PDA

View Full Version : New Discovery!! Broken Knock Sensor on 87 XJ


csr_011
July 15th, 2005, 14:04
Ok, found the knock sensor today, half of it in the block, the other half danglin and bouncing around with the wiring harness...

So, the question is??? Will this cause the jeep to stumble/cut out under load/stall on bumps and turns with it in two pieces and the wire dangling around??

(Pulled off the cold/open air set up today, put the box back on, was starting to suck in too much hot air and kinda defeat the purpose, think it was making it vapor lock when I got into the throttle, would choke it out sometimes and you could hear the loud suction, but the engine was not running??)

Thanks for all the help guys!

Cory

old_man
July 15th, 2005, 14:11
If the sensor bangs against anything, it will be sensed as a knock and the timing will be significantly retarded at least for a few seconds. Yes is can cause a stumble.

csr_011
July 15th, 2005, 14:31
Thanks for the prompt reply!!

Im wondering if this has been damaged for some time and gradually to the point where it separated; causing my headaches, intermittent as they have been, for A LONG TIME now...

Cory

martin
July 16th, 2005, 06:54
I replaced my knock sensor a few years ago. At that time only the dealer had a knock sensor for the 87-90 4.0 engine. I understand now there is an aftermarket sensor out there.

The sensor has been redesigned and looks like a doughnut instead of an oxygen sensor. You will find your XJ runs a little better after the new knock sensor is installed. What I recall the instructions with the sensor said to use lock-tite to hold the nut at the correct torque.

I recommend you change the MAP sensor next. Once you do that then do the Inlet air temp sensor, it is located in the intake manifold. Once you do those two sensors you will think you did a tune up.

Parts wear out, so doing a replacement program is a good thing. Did it on my 88 XJ and I am very happy. I ended up changing every sensor and installed a rebuilt engine computer. Worth every penny.

csr_011
July 16th, 2005, 08:54
Martin, good points and tips.

My motor only has 16k miles on it. Had it rebuilt a few years ago, the chasis however, has 266k, original tranny, NO REBUILD....yet and shifts even smoother now that the knock sensor has been replaced.

The new aftermarket knock sensor was the doughnut style one, very nice considering how close it is to the motor mount, also allows you to turn the connector location to a position you choose it to be. Only $55 at O'reilly auto here in Kearney, NE.

The MAP is the next and I believe the last sensor/switch...that can be replaced under the hood, EVERYTHING else has been replaced since the rebuild...over time anyways since i ran short on funds with the rebuild.

The power increase and smoothness in shifting, idling, and the amazingly faster start up time is immense. I finally feel the money that I shelled out for the rebuild!! Its like a new vehicle.

I do have the EGR solenoid removed and the vacuum line plugged, will that affect performance or is that just to keep my exhaust cleaner??

I think for safety measures, I'm going to run an additional ground from the dipstick stud to the chassis. Also, getting my A/C back!!! Getting it vacuumed and pressure checked for leaks next week, if all goes well, ordering a new Modine? radiator(closed system of course..:D) and then taking a few weeks to think about a lift kit...again!! Now that it is running GREAT, I have some enthusiasm to "upgrade" the suspension, no more than 3"...

Ok, thats enough blah blah....Did I mention I love my RENIX..again!!!

Cory

martin
July 16th, 2005, 09:51
If you want to upgrade you bonding of your electrical system here is what I did with the help of the Criag H web page.

Craig recommended a bonding wire be run from the NEG of the battery to the chassis. In the stock configuration the only path is firewall, head, block, NEG battery cable. So if you have one of those fail you have no ground.

I installed an 18 inch #4 wire from the NEG to the radiator support. I also repalced the NEG cable from the BAT to the block. I got my cables at NAPA.

Craig also said to upgrade the alternator wire to a #4 to replace the #10 or #8 it came with from the factory. I think I got a wire around 34 inches but it been 5 years since I did the upgrade. With the larger wire the current flows easier and will take a source of resistance out of your electrical system.

I wrote Craig and asked him why not a dedicated wire from the alternator to the chassis. He told me he meant to put that in the article but never got around to it. I found using a wire less than 24 inches did help, I tried a 36 in cand it made no difference on the dash board volt meter. The Delco Alternator has a 8 mm threaded hole in it for a bonding connector. I ran my wire to one of the studs on the coil mount. Remember is your alternator which makes all the power for your vehicle so why not make it easier for the current to return to the alternator.

Craig also talked of using a product called Ox-gard made by GB, you get it in the electrical section of any home center. It is a conductive grease and will keep corrosion from forming. So when installing the new cables then apply ox-gard to the terminals then assemble.

