94 L27 overheat: EGR path blocked ? or water pump impeller ?

[MikeESoc94]

94 Bonneville L27 naturally aspired engine:

Hopefully someone has had success with correcting a similar overheating problem since it doesn't make sense to me.

I don't put much mileage on this car (note: runs smooth at all road-speeds) ...
... currently it's something like 165K miles on the odometer ... bought it ~ year 2001.

Started to notice tendancy to overheat starting couple of years ago ...
and then ~ beginning of last summer the fans were repeatedly going on-off (and the temp gauge would swing from ~ 210F to 225F) ... including only when idling with the hood open.

First changed the thermostat to 180F but that didn't change the too-hot condition ... that thermostat is currently installed.

Then replaced the radiator and the transmission fluid lines (also POR-15 body-rust repaired the cross-structure the radiator sits on) ... and bled air from the coolant with a "burp funnel" set-up at the radiator neck in addition to using the OE bleed-screw on the thermostat housing.

Still the same runs-hot (between ~ 210F to 225F), so:
last weekend outside (outdoor temperature last weekend ~ 35F)
ran a Block Tester (blue fluid) test ... which apparently shows no exhaust gas in the coolant (yes, a good thing).

But the unexpected thing:
as part of the Block Tester test I siphoned Quart+ coolant out of the radiator neck (so coolant wouldn't get sucked into the Block Tester),
and,
with the coolant level then down ~ 7 inches in the radiator (coolant temp ~ 210F) and with the radiator cap off,
the coolant-level would quickly surge-up to ~ 2 inches below the radiator top when temp went to ~ 225F
and then when the fans turned on the coolant level would then fall-down to the ~ 6 to 7 inches below the top ...
... that cycle happened repeatedly consistently ... like clockwork with ~ 45 seconds for rising from ~ 210F to 225F.

Engine idled smooth during the entire ~ 2 hours we were watching all that (ran the Block Tester test a few times),
and,
simply by-sight the coolant flow-rate thru the radiator core appears good
(hmmm ... ? but what's "good" versus what it should be ?).

Engine oil and coolant don't appear to be mixing (no murkines in either oil or coolant).

No coolant loss, and, no noise from any of the serpentine belt pulleys (note: A/C pulley is bypassed via shorter belt).

What I'm currently thinking as a possible problem:

- water pump impeller has internally detacted from shaft (so, even though looks like good flow thru the radiator it may be below spec);

- EGR flow-path into the intake is blocked thereby causing the engine to run hot (but EGR valve itself is OK ... no MIL code ... engine runs smooth at all speeds);

- A/C condensor fins need to be cleaned (but maybe not, since overheating at idle with hood open when 35F outdoor temperature).

Next steps I'm planning:
- take the serpentine belt off and see if I can rock the water pump pulley by hand;
- take EGR valve off and run a length of mechanics wire down the pathway to see if any blockage to the intake.

Any other ideas for a possible problem ? ... thanks.

Here a prior post I started regarding the water pump pulley clearance concern that I had:
http://www.pontiacbonnevilleclub.com/fo ... ml#p456680

[J Wikoff]

Have you thought about trying a new radiator cap? I've never experienced it, but I think a cap that doesn't hold pressure will allow larger than normal temp swings.

[michtom]

From what I am reading, your radiator isn't overflowing at any point. The 225 temperature you see is normal in most cars, as that is when the coolant fan is commanded by the engine control module to turn on. The fan will turn off once coolant temp. reaches 200. The fans will continue to cycle on and off continually. If the fan doesn't come on by 227 degrees, the you need to be concerned.

[MattStrike]

I would also recommend starting with the radiator cap, or if you have access to it a coolant system pressure test. If that doesn't fix the issue then check the water pump. ACDELCO 252694 {#88926208} is the one that has the cast impeller (better design).

Your cooling system should have no issues maintaining the thermostat set point if it's 35F outside. Maybe verify with a thermometer that the gauge is reading accurately? I know my '93 the gauge is just a ballpark, scan data and my IR thermometer show 195 when my gauge reads 210

[MikeESoc94]

Thanks for all the replies ... here some more info and another question (regarding water pump impeller design):

I included a new radiator-cap when I replaced to a new radiator this past Summer
- the old cap didn't look bad ... but new is new ... and the radiator replacement detail specifically recommended a new cap.

