Knock

Machiner 55 · May 4, 2013

Here's for all you technical types who know the 3800 engine inside and out.

My recently aquired '90 came to me with an engine that wasn't running quite right.

It would miss and jerk when adding power while in overdrive. Classic symptoms of a much

needed Pagdett/Delco upgrade. Changed out the Magnavox with a Delco unit.

Replaced the plugs (AC Delco R45LTS6) with AC Delco Rapidfire #14's. The plugs in positions #4 & #6 were loose.

Turned them out with my fingers. At the time, I didn't think to tighten them in to see how far they had backed out.

Too late now. All the plugs were somewhat blackened with carbon. No one any more than the others.

Feeler gauge told me they were set to .060" + - .008". Think they should have been set to .045" with the Mag. coil pack.

The wires looked relatively new (as did the Magnavox. More on that later). So I didn't change the wires.

The engine started and ran like a champ and sounded great... until it warmed up.

After riding around for about 15 minutes, there arose a knocking sound.

Something like the noise that would be made if one were to take a plastic headed mallet

and tap on the inside of the valve cover. Kind of dull sounding as opposed to a tinny or pinging like sound.

Once the car has sat for some time and cooled down or left sitting over night, I can once again start the car and it

sounds tight and right until it warms up again. I don't have to drive it.

I can let it sit in the driveway idling and the sound will return again after warming up.

Opening the throttle only generates engine noise enough to drown out the knocking sound so

I can't tell if it gets worse or better at throttle up.

I've got a stethoscope and have been trying to use that to locate the source.

The knock seems to be coming from the rear bank (2-4-6). Seems close to the EGR valve.

Oddly enough, when I probe the power steering pump, Seems to be coming through there also.

Right now I'm waiting for the engine to cool down. I'm going to replace the wiper blades and the hood struts.

(Now I can give my wife her broom back. Hope she doesn't notice the shortened handle.)

Once reasonably cooled, I'm going to replace the tensioner idler pully. Noticed a bit of "growling"

with the steth. Once I have the serp. belt off, I'll run the engine just a bit to see if the knock goes away

with the PS pump disengaged.

I haven't dropped the oil pan yet but I checked the oil before driving it back and it was golden honey colored then

(engine had just been serviced prior to my purchase) and it still looks good after putting almost 800 miles on it. (550 for the trip home, the rest, local.)

Anyway... Should the knock still be there (almost positive it will be) what, in your esteemed opinions,

do you kind folks out there think it could be?

PS. The main reason (I believe) the Magnavox wasn't working properly was that one of the contact pins in the connector slot was bent and folded in such a way that it couldn't possibly be making contact with the plug. It was the second pin, on the left of the center screw hole when viewed looking directly into the slot when held at eye level.

I've since straightened the pin and, as time allows, I will put it back in and try it.

John F.

PPS No codes.

Edited May 5, 2013 by Machiner 55 (see edit history)

Ronnie · May 4, 2013

No knock until the engine is warm
Dull knocking sound deep in the engine
Drop in oil pressure
Knock comes and goes as engine speed is increased or decreased just above idle

Above are the classic signs of a rod knocking. Has the oil pressure dropped?

Here is how I test to determine if a rod is knocking.

Warm up the engine to operating temperature. Hold the engine about 1500 rpm or wherever you can hear the knock the loudest. While holding the rpm there remove one spark plug wire at a time and replace (repeat several times) while listening to see if the knock goes away or changes significantly. If it does you have likely found the cylinder that is knocking.

Next step is to pull the oil pan and check the rod bearing to verify your results.

Don't overlook the possibility of the harmonic balancer making the noise.

Edited May 4, 2013 by Ronnie (see edit history)

May 4, 2013

Sounds like problem occurs when going into closed loop. Main thing entering the equation is the O2 sensor. Can cause most of your driveability problems. Enter diagnostics and monitor cross counts using ED18. If that doesn't seem out of place, take a snapshot and post the results after engine is warm and in closed loop. That should give us a good place to see what's going on.

