Over Heating
The early Z car's
are known to have overheating problems. This is due to, among other things,
the long, heat trapping hood and the long exhaust manifold. Some attempts
to remedy overheating problems include the addition of stock heat vents
in the hood and after-market 4 core replacement radiators (where core number
indicates the number of "layers" of cooling fins: stock radiators are 3
core). Contained in these pages are my battles with overheating and the
different (finally successful) methods I used to solve the overheating
problem. Before launching into a test and solve discussion of problems
let me cite some possible problems and solutions (some of which I ran into).
These will start with cheap and easy fixes and continue to more difficult
and subtle issues. The symptom generally will be overheating, but in some
cases may be more subtle, for these cases the symptoms will be described.
Quick Fixes
Here are a
couple of things to check real quick - who knows you might get lucky.
1) Use method below to check actual engine temperature.
2) Exhaust bolt/nut under thermostat housing is loose.
3) Pull radiator cap after first 2 minutes of operation
and listen for escaping gas (blown head gasket).
Measurement
From experience,
I have found that with my '79 280 ZX the most important requirement to
solving an overheating problem is making sure that one exists. Most cars
today use a simple electronic method to measure water or head temperature
(I say most because I worked on a '72 Mercedes Benz 220 with an ether and
diaphram set up! I clipped the tube containing the ether thinking it was
a wire - ouch). There are basically three components to the temperature
measurement: a voltage regulator, a gauge and a sensor. First, because
a cars voltage can vary considerably depending on load (12 to 14.5 Volts),
it is necessary to regulate it. This is done with a voltage regulator.
While I have not dissected a Z earlier than my '79, I assume that the 5
volt voltage regulator earlier Z's employ is the same. It is contained
inside
the gauge housing (not on back like most cars) and it is hard wired to
the case behind the speedometer. Therefore, it is not a replaceable item
(one of the few poorly designed features of the '79 280 ZX). The regulator
consists of piece of ~45 ohm wire wrapped around a bi-metalic strip with
a contact at the end. When the contact is closed, current flows through
the wire heating up the bi-metalic strip and lifting the contact point.
When the contact point is lifted, surprise, no more current and the bi-metalic
strip cools. This means that the output from the voltage regulator is either
the 12 - 14.5 volts mentioned earlier, or 0 volts. The reason it
regulates is that if the applied voltage is higher, it doesn't matter,
it just heats up faster lifting the contact more quickly. After a moment,
the opening and closing of the contact "should" reach an equilibrium where
the average voltage is 5 volts. Unfortunately, because this oscillation
is so slow, you can not use an AC voltmeter to measure it, but would need
an oscilloscope or an analog voltmeter and a good guess.
Visit the
Voltage
Regulator Repair section for a discussion on how you can adjust your
voltage regulator and test it. For the moment, however, lets just assume
that you are getting 5 volts from it (if you were to try and measure it
however, your multimeter will bounce between 12 volts and 0 making it difficult
to read). The next part of the circuit is the temperature gauge. It is
an ammeter which is constructed similar to the voltage regulator. As a
current heats a wire, it bends a bi-metalic strip and the needle
moves. The gauge is in series with the temperature sensor, and as the sensor
heats up, it's resistance drops. This allows a greater current to pass
through the gauge and results in more needle deflection. As I have said,
this is a typical setup and one that works very well - until your voltage
regulator gets flaky. The way to by-pass the voltage regulator and gauge
is to measure the resistance of the sensor and convert it to temperature
yourself. Unfortunately, the conversion chart isn't in the manual (another
of my gripes with Datsun/Nissan). Therefore, it was necessary for me to
remove the temperature sensor and make the measurements myself. Here is
a chart showing the resistance as a function of temperature for the sensor.
Now you can measure the temperature of the car with a simple
voltmeter, but you must disconnect the wire going to the sensor
to make this measurement. Otherwise the 5 volts from the regulator will
change your voltmeters impedance measurement! Also, I had to shut of
the engine to get a reasonable measurement. Why? I don't know, maybe
I had a ground loop or something.