Regarding the heat that is being generated out of this L27 engine into the coolant:
when the new top hose is grabbed it feels notice-ably hotter to the touch
than other vehicles that have 195F thermostats (such as a few Jeep 4.0L models that I also work on).

I also replaced the bottom radiator hose when I installed the new radiator this past Summer
- however:
after I installed the new bottom hose I realized I likely installed it backwards
- that is:
the longer-leg of the S-shape should likely have connected to the radiator but I connected it to the water pump
(but I don't think that enters into any reason for the current too-hot running ... hmmm ... maybe I'm wrong about that).

The dashboard temp-gauge is definitely at ~ 225F (or thereabout) when the fans turn-on,
and yes:
one of the things me and my brother were talking about is using his point-n-shoot directional electronic-thermometer
to point at the top hose (but, we haven't done that yet).

Since owned this 94 Bonneville since ~ year 2001 and drove it many many times before this overheat behavior started,
I have to say without doubt it is currently consistently in the overheat
especially since last Summer
(and, as mentioned, I noticed it starting with a milder version of that overheat couple of years ago).

And correct:
when the radiator-cap is on and system full with coolant
it doesn't get extreme hot to point of coolant surging out of the overflow tank
but,
it's still running too hot relative to where it should be ... per past experience with driving it over the years
(and reminder: this is with a 180F thermostat).

Question regarding the ACDelco 252694 water pump:

Per RockAuto photos the ACDelco 252694 impeller looks like duplicate of the Gates 42095
(impeller looks like 10 short stubby fingers with a hook on each end);
but,
the Cardone Select 5513131 impeller looks very different (looks like 10 long curved fins without hooks on the ends).

Is the ACDelco 252694 continuing to be the better design, or, did that ACDelco impeller design recently change
(and now the Cardone Select impeller design is the original preferred design) ?

Thanks.

[MikeESoc94]

1 more post for today Tuesday 01/31/2017:

This is why I am looking at maybe the EGR flow-path to the intake being a possible problem (if the EGR flow is blocked)
for engine coolant overheat:

http://www.aa1car.com/library/egr.htm

Subject section:
"WHY EGR?
Exhaust gas recirculation reduces the formation of NOX by allowing a small amount of exhaust gas to "leak" into the intake manifold. The amount of gas leaked into the intake manifold is only about 6 to 10% of the total, but it's enough to dilute the air/fuel mixture just enough to have a "cooling effect" on combustion temperatures. This keeps combustion temperatures below 1500 degrees C (2800 degrees F) to reduce the reaction between nitrogen and oxygen that forms NOx."

The aa1car info-attached video states a common problem on 1997 (OBDII) Buick Century hi-mileage 3.1L GM vehicles
(although NOT focused on L27 3.8L)
is the EGR-fed flow-path into the throttle-body becomes clogged inside the throttle body with carbon build-up:

https://youtu.be/RggMxE3k4b4

https://www.youtube.com/watch?v=RggMxE3 ... e=youtu.be

And (as a question):
the OBDI 1994 3.8L L27 94Bonne may NOT have a diagnostic code for P0401
(? perhaps P0401 isn't an OBD1 MIL code ?).

[RJolly87]

Allow me to share some of my own W-Body L27 observations backed by scan data (and, likely, some expectations).

About the first 1/8th of the temperature gauge accounts for cold temperatures (100*-160* in my case), and then the scale is consistant from 160* to max. Approximate mid way point is marked at 220*.

At normal cruise speeds, or with the A/C running, I am consistently indicating 180-190*, or at about 1/4 of the gauge. I suspect it is currently fitted with an old stock 195* unit. In town 35mph cruising is usually enough to keep it at this point. When stopped, the temperature starts to creep up until it gets to just before 220* indicated (215* most likely), at which point, low speed fans kick on (or, in my case, 1 of 2 fans), and usually bring temps down to about 210*, then shut off. Soon after I am moving again, the temp starts dropping down until the thermostat starts closing back up to keep it from getting too cold. 35mph is usually enough for this.

