" at the end of the gauge.
I'm from the old school where I've always felt gauges should normally run halfway across the scale, or a bit more in extreme conditions.
Over the years I have tried almost everything: Water Wetter, Evans high-capacity coolant pump, radiator fan high speed override, auxiliary pusher fan in front of the A/C condenser, a 160F t-stat, and replaced the stock radiator with a two-core . Each of these mods helped, some more than others. The two-core with the 160F an d the fan override operating together do a great job.
After saying all af that there is one more thing I am currently in the process of testing out. Over the years I have read many other owners woes (like you) and have analyzed my own situation, boy have I analyzed it! This past winter I had an idea after I reading people's ideas about putting in chips or having their ECU re-programmed to lower the operating temps of the radiator fan. So I researched the service manual and studied how the cooling system is supposed to work and what makes it work. In the process of this research I found some really technical information on the "sending unit", or more precisely the ECT (engine coolant temperature) sensor. It is what is know as a thermistor, a device whose resistance varies non-linearly, or more precisely its resistance goes down as its temperature goes up. The service manual even had a chart of temperature versus resistance. From this chart and an explanation of how it operates to control the radiator fan being turned on at 221F and off at 200F, I applied electronic theory for two resistors in parallel (the total resistance of a pair of identical resistors is 1/2 of that of either). I took the temperature vs. resistance info from the service manual and plotted it on a curve. Then I plotted 1/2 the resistance values throughout the temperature range and found that the fan-on temperature should drop to 180F and the fan-off temperature should drop to 160F. Well, at least theoretically anyway.
So to apply the theory I bought a second ECT sensor and a stock plug connector from Ford, and a tee connector that I could screw into the crossover tube and then screw the two ECT sensors into it, then wired the second sensor in parallel with the original sensor's wiring. I did all of this over this winter while SNAKEYE was in hibernation, and just brought the car out for the season this past weekend and have just completed two days of road tests (driving to work): first with the second ECT sensor unplugged so that the system would function as stock, then with the second ECT sensor in parallel per my theory. I ran the same routes (60 mile roundtrip) to and from work both days with abient temps in the A.M. in the 45F-50F range, and home with ambient temps in the 72F-75F range. Driving speeds were from first gear bumper-to-bumper rush hour to sustained 65+ MPH on the freeways.
The results have just blown me away!
With just the one ECT sensor at 45F ambient and 65 MPH, the gauge was right inbetween the "R" and the "M". At 72F ambient and 65 MPH the needle was just touching on the "L". Then after exiting the freeway and running in local traffic on streets with stoplights, the gauge went to the far side of the"L". All about as I suspected after years of having to operate my high speed fan override system under similar conditions.
Now here's what blew me away. The next day, with both ECT sensors plugged in, over the same course with an A.M. ambient at 50F and the P.M. return ambient at 73F, the gauge just touched on the low side of the "M" and didn't vary more than a needle width for the 60-mile round trip!
This weekend I am going to get a freind to plug his scantool into my OBDII connector and read the exact fan-on and fan-off engine temperatures.
I know this has been a long reply. I hope it will be of help to those who suffer the same problem. I have been preparing a writeup on this with much more detail, pictures, part numbers, step-by-step instructions, and have been waiting to get this weekend's scantool readings of the temperatures before completing it. I was going to post an offer to all who might want the writeup to be able to do-it-themselves next week so as to be able to e-mail me for an e-mailed reply of the writeup. This thread kind of got me going a bit early. I believe all who are interested should be able to e-mail me via the forum with their e-mail address for me to reply with the paper attached.
That's about it for now.

97COB
05-04-2006, 10:43 PM
BIG HELP! SPENT THE DAY THINKING ABOUT IT AND DECIDED THE NEXT BEST THING FOR ME TO DO IS THE RADIATOR UPGRADE. I APPRECIATE THE POST AND WOULDNT MIND SEEING THE TRICKS FOR THE SENSOR UPGRADE.

ANY RECOMENDATIONS FOR DIRECT BOLT IN RADIATORS?

ROBERTrwo@AOL.COM

SNAKEYE
05-05-2006, 06:12 AM
BIG HELP! SPENT THE DAY THINKING ABOUT IT AND DECIDED THE NEXT BEST THING FOR ME TO DO IS THE RADIATOR UPGRADE. I APPRECIATE THE POST AND WOULDNT MIND SEEING THE TRICKS FOR THE SENSOR UPGRADE.

ANY RECOMENDATIONS FOR DIRECT BOLT IN RADIATORS?

