I have an Accord Coupe EX V6 with over 215K miles. About 1 year ago (July '08), I noticed some temperature excursion issues. While driving on the expressway, the temperature gauge would show normal (less than halfway mark). Upon coming to a stop or if traffic slows, I can see the temperature needle moved toward the red (max). When traffic began moving again, I can see the needle move back to normal operating condition.

The diagnosis at the time was a semi-plugged radiator so a new radiator was installed and the overheating problem appeared to have gone away. Not sure if the problem was completely solved, but with a new radiator and the Chicago fall/winter/spring driving, the cooler ambient temperature may have help hide the problem.

About 1 month ago (June), I noticed similar overheating problem as before despite having a new radiator. The initial thought is that the cooling fans were not starting properly. The fans run only when the A/C is running. It did not appear to start even when the temperature guage started moving toward the red. The fix was a replacement of the engine coolant temperature sensor. This solved the problem of the fan starting when temperature gauge on the dash moved beyond 75%. However, I still have the overheating issue as described previously.

I should note that the car will begin to overheat (temperature gauge touching the red, but no steam from hood) only after 60+ minutes of continuous driving. The overheating is "real" as the coolant overflow reservoir is full and the top cap on the reservoir is wet.

Sorry for the long winded posted, but I am trying to be detailed with the problem. Could this simply be a thermostat problem? Can the thermostat open partially and thus not allowing sufficient coolant to the radiator? Has anyone else has had similar overheating problems? Any help or guidance would be greatly appreciated.

Thermostat would be the next thing id look at.

Be sure to get an oem t-stat. The universal ones from the autoparts stores do not work nearly as well.

I was going to say thermostat too.....

Does the OEM t-stat from a dealer have 2 temperatures as well? I see some online with 180 Deg and 170 Deg. Which is OEM?

I saw this at rockauto.com for ~$26, but not sure if this is an OEM part.

BECK/ARNLEY Part # 1430707
Original Equipment Standard Temperature; TEMPERATURE=170

How easily can the thermostat be replaced and is there a diagram/instruction for this? Let's just say I am the less than "hands-on" type.

Thanks for the suggestions.

A good place to buy oem parts is www.collegehillshonda.com or www.bernardihonda.com . I'd call them on the phone though if you place an order.

That t-stat you listed does not sound like oem. From what I recall part of the prob with the universal ones is that they open all the way or not at all vs. the oem ones that can open partially. I'd have your mechanic replace it.

I'd also replace the pcv valve every 7 yrs/105k with an oem one. When it goes bad you will start to burn oil. Your car may have two of them...I'd ask the parts dept. about that.

The thermostat wouldn't show itself after 60 minutes, especially with the warm temps outside. In the summer the stat should be open fully in about 5 to 10 minutes max. If it stuck shut, you'd cook really fast. The fans won't have any effect on the temps when moving over about 20-25 mph. Sitting still, definitely.

Does the car have full heat when it's "hot"?

It can't hurt to change it, but I don't feel that is your issue.

What coolant has been run in this car? If you've got conventional green in it, that rad is done again. The reaction between conventional green or universal garbage will cause a reaction that makes the die castings corrode in a way that sends deposits around.

Was the cap changed when the rad was? Last water pump and was it a Honda pump?

The temp going into the red means the car is severely overheated. These gauges are "damped" so by the time it's up there, its cooked.

I'd be watching for bubbles in the reservoir. (not when it's super hot) When it's cold, drain off a slight amount to lower the level in the rad. Smell it, what does the inside of the rad smell like? Coolant or exhaust?

Your heading for block test time.

^^^ Che, since your a new member you may not be familiar with the great advice Fredsvt gives. I'd pay close attention to his post.

The stock thermostat begins to open at 176F and is fully open at 194F. The aftermarket stats, from what I've seen snap open at 195, or whatever they are rated at. It causes a greater degree of thermal shock going from hot (closed) to cold (open) especially in colder weather.

Fredsvt - Thanks for chiming in and the diagnostic.

When the new radiator was installed about a year ago, it was done at an idependent shop so the green stuff was used. Until I found this site, I did not realize the green vs blue Honda coolant had that much effect on the radiator.

The car does have full heat when the engine is warmed. I've had to turn the heat and fan to high on several occasions to keep the dash temperature gauge from staying on the red when in traffic. I've also observed the temperature gauge w/ car in park and can see that when the needle goes above 3/4, the cooling fan kicks on. With the fan on, the needle will slowly stop its ascension and work its way back to 3/4 and then the fan would cut off. This will cycle a few times with the temperature rising less after each cycle. This appears to indicate the car is once again cooling down (as long as no additional load is applied i.e. not driven but in park).

