[violet672]

Hi all

We have been reading/searching this site for days and I have finally decided to post my question as we have had no fewer than 5 mechanics (including the Honda dealership) absolutely stymied by our problem.....

1999 EX 185,000 kms. (Canada)
No problems at all with the van except the failed tranny 2 weeks into ownership 6 months ago at 176,500 km (yikes!) Last week my husband installed a new timing belt, tensioner and pulley, water pump (brand new), thermostat, p/s and serpentine belts and serpentine tensioner. All from NAPA, all aftermarket parts. Everything went pretty much like clockwork.

We refilled the coolant and ran through the Service Manuals instructions for reving the engine at 1500 rpm in 5 min. intervals alternating between cold and hot etc. The only thing we did wrong, in our ignorance, was perform this with the rad cap ON. Heat blew nice in hot in both front and rear of the vehicle, engine temp on dash gauge was normal. No problems. (or so we thought). 2 days later, as I was driving on the highway the heater gauge started to climb from just above mid-point to the red area as I approached my home. I slowed right down, turned a/c off and went very slowly...

So, now we have had 4 mechanics (including the Honda dealership today) and a 5th over the phone looking at our vehicle. The thermostat had been changed back to OEM with no improvement, we have tried 'bleeding' the system for countless hours....and now have finally removed the thermostat to try to find out WHY the top rad hose is consistently 40 degrees hotter than the bottom (and why the coolant wants to boil out of it when it's open).

With the thermostat installed, the top hose would reach 195-198 before it would drop at all, would only drop 10-15 degrees and the bottom hose never comes up above 155. The coolant would back out of the radiator into the large funnel and then eventually start boiling. We even installed a bleeder on the back top heater hose, checked for coolant flow to it (no problem), but the hose temps are a mystery and the van starts to overheat shortly after the rad cap is on even at an idle.

It's not the temp sensor, as he used a paperclip to create a continuous circuit and keep the fans both on (advice of non-honda mechanic), and the hoses are giving the same temp differentials with the thermostat OUT of the housing (that is, slightly cooler but still 40 degrees difference). Temps only ever reach 102 on all rear heater hoses, while the top rad hose off engine reaches as high as 198 (and if you check bleeder on back, there's fluid there).

Is the new water pump bad, do we have an extremely bad airlock (the entire system has now been drained and refilled 3 times), a possible physical block in the system somewhere, small leak, or .... eghads.....a failing head gasket? None of the mechanics think it's the head - there are no other signs of it, and the van never overheated to the point of steam etc. Just hit the red line that one time on the way home the other day.

It is now parked outside a different mechanic's shop with an extremely good reputation (the one giving us advice over the phone) awaiting monday morning. Honda was pretty much useless - including saying they couldn't change the oil for us as the bolt was on too tight! (Need a new $250 oil pan.....hmmmm....). $200 later in towing and diagnostics and we're no farther ahead (and 3 kids wondering when we can go to Disneyland!)

I think I'll ask for a pressure test first thing to rule out head gasket leak or other leak. Any thoughts/advice would be GREATLY appreciated. We've had this big holiday planned for months and are supposed to head out in the morning here. Not going to happen, and, like I say, we've been working on this for days!!!!!

Sorry for the long post - hoping someone has got advice, and that maybe someone else can benefit from it at some point too!

Thanks so much.
Bonnie

 

[0dyfamily]

Bonnie, some of those symptoms are indicative of a leaking head gasket. Yes, a pressure check looking for bleed-down (and the leak(s) that cause pressure loss) is a good a idea.

One question: when your hubby was performing the coolant refill procedure with the radiator cap on (installed), did he also do this with a full (to the top mark) coolant overflow reservoir?

The reason I ask: If not enough coolant is in there, when the cooling system goes through an expansion cycle and purges air from the top radiator tank (that plastic raised top end portion of the radiator), it goes into the reservoir and bubbles to the top.

As well, when the engine cools after shutdown (along with the coolant), the coolant contracts and withdraws coolant from the reservoir. This is important. If there was not enough coolant in the reservoir to keep the tube attached to the reservoir cap continuously submerged in the reservoir's coolant, then air is just simply being expelled and re-introduced into the engine cooling system. Air in the system increases risk of air lock.

I just did a full timing belt service on our 98 Accord, raising the front end using ramps and performing the shop manual refill procedure. It was tedious.

....**SNIP**.... So, now we have had 4 mechanics (including the Honda dealership today) and a 5th over the phone looking at our vehicle. The thermostat had been changed back to OEM with no improvement, we have tried 'bleeding' the system for countless hours....and now have finally removed the thermostat to try to find out WHY the top rad hose is consistently 40 degrees hotter than the bottom (and why the coolant wants to boil out of it when it's open).