I even went to the point of cleaning up the grounds, my headlights got brighter. I used a wire brush on my drill to remove the paint, applied gard to the terminals and reassemble. I even applied it to the ground strap attached to the firewall. The head light ground is on the driver's fender near the washer bottle.

I hope you add improvements Craig H figured out. If it was me I'd forget about the additional ground on the dip stick stud and in place run a wire from the alternator case to the chassis.

XJXJ
July 16th, 2005, 11:08
Engine off the MAP signal voltage to ECU should be around 5Vdc. Start engine and let idle stabilize, the ECU signal voltage should drop to between 0.5-1.5Vdc. The signal voltage is on terminal B (middle) of the MAP sensor. Input reference voltage (around 5Vdc) is supplied via terminal C.

If MAP sensor meets that criteria, it's probably okay. There's also a method to test the MAP using barometric Pressure value. A good MAP sensor will read within 0.60 inches (14 millibars) of the current Barometric pressure after it is adjusted to sea level. To adjust to sea level, subtract 0.1 inches for every one hundred or (1.1 millibars for every ten meters) above sea level.

I'm still running an OEM MAP after 270k miles with no problem passing SMOG or driveability issues. It's a fairly tough device and quite simple in operation. Most MAP problems are with the vacuum line connection to the TB.

The ECU looks at many inputs throughout stages of Key-on, cranking, cold start-up, hot operation. At about every one of these ECU input stages, it's looking at a signal from the EGR relay. Make sure you haven't disabled that because Renix does store some KAM codes based on those test conditions and they can and will cause issues if a sensor and/or input signal is not passing the ECU self-test routines. I'm 99% convinced some of the Renix high-idle issues people experience are a result of a failed hot-start ECU self-test routine.

Many other sensors play a role too so I suggest you get a manual (PM me if needed) and learn to qualify this system without "guessing" which sensors are faulty. Changing stuff is okay but what if you get a funky new sensor? IMO - You're better off with something that's burned-in and a proven track record of operation compared to something from a box that has probably never reached full operating temperature yet.

If something's bad - sure, replace it but most Renix sensors can be qualified with simple test tools and you'll eventually find the cuplrit. 8/10 times it's usually wiring/connector problems based on my '89 Renix experience.

martin
July 16th, 2005, 11:59
I will say up front after 10-15 years of service IMO repalcing sensors is just "routine maintenance". I think we all agree that sensors can be catagorized into 3 groups:
1) new, meeting factory specs
2) used
3) Failed, no longer operating

We all understand "new" and 'failed". It how the "used" sensor works which is what we are talking about. Over time a sensor degrades until one day it finally fails. I know the Factory Service manual gives test values for sensors. If you are like me you buy a Snap-on MT2500 off of ebay to check out the operation of the Renix XJ to see if there is a problem.

I am advocating replacing sensors as part of a "Service Life Extension" program. Remember the Renix system was only used 5 years and the availability of spare parts in the future may become problematic. Like try to find a "new" cruise control box, the yellow thing, they no longer are available.

The MAP sensor is piezo electric item, the sensor moves with the pressure to give a value to the computer. Over time the crystal has to degrade. Sure if you do a service manual test with a vacuum pump you will get a book value. I am advocating replacement because the dynamic response is not what it was when it was new.

I figure replace the part on my schedule, not on the sensor's schedule. Also I highly recommend if you own a Renix XJ to get the snap on scanner I purchased. It is the only scanner beside the dealer scanner which can read the renix computer. It very nice to use the scanner to set the TPS voltage, avoids the whole back probing of the TPS.

DrMoab
July 16th, 2005, 12:03
What is this Craig H web page you talk about? I just search here and it only pulled up this thread?

XJXJ
July 16th, 2005, 14:41
I will say up front after 10-15 years of service IMO repalcing sensors is just "routine maintenance". I think we all agree that sensors can be catagorized into 3 groups:
1) new, meeting factory specs
2) used
3) Failed, no longer operating

We all understand "new" and 'failed". It how the "used" sensor works which is what we are talking about. Over time a sensor degrades until one day it finally fails. I know the Factory Service manual gives test values for sensors. If you are like me you buy a Snap-on MT2500 off of ebay to check out the operation of the Renix XJ to see if there is a problem.

I am advocating replacing sensors as part of a "Service Life Extension" program. Remember the Renix system was only used 5 years and the availability of spare parts in the future may become problematic. Like try to find a "new" cruise control box, the yellow thing, they no longer are available.