The radiator cap was changed with the new radiator. I don't believe this radiator and cap are OEM parts either. The water pump and timing belt (likely not OEM) was also performed at the same independent shop in January 2009 (~6 months ago).

I'll check the coolant quality again, but when I checked the coolant reserve tank yesterday, I did not notice any particular oil or exhaust smell to it. It wasn't super green like unused coolant from a new container, but it didn't have oil or black crud either. I'll check in the radiator next time.

Thanks to SatinSilver's link, I ordered OEM thermostat & Honda blue coolant from Bernardi. It should get here midweek for my weekend project of flushing the cooling system and replacing the thermostat. I think I'm good with the procedures, but I'm not sure if I need to bleed air from the cooling system. Is there a bleeder bolt on the t-stat housing? I tried to feel for it with my fingers, but could not locate it. If anyone as a diagram or photo of this bleeder bolt, I would appreciate your sharing.

I'm not so concerned about the rad not being oem, many of the rads on the outside are just fine. I just hope that the conventional green stuff hasn't gelled and plugged the passages in the bottom of the core. If there's any concern about the cap, get a Honda one, I haven't seen an aftermarket one yet, that holds the same rated pressure or pops off at the same pressure as the oem one.

It's good that it still has heat, that shows, at least, there's some kind of flow and that there shouldn't be combustion gas in the coolant.

I hope they didn't install a rebuilt water pump, it may not be pumping fully, the impeller can be slipping.

The temp gauge should NEVER move, once up to that 1/3 or so of the way up, that's it. The fans should go on and off without even seeing a micron of movement on that gauge.

In order for the fans to come on, the coolant at the outside of the thermostat housing must make the temp switch turn on the fans. If the stat gets stuck shut, the fans most likely will never come on.

The smell of a badly overheated engine is unmistakable, you'll smell what seems like burnt coolant mixed with exhaust. You really shouldn't see any discoloration.

Make sure prior to draining that the heat is turned up all the way. Drain the rad, drain the rear block drain, and if you can do it, pull the forward block drain. You may want to pull the lower hose off the thermostat housing, then take the upper hose off and put it into a bucket, using a hose, run water through the rad backwards.

On the V6 there's no bleed procedure, the radiator neck is the highest part of the system, just fill it slowly, and it'll purge itself. once it won't take any more, put some coolant into the reserve, put your palm over the radiator neck, and squeeze the upper and lower hose. If it needs a little more coolant add some, cap it and start it and let it idle, you should get heat out the vents in short order. Once you do, shut the heater fan off, but leave the temp turned up all the way.

From almost totally empty you should be able to put in around 1.5 to 1.75 gallons.

FredSVT:

I strongly disagree with you, respective to replacement radiators. :thumbsdow

I installed a new, OEM NipponDenso radiator in my trusty '86, shortly after I bought it. Cost me almost $300. This, of course, was long before the Internet became widely accesible/utilized.

Fast forward, to Year 2001: The engine failed because of my FOOLISH lack of maintenance to the cooling system. Thermostat broke (pin came out of it), head gasket blew, warping the head. Almost crying, as I pulled it apart, I was vastly relieved it didn't damage the short block. :hug:

When I put the motor back together, in August 2005, I went with a Koyo "high perf" radiator. Unlike the OEM ND unit, it is solid-copper core, with TWIN flues. Delivered to my front doorstep, it cost me about $112. :)

Some four (4) years later, I have NEVER had a single problem. My temp guage seldom ever gets above 3:30 clock position, regardless. But I'm STILL going to let you kick my A$$ for installing GM Dex-Cool instead of Honda Type 2 (blue). I flushed it out about a month ago, replacing the thermostat, all the hoses, and the heater control valve. :thmsup: :)

If/when the radiator needs to be replaced in my '02? I'm going with a Koyo, or otherwise. On the other hand, if YOU want to buy me an OEM ND unit, I don't mind spending YOUR money, Sir! :D

My local Honda dealer will NOT install an OEM (ND) radiator, except upon explicit customer demand, in WRITING. NO. They pick up the "horn," call the NAPA dealer, right down the street, and save the customer 2/3 to 3/4 on an aftermarket radiator which works just as well, and lasts just as long as OEM. :thmsup: :yes:

Robert,
Nowhere did I say anything about putting in an OEM (from Honda) radiator.

I said that the rads available from the outside are fine. I install them all the time. These are usually Visteon, or Koyo OEM "style" rads.

However, I'd never install an "all metal" copper/brass rad in one of the newer Hondas with all aluminum engines. If our supplier sends us one I send them back.

Dissimilar metals, especially with copper/brass/LEAD (solder) and that much aluminum can cause all kinds of odd electrolysis issues. I've never liked them to begin with as in our area the cores rot out from the outside in long before the insides fail.

