AJ16 cooling system

[brian696]

symptoms were gauge reading between N&H and fluctuating

changed thermostat - tested before fitting (with toggle pin) specs 88oC opening - no change when fitted.
Removed radiator - had full re-core (cost £193) refitted it - no change when fitted
Ive removed water pump and waiting for replacement to arrive - BUT I cannot see what could go wrong with a water pump (wasnt leaking)

anyone had water pump failure before?

XJS 4.0 Celebration 31K

[ptjs1]

Brian,

Bearing in mind the less than exemplary reputation of the late facelift minor instruments, do you definitely know that the car is overheating? You might try to check the actual temp at the thermostat housing when the gauge is reading high, using a simple infrared thermometer, in order to pin down whether the fault might lie with the temp sensor or gauge.

Good luck

Paul

[greeban]

Like Paul I reakon you have a signal problem - gauge, sender, connection to sender, perhaps even an earth problem?
Good idea to have the rad recored, just had mone done by the 'old boy' in Caddington recommended on this forum - still to fit it, but looks like a super job. The fins in mine were falling out as I removed it. I have a picture which I will post when I get the time to download it!

[brian696]

Ongoing with my high temp reading - was'nt the water pump so I removed thermostat, blanked off by pass and did 20 miles - temperature only just got past blue on the gauge - checked system throu-
out with infra red temp sensor and got consistant temp reading on radiator, thermostat housing and cyl head. (66degC) Conclusion -water pump and circulation o/k
Removed instrument panel and checked connections - all looked o/k
Fitted replacement temperature transmitter - Item DAC11079K - reading went into red after three minutes - removed and replaced original sensor and reading back to where it was at the start of the saga.
Local jag man says sensors are colour coded and the new one has black insulation around the terminal - my original one is red.
Anyone know about this?
I have not been able to source replacement "red" unit as the original part has apparently been superceded with the "K" letter added to the part no.
With everthing fitted I checked the temp with infra red and the highest reading I got was 84deg C but the gauge still reads above normal. - Its annoying that I cannot sort this out even if it is only instrumentation.

All suggestions welcome!!!

[greeban]

As you have an infra red temp reader I wonder if you could calibrate yours yourself by adding resistance to the circuit? I would guess the sensor is negative temp coefficient i.e. resistance goes down with higher temp - your gauge is reading too high so the sensor resistance has dropped off too much i.e. you need more resistance in the circuit (placed in series) to keep the gauge reading lower. I do not know what the resistance of your gauge is, but if you could measure it and find out the resistance curve for the sensor it should be possible to work out what additional resistance you need to reduce the voltage at the gauge to provide the needle position you require!
The alternative is to get a replacement sensor from a breaker such as Grublogger who has been very helpful to
me in the past when trying to source parts - he seems to have an encyclopedic knowledge for xjs parts.

Kind regards
Graeme

[brian696]

thanks for that reply graham - Car away this week (gear box fix) When I get it back next week I'll do resistance check through the temp range and see what it comes out like. I'll post the results so if you could recommend a resistance from that I could easily fit it in.
I could even end up a happy bunny after all!!
thanks
Brian
Celebration 1996 4.0 - 31k (and struggling)

[greeban]

When you make your measurements can you also take the voltage drop before the sensor. Also, the ideal resistance to find is the one when the needle is in the middle of the gauge so that we can set the resistance in the circuit to this value when the temp is at normal running.

Has anyone else done this? I have heard of taking the gauge out and having it recalibrated, sometimes you can simply get a small screwdriver in behind and move the needle to a new resting position so adjusting its movement.
Are some instruments able to recalibrate by an adjusting screw or such like method?

[Mark_B]

Here's some data I have, I don't know if it's going to be helpful or not. It might tie up in some way.


Gauge reading verses varying sensor resistance input to it.

http://www.far-out.demon.co.uk/cardiy/sensrest.gif


temperature - sensor repsonse
>Deg C ------- Ohms
>
>10 ------- 1427 Cooled with Ice cubes in water
>15 ------- Not Recorded
>20 ------- 856
>25 ------- 767
>30 ------- 623
>35 ------- 512
>40 ------- 435
>45 ------- 352
>50 ------- 293
>55 ------- 256
>60 ------- 218
>65 ------- 183
>70 ------- 153
>75 ------- 136
>80 ------- 117
>85 ------- 98
>90 ------- 87
>95 ------- 75
>100 ------- 63


temperature vs gauge reading

90 deg = 3/4 of the way up from C to N
95 = Just on the bottom of the N
97 = fractionally higher
99 = the bottom of the line was level with the bottom of the N

Many xjs many xj6 have low-reading temperature gauge, these examples read 'N' when boiling over. Previous owners may had fitted a more responsive sensor.

