Temp Gauge rises after 30 mins - Bad IPC on 90 coupe?

[tjenkins]

In tracking down a temperature rise (after 30 minutes) problem for my 90 coupe, the dealer:<P>o Flushed the system with new coolant<BR>o Replaced the thermostat<BR>o Replaced the pressure cap<BR>o Replaced the coolant sensor<BR>o Exchanged the ECM, BCM, no change in symptoms<BR>o Checked and confirmed proper operation of both fans<P>The indications are that the first 30-40 minutes of highway or city driving is of normal temp. That is, the analog gauge reads straight up at 12 oclock. BCM data for coolant temperature, BD21, reads 90C (right at 195F) for this first half-hour.<P>After that, the gauge starts to creep up to 1, 2 and then 3 oclock, seemingly what would be a 'HOT' indication on the coolant.<P>The dealer has confirmed a normal 200F reading, even though the IPC analog gauge registers 2-3PM. The BD21 rises to 105C, then, 110C or so, but apparently does not match the high analog reading.<P>The dealer has all be eliminated everything up to the IPC with "Further, traced problem to IPC cluster assembly, TEMP gauge is faulty causing wrong higher temp reading."<P>Two issues:<P>1) Has anyone experienced this type of scenerio, and if so (or not), does it 'sound' right?<P>2) How much are these IPC's for the 90?<P>I'm going to remove it this weekend, and clean the back connector just to make sure...<P>Any other ideas?<P>--Tom

[Guest]

After sitting in 90 degree (F) traffic and having run the car for at least an hour I found the gage go to one less than the max. At this point I could note that the engine fan kicks on and the temp. does not rise any futher. It is a funny feeling, yet I have not overheated in the last couple of years. I am in te NJ area!

[tjenkins]

Thanks.<P>I would tend to agree, but since we bough the car in 1997, and have now owned it for more than 3 years, and put some 40K on it, the temp gauge has NEVER drifted left or right of straight up. Only during the last few weeks has it creeped up. Over the last 3 years, solid both in summer and winter, highway or city traffic.<P>--Tom

[Guest]

I have owned my Reatta about the same amount of time. My problem was there from the first year. Let me know how you resolve the problem and how much it cost<P>Joe

[Guest]

The ipc is very expensive and mine is doing it latley and my diagnosis was that too. Then the other day my heater core went out I have a 1990 and mabye your experiencing the same problem I am

[tjenkins]

Hmmmmm don't know about a heater core. Why would a bad heater core cause the temp to rise after a solid 30-40 minutes in traffic?<P>Same outside temp...either at 0600 or later on in the day - no difference.<P>I've ordered an IPC from Sharon Auto Parts in PA... should be here early next week - hopefully, I'll be able to put this to bed.<P>--Tom<P><BR>--Tom

[Guest]

Tom<BR>How did you make out with the ipc. How much?<P>Joe

[tjenkins]

Well, the good news is that I only paid $150 for a working unit from Sharon Autowrecking in western PA. <P>The bad news is that indeed, this one exhibits the same indications as the other one. <P>So, either I have two bad temp reading IPC's (possible, but not probable) -or- I have something else wrong.<P>I can't imagine that the temperature gauge(s) would not read correct. In other words, I would tend to believe the fact that it's normal for 30-40 minutes (90C), then, rises. With the replacement of the thermostat and sensor, I am looking at somehow checking the engine to confirm that it is either normal or hot.<P>Today, after driving home, I anticipate checking the ohmic value on the sensor, but even if it did read low (which would produce a higher than normal temp reading on the gauge), why is the coolant hotter than normal as well?<P>Should I look at a weak water pump? Just what is a weak water pump? If the blades were broken, I would have imagined that the mechanic would have discovered this when he flushed the system.<P>The level has never changed - and I can find no leaks.<P>Although the mechanic says the fan(s) are both good, I have yet to confirm that they run in parallel (highspeed) when the engine starts to get hot. I seem to recall that they should go from low speed (serially wired) to high speed (in parallel) for both relays after 210F... but I can't confirm that.<P>Any ideas?<P>--Tom