With the thermostat installed, the top hose would reach 195-198 before it would drop at all, would only drop 10-15 degrees and the bottom hose never comes up above 155. ....**SNIP**....

You should see a pretty stout temperature drop between the top and bottom radiator hoses, but I don't know what it's supposed to be. The low temperatures on the rear heater hoses is something I can't explain. Those two together, to me, possibly indicate low coolant flow, which may mean an airlock or a bad water pump.

The boiling worries me. Letting it slowly warm up at idle, if it's moving enough coolant through the radiator, it shouldn't boil. Are you sure this is boiling, or is it bubbles of air burping out with the warming coolant that will overflow when it expands without the cap on? If a cooling system in any car is full, and you remove the cap while cold, you'll see coolant up to the top of the fill neck where the cap was. Turn on the engine and run it, the warming coolant expands, and overflows out the top. If you are seeing air burp out, it's coming from somewhere. That is why I am thinking a possible leaking head gasket, or other source of air (I sure hope it is not a cracked block or cylinder head).

That said, the Honda water pumps are pretty simple and robust affairs. Every one I've removed for a timing belt service (I'm a DIY guy, owned Hondas since the early 90's) looked like it could have been re-installed, with good bearing endplay (none) and a good gland seal (tight). I can't think of any material factors that would cause a new water pump built for a Honda by any manufacturer to not move coolant. Its design is truly that simple and reliable.

Good luck, keep us posted on the pressure check.

OF

 

[violet672]

OdyFamily - Thanks so much for your reply (and taking the time to read my long post!)

Yes, the reservoir tank was full during the whole initial purge cycle with the radiator cap on. The level in the radiator never dropped between any of the cycles, however (when turning off the engine and checking the radiator level as indicated in the service manual). We didn't realize this may not have been normal.

It just seems so strange that everything was fine for a few days after the new install of everything - then this! If it IS a leaking HG, would it just be an extremely strange coincidence, or is it possibly something we did wrong?

The only other minor detail here that may have some bearing is that the coolant level in the reservoir had been topped off at the service station about a week before my husband did the TB install. It apparently was at the min line when the attendant checked my oil. Don't know how long that had been for - and do remember wondering at the time if it was normal for an engine coolant level to have dropped (I NOW know MUCH more about coolant systems than I ever wanted to know!! )

Thanks again - Bonnie

 

[gregersonke]

Make sure also that none of your coolant hoses are leaking. This includes the ones going to the heater core and the ones leading into the intake throttle body. If any of these have a pinhole leak you will have the temp problems you describe.

 

[violet672]

That gives me faint hope - how I wish it might be a pinhole leak!!

We'll definitely check for that, too. Haven't seen any visible coolant leakage beneath the vehicle, but I remember reading somewhere to run a paper towel or the like over the hoses to check for any small amount of fluid after the system is warmed up.... is that the correct procedure?

I'd have to say that would be a strange coincidence, though, if my husband hadn't touched these coolant hoses! hmm....

What about the fact that we also installed new NGK platinum sparkplugs? Could this (strangely) have any effect on the temp issues?

OdyFamily - I forgot to mention earlier that yes, the radiator is in fact reaching a boil when the cap is off and no-spill funnel is attached. Initially we see intermittent air bubbles, so big some small, but it will eventually start to boil - especially once the a/c is switched on. When this happens, the temp in the boiling fluid is actually 10 degrees higher than the top hose temp (which will have dropped to about 180 from high 190s).

 

[SuperDad]

There are any number of problems in what you describe.

I'm also thinking head gasket/cracked head is possible, but I'll explain a bit...

With the improperly bled coolant, it is likely that one or both heads were getting shock-cooled or not cooled, as the hot coolant sloshed around. This absence of cooling or shock-cooling is REALLY hard on the head.


The coolant shouldn't be boiling at anywhere below 210F. The antifreeze should increase the temperature of boiling, even without the radiator cap on. Once the cap is on, the boiling point should jump another 10-30F due to the pressure. I actually think bad coolant is the most likely cause of your woe.

Are you sure of the quality of your antifreeze, and that it is diluted(or not diluted) properly?

Your radiator temperatures seem fine to me. I haven't looked in the manual yet, though. (The manual states full-open on the thermostat is only achieved when it sees 194F) The flow in the radiator is doubtless top-to-bottom, so losing ~40F is expected. Maybe more is expected, but maybe the NAPA coolant has shredded or partially blocked your radiator(This is only a guess). In any case, it doesn't seem that unreasonable.


It is unlikely to be a bad water pump. There really isn't much to go wrong with the water pump. Their failure modes are typically seizing or leaking.

