Proper Radiator Cap Pressure Rating for 1994 Honda Accord EX?

I picked up a thermostat at a Honda dealer and installed it last night.
So far, it does not seem to have made any difference in my symptoms.
Maybe the damage is already done. I will continue to monitor things
just to make sure that all the air has been purged from the system.

I am also wondering what part the sensors for the radiator fans
play in keeping the coolant from boiling. I have noticed that
after starting the car, it seems to take about 20 minutes before the fans
come on, and when they do, both fans come on at once, never just one.
When they turn off, they both turn off simultaneously. This happens even
when the AC is completely off in the car. Is it possible that a sensor is
not
"telling" one of the fans to come on at a lower temperature? Also, are these
fans set up for one speed or two speeds?

Thanks for the help.

 

monitor that coolant level.

the wrong temp thermostat won't "damage" the engine, but it'll change
normal operations. it'll revert beck to normal with the correct one.

there's only one fan temp sensor, so unless one is triggered by a
separate circuit like the a/c, they'll both come on together.

 

It CAN damage the engine, and that's Gary's threat right now. If the
cooling fluid boils for long enough during the engine's operation,
localized overheating may warp the head and cause gasket failure. He's been
driving around with the wrong 'stat for a quite a while now.

Honda blocks are quite strong and do not often distort unless severely
abused. It's normally the head that gets damaged.

Gary's next step (to be performed **ASAP** and by a more competent mechanic
than the one that put the 90C unit in) is to get a pressure check done, or
a combustion gas analysis performed. Either will confirm or deny the
presence of head gasket failure. Usually failure is dramatic enough to
create bubbles under air pressure.

 

Does anyone know what temperatures the sensor turns the fans
on and off?

 

Above 88.5C to 91.5C (it's a range, not an absolute figure).

 

but a 'stat that opens at 90 simply means the motor runs too hot - it's
still well below boiling, especially on a pressurized system.

correct, but localized overheating is where it's boiled dry.

yes, because that bit boils dry before the block does.

mine would only bubble under extended full throttle. the only clue was
that there was gas in the system so it wouldn't suck back properly.
purged, it would suck back fine under normal driving, and would behave
itself for some time until the car got given its next good workout.
only with extended repetition of this process over many months [i.e.
nearly 2 years] did this finally start getting bad enough to be
"noticeable", and even then, it was only because i was gunning grapevine
every couple of weeks. if i were not a picky kind of person and/or a
more gentle driver, doubtless it would have survived many more months
before degrading to the point where it got to be a problem.

 

OK. I did some experimenting. I was suspicious that the new 16 lb radiator
cap that I put on the radiator in replacement of the old 13 lb cap was not
working
correctly either. So I spoke to the previous owner, and found out that the
current radiator
is an aftermarket radiator. So I installed another similar 16 lb cap, and
things seem to be
working slightly differently:

Starting with a "cold" engine in the morning on a warm summer day, it takes
about 17 minutes
of parked idling (w/ occasional 2000 RPM) before the AC fan and radiator
fans come on.
When they do come on, they only come on for about 13 seconds, and they seem
to only be coming
on every 4 minutes. Even though the temperature gauge is only 1/4 of the
way above the C, the
engine "seems" hot. There is bubbling in the reservoir. However, the level
of the coolant in the reservoir
does not rise as long as the car is sitting still. If I take the car out on
the open road and allow the natural
flow of air over the radiator, I find that the reservoir level rises about
1-1/2 inches. And when I stop
the engine and things cool down, it drops back down 1-1/2 inches. This new
cap seems
to be making that possible. Also, I had taken your advice awhile back and
removed about 1" of possibly
ill-fitting hose between the radiator cap and expansion tank. The success of
this possible improvement
may have been masked by the previous faulty cap.

I am thinking (hoping) that the bubbling is not combustion gas, but rather
coolant that is boiling because the fans are
not coming on when they should. Both fans work, and they have new brushes.
No intermittent problems involving
the fans themselves. I suspect that the fans are coming on at a temperature
that is in excess of the boiling point
of 50/50 coolant at 16 psi, which is probably well over the 88.5C to 91.5C
range that you cited. When I drive
down the highway, I'm thinking that the natural air flow does what the fans
fail to do, and makes it possible for liquid
to transfer to the reservior instead of bubbles.