Ronnie · May 4, 2013

It appears me and Mc_Reatta have different ideas on this one. I certainly hope he is correct. Rods knocking in a 3800 is rare but not out of the question.

May 4, 2013

It appears me and Mc_Reatta have different ideas on this one. I certainly hope he is correct. Rods knocking in a 3800 is rare but not out of the question.

For his sake I hope so too. I'm going by a mechanical issue usually wouldn't just show up when engine warms up and go away when cool. That's why I went with what changes when car goes closed loop.

Ronnie · May 4, 2013

For his sake I hope so too. I'm going by a mechanical issue usually wouldn't just show up when engine warms up and go away when cool. That's why I went with what changes when car goes closed loop.

Since he drove it a long distance home without problems, (I skipped over that part when reading his post), you are probably correct. However, I've had old cars with bad rod bearing that didn't knock until the engine got warm and the oil pressure dropped... I guess due to the oil thinning when it gets hot. I hope that golden honey colored oil wasn't from someone loading it up with STP or worse before he got it. It will be interesting to see what he finds.

May 4, 2013

What is the total mileage on the engine?

padgett · May 4, 2013

Sorry but rod knock often is masked by cold (thick) oil and appears when the oil thins out when warm. Need to know oil pressure at idle cold and hot and knock counts. If a HB it normally worst at idle. Also should be in closed loop within 2 minutees.

Machiner 55 · May 5, 2013

To all,

Thanks for the input, It's getting late and I'm really tired. Gotta get up early to start a new work week at 4:30am.

Need to tee the oil sensor port first to install a gauge. The graphic display in the '90 kinda sucks. Need the numbers.

Will get back with the results when I can.

G'night

John F.

Edited May 5, 2013 by Machiner 55 (see edit history)

May 5, 2013

Should be able to see the real reading for oil pressure by monitoring BD71 in diagnostics.

Machiner 55 · May 5, 2013

Richard,

Total mileage is 103,840. According to the maintenance records that came with the car the oil was changed regularly and often during the past 16 -17 years. Don't have any records for the first 6 years of it's life.

Machiner 55 · May 5, 2013

McR,

DoH! Still want to install a "steam" gauge 'cause that's just the kind of guy I am.

Gotta go to work now.

John F.

Howard · May 5, 2013

The sound you are describing is exactly the sound my Reatta was making. My Reatta engine cratered after about 150,000+ miles. Frankly, had a number of problems the engine once the harmonic balancer failed. It seemed as if everything starting going wrong after the balancer problem. Shortly before the "knocking" started, that car began to run badly and the gas mileage went out the window. A code was popped for the oxygen sensor. That was replaced, but the code returned before the car was back home. The knocking began shortly after. I took it to a mechanic I had been using and he told me without hesitation, it was a rod knock. He dropped the pan at that point and it was full of metal particles! I have no idea why that happened. I changed the oil at the prescribed intervals and never abused the car. A theory was put forth that one of the injectors was stuck open and continued to flood that cylinder with gas and washed all the oil out. The tech suggested that was why the gas mileage had dropped so much. I don't think that explained the oxygen sensor issue, but who knows. The estimate to replace the engine with a rebuilt unit exceeded $4000.00. I did not wish to invest that much money in the car. The mechanic said he would try to find a used motor. Since the engine was very common to many GM products, he said that he could probably find one reasonable priced. While my Reatta was at the shop, someone crashed into it. Thus, ending my 9 year ownership. Let's hope your story turns out better than mine.

May 5, 2013

One cylinder running super rich and maybe spurting raw gas at the O2 sensor probably cooked it.

Good thing these kinds of problems are rare in a 3.8.

Machiner 55 · May 6, 2013

ECM DATA: “Snapshots”:

1.) Codes clear, Engine cold. Not running.

2.) No Codes, Engine warm, running in Closed-Loop for about 5 - 10 minutes.

3.) No Codes, Engine warm, running in Closed-Loop for 25 – 30 minutes.

4.) 40 minutes running. Scrolled through and noted readings that changed or were changing. (Not Snapshot.)