 

[SuperDad]

As a test of both your thermometer and your coolant, boil some plain tap water on your stove, and check that its 210F or so.

Do the same with some of your coolant. If it starts boiling below 210, your coolant is just garbage. Its probably garbage if it boils anywhere below 225 or so, too.

(I know its obvious and insulting, but you didnt' use "Windshield Washer Anti-Freeze" in your cooling system by accident, did you?)


BTW, boiling coolant indoors isn't a great idea. Do it with lots of ventilation, or maybe outside on the grill.

 

[d-rex]

bad water pump-either the impeller shaft broke or impeller isnt spinning some how
the conditions you described are what happens with no coolant flow
i have had aftermarket water pumps do this to me before
the impeller is pressed on the shaft and works itself loose somehow
anyway you slice it the water pump is not pumping

 

[violet672]

SuperDad - we used Prestone long life pre-mixed from Lordco. My husband had checked with Honda that this would be ok, which they had said yes to. We even then double-checked on Prestone's website that it contained no silicate and would be compatible with the aluminum engine etc. Of course, at this point I'm REALLY wishing I'd made the 30 km trip into to the dealership to get their coolant, but it didn't appear from any info I was getting at the time that it should be an issue.

But what it sounds like you're saying is that possibly the improper way we purged the system initially caused some air pocket in the system to form overheating pockets in the engine.....would this be why it was fine for a couple of days before the overheating issue started?

d-rex - you said you had an aftermarket waterpump fail. Was it right away like this (day 1 or 2), and also, was it for a Honda vehicle that you were replacing? Nobody else seems to think that there should be an issue with the replacement part, although it does seem somewhat intuitive (although my husband did a very thorough side-by-side comparison of the 2 parts which were virtually identical).

I really appreciate everyone's responses here. The more we have to go into the shop with tomorrow when it opens, the better!

Would love to get heading out on the road.....

 

[SuperDad]

Yes, my theory is that the heads would be left with uncooled pockets of hot air, then you'd turn, hit the brakes, whatever, and the coolant would slosh into them, instantly quenching them from 500F to 220F. Do that enough and metal fatigue will occur.

I'm still not convinced the water pump is not working. If it was not spinning at all, the radiator would only get hot due to convection. You should be able to check this easily, if the thermostat is out of the engine.

Start the engine cold, and look for flow across the top of the radiator. If there isn't any flow, the water pump is suspect(may be hard to see the flow, which is the only reason you can't immediately call the pump bad).

If there is flow, the water pump is fine. Might want to rev the engine a bit to get the water pump spinning faster to see the flow.

 

[SuperDad]

I assume the DH checked that the water pump could spin freely after installation.

I recall a Peoples Court from years ago where a guy was suing because the new water pump he installed was defective and wrecked the engine because the impeller was contacting the engine castings. He lost the case because the dealer would have recognized the dragging impeller before putting everything back together.

 

[violet672]

Yes, we could see coolant movement across the top of the radiator last night once the thermostat was removed - not fast, but definitely movement. Of course once it started warming up, it started backing out and eventually the coolant boiled.

He definitely checked for free play of the bearings before the WP was installed, and then after it was installed we had done the minimun crank cycles noted for the crankshaft to check for correct alignment of the timing marks on the top cam shafts. There were no sounds etc. as you mention might occur with improper clearance of the impellers.

Sounds like we're leaning back towards the air pocket fiasco. I guess the pressure test in the morning will confirm/rule that out.

Just to confirm - are you saying that it is definitely possible for head damage to occur with no visible steam and no red-lining on the temp gauge inside until after the hg damage has occured?

Thanks again. Bonnie

 

[violet672]

One more thing here which may or may not have a bearing on the situation.

The reason my husband intially put on the rad cap at the beginnin of the purge cycle was because the coolant started splashing out quite heavily after the 'warm up' stage (where we waited for fans to cycle on and off twice). Not just little splooshes, but the hot, boiling-type of splashes. Since the service manual didn't specify anything regarding the rad cap until step 25 or something where it said to install it loosely (oops, read that step a little late!), we assumed you wouldn't want coolant splashing all over the garage so installed the rad cap.

Could this fact, combined with me having had to top off the coolant reservoir earlier that week, point to a pre-existing but not yet obvious head gasket issue? Just wondering........