I see references to ECT & TW sensors, and Auxiliary Fan Switch. Are they all
the same thing? What is the sensor that
turns on the fans on 94 Honda Accord EX (4 cyl, 4door)? I suspect I could
remove the sensor and check for
contnuity by placing the sensing part in boiling water, just to see if it
responds.

This car has only been used for short trips over country roads, with very
little stop and go. It has almost 300K, and I suspect that
the sensor (whatever it's called) is original.

Thanks,

Gary

 

<snip>

ok, if you're getting bubbling /after/ you've purged the system, it's
head gasket. sorry to be the bearer of bad news. the higher pressure
cap may delay onset a little, but it won't fix it. the good news is
that the parts are relatively cheap. if you want to do the job
yourself, it takes about a day if you want to do the job cleanly and
well. if not, the local shop will charge about one months worth of new
car payment. if the gasket's done cleanly [*NO* sandpaper or other
abrasive used to clean anything up], you should be able to get another
50k to 100k out of this car.

 

Seems to me this assumes the head will not need machining.
That's a mighty big gamble to take.

But more than likely will check the head surface and send it
out for machining as necessary.

 

no it's not. honda heads don't warp unless they've been badly
overheated, like what happens if you lose coolant on the freeway and
don't bother to check the gauges. a little bubbling won't cause warping.

don't assume the head is warped. yes, a lot of shops will send it in
even if it's perfectly flat because they make more money that way, but
it doesn't mean the work is necessary. you should know that by now elle
- that's why you do your own work - you want it done right and you want
it done without unwarranted expense.

 

We disagree. To back this up, a friend of mine kept her 99
Honda Civic out of the red zone for a month or so while
there was evidence of a head gasket leak (overheating; loss
of coolant).

The shop said the head was warped and had to be machined.
(But keep reading.)

It's a gamble.

Toss in that it might not be a head gasket breach but a
breach elsewhere in the block, and it's an even bigger
gamble.

Sure, that's possible.

You should know by now that do-it-yourselfers have limited
tools which may preclude the checks--and certainly the
machining--of which we are speaking.

IMO too often you tell amateurs to take simple problems to
a shop, or spend money for this or that when it's not
well-warranted. So telling someone that they can do a head
gasket (and it might not even be the head gasket) is kinda
reckless.

Either way, the OP has some opinions on this at this point.
<shrug>

 

Is there any chance that the bubbling is coolant that is boiling because the
coolant
temperature sensor is faulty and not turning on the fans soon enough?

I might be willing to give it a shot. Especially, if there was a specfic
manual available for my exact car.

Gary

 

elle, you've evidently never had a honda head off. the liners are held
by a thin alloy casting, then there's a huge water gap, then the block
starts again. because the casting is so thin around the liners, there's
very little thermal stress because there's very little heat
differential. and guess what, that's why honda blocks hardly ever
crack. but gaskets leak because the down side of this thin "free" liner
concept is that it allows the liners to move slightly relative to the
head. but that's a different story.

regarding warpage, it's real easy to test. get a metal sided spirit
level or even a steel straight edge, and a set of feeler gauges. it's
immediately apparent if there's warpage, and if there is, you can
measure with the gauges the extent and whether it's within limits.

measuring warpage is real easy. if it's warped beyond limits, take it
to the machine shop! but if it's not, don't waste the money. indeed,
/definitely/ don't waste the money because there's more to it than just
the cost - there's the effect on compression ratio and valve clearance.
machining warped heads is questionable too because of the
considerations of running a straight cam in a warped head - most head
shops won't even /attempt/ to address this.

if you mean my advice to that pillock red cloud, that guy shouldn't even
drive, let alone try to mess about under the hood. the /only/ advice to
him, short of taking the bus, is to take the vehicle to someone that
knows what they're doing. he doesn't have a clue, and based on
experience here, never will.

it's involved and it takes time, but it's not hard. especially not if
you have the book in front of you. the only special tool you need is a
pulley wheel holder, but you seem to have no problem recommending
kludges for that.

and that's what i'm trying to address. he's doing the right thing in
that he's eliminated every other possibility, but at this point, the
conclusion of where the fault lies is inescapable, so he needs to move
to the next level. he's clearly concerned about the mileage vs. cost of
repair, so perspective is required - all i see you doing at this point
is muddying the water with theoretical fears vs. hands-on experience,
and that helps no one.