One concern is that the fans shouldn't let it get to 225* under normal conditions. If low speed is not working (Low speed on the main fan, with secondary fan running), you may wish to start there. 225* is usually the point where the computer commands the main fan to switch to high speed. So, if you have no fans running 215*-225* at all, check out the low speed fan relay to confirm it is getting powered. Also, when it hits 225, and the main fan hits high, there should be no question in your mind it is running on high, because it gets quite noisy. I had a main fan failing for 2 years that didn't cause a tornado when it hit high in the end, and I only knew it was an issue because the motor squalled when it was humid, and because it vibrated moreso than usual during the summer, and simply didn't move as much air as it should have. It also let the A/C go warm when it was hot out and I was standing still. To the untrained folks, they would have never known there was an issue until it made noise.

Also be aware, that a 50/50 water/antifreeze mix boils at 223* at sea level with no pressure. It's possible your coolant may have been starting to boil when you were running it with no cap, and would stop boiling once the fans brought the temp down.

If you are getting all of the proper fan action, and building proper system pressure, I would clean the condenser out really good and see how that does. It's a cheap fix, and can only help things in the end.

I hate to say it, but I think you may be overthinking it. Usually if an issue is affecting the engine bad enough to cause an overheat not related to the cooling system, there is usually other symptoms that go along with it too (Catalytic converter in my case).

[MikeESoc94]

Thank you for that immediate-prior reply, especially "... low speed fans kick on (or, in my case, 1 of 2 fans) ..." and "... if you have no fans running 215*-225* at all, check out the low speed fan relay to confirm it is getting powered. Also, when it hits 225, and the main fan hits high, ...".

That is:
the fan operation on this 94Bonne L27 is that either both fans are off or both fans are on, and, it looks like there is only 1 speed for the both fans on ... as I had mentioned in prior posts the fans turn-on temp is ~ 225F (approximate according to the dashboard gauge) and then at-idle with hood open and 35F outdoor temperature the both fans cool the system down in ~ 30 to 40 seconds
- but when Summer-time outdoor temperature and hood-open at-idle then both fans stay on considerably longer.

So, a question regarding the "... low speed fan relay ...":
can you give me a brief pointer as to where I find that (under the hood or wherever).

Also note:
there's a re-manufactured (Cardone) computer in this 94 Bonne L27 because from ~ 18 months ago to 3 years ago I was having a consistent problem with "shaking" at highway speed under load (such as going up an incline or accelerating) ... and ... I changed a lot of parts (examples: spark plug wires, O2 sensor, ignition control module, ignition coils, MAF sensor, air temp sensor, engine knock sensor, oil pressure sensor, camshaft position sensor) until it was the computer that remained to be replaced (which cleared the problem ... now runs very smooth at all road speeds)
- and perhaps interesting:
when I took the old computer out from the glovebox area there were couple of resistors that immediately fell off the board ... which was a sign that the old computer function had decayed.

And regarding "... I would clean the condenser out really good and see how that does ..." - yes, I picked-out a number of pebbles (whatever) from the condenser fins last Summer before I replaced the radiator ... but I didn't use a detergent with water spray to clean the condenser fins (but I do have a spray can of "Coil Cleaner" and using that on the condenser fins was 1 of the 3 things I was thinking of doing ... listed at-bottom of my 1st post).

Again thanks ... since now I have another clue: the fans may not be operating correctly.

[MattStrike]

on the water pump:
viewtopic.php?f=12&t=15970

[J Wikoff]

http://www.pontiacbonnevilleclub.com/fo ... 27439.html

Here's a shot from a factory service manual labeling the relays under the maxi-fuse relay center cover on the fire wall.

[RJolly87]

On my '93 PA, the secondary fan is single speed, and the large main fan will run at 2 speeds. The secondary fan runs on both high and low. Low speed is simply feeding power through the giant resistor on the fan shroud before getting to the fan motor.

You can likely switch the low speed fan relay with another relay (the horn sounded like a good idea) to see if it may be a bad relay.

[RJolly87]

And be warned, the condenser cleaner should't be used around painted surfaces, because there is a chance it can damage the paint. Some general degreaser (simple green, purple power, etc) can go a long way. Save the big guns for when they are absolutely necessary.

[MikeESoc94]

Thanks for above prior post from MattSrike for the link into the water pump options discussion ... good stuff
- note: if I determine I need to replace the water pump on this 94Bonne L27 then I may be back to this post with 1 more question on that,
since the RockAuto current photo of recommended ACDelco 252694 looks like the impeller is stamped steel (? or maybe I'm wrong about that ?)

And thanks to J Wikoff for the above prior post for the link into the real-clear diagram for the radiator-fan-speeds relay area ... much appreciated ... I think I'm going to get into that as 1 of the 1st things this coming weekend (weather and time-available permitting).