ROBERTrwo@AOL.COM
I got mine from Steeda: www.steeda.com (http://www.steeda.com). It fit in just fine except for a small mod to the fan shroud to clear the air intake tube to my Vortech.

ausie
05-05-2006, 07:09 AM
Very interesting.. but there may be an issue if the ECU uses that temperature measurement for A/F altering or timing retard.... As you stated, a thermistor is a variable resistance which has a negative temperature coefficient (means the exact same thing you stated as it drops in resistance with increase in temperature). Basic ohms law will work in this application but the Thevenin resistance value will not be half of that. The equation you should use is this:

R1= hot sending unit temperature resistance value
R2= second sending unit temperature resistance value

Rth (what the ECU will see) = [ R1 x R2 ] / [R1 + R2]

If the resistance values are the same, the Thevinin resistance (effective parallel resistance) becomes half of the value. However when they are not the same, the smaller measured resistance will be the dominant value.

Some hand held programmer tools may allow you to alter the fan temperature settings.

Since I have an 04 and had an 01 (not sure how much differece there is in the cooling system of the other model years), I discovered how the cooling flow actually works (which is not what I expected). Please correct me if I am wrong on this: When the Tstat is closed, coolant circulates throught the radiator and block. When the Tstat is open, coolant bypasses the flow throught the radiator and just circulates back throught the block. The fan is used to cool the trapped coolant in the radiator until the specified temperature is reached and the fan turns off. Once the thermostat closes the coolant flow continues as previously stated. Some of this is beginning to make sense now. The Tstat is just an over temp bypass to allow the radiator to cool down. It seems that having a more efficient radiator would be more important than changing the thermostat temperature. Also having the fan run most of the time would aid in the heat transfer. I used to think that the coolant flow was different but considering where the thermostat is on the Cobra it became more apparent that what I was told along time ago does not apply. ( used to think that the coolant circulated through the radiator but not the block at the same time, and that the thermostat opened when the coolant in the block reached a certain temperature and only circulated from the block when the tstat was opened.) I would not change the tstat temperature range since this is just a bypass for the radiator. Droping it to a lower value would not be very effective since it would probably be open longer than it should.

SNAKEYE
05-05-2006, 04:03 PM
For all practical purposes I would consider that the Ford replacement ECT sensor that I am using is as nearly identical to the stock unit as manufacturing tolerance would allow. If it were otherwise, then it would not be a practical replacement for a stock ECT sensor that went bad. The range of resistance is something like 1K-ohms to 60K-ohms. Both sensors are screwed into the same metallic tee connector and emmersed in the same pool of coolant. I call it a pool because there is no actual coolant flow across the ECTs' business ends. The coolant in the tee connector gets it heat from the coolant flowing by in the crossover tube and by conduction through the metal parts. So there should be no appreciable difference in the resistances. Thus the total resistance of the paralleled pair as seen by the voltage divider circuit of the fan switching circuit of the ECU is 1/2.

According to the service manual, "Group 03: ENGINE Section 03-03: Engine Cooling DESCRIPTION AND OPERATION Thermostat, Water When the coolant is cold, the water thermostat is in the closed position and the coolant flow is restricted to the cylinder blocks, cylinder heads, intake manifold and heater core. As the temperature increases, the water thermostat opens, allowing a portion of the coolant to pass into the radiator." Exactly how you interpret that rather poorly worded statement is up to the reader.

Tomorrow's extraction of on-board data will most certainly add to the knowledge of how it works and what is happening. I expect to not only get fan-on and fan-off temperatures but A/F mixture data also. I have no idea what the A/F mixture has been. I have put 30K miles on the car since taking ownership, probably 20K of it with the high speed fan override modification in operation. Granted, that mod would appear to not alter ECU input. Or is grounding a certain input to the ECU amounting to telling the ECU the engine has hit some high temperature and the fan is needed to cool things down and, oh by the way, adjust A/F accordingly? I do know this: the exhaust pipes are coated with carbon which indicates that the engine runs rich. The car has never thrown a code for fuel mixture related reasons. It always passed PA inspection under the old tailpipe sniffer program, and has since with the OBDII plugin program. Of course, the car performs very well with the Vortech S/C, and continues to do so with the ECT mod, but then i've only put about 200 miles on it since installing the second sensor. Until tomorrow (and beyond) then.

ausie
05-06-2006, 08:58 AM
If you have both sensors in the temperate zone than it will work as you described. That does not sound like it would cause any problems. My mistake, the A/F ratio is directly effected by the inlet air temperature thermistor not the coolant sensor. However there is some A/F alteration based on engine temperature especially at start up (runs a bit rich, choked when cold).