 

unlikely. antifreeze boils higher than water, higher yet with a 13lb
cap and higher still with a 16lb cap. it'll /never/ boil in service
unless there's something seriously wrong.

there sure is!
http://www.helminc.com/helm/product2.asp?class%5F2=AHM&mk=Honda&yr=1994&md=Accord&dt=Shop%2FService+Information&module=&from=result&Style=&Sku=61SV400&itemtype=N&mscsid=F04JGJEPXNAD9KP7M2RTCTTVB9858U3B
factory honda manual. there is no finer.

 

Yours is a non sequitur. I said nothing about the frequency
of the blocks cracking. You yourself are admitting it's a
gamble.

I think you bullshit your way through way too much here.
AFAIC it's at the point where anything you say is suspect.

Nope, I mean the crap you post to many people here; the
belittling remarks you make to newbies just trying to get
some help (like they need more trouble when their car is not
running right; etc.

Daddy didn't give you enough attention, did he?

Typical assertion of opinion as a fact. It's why you have no
credibility with me: Your under daddy-fed ego gets in the
way of truth.

Good day, boy.

 

searching at www.groups.google.com for {(Honda OR Civic OR
Accord) cracked (head OR block)} turns up numerous reports
(not mere speculation but actual) of cracked heads and
blocks for a number of years of Honda models.

 

Gary, you're being unnecessarily analytical about this. By continuing to
drive the car with a suspected failed head gasket, you are doing two
things:
1) Risking corrosion of the block sealing surfaces, and
2) gas-cutting of same.
Either will make the engine uneconomical to fix, and greatly hike your
repair bill, from several hundred to a couple of thousand.

You _MUST_ _immediately_ take the car to a garage to confirm the integrity
of the head gasket before doing anything else. The cost of such a
confirmation will cost you less than $100.

And if the gasket IS blown, it would be foolish to spend all that money to
fix it, and not spend the extra $50 to have the head machined. Yes, it's
slightly possible the head is not warped, but the garage will likely refuse
to warrant the repair unless you agree to the machining. If there is slight
warpage, this increases the probability of early failure of the new gasket.

IF, and ONLY if, the gasket is still intact, then you may noodle around
trying to fix the bubbles. But even then, bubbles suggest possibly
localized boiling around the combustion chambers, which WILL eventually
damage the head gasket anyway. Every mile, every day that goes by that you
do not get a professional to look at this increases the chance or
irreversiable damage. There is a time to call it quits and leave things to
the pro's.

 

Seems like your concern about the fan sensors is not being
fully addressed. For your reference, a lot of heating before
the fan comes on from cold engine is usual. On a mild summer
day, my 91 Civic's rad fan does not come on for 40 minutes
or so. This I have timed on a few occasions in the past.
Then it stays on for IIRC something like 15 seconds. Goes
off. Something like five minutes later it comes on again. (I
performed this check during air purges of the cooling
system.) The car runs great. No overheating problems
whatsoever.

One could research the details of when the fan comes on, but
I think the potential conflicts of questionably matched
parts (thermostat, rad cap, unknown radiator and fan makes,
which might mean little if anything) makes such an analysis
a bit of a crap shoot as far as deducing anything.

Toss in that you're just now learning where key components
like the bleed nipple are, and I think maybe you're in a tad
over your head, Gary. You're a sharp enough sounding person,
and I think you're going to nail this sooner rather than
later, but being grounded in the experience of what most
often causes whatever problem in a car tempers judgement
calls on what is most likely "Thee Problem."