Your '86 should have a cast iron block, correct? I wouldn't be as concerned, as you already have a large amount of dissimilar metals in contact with one another. That being said I'd never go any more than 30k on coolant changes in that car.

FredSVT:

Fair points, Kind Sir! :yes:

The "A20A3" engine in my trusty '86 involves a cast-iron block, with an aluminum head. Despite other (chemical) factors, IRON and COPPER tend to "get along" quite well, Sir.

But with the all-aluminum engine in my '02? I will DICTATE to Koyo (or other vendor) that they furnish me an ALUMINUM-core, if/when radiator replacement is necessary.

The iron and copper get along fine, yes, but the lead in the solder sure doesn't, and the brass is pretty much sacrificial.

I've seen many of the "all metal" rads that were put in elsewhere and the top and bottom (or side) tanks end up paper thin.

Older all cast iron engines like many of the diesels I see have original copper/brass rads at 20 years or more.

I'm trying to remember the last alum/plastic rad I changed on a later model Honda. Only two comes to mind, one a kid who tried to change his own oil, and didn't use the proper jacking point and crushed the bottom of the rad.

The other a '93 Civic with a plugged rad. It would try to overheat if you kept it at 5000 rpm or more for about 20 minutes. That girl used to drive like a maniac on the highway.

FredSVT:

I would point out that, from what I've seen available (for either of, say, my cars), the only construction is either aluminum or copper, with plastic end tanks. I haven't seen a brass radiator in 100 years. :D

The last all-metal radiator I can recall was on my 1983 Civic, and I think it was copper. :scratch:

The Koyo radiator I installed in my trusty '86 really "Super-cools." In fact, it slightly overcools during the Winter. No joking, I have to let the car warm up about 5 to 10 minutes, before the thermostat fully opens, and I can heat the interior. This particular radiator was made in Malaysia, and I am really impressed with the quality/performance/price. Thus, I'll give Koyo a big thumbs up! :thmsup: :)

Hi Robert,

I haven't seen a copper cored, plastic tanked rad in a long time. Last one was on a Toyota. They just don't last in our area with the winter crap thrown on the roads.
The radiator places here will still offer "all metal" rads for cars all the way up to current. No thanks.

Um, you said something that isn't likely, your '86, there should be NO coolant flow through that rad when cold. (stat closed)

If the stat is closed, all the coolant should go through the bypass circuit, and if the heat is on, it'll go through the heater core too.

If that stat is open, it'll dump the coolant into that rad, and then the car would never warm up. If it does actually take the stat to open and allow heat to the radiator and then you get heat in the cabin, something is not right.

Remind me, memory fails, which hose has the stat on the '86, upper or lower?

Well I have a slightly different take, I think the t'stat might be slightly stuck open/ not closing all the way. Every symptom you have posts point to that IMO. If it is slightly stuck open or not closing enough, the car will circulate the water too fast. If the water doesn't have a chance to sit in the radator and cool down before it gets to the motor it will overheat, it just takes some time. Exactly like your describing. Just my 2 cents.

FredSVT:

On Gen Three "A20A3" engines, the bottom radiator hose connects to the long, metal tube running along the back of the block, feeding the water pump. The top hose runs from the thermostat housing to the top of the radiator.

Gently disagreeing with both you and 09^CBT^6MT, allow me to explain why. Let's assume it's January/February, and 22 F outside. As you say, the coolant is running in bypass mode until the thermostat begins to open. However, that Arctic blast surrounding/flowing through my twin-flue Koyo significantly increases the time factor for warm up. I know this fact, from direct experience, comparing (in my fading memory) how the car used to operate with the single-flue, ND radiator I previously had. Previously, the temp gauge would sit around 3 o'clock-3:30, with the ND radiator, after reaching NOT. Now it sits around 4:30-5 o'clock, with the Koyo radiator, after reaching NOT, even in blistering August heat. With a "hot soak," while extended idling (say, in traffic), I get a "fan kick" around 3 o'clock-3:30.

BTW, I should mention that this Koyo radiator is copper core with plastic end tanks. I found it extremely well made, it works very efficiently, and it's been in the car four (4) years, as of next month. I think its intended application is for a "Ricer" motor, running a "Bottle" (NOX) and/or a Turbo setup, involving high-heat concerns. Regardless, I would not hesitate to buy another Koyo radiator.