I have given the coolant flow some serious thought which I can relate to the 01 and 04 models (which probably has not changed much until 05). Since the Tstat is housed in a cast metal T section located outside of the block, there is no way that ford is accurate with their statement. The entire Tstat assembly basically forms a coolant shunt to the radiator. If there is inblock valving of coolant flow based on engine temperature than I might buy into Fords description. I think that the technical writer was sniffing glue at the time they wrote it. There is something missing in that.

As for replacement radiators: Fluidyne, Becool, and there are a few others as well.

There are a few things I can think of that will cause the running hotter than it should.
1. scale or calcium build up within the radiator or block. Also oxides may form within the block if it is cast iron. As for calcium or scale buildup it does not matter what the base metal is, it will form due to the break down of the antifreeze. Antifreeze has a life span and should be changed every two years and has a tendency to turn acidic over time.

2. Ratio of water to coolant has changed. Water will eventually evaporate from the system. It is the water that is the vehicle for thermal energy transfer and not the antifreeze. The only benefit from antifreeze is that it lowers the freezing point of water. There are some other benefits but the main purpose is to prevent damage to the engine from freezing. When I flushed out my cooling system in the 04, I ran the car for a few days with just water in it. The engine ran better with just water than the 50/50 mix of the coolant.

What I noticed with the on board diagnostic (HEC) display: the fan will turn on at temperatures above 100C (212F) and turn off when temperatures fall below 95C (200F). I am not sure if this is availalbe for model years prior to 99 but it is there for 99 + .

Here is the link to the diagonstic:

http://www.mustangworld.com/ourpics/fcar/dtcodes.htm

I am not sure if this will work with a 97 model year. If it does not the only thing it will do will just reset the trip odometer.

SNAKEYE
05-06-2006, 10:40 PM
First, prior to '99 the odometers were electro-mechanical, not digital, with input from a pair of wires from a module plugged into the transmission where formerly a speedometer cable would have been installed. The trip meter is reset by a pushbutton, but it is purely mechanical.

Second, just yesterday, on my way home from work, my odometer rolled over (literally) 50000.

Third, today my battery died while playing with my friend's AutoTap OBDII diagnostic tool! Such is life. It was the original OEM battery going on some nine years old, so it was to be expected.

Fourth, with regard to things that can cause the car to run hotter than it should, another, and I don't know how to describe it exactly, is the inability of air to flow naturally through the radiator core as the car is moving. I saw this (I think) in my test drives. Both test drives were in similar ambient temperatures (72-73 F) and at similar (60-65 MPH) vehicle speeds. In the test run with the stock single ECT sensor the temperature gauge almost went to the "H", while in the test run with the second ECT sensor in parallel with the stock unit the gauge stayed at a steady position just coming on to the "M". The only difference: THE FAN WAS MADE TO RUN! And to believe that Ford actually turns the fan off at speeds over 45 MPH. For what: to save a fraction of a horsepower and gas? Something is happening here that I have gotten myself wrapped around yet. Maybe the aerodynamics are such that air cannot enter without some kind of puller or pusher fan to upset the aerodynamic stalemate? In the good old days the fan was ALWAYS pulling air in. Then came thermostatic fan clutches, flexible fan blades, electrically operated fans, and practically sealed engine compartments, and the like. Nothing has been the same since!

Fifth, access to the A/F mixture was not available to us with this tool, so I have no data to report. I feel that there is little or no consequence with regard to A/F and reduced operating temperature, except during engine warmup when A/F is enrichened like a choke does with carb. I feel that by 170-degrees the enrichment is reduced to nearly, or completely, nothing.

Sixth, what the AutoTap did tell us confirmed my theory of operation of the ECT sensor modification. Here's how it went. With both ECT sensors plugged in (the modification) the temperature reported at the moment the fan turned on was 208-degrees F. I expected this indicated temperature reading to be this high because the (2) ECT's in parallel are representing the resistance equivalent to that temperature. Just as the fan came on I unplugged one of the ECT sensors, thus reverting back to stock. The indicated temperature reading immediately dropped to 181-degrees F (the true coolant temperature), and the fan shut off. I then plugged the second sensor back in, the fan came on, the indicated temperature reading went back up (to 214- degrees F actually). Then we waited and watched as the indicated temperature reading started to drop. At 199-degrees F the fan turned off. I immediately unplugged the second sensor and the indicated temperature reading had dropped to 162-degrees F (the true coolant temperature).
Based on all of my research, calculating, and plotting, I had estimated the fan-on temperature would be 180-degrees F and the fan-off temperature would be 160-degrees F. I pronounced the whole project a complete success, unplugged the AutoTap diagnostic tool, thanked my friend for his assistance, and called it a day.
Now all I have to do is finish my writeup and anyone with about $60, some basic wrench-turning skills, and 30-minutes of installation time, can get their engine temperatures down to where they, like myself and many, many others, feel they should be.