A friend of mine paid under $50 at the local Honda
dealership not long ago to get the test Tegger and the
Beamer describe: Pressure test the cooling system and check
for exhaust gas products in the coolant. It quickly
ascertained that the head gasket failed. She had them do the
repair. Sure the machining can always be questioned but the
fact is on a roughly $1000 repair job an extra $50 or so
IIRC for the machining so one can drive knowing the car is
reliable isn't such a bad deal. Car runs fine now.

Of course, then the issue becomes what was behind this.
Aftermarket radiators are very common in Hondas and don't
seem to be a problem. Aftermarket thermostats are
questionable, by reports here and IIRC the Honda manual. I'd
say not to mess with aftermarket rad caps, too.

I'd go with the results of the cooling system test, trust
the radiator and fan (and fan sensors) for now, but as I
think you already did, replace the thermostat with OEM and
also the rad cap.

I trust you found the free online manuals at
www.autozone.com (a Chilton's manual but with many excerpts
from the factory service manual) and
http://www.honda.co.uk/car/owner/workshop.html (has a 91-93
Accord manual which should nonetheless have some useful info
for your 94 on this). You can bounce part numbers for your
94 against the 93's using online parts sites like
www.slhonda.com and www.hondaautomotiveparts.com , to see
how similar they are, for one.

 

:

When my head was skimmed, about .002"-.003" was taken off. My mechanic says
(at least on the B18) you can skim up to 1/16" off before you run into
valve interference and compression issues. Up to that point, you can
compensate. If you do not machine the head for warpage, there is an
increased probability of early gasket failure down the line.

However, he also tells me Honda heads rarely warp much unless badly abused,
so you rarely need to take more than a few thou off for warpage. Machining
beyond that is often done to address *corrosion*, not warpage. If the head
is badly warped on account of severe overheating, he normally advises just
to junk the engine, since head warpage is only a small part of the damage
that will have occurred.

Finally, he tells me Honda blocks are really strong. So long as they are
not severely overheated or the victim of neglected coolant changes, the
block faces are almost always well within flat tolerance.

 

1/16" or ~1.6mm is a /lot/. you'll never find a gasket that much
thicker, so you're automatically into higher compression ratio
territory. if you know what you're doing, that may be fine, but for a
stock motor with stock componentry and regular gas, that's not a good
place to go.

if it's slightly warped, yes, but it doesn't mean it is.

i have two comments on that.

1. this head is bolted to an alloy block subject to a similar thermal
regime to the alloy head. in the old days of alloy heads and cast iron
blocks, this may have been more of an issue, but the all-alloy
engine/head combos often "warp in sympathy". straightening a head may
not be a great plan without paying attention to the block as well. and
suddenly, that's starting to get to be a much more expensive proposition.

2. skimming heads is just like skimming brake disks. most shops do it
because it's "recommended", and means no personal labor involved in
cleaning and prep for re-fitting. but it doesn't mean it's even halfway
necessary. when i did my own head gasket recently, it took me 7 hours,
of which, 3.5 was carefully cleaning up the head/block surfaces so as to
not scrape or gouge them. there ain't no commercial operation i know
that's going to bother with that - it's straight off to be skimmed and
that's that!

perfectly valid reason, if present.

absolutely. for older vehicles, i think the economics of replacing the
whole shebang with a low mileage low cost jdm engine has got to be
considered.

absolutely.

i guess my biggest deal in all this is that i've seen machine shops make
real bad jobs of head skimming. but the thing just gets slapped back on
the customer's car anyway because no one is going to fess up to having
just trashed [replacement cost] $2500 dollars worth of componentry. i
therefore do /not/ recommend head skimming unless the head is out of
spec., and even then, i'll look at the block as well to see how that is.
if the two "warp to fit", i may well take a view on replacing the head
unmachined. but i guess i get to choose because it's my car and i do
the work. in a world where the public are brainwashed into thinking
that all heads need to be skimmed and all brake disks need to be
skimmed, who's going to take a chance on not giving the customer what
they expect?