The MAP sensor is piezo electric item, the sensor moves with the pressure to give a value to the computer. Over time the crystal has to degrade. Sure if you do a service manual test with a vacuum pump you will get a book value. I am advocating replacement because the dynamic response is not what it was when it was new.

I figure replace the part on my schedule, not on the sensor's schedule. Also I highly recommend if you own a Renix XJ to get the snap on scanner I purchased. It is the only scanner beside the dealer scanner which can read the renix computer. It very nice to use the scanner to set the TPS voltage, avoids the whole back probing of the TPS.

LOL... Yeah, you're really proud of your scanner and all that but there's nothing I have dealt with that says you MUST use a DRB scanner and/or change perfectly good sensors for "preventitive" measure. The Renix sensors that may fall in that category are 02 and perhaps the TPS although it usually lets you know when it's failing in plenty of time to repair. It's also easy enough to tell if TPS has a bad spot. 02 switching signal ain't hard to check either. So even those aren't "needed" for replacement although they will probably be the 1st ones to go based on my 1st-hand 190k miles (10+yrs) for a total 270k on the '89 XJ.

Working on a electronic repair bench for 15 years I've seen my share of "new" components fail at the 1st burn-in. Under my hood, if the sensor physical condition is sealed, solid and functioning properly, I'll take a burned-in and PROVEN component over a new one any day.

Also can't wear-out a piezo crystal device. It either works or it doesn't. As with many electronic devices 5-10% or so tolerance is gonna be fine because the designers expect that and will provide measures to deal with slight component tolerances in the circuit. Why do you think the manual gives 0.5-1.5Vdc range for MAP idle reference voltage?

Blindly replacing a bunch of sensors only puts new "untested/unheated" sensors into your system. As simple as the Renix is even without a DRB scanner, it's not hard to find failed sensors and/or wiring paths. All it takes is a little electronic rework learning and Renix can be the finest performer at SMOG time or for excellent driveability. It's also nice for managing a bigger engine than stock.

5-90
July 16th, 2005, 15:54
All it takes is a little electronic rework learning and Renix can be the finest performer at SMOG time or for excellent driveability. It's also nice for managing a bigger engine than stock.

I've really got to go with you on that point - after I finally get the smog guy to understand that, "No, my truck doesn't have a MIL - it's pre-OBD! Yes, I know it's an 87/88/89..." I then get asked what I've done to make emissions so low. It seems that my 88 manages to keep lower emissions that most cars ten years newer! My 87 is PNO, and my two 89's run quite clean (although I've got to run through the new 89 - I've only had it for a few months... Gimme time.)

Overall, I'm really happy with the RENIX system - it's been rock solid.

Oh - I know the knock sensor is a piezo, but how can the MAP sensor be a piezo? Piezoelectric effect is the generation of a voltage pulse by a crystal when subjected to impact or mechanical shock - and quartz is commonly used for this purpose. Piezo sensors are good for intermittent or instantaneous readings on things, or the effect can be reversed to make a buzzer (like in cheap alarm clocks and some warning buzzers.) Problem is, a MAP sensor has to present a constant variable reading, and it does so in response to external pressure/vacuum. Being bolted to the firewall woud make for an erroneous reading each time you hit a bump or something, if it was piezoelectric - and it's usually wired up as a voltage divider (like you'd do with a potentiometer when creating a reference signal - like for the LM317T Variable Voltage Regulator.)

Granted, I haven't taken one apart yet, but I'd likely expect to find a carbon pile, which is variably compressed/expanded by the pressure being monitored. Altering the compression upon - and therefore the spaces between - the carbon particles in the pile would create a variable resistance. Calibration of this resistance at varying levels of absolute pressure (meaning, not accounting for atmospheric pressure when measuring) would provide a reference signal when supplied with a known voltage, which is then tapped off of the variable resistance terminal. I don't think that can be done with a crystal.

Therefore, as the carbon pile is subjected to mean pressures (say, at cruise and/or idle and/or 1 atmosphere,) I could see the MAP sensor generating a small number of "flat spots" in its response curve, which would create erroneous readings - and corresponding errors in fuel metering. Thermistors have a similar problem - which is why IAT's and CTS's can also develop flat spots.

A similar condition exists with the TPS - however, this would be more like an actual potentiometer, with a circular or semicircular resistance element and a "wiper arm" for the variable resistance. Leaving the thing sit at idle or at part-throttle (when consistent) eventually causes worn spots on the resistance element, which then indicates replacement.

While it's true that knock sensors don't usually wear out (the crystal is fine - although you might lose the connection after a while...) the key sensors for fuel metering are subject to wear and require replacement.