The statements in red explain the conditions under which the red numbered columns were taken.

******(1.)*********(2.)**********(3.)*************(4.)

ED01 – 0.38 --------- 0.38 ----------- 0.40

ED04 – 30 ------------ 68 -------------- 100

ED06 - 41.9 ---------- 4.8 ------------- 4.2

ED07 - 0.45 ---------- 0.79 ------------ 0.11 ------------.13 - .82

ED08 – 70 ------------ 20 --------------- 19 -------------- 20

ED10 – 11.8 --------- 14.3 ------------- 14.1 ------------ 13.9

ED11 – 20 ------------ 830 -------------- 748

ED12 – 0 -------------- 0 ----------------- 0

ED16 – 0 -------------- 0 ----------------- 0

ED17 – 0 -------------- 9999 ------------ 251 ------------ Holding steady at 251.

ED18 – 0 -------------- 3 ----------------- 0 ---------------- 1

ED19 – 128 ----------- 129 -------------- 131 ------------ 124 - 133

ED20 – 128 ----------- 130 -------------- 126 ------------ 127

ED21 - .0 -------------- 6.5 --------------- 5.2 ------------- 5.0 – 5.2

ED22 – 150 ----------- 60 ---------------- 44

ED23 – 24 ------------- 26 ---------------- 34

ED98 – 51 ------------- 51 ---------------- 51

ED99 – 8064 -------------------------------------------------

Note: After about 2 minutes of run time, it enters Closed-Loop and immediately, the idle becomes unsteady and the RPMs drop off about 50 RPM. (don’t have those exact RPM numbers at this time.)

Oil Pressure (BD71) readings taken after 5 – 10 minutes of run time:

32 – 35 @ idle.

43 – 46 @ 1500 RPM

40 – 46 @ 2000 RPM

Knock heard now after running for 15 – 20 minutes. BD71 reads 18 – 32 PSI at idle after running for 30 – 40 minutes.

Can hear knock underneath the car while lying on ground.

That's about it for now. We have company over for dinner now.

John F.

Edited May 6, 2013 by Machiner 55 (see edit history)

padgett · May 6, 2013

That is lower warm idle oil pressure than I like to see, even the high side. That said the senders are notorius for being inaccurate, even the factory says when in doubt use a mechanical guage. However loose bearings and low oil pressure go together.

All of that said I have replaced bearings without pulling the motor or the heads before (on a GM 6 but not a 3800). Next step is to get some pastiguage and a mike, pull the pan, and see if any rod caps wiggle.

Machiner 55 · May 6, 2013

Padgett,

(Company's gone now.) Yes, I agree with the pan drop. It's good to know that someone else has done the bearings without pulling the crank/engine. I'd considered this very action earlier today when it became apparent where and how the noise was being generated. (Yesterday, I went so far as to change out the PS pump to rule it out.)

I guess so long as the journals aren't scarred I'll be OK.

Also, today I gathered the materials to construct a remote "steam gauge" oil pressure gauge so I can compare the readings with the on-board sensor readout. Anyway...

What do you think of the EDxx data and the fact that the idle becomes irregular at close-loop operation?

And... what the heck is the Oil Pressure reading supposed to be when all is well? Scoured the FSM and couldn't find the number.

John F.

Edited May 6, 2013 by Machiner 55 (see edit history)

Ronnie · May 6, 2013

I hate to always seem to be the bearer of bad news but... a rod bearings that fails and starts knocking normally isn't due to the bearings gradually wearing out to the point that the clearance is so great that the rod knocks. I simple bearing replacement rarely is a permenant fix. Usually something happens that roughs up or scars up the rod journal ruining the internal surface of the rod bearing which causes the excessive clearance. Insufficient oils supply (pressure) or metal particles from some other item wearing out, like a timing chain tensioner for instance, getting into the oil supply are the likely suspects that cause bearing failure. When something like that happens things go down hill quickly if the engine is allowed to run very much. If you have reached that point a crankshaft kit being installed is your best bet. I know this is not good news. Just an observation from a mechanic who has been down this road before.