Being PARANOID :paranoid: about overheating issues....after all, it was a faulty thermostat that wiped out the top end, in late 2001...I keep a spare, OEM Honda thermostat and gasket in the glovebox, at all times. If the temp gauge ever again goes above 60 percent, regardless of a "fan kick," I'd instantly shut it off, replace the thermostat on the spot, else have it towed back home (else to my dealer). I do NOT care to repeat a $2,000 mistake! A new cylinder head assembly, ALONE, cost me some $1,300, over four (4) years ago. (In mid-2005, Honda's stock position indicated only eight (8) units left in their California warehouse. I considered myself quite PRIVILEGED to get a new one!) I got off lucky on the bottom end, having only to lightly hone the bores, fit my pistons with a "standard" ringset, and I kicked in a new set of rod bearings, just for the Hell of it, since I was "already there." Other than end seals, I've never touched the crank.

Anyway, as usual, I digress. When I put the motor back together, I wanted all the additional cooling protection I could get. Explaining what I wanted to a rather pleasant fellow at Partswarehouse.com, he made my wish come true. Brown Truck delivered the Koyo to my front doorstep for about $114.00, as memory serves. :)

My "Old Girl," she runs like the proverbial $20,000 Swiss watch! Whirrs and purrs endlessly, with ZERO fluid loss. I have found Honda's "A" and "B" series engines to be very solid, reliable, "Work Horses." :yes: :hug:

Well I have a slightly different take, I think the t'stat might be slightly stuck open/ not closing all the way. Every symptom you have posts point to that IMO. If it is slightly stuck open or not closing enough, the car will circulate the water too fast. If the water doesn't have a chance to sit in the radator and cool down before it gets to the motor it will overheat, it just takes some time. Exactly like your describing. Just my 2 cents.

Sorry, but the recent posts have been about Robert's '86 radiator so I'm not sure if you are responding to my orignal '99 Accord overheating problem...but I think you are.

In any case, I'm learning alot about the cooling system of my car. Can someone please clarify another point for me? My overheating issue appears when the car is stopped or stuck in traffic (going less than 20 mph). When the car starts moving above 20 mph again, the temperature goes back to normal operating.

Question: Does the water pump pump harder (higher pressure) or faster (higher flow) as the car speed increases (i.e. pumps less below 20 mph and pumps more when above 20 mph)?

If so, this may explain how the car returns to NOT when driven above 20 mph if the faulty thermostat is not allowed to fully open. My reasoning is that more flow will allow for sufficient cooling therefore a return to NOT. Higher pressure will overcome the restriction created by a semi-open t-stat thus allowing more flow for heat exchange.

In regards to requiring more residence time in the radiator for cooling, I would think the radiator's surface area would have a greater influence and more effective cooling than residence time. Just MHO.

Can someone please clarify another point for me? My overheating issue appears when the car is stopped or stuck in traffic (going less than 20 mph). When the car starts moving above 20 mph again, the temperature goes back to normal operating.


The fans won't have any effect on the temps when moving over about 20-25 mph. Sitting still, definitely.


It sounds like the fans not coming on at idle or going slow maybe the problem or part of the problem. On a different thread you mentioned your still running on the original plugs and wires. Prob original cap too. Not sure if any of this is related to your problem or not.

To All:

Cooling system pressure is governed principally by the radiator CAP. It incorporates a two-way valve, to allow overage to flow into the coolant reservoir, after the coolant is heated and expands. Conversely, the cap allows coolant to flow back into the radiator as the coolant cools off and contracts.

Salient Point: A faulty radiator cap (bad seal and/or spring) WILL cause a cooling system malfunction.

Depending upon the engine, the cooling FANS are governed by thermosensors (switches) located in the radiator, the thermostat housing, possibly the head and/or the cylinder block. The ECU, since it's taking a feed off of the temp sensor(s), can engage the fans, regardless of what the engine is doing. I KNOW how all this works on my trusty '86, but I haven't yet studied all the electrics on my '02.

Salient Point: If you have a defective thermosensor, the cooling fans will likely not work properly. And you'll probably get a CEL.

The engine thermostat regulates cooling system temperature. The DEFAULT position is CLOSED. It gradually opens, as the head/block warms the coolant, to control system temperature, in a cycling effect, depending upon engine load.

Salient Point: If you have a defective thermostat, coolant temperature is not properly regulated, and it is most likely causing the thermosensors to feed misinformation to the ECU. If the ECU gets whacky feeds, then it tries to compensate by modifying ignition and fuel events. It may also turn on the radiator fans.

(I KNOW this fact, firsthand, as a defective engine thermostat in my trusty '86 was causing a fan kick, after I shut the engine off. Thermoswitch "A" or "B" (in the thermostat housing) was overheated, completing the circuit to the fans. After replacing the thermostat, the problem instantly went away! :D)

Cooling system issues on the old cars sure were a Helluva lot SIMPLER and CHEAPER to fix! :thmsup: :lmao:

It sounds like the fans not coming on at idle or going slow maybe the problem or part of the problem. On a different thread you mentioned your still running on the original plugs and wires. Prob original cap too. Not sure if any of this is related to your problem or not.

