The Wierdest Thing!!! HCPA Coolant Temp Sensor

[StephenA]

Since it's been too cold and snowy to work on extracting my ^*(&^(@@ frozen turbo crossover bolts, I picked up a new HPCA coolant sensor - the one that goes in the coolant crossover and turns on the cold advance and fast idle solenoids on my DB2-4911 mechnical IP in cold weather-

and I've run into the WIERDEST THING!!:

1. I have 12v going into the sensor connector, which, when the two terminals are shorted to each other, delivers enough current to easily power both the cold advance and the fast idle solenoids. Engine advances properly and speed up from 700 to 800 RPM, per specs on manifold lable.

2. Old "inoperable" sensor switch shows proper resistances from 0 F to 220 F and correctly passes 12v current when below 95 F which is spec listed in GM 6.5 Factory Training Manual. In other words, it's working properly when removed and powered with separate battery.

3. Neither the old or new sensors pass any current to either solenoid when connected and installed in coolant crossover, engine temp is cold -about 30 F,- both batteries at 12v & brand new!!!!! Truck starts easily, despite lack of solenoids.

4. I replaced the old connector with a new one, and can still power both solenoids by crossing the terminals... - but not with either the old or new sensor!!!!

5. Per my GM 6.5 Factory Training Manual, there are no other sensors in this circuit, just the gage fuse, which is obviously working, or I wouldn't get power at the solenoids by simply jumping the connector terminals...

6. Terminals on switch and connector are brand new and clean - I even bent them over a bit to insure good contact. Connectors and spades on both solenoids are clean and pass plenty of current when sender connector is jumpered.

7. I've checked with 4 dealers and I have the right part number (15326386) and each dealer confirmed the crossover location of this part.

8. I think I hear someone/thing giggle everythime I scratch my head... (just kidding).


WHAT IN THE SAM HILL CAN IT BE!!!!!!!!????

[gmctd]

Shouldn't show any resistance, Stephen - HPCA Temp Sw is a temp controlled switch, open at warm or closed at cold - no in-between.

If you're measuring varying resistance based on temp, that is a gage sender you're troubleshooting.

The two solenoids draw ~2-5amps, collectively, iirc.

'91 HPCA Temp Sw was in right head, rear - the resistive sender in the t-stat\crossover was for the 4L80E TCM, also iirc.

Either case, the HPCA sender is a switch, not a resistive element.

[StephenA]

Hey jd! Great to hear from you!! I was hoping you'd make a posting -you're knowledge and application is always so vast. I also suspected that my switch should be a simple temperature controlled reed switch or whatever with simple open or closed terminals, but according to the following posting (as well as my tests on the original stock temp sensor), it appears to be a resistance switch where by a voltage or current differential is used to close a transistorized switch which passes a full 12v current or not according to a set threshold. Same result- i,e, full 12v is delivered only when temps are below 95 F:

This is topic- How do I test the Coolant Temp Sensor? in forum 6.5L Turbo Diesel at TheDieselPage.com Forums.


To visit this topic, use this URL:
http://forum.thedieselpage.com/cgi-bin/ultimatebb.cgi?ubb=get_topic;f=1;t=009979

--------------------------------------------------------------------------------
Posted by Dennis Galligani (Member # 10379) on 01-30-2006 06:15 AM :

Concerning the Coolant Temp Sensor in the T-stat housing - can it be Ohm tested? What would the appropriate reading be? Will the reading vary depending on temperature?

I recently replaced one and immediately got better fuel mileage, but I was just wondering how to actually test the sensor.

[ 01-30-2006, 07:18 AM: Message edited by: Dennis Galligani ]

--------------------------------------------------------------------------------
Posted by JohnC (Member # 2493) on 01-30-2006 11:55 AM :

T(F) resistance
210.....177
194.....241
176.....332
158.....467
140.....667
122.....973
104...1,459
86....2,238
68....3,520
50....5,670
32....9,420
14...16,180
-4...28,680
-22..52,700
-40.100,700

[ 01-30-2006, 10:25 AM: Message edited by: JohnC ]


This threw me for a loop at first, because as you can see, the resistance rises as temp drops, which would pass less voltage/current if not for the fact that it actually acts as a transitorized switch, when hooked up to a 12v battery. By actual tests, both my new and old sensors return about the same resistance curve when measured at cold & hot temps. And when hooked up to 12v, they both allow no current at temps above 95 F and full 12v at 5 amps at any temp below 95. My surmize is that an IC oriented switch was engineered because of the mean time to failure ordinarily encountered with mechanical switches.

I also read that the coolant temp sensor was moved from the 1985 block, cyl 8 position to the coolant x-over/T-stat housing in the 92-93 version, which was a blessing, since reaching the other was a real PITA. In any case, it's definitely on the x-over on my truck, and the gage sensor is on the head under the cyl 1 glow plug.

Before kickoff today, I intend to take out the new sensor and it's accompanying harnness/pigtail and bench test the whole rig, Gremlins tend to resent this sort of an approach... Could be the new connector, but this doesn't explain why the old switch was inoperable, unless the old contacts were too oil & debri soaked to pass the current. Will report after game...

If I actually have the right part, and it performs correctly on the bench, then perhaps I have an actual gremlin, which if bottled, could be patented and sold to Iran for use in their atomic program...
Steve

[BobND]

As the other poster said, your '92 with mech IP will use a simple "on or off" temperature switch for the HPCA and fast idle control.

It MAY use a thermistor sensor for the emissions computer, as it does for the heat gauge.

If I recall correctly from your first post of this thread, when you jumper the wires at the sensor, the solenoids activate.

If this is true, you are at the correct sensor location, but have the wrong part # sensor installed.

Simple on or off. NO thermistor with varying resistance.

[gmctd]

Right - that post is correct for a '98 OBD-II PCM EFI system, where the crossover sender is PCM temperature ref for FI control.

Also same sender for OBD-I system, replacing that one could alter fuel mileage in cases where resistance did not reflect actual coolant temps.

And it is a negative temp coefficient probe - resistance is hi at low temp, low at hi temp.

If yours matches that post, then it is for the 4L80E temperature input for emissions-control - different shift-points and pressure when cold.

Can also be as you state - resistive sender for an electronic switch module - but you'll need to pull-trace the green HPCA Sol\Fast Idle Sol wire back to it's source in the wire loom, where you'll likely find a rectangular relay-looking module.

The resistive temp sender wire colors are yellow\black, iirc

[ 02-05-2006, 12:34 PM: Message edited by: gmctd ]

[More Power]

Stephen, We see this sort of behavior occasionally with the Oil Pressure Switch. The switch will read battery voltage correctly at the lift pump connector when unplugged, but not when it see the load of the electric lift pump. Bad/burned contacts typically cause this.

In your case, since you're using a new HPCA thermal switch, you may need to go further back in the circuit - like the ignition switch or fuse panel, to find the source of the problem.

If you can power one side of the thermal switch with a direct connection to a battery, and the HPCA operates the way you expect, then you've more or less discovered the cause of your problem as being a bad ignition switch or bad fuse, etc.

Jim