ausie
05-07-2006, 01:21 PM
Good Job man! looks like you found the sweet spot in coolant temps! :thumbsup:

The more I think about the A/F being effected by water temp, more than likely it is not a critical issue. From what I recall the ideal water temp should be around 160F.

Sorry to hear that your battery is toast...
It was worth a shot in the dark if the HEC would have worked. Bummer that it will not.

I think you are on the money with the closed off engine compartment issue. It is always nice to have a vented hood that works the way it should. The only thing that would aid in air flow other than a functional shouded fan arrangement would be to install a modified front bumper cover that has an air dam build into the lower part to draw more air towards the radiator. There probably is not much else that can be done other than changing the hood to improve air flow. But for the cost, it will be far more than the mod you suggest. For $60 and 30 minutes of time, sounds like a success to me. :cool:

dewone
05-08-2006, 08:22 AM
Just curious if you calibrated the read out. In other words taking a decadent meter and applying a resistance and comparing that to the read out. The system is operating similar to RTD's in motor windings or more like a transfromer dry well system. With the decadent meter we calibrate the read outs by the set resistance up and down the scale. Just a suggestion to confirm your results that what the gauge says is what the engine temp is by comparing what Ford says the values should be. ie 60K ohms=212f and of course we can take a multi meter and read what the resistance of the sensor is at different temps to complete the test loop.

SNAKEYE
05-08-2006, 12:57 PM
Just curious if you calibrated the read out. In other words taking a decadent meter and applying a resistance and comparing that to the read out. The system is operating similar to RTD's in motor windings or more like a transfromer dry well system. With the decadent meter we calibrate the read outs by the set resistance up and down the scale. Just a suggestion to confirm your results that what the gauge says is what the engine temp is by comparing what Ford says the values should be. ie 60K ohms=212f and of course we can take a multi meter and read what the resistance of the sensor is at different temps to complete the test loop.
We didn't calibrate anything. Just plugged in the AutoTap and watched the laptop as the engine idled and heated up.
Before installing it I did throw the second ECT sensor into a pot of water and "cooked" it on my kitchen stove with my digital meter hooked up to it. Then I plotted the service manual's tabulation against my data. The two curves were nearly indistinquishable. For as unscientific as my method was with collecting this bit of info, I'm happy.

ausie
05-09-2006, 07:04 AM
I do not think your method was unscientific. Although the stove top may be a bit crude but it is a source of heat that can be controlled to some degree. Since you have some acces to data from the vehicle, can it be accessed while driving to confirm your results from the idle test (you probably already did that). Also I would correlate the data to the ambient air temperature to see if the coolant temperature stays within the ideal range. You would probalby need several tests at various temperatures as well as a winter test.

dewone
05-09-2006, 10:36 AM
Okay great job I liked the method of test, and it corelates to the info already availible. Now to get the answer to if the gauge is reading correct, and not having a decadent meter we can get a vairiable resister usually a wire wound resister with a sliding tap. (Try radio shack) It won't cost much $10 at most. Be sure to get one with the values around the center values of the table. Connect leads from the sensor input to the resister one lead on the end of the resister one to the tap. Then you should be able to slide the tap up and down the resister and watch your gauge movement with the change. Take the multi meter and set the resistance to a set value on the scale and see if that looks correct to the table.