Ronnie · May 6, 2013

BTW, It would make me very happy to find out I'm completely wrong and you find out your problem is something simple that can easily be fixed.

Machiner 55 · May 6, 2013

I hate to always seem to be the bearer of bad news but... a rod bearings that fails and starts knocking normally isn't due to the bearings gradually wearing out to the point that the clearance is so great that the rod knocks. I simple bearing replacement rarely is a permenant fix. Usually something happens that roughs up or scars up the rod journal ruining the internal surface of the rod bearing which causes the excessive clearance. Insufficient oils supply (pressure) or metal particles from some other item wearing out, like a timing chain tensioner for instance, getting into the oil supply are the likely suspects that cause bearing failure. When something like that happens things go down hill quickly if the engine is allowed to run very much. If you have reached that point a crankshaft kit being installed is your best bet. I know this is not good news. Just an observation from a mechanic who has been down this road before.

Ron,

I don't disagree. I really won't know for sure what I'll find until I get in there.

I'm a Mechanic. Have been for most of my life. I work on an assortment of machines but not ICE machines.

I've never had an issue with a car such as this (just lucky I guess) so I'm treading in new territory. Not Rocket Science.

Just different from what I'm accustomed to.

Never the less, thanks for your input.

John F.

Machiner 55 · May 6, 2013

Howard,

Good to hear from you.

Always wondered what happened to your car.

Thanks for sharing.

John F.

Ronnie · May 6, 2013

John, I hope it turns out not to be too bad. One thing you have going for you on the Reatta is the oil pan is easy to get off to check things out.

Machiner 55 · May 6, 2013

Ron,

Thanks.

Relatively speaking, getting the pan off and out is a piece of cake. Especially when compared with the one on my '91 Crown Vic. I hope I never have to pull that one. I wouldn't classify it's structure or placement as one of Ford's "Better Ideas". It's got a hump in it to clear a cross-member and because of that, two oil change ports. Yah, that's right. Count them. Two. Two plugs.

The first time I went to change the oil, I, initially, couldn't figure out why the amount of oil drained didn't equal the pan's capacity.

John F.

PS.,

Any comments on the numbers posted in post #15.... Anybody?

What's up with that 9999 for ED17? And the roughness when it drops into Closed-Loop?

padgett · May 6, 2013

"C" spec is 40 psi @1850 rpm on 5W-30 at operating temperature.

My thought is that the knock (OldPA3 - 9999) is triggering retard. Have seen a bad HB do the same thing.

ps the one I did without pulling the engine was a Pontiac OHC 6 in a '68 Firebitd and it ran fine until I sold it (was my daily driver) without any problem. All the surfaces looked and miked good, it was just worn out.

Edited May 6, 2013 by padgett (see edit history)

May 6, 2013

In an early post you mentioned that the noise seemed to come from the rear or transmission end of the engine. Have you checked the flex plate to torque converter bolts for looseness? Happened to me once when engine went to warm, low speed idle.

Machiner 55 · May 7, 2013

Padgett,

Thanks for the info. I'm hopeing for the same.

********************************************

Jimmyk,

Good info there. I'll keep it in mind but, not the rear/transmission end but... "The knock seems to be coming from the rear bank (2-4-6). Seems close to the EGR valve." end.

John F.

Edited May 7, 2013 by Machiner 55 (see edit history)

Ronnie · May 11, 2013

Anything new to report on the knock investigation?

padgett · May 11, 2013

"How does the help desk know the problem has been solved ? The customer stops calling."

Machiner 55 · May 11, 2013

Ron,

I appreciate the interest in the issue of the engine knock and all the suggestions and info provided by you and others. But, I've yet had sufficient time to get under the car and drop the pan. I might be able to get to it later today but it's doubtful that I will.