That is correct...215K+ miles and still running on original tranny, spark plugs, wires, and thermostat. Car starts w/o issues (starts on first crank) and still getting 25-28+ MPG with ~70% highway driving. I think this is still considered good MPG for a 3.0L V6.

The initial diagnostic was that the fan is not kicking on due to a faulty engine coolant temperature (ECT) sensor. The ECT sensor was replaced and the fans can be verified to kick on and off accordingly. Also, there are currently no CEL's.

The radiator and radiator cap was replaced about 1 yr ago as a result of similar overheating problems (i.e. overheats when stopped or in traffic, but otherwise no overheating). After the replacement radiator, I did not notice any additional overheating issues until about 1 month ago. Hence, I thought the problem was fixed with the new radiator.

My project for this weekend will be to flush the cooling system and replace t-stat and use OEM t-stat and Honda blue coolant. I am skeptical that this will resolve the overheating problem, but this is a simple and cheap attempt to resolve the issue. I'll also replace the radiator cap as well with an OEM cap.

Damn bro If I was you, I was also swap the cap, rotor, plugs and fuel filter and you might get a nice bump in MPG. You MUST take some pics of the plugs if you do pull them. I am very curious to see what they look like after 215K miles.

27-28 is a little low for all highway mileage. The '98 here with 233k on it gets 33-34 on highway runs.

I'm glad you mentioned that it's only getting hot at slow speeds, I guess I missed it a long time ago. Are you sure the fans are coming on when they are supposed to? It shouldn't get over 200 before the fans come on and go off at 180. The switch for the fans is in the thermostat housing. I hope they didn't replace the sensor for the computer which has nothing to do with the fans operating.

Number 14 in the following picture is the fan switch. The other two are the gauge sender and the computer sender.

http://www.hondapartscheap.com/southbay/jsp/prddisplay.jsp?hidSwitch=Switch&hidIrno=|014|015|013&catcgry1=ACCORD&catcgry2=1999&catcgry3=4DR+EXV6&catcgry4=KA4AT&catcgry5=WATER+PUMP-SENSOR+%28V6%29&vinsrch=no&ListAll=&prdrefno=&act=&count=0&quantity=0&qty0=1&qty2=1

To All:

Cooling system pressure is governed principally by the radiator CAP. It incorporates a two-way valve, to allow overage to flow into the coolant reservoir, after the coolant is heated and expands. Conversely, the cap allows coolant to flow back into the radiator as the coolant cools off and contracts.

Salient Point: A faulty radiator cap (bad seal and/or spring) WILL cause a cooling system malfunction.

Depending upon the engine, the cooling FANS are governed by thermosensors (switches) located in the radiator, the thermostat housing, possibly the head and/or the cylinder block. The ECU, since it's taking a feed off of the temp sensor(s), can engage the fans, regardless of what the engine is doing. I KNOW how all this works on my trusty '86, but I haven't yet studied all the electrics on my '02.

Salient Point: If you have a defective thermosensor, the cooling fans will likely not work properly. And you'll probably get a CEL.

The engine thermostat regulates cooling system temperature. The DEFAULT position is CLOSED. It gradually opens, as the head/block warms the coolant, to control system temperature, in a cycling effect, depending upon engine load.

Salient Point: If you have a defective thermostat, coolant temperature is not properly regulated, and it is most likely causing the thermosensors to feed misinformation to the ECU. If the ECU gets whacky feeds, then it tries to compensate by modifying ignition and fuel events. It may also turn on the radiator fans.

(I KNOW this fact, firsthand, as a defective engine thermostat in my trusty '86 was causing a fan kick, after I shut the engine off. Thermoswitch "A" or "B" (in the thermostat housing) was overheated, completing the circuit to the fans. After replacing the thermostat, the problem instantly went away! :D)

Cooling system issues on the old cars sure were a Helluva lot SIMPLER and CHEAPER to fix! :thmsup: :lmao:

Hi Robert,
I think you're also seeing a change in operation (proper) now that the cooling system is clean. You mentioned that the system wasn't maintained prior to the failure, with rusty, dirty coolant, it can't dissipate heat as well. Is your '86 injected?

My Integra GSR, would warm up at a much slower rate in the winter. The thermostat would never open in my daily drive in temps below 40, the aluminum block and heater core would dissipate so much heat that it just couldn't there. I ended up putting in a block heater to speed warm up and improve winter fuel economy.

The '04 Accord V6 is made like many newer cars, a slave to emissions, it's made to restrict coolant flow with the stat closed to force it to heat up faster. Then the importance of the stat is paramount as it's got to open to keep the engine cooled properly.