SNAKEYE
05-09-2006, 09:16 PM
I do not think your method was unscientific. Although the stove top may be a bit crude but it is a source of heat that can be controlled to some degree. Since you have some acces to data from the vehicle, can it be accessed while driving to confirm your results from the idle test (you probably already did that). Also I would correlate the data to the ambient air temperature to see if the coolant temperature stays within the ideal range. You would probalby need several tests at various temperatures as well as a winter test.
I could take the AutoTap and laptop for a ride. I believe I could also have the AutoTap/laptop generate a time vs. data file. I personally do not know how to setup the system (it belongs to my friend). He says I could use it with my laptop, but my laptop complained vehemently about what AutoTap's software might do to it if I continued to install it, so I canceled the install. I don't need the grief. The best I might be able to do is arrange with my friend to hook up his laptop with the Autotap in my car when we travel together to either Valley Forge Mustang Club's show at Kimberton over Memorial Day weekend, or when we cruise on over to All Fords at Carlisle via C J Pony's Pony Trail on June 2. I'll have to pursue this line of thought with my friend. Daytime ambient temps may be in the 80's until then. Hurrah!
While we were playing around the high speed fan did come on. I don't know if it was the car or our manipulating the settings of the PCM with the laptop. I do not think that this modification will ultimately lower engine temps on a 90+degree ambient temperature day. That is most like going to depend on radiator capacity (which I have with the 2-core). However, with the fan coming on at 180 instead of 220, the cooling will begin at a lower coolant temp and perhaps slow the temperature rise to something reasonable.
As for winter-type temps, the 160F t-stat (which I've had installed for years) provides enough heat. If the fan does ever come on, it won't be for long. Anyway, SNAKEYE hibernates for the worst of winter, so I may never know.

SNAKEYE
05-09-2006, 09:30 PM
Okay great job I liked the method of test, and it corelates to the info already availible. Now to get the answer to if the gauge is reading correct, and not having a decadent meter we can get a vairiable resister usually a wire wound resister with a sliding tap. (Try radio shack) It won't cost much $10 at most. Be sure to get one with the values around the center values of the table. Connect leads from the sensor input to the resister one lead on the end of the resister one to the tap. Then you should be able to slide the tap up and down the resister and watch your gauge movement with the change. Take the multi meter and set the resistance to a set value on the scale and see if that looks correct to the table.
If you are referring to the stock gauge in the car, you can forget about any kind of accuracy. Besides, it doesn't even have any numbers on it! I look at the stock gauge as a "relative" gauge. If it reads unusually high there may be a problem, check it out. The fact that it nearly went to the end of the dial with just one ECT sensor on-line and stayed up the middle with the second ECT sensor in parallel is "relative" proof enough that the system is doing what I want: cooling things down to a much more normal range.
If I can arrange to setup my freind's AutoTap and laptop when we go to Kimberton or Carlisle together (he in his '99 and me in my '97), I could cruise while collecting data with one ECT sensor on-line, then do a quick pitstop to plug in the second sensor, and continue on the trip collecting data with two sensors on-line. Sounds like a plan!

ausie
05-10-2006, 06:38 AM
The best I might be able to do is arrange with my friend to hook up his laptop with the Autotap in my car when we travel together to either Valley Forge Mustang Club's show at Kimberton over Memorial Day weekend, or when we cruise on over to All Fords at Carlisle via C J Pony's Pony Trail on June 2. I'll have to pursue this line of thought with my friend. Daytime ambient temps may be in the 80's until then. Hurrah!

Actually the ultimate test would be to take it on a road course track. Drag strip may not be the ultimate test but it does put a different strain on the motor. The Valley Forge Mustang Club show? I know where Valley Forge is but Kimberton is a miss.... However the Fords at Carlisle show may be of interest.

SNAKEYE
05-10-2006, 09:36 PM
The best I might be able to do is arrange with my friend to hook up his laptop with the Autotap in my car when we travel together to either Valley Forge Mustang Club's show at Kimberton over Memorial Day weekend, or when we cruise on over to All Fords at Carlisle via C J Pony's Pony Trail on June 2. I'll have to pursue this line of thought with my friend. Daytime ambient temps may be in the 80's until then. Hurrah!

Actually the ultimate test would be to take it on a road course track. Drag strip may not be the ultimate test but it does put a different strain on the motor. The Valley Forge Mustang Club show? I know where Valley Forge is but Kimberton is a miss.... However the Fords at Carlisle show may be of interest.
I don't drag race (the car is a convertible and would require a rollbar because it would be too fast with the supercharger) or road course, so the best I can do is a cruise to somewhere.
Check out Kimberton at www.valleyforgemustang.org (http://www.valleyforgemustang.org).
Check out Carlisle at www.carsatcarlisle.com (http://www.carsatcarlisle.com).
Check out C. J. Pony's Pony Trail on June 2 on their website www.cjponyparts.com (http://www.cjponyparts.com). They are located west of A-town along Rte. 22 near Harrisburg. Two of us are doing the Trail and then going over to Carlisle for the rest of the day.