And I also re-read the above 1st post from RJolly87:
wow ... very good info as a road-map for detailed understanding ... especially the main-focus 3rd and 4th paragraphs
- and regarding the 5th paragraph
"... a 50/50 water/antifreeze mix boils at 223* at sea level with no pressure. It's possible your coolant may have been starting to boil when you were running it with no cap, and would stop boiling once the fans brought the temp down. ..."
- yes that explains the operation behavior I described in my above 1st-post:
since am located ~ quarter-mile from the bay directly into the Atlantic Ocean.

And since the above posts have directed part of my focus to the cooling fans:
I am now remembering that when I removed and re-installed those OE fan assemblies during the radiator-replace last Summer
that I thought to myself that the external wiring on the larger driver-side fan assembly was looking a bit decayed
- that is:
the connectors for the wires to-from the white blocks in those external circuits (I assume the white blocks are high-load resistors)
- point being:
I am now thinking I should also remove that driver-side fan assembly again
to up-close detail-review that wiring for possible breakage at any point
- which goes to what RJolly87 said in 2nd-above post:
"... Low speed is simply feeding power through the giant resistor on the fan shroud before getting to the fan motor. ..."

Well ... all that will need to wait for this weekend (at best) ... and thanks again to all.

And 1 more thought (within past few seconds):
if I wire bypass the large white resistor on the driver-side fan external circuit,
then when the low-speed relay engages then the fans operation will be high-speed
- that is:
currently it's Winter outdoor temperature
but during the Summer is where my concern changes to feeling the overheat is an issue
(so perhaps bypassing low-speed and directly always going to high-speed is perhaps the way to go)
- don't know if there are any thoughts about that from others with more experience than me on this.

[RJolly87]

While the wiring could be suspect, honestly I would start back at the relay to confirm it's even being commanded on, and working your way down the line. You can easily track down where you are not getting power usually, or where you are losing it once you have confirmed the relay is working and being triggered.

[MattStrike]

I think that the picture on RockAuto for the water pump is wrong. I haven't ordered one in the last year, but last time I did it had the cast impeller.

[MikeESoc94]

ok ... after a 2 weekends delay (2 weeks ago replaced severe-leaking radiator in a 97 Grand Cherokee 4.0L, and, last weekend was too much snow on the ground) ... but yesterday I tested the fan low-speed power-circuit.

1st: tested the horn ... worked ... then swapped horn and low-speed relays ... horn still worked ... so ... both relays good.

2nd: alligator clipped voltmeter to circuit input-side of the 2 block-resistors on the coolant-fan assembly and started the engine ... when engine coolant temperature (per dashboard gauge) looked like it was at that ~ 225F mark (that I had reported ~ 2 weeks ago) both the VOM displayed 13+volts and the fans turned on ... so ... to confirm it wasn't the high-speed circuit voltage backtracking into the VOM (thru the block-resisters path) I pulled the low-speed relay and the fans turned-off ... then re-connected the low-speed relay and the fans turned on again ... so ... confirmed: the low-speed circuit was powering the coolant fans.

I suppose that's good news (that the low-speed circuit is flowing) ... but less than good news that the dashboard gauge is showing ~ 225F when the low-speed fans engage ... and it appears to be that it is ~ 225F that the low-speed is engaging per the observation that I had reported ~ 2 weeks ago about with the radiator-cap off and the coolant-level down ~ 7" (intentionally for the blue-chemical "Block Test") that the coolant-level would then rise to ~ an inch below the radiator-neck and then the fans would turn-on and the coolant-level would then go back down ... that is (per RJolly87's post): at-sea-level (which is ~ where I am) that 50-50 coolant mix will start to boil at 225F (without radiator-cap pressure).

So:
next thing I am going to look at is the engine coolant temparture sensor .... what I mean by "look at" is look at how easy or difficult it is to change on the 94 L27 engine ... that is: the current thought I have is that the next objective is to get the low-speed fans to engage at a lower engine temperature ... and ... being it's likely it's the 1994 OE coolant temperature sensor that currently in there I might as well replace that sensor (since I've already replaced so much other stuff on this 94 Bonneville over the past 3 to 4 years).