Just a thought.
If the water pump was not working good, the water flow at slower speeds, mite not be high enough. Do you know any one with a scanner that dose live data mode (allows you to look at the sensor data, mite give you a hint what going on).

I'd want to compare what the PCM is seeing vs what the temp with an infrared gun is.

The temp gauge shouldn't move at all before the fans come on.

If the water pump isn't moving enough water at idle, it really shouldn't be moving enough off idle either. Unless the impeller has been eaten away by cavitation, which is a possibility using the wrong coolant. Some will move water at idle, but at higher speeds they stop. The impeller slips on the shaft.

On my Toyota Truck, my wp disintegrated (I did not know at the time). It ran OK. Hotter at a stand still, not over heating, bumper-2-bumper, it was getting hot, slow steep grades, it was getting hot (hotter than normal, but not seriously over heating). It was normal driving on the free way or in town. The wp was making noises, so it got replaced. Once it was off the truck, there was a shaft, but no impeller (or any pieces found). Ether convection or the spinning shaft was allowing the cooling system to almost work OK???
BTW; I have a 4-row copper/brass real radiator...

27-28 is a little low for all highway mileage. The '98 here with 233k on it gets 33-34 on highway runs.

I thought for a 3.0L V6, the 27-28 MPG was good. 33-34 MPG sounds really high for a V6. I don't think I ever got anything higher than what I'm getting now even when the car was new. I was even told that I must have OCD for jotting down the mileage after ever fillup for the past 10 years.:yes: Maybe age is catching up to me and I forget...


I'm glad you mentioned that it's only getting hot at slow speeds, I guess I missed it a long time ago. Are you sure the fans are coming on when they are supposed to? It shouldn't get over 200 before the fans come on and go off at 180. The switch for the fans is in the thermostat housing. I hope they didn't replace the sensor for the computer which has nothing to do with the fans operating.

I'm not sure at what temperature the fans come on/off or if they are coming on/off at the correct setting. However, I have witnessed the fans starting and stopping w/ the car in park. Maybe the fans are not coming on soon enough or does not come on until the car overheats. Will replacing the fan switch (part# 14) affect (i.e. fix) when the fan kicks on or off or is its function just like a typical switch...either on or off?

Number 14 in the following picture is the fan switch. The other two are the gauge sender and the computer sender.

Part Number 15 is what got replaced. It was described to me as the engine coolant temperature sensor. If this sensor sees a high temperature, it sends a signal to the fan switch (part# 14) which engages the fans on or off. Is this description even close to what the function of this part# 15 does?

BTW, I want to thank everyone for your continued interests and contributions to this thread.

Che,

Have you checked to make sure both fans are coming on when in park? Was an oem water pump used when it was replaced? I think Fred asked that earlier but I didn't read your answer to that question.

I thought I saw both fans come on. But the fan on the passenger side definitely came on and off as the temperature gauge went over 2/3 or 3/4.

The water pump and timing belt were changed in Jan '09. I don't believe these were OEM parts. I didn't have any overheating problems after the water pump/timing belt change. Maybe the midwest winter ambient temperatures were sufficient to hide the overheating issue. It wasn't until June (when ambient temperatures are >75 Deg F) that I noticed the temperature excursions. I would think if it were the water pump, I would have overheating issues regardless of what the ambient temperatures were.

Can someone help explain how the water pump works or direct me to a source where I can read up on this? Is the pump driven by belts that correspond to the speed/load on the car? More specifically, does the pump run faster if the car is moving faster or does the pump run at a fixed speed/rate and pumps the same flow at a fixed head pressure regardless of how fast the car is moving?

After looking at some diagrams for your WP, it looks to me that it is a direct drive. The pulley is connected to the pump on a common shaft and the pump is spinning at the same speed at the accessory belts. So if the pump isn't leaking anything, it should be good. The design doesn't fail very often, unless water starts to come out of the weep hole by the bearing and seal. BTW I just google imaged your water pump to find info on design. Not sure on your model, but if you turn on the ac on, isn't one of the fans designed to kick on?

With A/C both fans should run.
The pump is driven by the timing belt and it runs slightly slower than the crank as the pulley is somewhat larger.

The impeller is still pressed on the water pump shaft and if they used a reman pump (YUCK) the chances of a slipping impeller is quite good.

The impeller is still pressed on the water pump shaft and if they used a reman pump (YUCK) the chances of a slipping impeller is quite good.

There is no way to test this without the pump coming off correct?

On the V6, yes and no. You can run a flow test with clear tubing in place of one of the hoses, which is a pain trying to find such hose.