[Guest]

Tom have You tried sticking a thermometer in the radiator or using a thermosense gun on it to verfy its temp? When I first read Your post it sounded like it was running pretty close to normal.Just the analog gage was erratic. If it is heating up slow I would think that You may have a clogged radiator. Or should i say partially clogged. I know shops like to just flush them out but once it has deposits in the tubes, the only thing I know of that cleans them out is to dissasemble the radiator and rod it out. Not many places do that any more. What happens is as your car runs for a while and gets hotter the radiator isnt as efficient so it cant dissipate the heat as fast. It will work fine until it heats up. I have also seen leaking head gaskets cause somewhat the same problems but it usually happens faster. Just for a easy way to check it, if You take Your radiator cap off when it is cool and run it till it heats up, it will look like it is boiling bad in the radiator. Again I dont think that is the problem tho. If I am missreading the post let Me know and I will check the diagnostics for Your car.

[grm]

Double check to see if the thermastat is not upside down. I have seen when this is done you get different readings. good luck. grm

[Guest wally888]

Wonder if nothing would change, all stay the same, after it heated up, w/o the Thermostat installed?<BR> Bet there is someone out there driving w/ no or a broken, open Thermostat. Wonder how his/her's reacts to hot wx?

[Guest]

The bad part of running without a thermostat is there is no control over temp. New systems must run a certain temp. Hence the 195 thermostats, otherwise Your fuel and emmissions system will not work properly. The reason I did not give the thermostat much thought was because Tom had it replaced and it is doing the same thing.

[Reatta Man]

All,<P>If your car's temp starts creeping up, and you have approximately 100,000 miles on it or more, you MUST have the radiator rodded out--not just flushed. <P>Since the EPA had tightened rules on harsh cleaners, all flushing does is replace the coolant and remove any loose debris. In order to have 11-14 years of mineral deposits removed, the end tanks must be taken off and the core given a dip in a strong industrial cleaner or an acid solution to soften the deposits. Then metal rods are run through each row. If you have the time, watch the radiator shop do this on your core--you will be shocked at what they remove. <P>One other way to tell if your radiator is tired is to see if it is bowing. When you take it out of the car, step back and look at the rows to see if they are straight. If it seems to be bowing out, especially in the center of the core, this means that the radiator is bowing or flexing when the temperature and pressure build, but it is so clogged the coolant can't flow through it fast enough, so it bends. <P>Also, recheck the fan relays located in the relay center box in front of the air cleaner housing. There are three relays in there, and I just had one fail last week. My "pusher" fan in front of the radiator was not working at all, so I was getting the temperature creep as well. <P>Good luck!<P>Joe

[Guest]

Tom, There may be two coolant temp sensors on your car. One for the ecm, the other for the gage. If the gage has a dedicated sensor, the dealer may have replaced the wrong one.

[tjenkins]

Interesting and thought-provoking comments.<P>Btw, I was under the impression that there was only one sensor at the top, drivers side, with a black and yellow wire.<P>The flushing vs rodding (or replacement) of the radiator didn't cross my mind, since the dealer was telling me that the car was not hot. Although I haven't yet confirmed the resistance (ohms) reading of the sensor at a given temp reading on the cluster, I would tend to agree with the above comments regarding the radiator.<P>Assuming that the car is in fact NOT of NORMAL temp (and the sensor/ECM/BCM/IPC were good), it would make sense that the radiator was somehow limited in functionality.<P>I talked to a radiator service today, and they confirmed that 10 years is about average for a copper core/plastic sidewall unit. For the radiator only, the dealer wants $577, NAPA is $190, and the radiator service (complete job with part and labor) is $260.<P>I'll call the dealer one last time to confirm that they are standing by their statement of the 'normal' temp, but I would be inclined to R/R the radiator at this time.<P>Film at 11.<P>--Tom<p>[ 06-05-2001: Message edited by: tjenkins ]