Any recommendations for a replacement coolant temperature sensor ?
... via RockAuto I see ACDELCO 213928 and AIRTEX/WELLS 5S1018 as possible choices ... any other part# recommendations ?

[MikeESoc94]

ok ... found where the coolant temperature sensor is ... not pretty ... in terms of access.

Sample view of the sensor location (but without all the surrounding parts bolted in):
https://www.youtube.com/watch?v=FrPM31q ... 200.359982
- starting approx 3:18 into this video shows the coolant temperature sensor location.

When I look under the hood of this L27-engine 94Bonne,
looks like there is a 3-nuts support bracket that maybe (or maybe not) can be unbolted & taken-off
without also removing the throttle-body and surrounding components
- which:
without that 3-nuts support bracket in the way,
then maybe (or maybe not) a socket can be fit onto the temperature sensor to remove/replace.

If the throttle-body & surrounding parts also have to be removed to replace the temperature sensor,
then I'm a bit annoyed (no other description for it)
because I had that throttle-body off ~ 4 year ago to replace the throttle body leaking coolant gasket
(annoyed because I have to do all that again).

Regardless:
as possible alternative approaches I am thinking about the following
(since replacing the temperature sensor may not get to the objective of turning-on low-speed coolant fans at an ~ 10F lower coolant temperature .. instead of what is appearing to be 225F):
A:
wire-up a relay (direct from the battery) with trigger to a DIY-switch into the passenger compartment
to direct-feed from the battery into the input-side of the double-block resisters in the fan assembly
(I can then manually control when the low-speed fans turn-on and turn-off);
and/or (depending what I see is the cooling behavior this Summer);
B:
instead of wiring that DIY-switch-relay into the circuit for the low-speed double-block resistors,
maybe wire the DIY-relay into the 2nd of the 2 block resistors (for a "medium" speed option)
or directly into the fan motor wire (for "high" speed).

That is (not for nothing):
I am annoyed that the GM engineers decided to place that coolant sensor in such an inaccessible location that I am inclined to re-engineer with a DIY-switch-relay direct to the coolant fan wiring.

Yes, usually I am the only person who drives this 94Bonne (at least over the past ~ 8 years),
regarding then knowing that the DIY switch needs to be used sometimes.

Note:
currently the coolant low-speed fan doesn't turn-on when I switch the dashboard-control to A/C
(? maybe because I have a shorter-length serpentine belt installed that bypasses the A/C compressor pulley ? ... the PCM doesn't "see" the A/C compressor turning ? ... whatever)

Of course ... any feedback is welcomed.

[MikeESoc94]

Regarding immediate-above post
"... re-engineer with a DIY-switch-relay direct to the coolant fan wiring ...":

Over the past week, my 1st inclination for re-engineer was 2 aspects:
A:
butt-connect the low-speed wire from the up-front fans harness into the harness hi-speed wire path (so hi-speed would auto-on at 216F and hi-speed would continue auto-on above 225F);
and:
B:
set-up a relay (with fuse) from battery to low-speed 2 block-resistors on the fan structure with a relay-trigger from an engine-run circuit.

That is:
more I thought about it more I don't understand why the fans circuits weren't OE-designed to always run low-speed with ignition-on
(altho winter-time very cold weather temperature with possible low-power in the battery would be a concern for engine start-up),
or,
at least OE auto-on the fan low-speed when ~ 205F instead of 216F.

As I was looking around (googling) for a possible discussion regarding an easy-access engine-compartment ignition-on circuit to tap-into for the low-speed new relay (new circuit) trigger,
I came across the below 4 links (via gmforum.com).

So now perhaps I will instead install the discussed (below links) manual-switching the fans OE relays 86-ground wires that is explained in the 4 links;
- OR:
maybe (maybe not) more simply just always-ground the low-speed OE relay 86-wire so that the fan low-speed will always run with ignition-on (altho again: concern I have with doing that is winter time very cold weather temperature with possible low-power in the battery at start-up time).

Radiator fans over-ride switch install with photos:
http://www.gmforum.com/1992-1999-91/fan ... de-220002/

and regarding same subject:

http://www.gmforum.com/performance-brai ... od-207063/

http://www.gmforum.com/performance-brai ... -a-207708/

http://www.gmforum.com/1992-1999-91/wir ... ch-205533/

Anyway:
likely this is the end of the trail for me regarding the subject of these specific postings
- thanks to all who provided feedback (above) few weeks ago.