The reason I haven't gone into HEGO failures is simple - it's really fairly obvious. The HEGO doesn't generate a reliable output signal until the sensor tip gets up to about 1000*F - which is why the later ones are heated - and any ceramic subjected to elevated temperatures over extended times will eventually embrittle and require replacing. This is especially true of electrical ceramics - like the titania and zirconia used in HEGO sensors.

Bear in mind that most of the HEGO sensors are set up to reach operating temperature in less than thirty seconds, then you can imagine the amount of heat being applied to get a delta-temp of about 930*F (mean.) Might as well take a torch to the thing.

The same thing causes catalytic converter breakdown - the ceramic matrix degrades to the point where it can no longer maintain structure, and starts to fall apart. The catalyist itself (a combination of "rare earth metals" - usually palladium and rhodium) is still good, but the ceramic "honeycomb" matrix has fallen apart and is no longer allowing free flow of exhaust gas. Result? Go get a new one.

Remember that understanding how something works is the key to making it work better...

:lecture:

5-90

martin
July 16th, 2005, 16:10
DrMoab,

Here is the Craig H write up I was talking about, he an electrical engineer who figured out weaknesses in the sytem: http://www.olypen.com/craigh/charge.htm



XJXJ, I guess we have different maintenance practices. I grew up where my Dad ONLY did on condition maintenance. There were countless times in high school I had a car broken down. I don't want to be borken down 30 miles from a town with a parts store again, let alone the times I had to leave the car and walk home in high school. Needless to say I now do preventive mainteance to avoid unscheduled breakdowns.

I also purchased a scanner to gain some insight into how my XJ works. I have also helped buddies with late 80s GM vehicles caribrate their TPS to the correct value. Everyone of them used the mulitimeter method and the value the computer saw was low. On a buick 3800 guys had .20-.25 volts when they thought they had set it to .40 using a mulitmeter. If you were close by XJXJ I'd offer to hook up my test set and see how your XJ reads. It pretty cool to see the injector pulse width, see the data stream of all the sensors. It is also good to confirm the AW4 is going into and out of Lock up.

I also work in the aviation industry and replacement of components due to age (flight hours) is a normal practice. To me replacement before failure is "normal practice." The purpose of replacing components is to increase the overall reliability of the vehicle. To me buying replacement parts for the XJ are cheap compared to the price I see paid for Aviation spare parts.

I will say I spend 18 months changing out one thing a month, then drive it some more and then find another item to replace. I also kept the old part for quite a while to make sure I was over the infant mortality of the component.

Also, when I'd change a part I'd expect no different in how the vehicle ran. I'd drive it after I changed the item and notice better throttle response, better idle. All stuff me as the owner would notice.

When I had a new distributor installed, I had already installed the high alt CPS, and I noticed I had more power. I talked to the mechanic who installed it and told him about more power. He told me as the bushings wear in the distributor the timing retards. The new distributor took out the wear and gave me more zoom. The XJ ran fine, just saying new parts can restore power you have lost over time and you do not know because it been so gradual.

I am talking from my experience and as I stated before I thought it better to replace components while they were still available. The fleet size of the Renix is small compared to GM or Ford ignition systems. If you don't drive more than say 25 or 50 miles from home the the "on condition" maintenance generally works. I find myself on 500 -600 mile one way trips to Chicago or Missouri and the hassle of breaking down can be real bad. There plenty of open spaces in IL, IN, MO, KY, TN, and MS where finding a parts store, let alone one open with the part in stock is difficult.

XJXJ
July 17th, 2005, 07:28
Here's some background about the MAP -

Purpose:
The Manifold Absolute Pressure (MAP) sensor is used to monitor intake manifold pressure (engine load). It sends voltage signals to the Powertrain Control Module (PCM) that represent the engines varying load conditions.

Theory/Operation:
PCM supplies 5 volt sensor reference voltage. The sensor, connected to manifold vacuum at throttle body, converts intake manifold pressure into voltage.

A silicon crystal in the MAP sensor senses changes in manifold absolute pressure. This crystal changes the resistance of the sensor depending upon the manifold absolute pressure acting upon it, and the change in resistance affects the amount of voltage that the sensor allows to flow back to the PCM.

Manifold absolute pressure and voltage to PCM are directly proportional (manifold absolute pressure increases, low vacuum, voltage to PCM increases and vice versa).

Sensor resistance and manifold absolute pressure are inversely proportional (as manifold absolute pressure increases, (low vacuum), sensor resistance decreases and vice versa).

Handling No Trouble Code Problems

http://autorepair.about.com/library/graphics/41924011.gif
MAP Sensor (Typical)

Typical Readings:
Sensor output voltage range is 0.5 to 4.5 volts.