[Reatta Man]

Tom, <P>If NAPA or another shop gives a lifetime warranty on a new radiator, I'd consider going that route. <P>If the radiator shop is a class act, they will tell you if your core is shot or if it can be saved. If its too far gone, I'd go with an complete new radiator rather than have them put old tanks back on a new core. Back in the old days, we could put the old metal tanks on a new core and go on down the road for less than a new radiator. However, these days, NO plastic tanks last forever, so just getting a new core may not fix the problem for long. <P>If you don't have many choices to shop for a new radiator in Hawaii, and you can do without the car for a few days, check the online sources and have them air express a radiator to you. <P>By the way, just remember, the shop will likely be checking the heat measurement on your car inside the shop, with the hood open and the engine idling. That is NOT the real world, and sitting on a 100 degree street with the A/C on in the blazing sun is a totally different situation. That may explain why they say it is not overheating. Been there, done that, have the blown head gaskets to prove it. <P>Joe

[ductune]

Does the HOT warning light come on? The gage gets information from the BCM but the signal originates at the temp sensor and goes to the ECM. The ECM sends the signal through a data link to the BCM. That link may your problem. I think the ECM converts the signal to digital. If the gage reads hot but the hot warning light does not come on it may be a BCM problem. If you have a manual, it might be worth your time to check the link from the ECM to the BCM.

[Guest wally888]

Tom. I knew if this went on long enough, it would get over our heads!

[tjenkins]

Ah yes, been there, done that, got to the T-Shirt <P>Although the HOT light/indicator has yet to come on, the temp gauge, and cooresponding ED04/BD21 computer readings indicate that a higher-than-normal (90C) reading exists at the single sensor. Typically, after 30-40 minutes, it'll rise to 95, 100, 105, 110 or so. Haven't had it on after an hour, although I'd suspect that it would rise correspondingly.<P>Recall that the dealer had R/R'd both ECM and BCM, which all but rules out these two modules. He also replaced the sensor and thermostat.<P>So, with all the replacements, I need to actually CONFIRM that the engine is either normal 90-95C, or indeed running hot (>110C) first, THEN, fork at either a radiator or electrical problem.<P>Thoughts??<P>--Tom

[Guest]

Tom,<BR>I'm in FLA with 91 coupe with 138,000 miles i've had for over four years. I posted this problem about two years ago.<P>I have replaced the radiator with a larger capacity three row new radiator, checked the fans and relays. Normal reading is still mid-way, but the guage still rises, during stop and go under 35mph, to between 2 to 4 o'clock. The guage drops pretty soon after speeds above 35mph.<P>I have bought, the last one in stock, an adjustible auxillary temp sensor/switch from Discount Auto Parts that mounts between the fins. You then splice into the fan wiring that then turn the fans on at any desired temp. I have not installed it yet. I am consulting with a electrical/instrumentation engineer for his advise on the splice as it may fowl up the computer to read current flow when there shouldn't be any flow.

[tjenkins]

Intersting... was it always this way? Did the temp rise on the gauge happend since you owned the vehicle?<P>Also, did you ever check the temp (either from BD21 or ED04 from the on-board computer?)<P>Have you tried other things such as the sensor, the thermostat, the BCM/ECM combo, and even the IPC?<P>I recall that the sensor is essentially a variable resistor - when cold, there is high resistance, and when warm or hot, the resistance drops. The ECM/BCM combination turn this analog signal into a serial digital signal which in turn feeds the IPC. So, if a corresponding drop is resistance is sensed, the IPC reads high(er than normal?).<P>As I understand it. <P>I'm going to make it a point to actually check the resistance of the sensor in a day or so to see if in fact it does equate to a warmer than normal temp.<P>From the manual, temp vs resistance on the sensor:<P>-- F ---- C ---- Ohms<BR>- 210 -- 100 ---- 185<BR>- 160 --- 70 ---- 450<BR>- 100 --- 38 -- 1,800<BR>-- 70 --- 20 -- 3,400<BR>-- 40 ---- 4 -- 7,500<BR>etc etc<P>--Tom<p>[ 06-05-2001: Message edited by: tjenkins ]

[Guest]

Tom I would not change anything until I verified that the engine is or is not running hot. If it is not hot then You can eliminate the mechanical and focus on the electrical. I dont have a schematic for Yours or I would try to look it up. If there is more than one senser it aould give You someplace to look. I will say that I can see how it could have one for the dash gage and one for the ECM. I would get a thermometer on Your engine and radiator.