My situation is as follows. I work 12 hour shifts. Four days on, three days off. Then, three days on and four days off. Occasionally, I work extra days (overtime). And sometimes "straight through" (3+4+3=10). Don't do that very often any more). On the days I work, I don't do much of anything else. Too tired and not enough time to get involved in anything. Add to that, the fact that my attached two car garage is really only big enough for one car (my Wife's car) and all my "STUFF". (See attached Photo) So, I have to be judicious on how I allot the space/time (continuum) in the space/time available on my days off. Also, because I don't get much done on the days I'm working, the "to do" list gets back-logged and I play catch-up (are all these words really hyphenated?) on my days off which pushes my Reatta projects to the back-burner (there... there it is again, another hyphenation) So.....................

The reason I've gone to this length to explain is that I know that there are some who post questions here and Forum members bend over backwards to help. Then

That's right, Nothing. Nada. They fall off the earth and we never know how, what, or if, there was any resolution to the problem. If my (failing) memory serves me, I think there's only one problem that I posed where I didn't post the resolution. That's my issue with the VSS and repeated code E024. The '89 that, that happens on, is being driven by my son who is seldom here. And, due to the trial and error method I'll have to use to help with diagnosing the problem requires a block of time and good weather to do properly, I've had neither. I bring this up 'cause I don't want the guys who took the time to respond to my query to feel that it was a waste of their time to do so. I just haven't had the opportunity to attack the issue properly. Life keeps getting in the way.

So again, thanks for the interest and the input. As soon as I can get into it and can report something new I'll post the news. Believe me, I'm just as anxious to find out what's going on inside there as you all. I've never experienced an engine knock and this is new ground to cover. The only thing to report is that it's confirmed that after the car has sat for a day or so there is no knocking sound discernible after engine start and it purrs like a kitten. That is until it's warmed up (15-20 minutes)

P.S.

One mo ting... It takes me FOREVER to type stuff. Often times the forum page times out before I'm done. Since I'm already time constrained, sometimes I just don't bother.

John F.

Sorting through the pile formerly known as my '88. Wife's car removed for clarity.

Edited May 11, 2013 by Machiner 55 (see edit history)

padgett · May 11, 2013

Well, stopping knock is one of the uses for STP (engine honey), kick the viscosity up until knock goes away. I used to run a can in my B/P SBC after an oder aned wiser person I trusted said that if oil pressure was lost it could save the engine for 4 seconds.

I have heard of (and experienced) rare oiling problems with 90 3800s before but most involved spinning cam bearings - no oil pressure but otherwise quiet engine. These usually show up in the first 100k.

That said the rod bearings get oil last and on a cold start it may take a few seconds to build pressure there - I cringe when someone revvs a cold engine on TV, am I the only one who always idles a cold engine for a while, fills the oil filter on change, and spins a newly rebuilt engine or one that has sat for a while to bring pressure up before starting (one of the reasons everything gets started and idled for a bit at least once a month) ?

Ronnie · May 11, 2013

Thanks for letting us know. Have you considered pulling plug wires to find the knock as described in an earlier post to find the knock? It usually works but not in all cases. I guess it depends on how much excess clearance there is in the bearing.

Machiner 55 · May 11, 2013

Ron,

No but, thanks for reminding me of the plug pull test. While Googling this issue, that came up several times. Will do that first (when I have time).

(If I could just break away from this freaking laptop (should lap-top be hyphenated?) I could actually get something done around here)

**************************************************

Padgett,

am I the only one who always idles a cold engine for a while, fills the oil filter on change, and spins a newly rebuilt engine...

Quite possibly!

John F.