Then there are two other more invasive ways, pulling the pump. Or pulling the water crossover and getting the water pipe out of the valley, then you can access the hole at the back of the impeller and try and hold it while HAND cranking the engine.

To me, pulling the pump would be easier.

OK...this weekend I finally got around to flushing the cooling system and replacing with the honda Type 2 (blue) coolant. I also replaced the thermostat with and OEM T-stat. Everything seemed fine after the repair. I even drove it for an hour (~25 miles total), mainly city driving between 25-40 mph with intermittent traffic light start and stop. I was also on the expressway traveling 60 mph for roughly 15 minutes during this 1 hour drive. The car drove normally and no overheating was observed throughout. Also did not overheat when the car was parked after the ~1 hour drive.

Today I drive the car to work (~50 miles) for a duration of ~65 minutes with mainly highway driving (mostly staying at 60-65 MPH). (The difference with today's drive is that the car was driven at a higher speed for a longer duration). When I reached the parking lot, I placed the car in parked and began observing the temperature gauge on the dash. Within 5 minutes in park, the temperature gauge began to rise. I can hear the fan kick on and slowly I can see the temperature needle decreasing. When the fan kicked off, the temperature rose and the fan kicked on again. The on/off fan cycle continued and with each cycle the temperature rise was less.

Long story short, these symptoms are identical to what was happening prior to the coolant flush and T-stat replacement. I'm at lost as to what could still be the problem. Any other thoughts?

Thanks in advance.

OK...this weekend I finally got around to flushing the cooling system and replacing with the honda Type 2 (blue) coolant. I also replaced the thermostat with and OEM T-stat. Everything seemed fine after the repair. I even drove it for an hour (~25 miles total), mainly city driving between 25-40 mph with intermittent traffic light start and stop. I was also on the expressway traveling 60 mph for roughly 15 minutes during this 1 hour drive. The car drove normally and no overheating was observed throughout. Also did not overheat when the car was parked after the ~1 hour drive.

Today I drive the car to work (~50 miles) for a duration of ~65 minutes with mainly highway driving (mostly staying at 60-65 MPH). (The difference with today's drive is that the car was driven at a higher speed for a longer duration). When I reached the parking lot, I placed the car in parked and began observing the temperature gauge on the dash. Within 5 minutes in park, the temperature gauge began to rise. I can hear the fan kick on and slowly I can see the temperature needle decreasing. When the fan kicked off, the temperature rose and the fan kicked on again. The on/off fan cycle continued and with each cycle the temperature rise was less.

Long story short, these symptoms are identical to what was happening prior to the coolant flush and T-stat replacement. I'm at lost as to what could still be the problem. Any other thoughts?

Thanks in advance.
Maybe im confused but it sounds to me like the car is doing EXACTLY like it's supposed to do. When the car temp rises, the fan cuts on to keep the engine at a nominal temp. When it gets there, the fan shuts off until it needs to cut on again. Just like any cooling system would do. Your home does this as well. Or did i miss something from your post?

The fans should and does kick on and off. However, you should never see the temperature gauge go higher than normal operating (which is 1/3 of scale). My temperature excursions go from NOT (1/3) to the maximum or the red area of the temperature gauge. This is not typical for a car w/o overheating issues. In addition, I also observed that the coolant reserve tank is full when the temperature excursions are noticed. This implies that the coolant inside the radiator is too hot and the radiator cap (relief valve) is opening to let the hot coolant into the reserve tank.

On Honda's you have to bead the cooling system, did you do this?

On Honda's you have to bleed the cooling system, did you do this?

:thumbsup:

I think Fredsvt indicated that there are no bleed valve for the V6. I'm not sure how else to bleed the system.

After filling the radiator with coolant, with the radiator cap still off, I started the car with the heat on high and fan on max. I let the car warm up and waited for the fans to cycle once. I then squeezed the upper radiator hose (which was hot at this point) to force air/coolant out before putting the radiator cap back on. Did I forget to do something or is there another way to bleed the air from the coolant system?

The V6 needs no bleeding, the rad is the highest. Doing as you did will make sure it's free of air.

Get a fan switch in the thermostat housing and replace it. I think it's not reading properly and turning the fans on late.

Were the ground wires clean that go to the stat housing? I believe the V6 also uses them.

The V6 needs no bleeding, the rad is the highest. Doing as you did will make sure it's free of air.

Get a fan switch in the thermostat housing and replace it. I think it's not reading properly and turning the fans on late.

Were the ground wires clean that go to the stat housing? I believe the V6 also uses them.

Fredsvt - Thanks for the confirmation on the air bleeding issue. You are correct in regards to the V6 having ground wires on the t-stat housing. This ground wire is located at the bottom center of the housing, straddling between the 2 housing assembly bolts. The ground wire was clean, free of any buildup. I had to loosen and remove this bolt and wire in order to take the t-stat housing off so I had a good look at it to verify cleaniness.