[Guest]

Sorry for the confusion. I checked the manual and it shows only the one sensor. I thought the first post said the dealer found that the reading in diagnostics showed the temp to be normal but the gage reads high.

[Guest]

I have the same problem with my '89. I have yet to find the source of the descrepancy but feel it is the link between the BCM and ECM. My CRT screen registers over 250 degs. at times and most of the time will be at 237 deg. The motor is not hot, the coolant is moving thru the system just fine but the temp gauge will show a problem. That coolant sensor is tied to the Airconditioning Low and High sensors and to the MAPS and TPS sensors and well as the sensor on the tranny. I have inspected each and every one of these sensors to no avail. All systems check normal but because of the high temp the ECM changes settings for fuel and air mix etc. so my mileage drops off. I have disconnected the sensor and mileage has increased and have been driving it that way for over a year. The MOTOR is not HOT just a system problem that someday I may trace to a final solution.

[Guest]

I have the same problem with my '89. I have yet to find the source of the descrepancy but feel it is the link between the BCM and ECM. My CRT screen registers over 250 degs. at times and most of the time will be at 237 deg. The motor is not hot, the coolant is moving thru the system just fine but the temp gauge will show a problem. That coolant sensor is tied to the Airconditioning Low and High sensors and to the MAPS and TPS sensors and well as the sensor on the tranny. I have inspected each and every one of these sensors to no avail. All systems check normal but because of the high temp the ECM changes settings for fuel and air mix etc. so my mileage drops off. I have disconnected the sensor and mileage has increased and have been driving it that way for over a year. The MOTOR is not HOT just a system problem that someday I may trace to a final solution.

[tjenkins]

How did you confirm the motor was NOT hot?<P>--Tom

[Guest]

I checked it with a old style temp gauge. You know, the fifties technology that has been all but lost.

[tjenkins]

50's technology - now THERE were the days.<P>Do you remember taking number 1 spark plug wire off at night, and holding near the crank pulley to extract a spark to adjust the timeing for TDC?<P>Gads..<P>In any event, I was going to wait until tomorrow to post to the group regarding my radiator replacement, but I've got some news.<P>$5 in Radio Shack parts and I was off to the local radiator shop today.<P>I reasoned this - if the coolant sensor, ECM/BCM and IPC were good, if I were to substitute a known good sensor, and compared the readout on the IPC with the manual values, and they matched, then I could assume the eleectrical connectivity was good. I went to Radio Shack, bought a variable resistor (5K audio taper potentiometer), and set it to 180 ohms with an ohmeter.<P>I ran the engine up to operating temperature, confirmed the 90C on the IPC, removed the plug on the engine coolant sensor, substituted the resistor, and viola - the IPC read 90C. I reconnected the IPC, warmed to 105C, disconnected the plug, reset the potentiometer to 110 ohms, and the IPC also read 105C. So, assuming that this logic is not faulty, I'm off to the radiator shop to give them $240 (Buick dealer wanted $700).<P>Will let you all know how it turns out.<P>--Tom

[Guest wally888]

I suspect this will be a hot topic this summer and maybe hotter next summer. Most overheat problems w/ Reattas probably won't show up here, on the Forum, and the percentage of Forum participants who have/report problems, will probably be low compared to overall owners heat problems because of care and awareness.<BR> I have fought a slight heat problem for several months. Flushed twice and replaced two thermostats (1st was defective) attempting to get the car to exhibit the same reading as my previous 89.<BR> Wonder if we can establish a norm for temperature readings. Maybe there is one?<BR> My previous 89 read 195* under almost all conditions after warm up unless:<BR> If it was cold outside less than 45* (hey, that's cold here) some times it read a few degrees lower.<BR> If it was more than 50* it read 195*<BR> If it was 90+* it read 195* but would warm a few degrees (197-200) when slowing speed from a 65+ mph run, then return to 195*.<BR> At 100+* it ran at 195* but would easily/quickly go to 206/209/213/ when I slowed or sat at a light. The fans always reversed the rise immediately.<BR> My current 89 almost always runs a little hot in comparison. At least 197* ,199/203/204 on the highway , upon slowing 206/209*. I suspect I will soon visit a radiator shop. May have to stand in line.

[tjenkins]

I've made some initial tests (below) and would like to share them among the group. Bottom line - replacement of my radiator did help, but problem still exists (albeit at a lower temp).<P>After opening up my wallet to the tune of $240, I rec'd a new radiator, with plastic tanks and a flush. The guage does read 88-92 during 'normal' highway driving. In traffic, after warmed up for 30 minutes, it starts to creep up to 95, and up to 105ish or so. Seems to settle around that temp, and the analog gauge bounces between 2 and 3 bars to the right of straight up.<P>Although there doesn't seem to be any restriction, I can't figure out why the coolant system 'seems' to work only when external air is driven through the core (like driving on the highway - confirmed in town at 105, then dropped to 88 on the highway within 5-10 minutes).<P>Tonight, I took the resistor and ohmeter out, and made some initial checks. It does appear that the analog and digital gauges are indeed logrithmic (not exponential), and drop/rise at a steady rate. <P>To check, I entered into the ECM computer, and displayed ED04 (coolant temperature in C), and with ignition OFF, measured resistance from yellow wire [active side of coolant sensor] to ground.<P>Temp - Resistance<BR>85C - 261 ohms<BR>90C - 233 ohms<BR>95C - 197 ohms<BR>100C - 168 ohms<BR>105C - 164 ohms<BR>110C - 126 ohms<P>Factoring in 10%, it's easy to see that the low to high readings are related. In fact, actual resistance from the sensor seems to confirm this:<P>Temp - Resistance<BR>87C - 253 Ohms<P>I'm not yet sure why the 105C reading is higher than expected - I'm going to recheck. I'm also going to rerun the tests with my other IPC. Will post results.<P>--Tom

[padgett]

Sounds like what is needed is proper instrumentation/recording using a scan tool. I have seen steady slow climbs caused by a clogged radiator but a sudden change 30-40 minutes in just makes no sense unless something else is changing (like a sudden reduction in spark advance).<P>For something like this I use Diacom sodtware (www.rinda.com - plug) on a laptop connected to the ALDL. This lets me do two things:<BR>1) look at the actual readings from all of the important sensors at the same time<BR>2) Record series of frames for later comparison.<P>If you are really seeing a sudden change 30-40 minutes after full warm up (closed cycle, thermostat open) then I would assume the coolant sensor is correct and "something else" is changing. Just for S&G, have you changed the O2 sensor ? - cheap and simple. Had one 3800 that two minutes at 70 mph and the SES light would turn on. Was an O2 sensor that worked fine below sixty, was erratic 60-70 mph, and failed open at 70.<P>As a consequence I would look for something else (advance, O2 cross counts, inj puse width) that changes at the same time the temp starts up.

[Guest]

This is very interesting. My 90 Reatta coupe has just developed a similar problem, except in my case, when the temperature gauge approached the three o'clok position, the ABS brake and brake warning lights came on, and the brakes behaved as if ABS was active all the time. Shut off the A/C and when the temp gauge came back to near the twelve o'clock position, the warning lights went off and the brakes again felt normal. Kind of scary for a few moments. Any thoughts about what may be going on would be greatly appreciated!