 

[d-rex]

air pockets would have been bled out by now i think-why did this happen a few days after?air pockets make the guage fluctuate rapidly as they burp through the system
i think the little bit of flow you see is the water pump barely pumping-what happens is the "blade" of the pump is not locked onto the pump shaft and thus will spin slowly but as engine load and speed go up"driving" the water pump shaft spins but the impeller just kinda lazily spins -not enough to pump the reqiured flow of coolant-which-causes hot pockets of boiling coolant in the engine which explode and burp through th easiest outlet{rad cap}THEN as the coolant boils you introduce air pockets in the engine and the cycle repeats

 

[mem]

violet672,

Don't worry so much about the bleeding procedure. If you had air in the system when done, then it would show up as as elevated temp reading and/or some transients in the temp reading. Unless you radically under filled things (hard to imagine) then you (or, should I say, your husband) didn't cause any harm.

And the "bubbling" during bleeding - after the fans came on - might just have been significant air in the system making its way out after the thermostat opened (that air and/or steam was being held in the water jackets by the pressure of the water pump, and the small bleed hole on the thermostat couldn't deal with that). No big deal. And most of that air probably exited before DH put the cap on. Over the next couple of days, any remaining air likely got moved over to, and out of, the expansion bottle...

Anyway, reading through this thread I can't help but believe that d-rex's conjecture is quite plausible. All he is saying is that the impeller may not, at present, be properly affixed to the shaft. Perhaps it was OK in the beginning, but that didn't last long. Here is an example of a case where the impeller was loose from the get-go (see the LAST post in this thread):

http://www.hyundaiperformance.com/f...9511-water-pump-pulley-shaft-broke-today.html

And the "symptoms" you are seeing certainly are not inconsistent with a water pump being unable to do its job.

Could it be something else besides the water pump? Sure. Perhaps it could be the head gasket (unlikely - at least let's hope so) or an obstruction. An obstruction is hard to imagine - unless someone put a rag in an opening to keep coolant from leaking out (and then forgot about said rag) or if the lower radiator hose is collapsing. If it is a Honda hose then the latter is extremely improbable. The former is also extremely improbable since it would show up quickly. Something else? - maybe.

Anyway, it seems to me that what d-rex is suggesting should not be dismissed out-of-hand. Best of luck.

 

[SuperDad]

I have only a few answers...

Yes, you could have HG or head damage without reading a high temp on the gauge. The gauge is reading the water temperature, and if the water isn't in contact with the entire engine(head), there could be areas which have a very high temperature getting quenched every now and again.

I happen to have done my own timing belt about two months ago, and had absolutely no issues with the refill/purge process. No major eruptions, no nothing. Added coolant, ran the front and rear heaters, let it cycle a couple times, moved it to a slight nose-up incline after a while, repeated. Checked rear heat temperature output was within spec, went to bed.


The free-spinning-impelller theory seems to be gaining credence, but I still have a little trouble believing that it would fail so fast.

I'm still a little concerned about your reported temperatures of boiling antifreeze, because bad antifreeze could probably also cause some of these symptoms.

 

[d-rex]

The free-spinning-impelller theory seems to be gaining credence, but I still have a little trouble believing that it would fail so fast.


its a napa water pump-its possible-i have seen it happen before

 

[0dyfamily]

One more thing here which may or may not have a bearing on the situation.

The reason my husband intially put on the rad cap at the beginnin of the purge cycle was because the coolant started splashing out quite heavily after the 'warm up' stage (where we waited for fans to cycle on and off twice). ....****SNIP****we assumed you wouldn't want coolant splashing all over the garage so installed the rad cap.

I did the exact same thing when I did the Accord's timing belt this summer...by then, it was full enough, and all I had to do was top off the reservoir.

In short, it sounds like he did it correctly, and his coolant fill/purge actions do not seem to be the cause of this problem.

I think you're best bet, still, is to do that coolant system pressure check first. That will tell us something (rule out some things, or tell us that some things are still possible.)

Hang in there.

OF

 

[d-rex]

here what ya do -start the engine cold in the am-take off rad cap
and ck for coolant drop in radiator-if after a minute or less you see bubbles and excessive turbulence its prob exh gas {as in cylinder head failure}but im still pretty sure its the pump

 

[violet672]

Thanks so much for all your reponses and advice.

The van is in the parking lot of the shop awaiting opening in the a.m., so I guess we won't know anything more until then. It's been a LONG few days, and we have quite a few people wondering what on earth is going on here....

I like the water pump theory, as it is obviously an easier fix (and I would be so happy if it was - although happy is very relative at the moment ) BUT the overheated/failing HG unfortunately makes a lot of sense, too.

At least we have it narrowed down to 2 main potential problems. I feel better heading into the mechanics with this information and at least hopeful that we might make it out on the road within a few days at most (assuming parts aren't a nightmare).

Do any of you EVER recommend going with non-OEM parts, or should we go with strictly Honda parts from here on out?

I will keep you all posted - thanks again.