Output voltages between 0.5 and 1.5 volts indicate a high vacuum (low pressure) situation, such as idle or deceleration.

Output voltages between 1.5 and 3.0 volts indicate a medium level of vacuum (pressure) such as a cruise or slight acceleration condition.

Output voltages between 3.0 and 4.5 volts indicate a low vacuum (high pressure) situation such as hard acceleration or a mechanical failure.

Any reading of 0 volts or over 4.5 volts indicates a problem.

Handling No Trouble Code Problems

http://autorepair.about.com/library/graphics/41909659.gif

NOTE: The following procedure tests the MAP sensor only.

1. Inspect the rubber nipple (fitting) from the MAP sensor to the throttle body. Repair as necessary. CAUTION: When testing the MAP sensor, be sure that the harness wires are not damaged by the test meter probes.
2. Test the MAP sensor output voltage at the MAP sensor connector terminal B. With the ignition switch ON, and the engine OFF. Output voltage should be 4 to 5 volts.
3. Test the MAP sensor output voltage at the MAP sensor connector terminal B at a hot, neutral idle speed condition. The voltage should drop to 1.5 to 2.1 volts.
4. Test MAP sensor supply voltage at sensor connector terminal C with the ignition ON. The voltage should be approximately 5 volts (▒O.5 V).

As always, check the service manual for the proper procedures and specifications for your particular vehicle.

5-90 - Yep, my SMOG tests always fall in the bottom percentage of similar XJ's. That's at 270k miles and no engine work. Stock CPS, MAP, CKS (cam position) and OEM thermal sensors. Had a few 02's and a couple TPS over the life I've driven my '89 but that's it for sensors. Wiring and connectors are another matter with Renix but I don't hesitate to cross the Mojave and visit family in AZ/NM on a whim.

M - However you wish to do it is fine with me. But, I wouldn't be so quick to brag about buying an expensive DRB scanner for a system that doesn't actually need it. IMO - A semester in electronics school would be a much better investment for any Renix owner/operator.

A new sensor has never been HOT so that only insures the risk of instroducing un-proven components into your system. That's fine if the component is faulty but if you're changing it only because of "moon phase", good luck and let us know how that works out for you on some deserted stretch of highway?

As with Renix and most modern EFI systems - knowledge of electronics will set you free!

rockwerks
July 17th, 2005, 07:46
From what i have seen on the Renix and ODB system they all use the same map sensor. or at least testing shows this. On a lark I have interchanged them and gotten the same readings. I like the older renix system a little more straight forward and the 4.0's seem to llive longer

martin
July 17th, 2005, 08:39
[QUOTE=martin] XJXJ, I guess we have different maintenance practices. [QUOTE]

As I said before we maintain our vehicles differently. I'd appreciate you'd to chill out, this post was about CSR_101 getting his rig running again. I wish you'd get over the envy of me owning a scanner. I've been posting on this forum since 1999 and sorry you missed out on the post when I talked about finding the good deal on Ebay. I also mentioned the scanner so new people reading on here will know the $50 or $150 model at Autozone will not work on their Renix XJ. The aftermarket scanner people have forgotten about the Renix.

I think we have discussed this issue to death hasta

This forum is to exchange information, people can read what we have posted and make up their own minds. I had an electrical engineer room mate in college, the problem with the electrical engineer types is they never understand the principal of Entropy. I also took a circuits class and an electronics class, could talk about op-amps, zener diodes, meaning of band colors on resistors, or current dependent current sources but I won't.

Main Entry: enĚtroĚpy
Pronunciation: 'en-tr&-pE
Function: noun
1 : a measure of the unavailable energy in a closed thermodynamic system that is also usually considered to be a measure of the system's disorder and that is a property of the system's state and is related to it in such a manner that a reversible change in heat in the system produces a change in the measure which varies directly with the heat change and inversely with the absolute temperature at which the change takes place; broadly : the degree of disorder or uncertainty in a system
2 a : the degradation of the matter and energy in the universe to an ultimate state of inert uniformity b : a process of degradation or running down or a trend to disorder

outlander
July 20th, 2005, 23:37
so....how much did you get the scanner for?Does it read "fault codes"(stored information) or active data?

XJXJ
July 21st, 2005, 07:48
Martian - You asked that we drop this and yet you continued writing a novel? What is it you REALLY want from this forum and why would I be envious about something that doesn't matter to me? IMO - Electronics training is so much more important than either DRB scanner available for Renix. So, you go ahead and "pat yourself on the back" because you have a extremely expensive scanner and that makes you so much better than everybody else. O&O...