In one of my previous post (#30, requoted below), I mentioned part 15 (ECT sensor)was replaced. Are my functional description of part 14 and 15 accurate?

I'm not sure at what temperature the fans come on/off or if they are coming on/off at the correct setting. However, I have witnessed the fans starting and stopping w/ the car in park. Maybe the fans are not coming on soon enough or does not come on until the car overheats. Will replacing the fan switch (part# 14) affect (i.e. fix) when the fan kicks on or off or is its function just like a typical switch...either on or off?

Part Number 15 is what got replaced. It was described to me as the engine coolant temperature sensor. If this sensor sees a high temperature, it sends a signal to the fan switch (part# 14) which engages the fans on or off. Is this description even close to what the function of this part# 15 does?

Does the switch determine when the fans come on or is it the car's computer that sends a signal to the switch that determines when the fan comes on or off?

Ok, I had to go back and look.

There are TWO fan control switches that report to a radiator fan control module. The sensor, 15 is for the PCM only. The PCM input and the fan inputs are completely isolated.

Both switches are there to activate the fans.

Switch A, at the thermostat housing should turn the fans on at 199*, and they should shut off at around 180*. If the gauge climbs, switch A could be bad.

Switch B turns the fans on at 225*, this WILL make the gauge climb.

(pictured) http://www.hondapartscheap.com/southbay/jsp/prddisplay.jsp?hidSwitch=Switch&hidIrno=|010&catcgry1=ACCORD&catcgry2=1999&catcgry3=2DR+EX+V6&catcgry4=KA4AT&catcgry5=FRONT+CYLINDER+HEAD+%28V6%29&vinsrch=no&ListAll=&prdrefno=&act=&count=0&quantity=0

I'd say that switch A is bad, a bad wire or the fan control module is bad. To check fan switch A wiring, unplug it at the t-stat housing, KOEO, jumper the green wire to ground, the fans should come on.

The module is under the dash, 8 pin connector.

Here's a pinout of the module (lock side of connector up) KOEO (a/c off), leaving connector plugged in.
1234
5678

1- 12v (yellow wire)
2- 12v (blk/yel)
3- 12v (yel/wht)
4- ground, make sure there's zero volts between it and body ground.
5- 11v or more at temps less than 225* (wht/grn)
6- 12v (white)
7- 12v (blk/yel)
8- connect a jumper wire from this to body ground, both fans should come on. (green)


2-

Sorry Fredsvt, but I have yet to follow through on your suggested tests with the fan swtich and fan module. I'm sure I still have the overheating problem, but now another issue has arisen. I have not driven this accord for the past 1-2 weeks since the discover of this "other" problem.

I don't know if it is a simple fix or not or even if it is more critical than the oveheating problem, but my speedometer has gone crazy. The speedometer mostly read correctly, but then it would erractically jump to 100 mph and back to normal or to zero. I can't imagine this is related to my recent work on the thermostat replacement. Additionally, I also changed out the air filter when I replaced the thermostat. Not sure if this problem is related to the air filter & thermostat replacement or if it is related to the overheating problem.

If this new problem is in no way related to the overheating, I will start a new thread with this speedometer specific issue.

Thanks,

I don't think the speedo issue is related. If you have access to a scanner, watch vehicle speed on it. If it is as erratic as the speedo, more than likely the VSS is flaking out.

If it doesn't then possibly the speedo head could be bad. I'd have to look into seeing if Honda filters the car's speedo reading through the PCM or not.

To Che and FredSVT,
Thank you both for continuing the thread about the overheating issue. It helped me a great deal today.
I had a similar problem which strted 2 days ago.
I was overheating real bad, and discovered a busted heater hose(inlet side). Took me about 3 hours to take everything apart to get to the hose, (i usually don't go beyond simple maintenance tasks like oil changes) but I finally got it, replaced the hose, and began a coolant flush. While the H2O was running thru the system, I noticed the temp gauge was climbing past the half way mark. (like you said, it's supposed to hover around the 1/3 mark) On a hunch, I drove to the dealer and picked up a new thermostst. Oh, btw, I forgot to mention that my fans were not coming on either, and only one fan would switch on when the A/C was turned on.
Installed the thermo and let the car idle for about an hour. temp gauge stayed where it should, and the fans (both) kicked on and off every few minutes or so.
Tomorrow I will be completeing the flush and re-filling with Honda fluid after your advice.
Thanks again for helping, even though you didn't even know you did.

P.S. Fred,
I was born and raised in Morganville. Roosevelt Ave, off 79.