If someone
else has made these measurements let me know. Specially if you have different
values! I have marked 195 degrees Fahrenheit because this is the center
of the temperature gauge. What this chart tells you is that if you measure
a resistance across your temperature sensor of 31 and greater, you are
fine. For instance, my temperature gauge in the car was reading 215 degrees,
but my resistance was ~ 37.3 ohms or 185 degrees! I was stressing for no
reason. I have also put a thin thermocouple into the thermostat housing
and it has reported the same temperature to within a couple of degrees.
Verification is always nice :-) Instead of adjusting my voltage regulator,
what I should have done was put a 31 ohm resistor in the circuit in place
of the sensor and adjusted the temperature gauge. As it turns out, when
I actually tried to do this, it turned out to be more difficult to do than
adjust the voltage regulator. If you do try and re-calibrate the temperature
gauge - be warned, it is not easy to calibrate, but it easy to break!!!
Thermostats Thermostats
are cheap. If you think that the thermostat is your problem (car immediately
overheats in city or on highway) then replace it. However, some
thermostats are cheaper than others. Here I have a picture of an OEM Nissan
thermostat and a Stant thermostat side by side.
The differences in their construction are obvious and
surprising. First, and most importantly, the aperature on the OEM thermostat
is around 30% larger! Assuming that they both open the same distance (I
actually think the OEM opens further since it is taller) means that
the flow through the OEM thermostat is much greater. Further, as you can
see from the photo, the OEM has a rubber seal as opposed to the Stant's
metal to metal seal (although some OEM thermostats are now metal on metal).
This improves heatup and regulation by reducing leakage through the thermostat.
I have found that with a Stant thermostat in my 280ZX the engine temperature
varies considerably (greater than 10 degrees Fahrenheit variation) while
the OEM thermostat regulates the temperature very well (greater than 5
degrees Fahrenheit variation). Generally, the thermostat is the problem
with relatively new cars (i.e under 100,000 miles) if you belong to the
100,000+ club like the rest of us, then replace the thermostat, but don't
hold your breath.
There are
three different thermostats which you can buy for the '79 280ZX. It seems
to me that this is a testament as to how poorly the car can regulate its
temperature. I have tried both the tropical 76 degree celsius one and the
normal 82 degree celcius one, and, obviously, my car ran cooler with the
tropical thermostat. How much cooler I can't say since I only recently
started actively monitoring the engine temperature with a chromel-alumel
thermocouple. Since my car is still running at about 185 degrees Farhenheit
with the normal thermostat, I may try the fridged 86 degree celcius one.
I don't know though, I would rather run slightly cool than slightly hot.
It seems that the lower temperature thermostat allowed the car to vary
in temperature more than the normal one. I assume that this is because
regardless of the temperature minimum your thermostat opens at, the engine
and cooling system are design to allow optimal cooling at 205 degrees Fahrenheit.
Radiator For cars over
100,000 miles, the radiator is usually the culprit of an overheating problem.
There are a couple of indicators for a bad radiator. The first one you
will notice is overheating in the city, but not at highway speeds. If this
is occuring and you think that it is a radiator there is a simple test
that you can perform. First, while the engine is cold, remove the radiator
fan. This is necessary regardless of whether you will check from the front
or back of the radiator (those of us with AC must check from the back or
have very skinny hands). I find it easiest to remove both the clutch and
the fan and hold the water pump pulley in place with some metric bolts
or nuts (don't forget lock washers), but you can probably just remove the
blade from the clutch. Then start the car. It is important that while you
do this you regularly monitor the engine temperature or have someone else
monitor it. Don't let the engine overheat! As the engine warms up
feel the surface of the radiator; it should heat up evenly. If you find
cold spots, this indicates that passages are blocked. I have been told
this typically occurs in the center of the radiator. If you do find cold
spots then it is time for a new radiator or a recore. Removing the fan
helps increase the temperature difference between the hot and cold spots
making it easier to diagnose.
Motorsports
offers a 4-core radiator for not much more than you would pay for the stock
one. This will improve your cooling capabilities by increasing the
surface area available for heat dissipation. Be warned, however, that this
places the surface of the radiator dangerously close to the fan and clutch
assembly. In a hard stop I actually had fan-to-radiator contact. Fortunately,
I didn't puntcure the radiator, but I did mar the surface. It was necessary
for me to remove my fan and clutch assembly and replace it with a set of
electric fans (more on that in the next section). Now that I have tracked
my heating problem to a faulty regulator I wish I had simply put the stock
radiator into the car and saved myself a lot of headaches. Before deciding
to purchase the 4-core upgrade, consider how you drive the car and how
much modification you want to make to the engine compartment to get it
to fit. If you don't have AC and/or turbo I wouldn't even consider the
4-core replacement.
Fan Clutch, Airflow,
Headers
Invariably,
one of the first items that gets replaced when trying to fix an overheating
problem is the fan clutch. Unfortunately, it rarely does any good. Whatever
you do, DO NOT ATTACH THE FAN DIRECTLY TO THE
PUMP SHAFT! The first instinct is to remove the clutch and see
if the overheating problem goes away. If you remove the clutch and attach
the fan directly to the pump shaft there is a very good chance that the
pump shaft will shear and you will be left with a fan through the hood,
or more likely through the radiator. I have been warned repeatedly about
this. What I believe is that the clutch buffers the angular momentum of
the fan during hard acceleration/de-accelerration of the engine. This permits
the use of a smaller diameter shaft in your water pump. I am not sure about
the reasons, just about the results. If you want to test your fan clutch,
it is not even necessary to remove it. While the car is cold, and turned
off, try and turn the fan. It should turn easily. Then start warming up
your car. As you stop the engine and test the fan, it should get progressively
harder and harder to turn freely. It will never lock up completely, but
you can tell that the clutch is working.
One of the
first questions you will be asked when talking to a mechanic about an overheating
problem (after going through all the usual problems) is if the fan shroud
is in place. On late model RX7's the shroud and "belly pan" are necessary
and the car will overheat without them. On the 280ZX they will improve
cooling, but I have never tested how effective they are. Dick Denno (IZCC
#452) has
an article
posted
on the Z Home Page
addressing the shroud and belly pan, but the only modification he states
had an effect was the addition of a front air dam. My opinion is that if
you have the parts then use them, they cannot hurt. The shroud will help
improve efficency by reducing the amount of air re-circulating around the
fan tips and by ducting air through the entire surface of the radiator.
The belly pan will improve passive flow at highway speeds. Dick Denno also
mentions how hood vents helped reduce trapping the heat from the exhaust
manifold. Since the hood vents were stock on my car, this was not an improvement
option for me. However, I have considered using headers with header wrap
to both improve my horsepower and try to reduce the under-the-hood temperatures,
but have not done this yet. I have read somewhere (Motorsport?) that header
wrap can significantly reduce radiation from the header, but it seriously
degrades the life time of the headers making them last only 2 years. One
thing that you should always keep an eye on is the
exhaust bolt/nut directly under the thermostat housing. On my '79
280ZX this tends to loosen up and allow hot exhaust gas to breathe onto
the thermostat housing causing false temperature readings.
Spark Timing
and Mixture
Suprisingly,
spark timing can effect operating temperature of your Z. This is specially
true at highway speeds (and above...) and not as important for around the
block driving. For a stock 280 ZX, the spark should occur at 12 degrees
before top dead center (for those of us who have degree'd an aftermarket
competition camshaft, this is subject to change..;-) Retarding the spark
(say 8 degrees BTDS) effectively makes the air/fuel mixture richer. While
advancing the spark (16 degrees BTDS) effectively leans the mixture. An
overly lean fuel mixture can make an engine run hot. This brings up another
point, and that is what your true mixture is. If you have installed a K&N
air-filter, a performance camshaft and/or headers and an open exhaust,
your car may be running lean. You can increase your fuel to air mixture
by adjusting your mass-flow meter, but this is not for the faint of heart
or typically for someone who needs to pass emissions.
Exhaust
Gas Recirculator (EGR)
The EGR valve
bleeds hot exhaust gas back into the manifold after the car has warmed
up. Counter-intuitively, the addition of hot, pre-burned gas actually lowers
the temperature of the combustion process. This is how it reduces the car's
emissions. By lowering the combustion temperature, you reduce the amount
of nasty radicals (pollution) which can form at higher combustion pressures/temperatures
(this is also why high compression engines aren't produced any more).
I have been told by an ASE Certified Master Technician that this process
can actually lower the burn temperatures up to 1000 degrees Farhenheit!
While I find this difficult to believe, it makes obvious sense to check
your EGR valve if you are having overheating problems. I have found that
it is actually the temperature sensitive vacuum switch which usually fails.
An easy way
to quickly check the EGR valve is to connect a hose to the vacuum nipple
and suck on it (no sexual implications...) If air moves easily, the EGR
valve is punctured and needs replaced. If you can pull a good vacuum and
can watch the diaphram of the EGR move, it is fine. This does not mean
that your EGR valve is working properly, however. As I mentioned earlier,
the EGR is only allowed to function when the car is at operating temperature.
To prevent cold operation, there is a temperature controlled vacuum switch
which is controlled by the water temperature. Following the vacuum line
back from the EGR, it immediately encounters a vacuum amplifier (rarely
fails) and then the vacuum switch. The vacuum switch is mounted under the
fast idle by-pass valve and has hoses connecting it to the thermostat housing.
The way you test this is just the opposite of the EGR. Connect two vacuum
lines to the nipples on the vacuum switch and put one of them into a glass
of water. Before warming up the engine, blowing on the free vacuum line
should not produce bubbles in the glass of water. After you warm up the
car to operating temperature, however, blowing on the free vacuum line
should produce bubbles in the glass of water. If you do not see bubbles,
the vacuum switch has failed - you can replace or by-pass it.
Head Gasket
One of the
overheating problems that I encountered was a blown head gasket. It was
only 'slightly' blown, however and the usual test of starting the car,
allowing it to run for a couple minutes and checking for pressure in the
water jacket didn't disclose anything. Usually, if you have a blown headgasket,
during this test you will hear gas escaping the radiator as you remove
the cap. The indicator, I felt, was the overheating at highway speeds,
but not at around town driving. I would not
suggest that if your car is overheating at highway speeds that you pull
the head and look for a blown headgasket though!!! My final indicator was
the radiator fluid leaking out from between the head-block surface.
Electric
Fans and Thermo-switches You might
wonder how I discovered that the voltage regulator was flakey and my temperature
reading was way off. Since I was combating an overheating problem, my final
solution was to install two electric fans to cool
the car; a 14" fan and a 12" fan. Since it is really only necessary to
cool the car once it gets up to temperature, I adapted a thermo-switch
(fan switch) from an '82 Honda so that it would switch the 14" fan on at
190 degrees Fahrenheit. After I set the whole system up (which I will describe),
the fan would not switch on. After spending days trying to figure out a
wiring problem and double checking the operation of the switch I finally
realized the fans were not switching on because the engine was not reaching
190 degrees!
First, the
fans. I originally purchased a 12" dynaflow fan from Autozone
and used the nylon mounting kit to attach it in a blow configuration (or,
you can do what Salman
did) in front of my original 3-core radiator and pulley mounted fan.
This did not help reduce what I thought was overheating. I then purchased
and installed a Motorsport 4-core radiator along with the original pulley
mounted fan. Boy, was I irritated when this didn't end my overheating problems!
After, a hard stop caused fan radiator contact, I finally decided that
the pulley mounted fan had to go and that I needed to properly mount two
electric fans behind the radiator. I constructed a metal frame which attached
to the radiator mounting bolts (not a good picture)
and put the 14" fan on the side of the radiator closest to the AC compressor
(greatest amount of air flow would cool AC Compressor and exhaust headers).
The 14" fan was tied into the thermoswitch while the 12" fan was always
on. Here are some pictures: Fan3, Fan4,
Fan5.
As you can see, the whole thing fits rather tightly, but it is not difficult
to remove. The only way I was able to get the 14" fan into this space is
because it is off to the drivers side of the radiator. Otherwise it would
fight with the water pump pulley. Two words of advice about both fans:
first, put them on relays. Not only do they require a lot of current, but
if they are tied directly into a switched circuit, the back emf can cause
strange things (like radios staying on for a minute after the car is switched
off.) Second, make sure you have a VERY GOOD alternator.
Next
is preparing the thermo switch. Below is a picture of one of these taken
from an '82 Honda (any model).
They are found on the lower right corner of the radiator
and will cost you close to nothing at a junk yard. Pick a couple up. They
are set to switch on at 210 degrees which is a little high for our application,
so it is necessary to adjust them. They can be taken apart by unscrewing
the sensor part from the switch body. DO NOT grab the bulbulous, ether
containg part of the sensor, but the neck (see picture below).
After you have it apart, you will see the switch body, a
thin, but long piston and the sensor body. If you look down into the sensor
portion, there is a diaphram behind which I assume is ether.
What you want to do is increase the length of the piston
so that less movement by the diaphram (i.e. lower temperature) will trigger
the switch. I found it necessary to add about 40 thousanths of material
to the switch piston. I used a punch and some stainless steel sheets to
make little spacers that fit down the neck of the sensor, towards the diaphram.
After I added spacers, I would test the switch by first connecting it to
an ohm meter and then suspending it into some water. I would slowly raise
the temperature of the water while watching a thermometer (230 degree thermometer
not allowed to touch bottom of container) until the switch turned on. Depending
on where it switched at, I would add or remove spacers.
Next, is finding
a place for it in the water jacket. I had originally removed the water
pre-heater line that went to the EGR vacuum-switch under the air-bypass
valve. It never worked right, allways leaked and was just basically a pain.
Upon removing it, however, the thermostat housing now had a perfect place
to mount the thermo-switch. A picture shows a thermostat housing from a
'78 280Z and the thermo switch.
It is necessary to mill the top of the area where the water
line came out, and then tap it to match the thread on the thermo-switch.
What thread is it? I don't remember, but I will post it when I get a chance
to measure it. Here is a picture of it mounted:
While it looks like it is crowding the distributor, there
is actually plenty of space. The wiring diagram for the 14" fan is as follows:
As noted in the picture, the 12 volts being switched by
the relay is delivered by a 10 gauge wire directly from the battery while
the 12 volts which throws the relay can be delivered by a switched (i.e.
when key is in on position) or un-switched (tied to the 12 volts from battery)
source. In the switched case when the car is turned on AND when the engine
is above 190 degrees, the fan will come on. In the un-switched case, when
the water is above 195 degrees the fan will stay on regardles of the ignition
key. Since the thermo-switch is in the water jacket and not the radiator,
the un-switched set-up may drain the battery. This set-up has worked extremly
well for me. While not shown, the 12" fan is on a similar circuit with
a switched power supply and no thermo-switch; the coil leads to ground.
In such a configuration the relay is thrown when the car is turned on and
the 12" fan is on while the car is. Not pictured in the circuit is a fuse
that is on the +12 V battery line. I stole the main curcuit fuse from an
'82 Honda (the Honda battery hooks directly into one of these). The switched
+12 V comes from an allready fused ignition circuit.
And there you
have it. If you still have overheating problems after checking all of this,
then your problem is beyond my help. Good Luck!
|