[poprivet]

Faulty gauge. However, to make certain, get three fixed resistors close to the ones below and substitute them in turn for the temp sendor.
These were what I needed to replicate the instrument readings of the 'barrel' water temp gauge on my '88 when I checked a few years ago,

H 22 Ohms
N 68 Ohms
C 150 Ohms

I don't know if the 'round' gauges are different though, An easy way to check would be to measure the resistance of your sender at a known temperature, mine was approx:

20 DegC 850 Ohms
30 DegC 620 Ohms

With these figures in mind, remember the XJ-S tendancy for the gauge to read cooler then the actual temperature! also it may not be able to 'calibrate' a sensor by adding resistance due to the non-linear nature of resistance V temperature. A better way, once the gauge is working correctly, may be to manually tweak the needle.
Pop

[MARKPEARSON]

these are on special offer at jdht web site at £185 +vat for new genuine part

[brian696]

Ongoing chasing the high temp gauge reading

Removed transmitter from car (colour - red type). Boiled pan on cooker and immersed transmitter. Muliti meter connected. Readings checked as temperature dropped. Readings as follows
All temp readings in Deg C.
temp. Ohms temp Ohms temp Ohms
98 86 89 110 80 150
97 87 88 115 79 153
96 88 87 118 78 157
95 91 86 120 77 162
94 95 85 125 76 168
93 98 84 129 75 175
92 100 83 133
91 103 82 138
90 106 81 143

as temperature dropped resistance jumps increased expedentially.

Refitted transmitter to car and went for run. When gauge was spot on "N"
reading was 100 Ohms -ie 92 deg C using above readings.

I could not get a reading when gauge was higher than "N" because of delay in getting out of the car and getting the meter on allowed the gauge/temp to drop back to around the "N" mark. On tick over after the run the reading averaged 110 Ohms (89 degC).

However, the gauge did increase to a higher position on the road (original reason for the concern) but very little coolant had passed to the radiator as this and the expansion tank remained quite cool even after a longish trip.
The engine is definately not overheating - the gauge still reads higher than it should but very little flow is going to the radiator (ambient temp was around 11 Deg C). The new thermostat I fitted was checked against temperature for opening and closing and worked fine. (At a running temp of 92 deg I wouldnt expect much to go to the rad)

From all of this I still dont have a definative answer to why the gauge reading is high. I may fit a replacement gauge and see what happens.
Any other suggestions welcome.

BC.
1996 Celebration auto - AJ16- 32K miles

[greeban]

Looks like you now have a lot of good feedback!

This link may be of interest?

http://www.lectriclimited.com/frequentl ... ns.htm#E02

[brian696]

Success at last - I think
Finally found time to do long run after gearbox fix and to see how the temp reading was doing after all the tests etc -
Did 500 miles this weekend and the gauge reading was just about spot on N reading all of the time - I can only assume by taking the sender through the full range of temperatures to check resistances it has settled down to give correct reading on gauge - either way I'm a happy bunny at last and my cooling system has had a complete overhaul chasing a duff reading - cost me a few bob but long term it was worth it.
Thanks for all the suggestions and information on the way - exactly what the Forum is about.

Wonder what will happen next?

Brian -
Celebration 4.0 coupe 1996 - 32.5K now and looking good!!

[greeban]

Just fitted my re-cored radiator and noticed the six was not silky smooth, felt like an engine mounting was on its way out - after a few days found out what the problem was when the temp gauge headed to the red - the water pump pulley was knackered, threw the belt off and the car began to overheat very quickly!! Managed to get it home and sourced a second hand pump from Gribblogger - graham there is a splendid chap to do business with.

Fitted new pump and all is well - apparently these pumps are prone to throwing the pulley as it is an interference fit - mine worked loose and according to Graham it is very difficult to get the pulley and shaft lined up again.

[MichaelSB]

Hi All
My car has the same symptoms of gauge reading between N&H and fluctuating H.A Fox Jaguar main dealer could find no fault so I think I will buy a new sender unit and try swapping it myself, hopefully less work than Brian had.
Michael