TexasJohn55 · May 12, 2013

Interesting thread, but the suspense is getting to me. Did you ever get around to running it with the belt off? While the belt is off, you can grip the crank pulley and see if it will slip either way without turning the engine. If the rubber deteriorates and breaks loose it will bang against metal to metal and will be most pronounced at idle and low rpm. Knock may change without belt pulling accessories. Nevertheless, it should knock on startup. I mention this because I replaced the harmonic balancer on my old '88 98 a couple years ago for knocking and just replace one today on friends '95 Olds 3800 for knocking, the rubber was shot. You may have already eliminated this possibility but I couldn't determine that by reading your posts. By the way, My Olds was prone to misfire due to plug wire and coil pack issues. I believe that the bad plug wire misfire under load killed my coilpaks. I have replaced it twice ( with used ones ). Here is another note: on the '95, when I pulled the oil fill cap while idling, there was a terrific rush of air from high vacuum in crankcase, engine stumbled and nearly died. I asked what had been done recently and he said oilchange, fuel filter and pcv at a shop. The pcv on this one is in the intake manifold at front end of engine under a hex headed plastic cap. I found a new pcv valve rattling around in the hole without an oring on it, pulling way too much vacuum around it. Put an oring on it and capped it. Just some thoughts to ponder. TexasJohn55

TexasJohn55 · May 12, 2013

PS: You could also put a socket on an accessory like the PS and try to turn the engine back and forth with the belt and watch the pulley to see if it slips on the crank member.

Machiner 55 · May 12, 2013

John,

I've pretty much ruled out the HB. I've changed a few in the last couple of years and I don't think the symptoms I'm experiencing point towards that being the cause. Not a bad idea though. Thanks.

John F.

2seater · May 12, 2013

I've never tried disconnecting one plug wire when running, but don't you lose both cylinders on that coil when you pull one? I'm really not sure, but it would seem that the way it is made, you would. If so, that might lead to a misleading result?

TexasJohn55 · May 13, 2013

2seater, the coil pack fires both cylinders at the same time so just don't ground the wire you pull off. TexasJohn PS; It is easiest to pull it at the coil pack anyway and you will know if it tries to arc to ground but it shouldn't.

Edited May 13, 2013 by TexasJohn55 (see edit history)

TexasJohn55 · May 13, 2013

John F., OK, I CONCEDE, PULLING THE OIL PAN IS AN EASY JOB ANYWAY, I GUESS I WAS JUST HOPING AGAINST ALL ODDS THAT YOU DIDN'T HAVE A BEARING FAILURE. WON'T TAKE LONG TO CONFIRM THAT WHEN YOU DROP THE PAN. TexasJohn

2seater · May 13, 2013

I do understand it fires two plugs at a time, however they are reversed polarity from each other. I guess it would be easy to find out with a timing light to see what actually happens. As far as killing the power stroke on individual cylinders, it can also be done in diagnostic overides by turning the individual injectors on and off.

Ronnie · May 13, 2013

2seater, I looked at the schematic for the coils and I understand what you are saying about each coil firing the way it does. I believe pulling one plug wire at a time would still be a valid test for finding which cylinder has a rod knocking, but I don't know for sure because I have not tried it on a 3800 engine. The reason I believe it would is I'm pretty sure that only one of the cylinders connected to each coil is on the compression stroke and the other is on the exhaust stroke thereby only firing one cylinder at a time per revolution of the engine. I have used this test many times to determine which rod is knocking on engines with a distributor but I have not had the opportunity to try it on a distributor-less ignition system. I hope I never do.

Testing for a rod knock this way isn't just a matter of removing the spark plug wire to kill the cylinder and then listening to the engine so I don't believe killing a cylinder by shutting off an injector would be a good way of doing it. In my experience for it to be a valid test you must remove the plug wire from the distributor and then move it close to the distributor terminal in a manner that makes the spark arc erratically while listening to the sound of the engine. I don't know if that makes the engine fire slightly out of time in a way that changes the way the slack taken our of the bearing or actually what is going on. It's just a seat of the pants test that I learned from a mechanic who owned the garage I worked in many years ago. This test is pretty common knowledge with older mechanics.

Edited May 13, 2013 by Ronnie (see edit history)

Knock

Recommended Posts

Link to comment

Share on other sites

Link to comment

Share on other sites

Guest Mc_Reatta

Link to comment

Share on other sites

Link to comment

Share on other sites

Guest Mc_Reatta

Link to comment

Share on other sites

Link to comment

Share on other sites

Guest Richard D

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Guest Mc_Reatta

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Guest Mc_Reatta

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Guest jimmyk

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites