Engine temperature gauge reading high ?

[Ralph Whitaker]

Not being able to afford all new "correct" gauges, I used a set from a spare vitesse dashboard. Unfortunately the temp gauge was duff, so I invested in a brand new one.

Originally I used the original sender that was in my engines thermostat housing (TR4a?) and it seemed to read high, so I bought a new one GTR104 as listed in Moss book, and it read the same, so presumably the original was OK. After a bit of research I got a GTR101, identical to look at, but listed for early Minis. I have just fitted that and it reads the same.

When checked at room temperature the original sender gives me 360 ohms, the GTR104 gives 660 ohms, and the GTR101 gives 916 ohms.

On the basis that the higher the resistance the lower the current flow I thought I was on a winner with the 101, it seems to start off slower but ends up in the same place.

The electric fan has a 86 degree switch, and as that is fitted in the pipe back from the rad to the pump it is reasonable to suppose that the water entering the rad is about 10 degrees hotter, ie around 100 degrees C, but the gauge, as far as I can work out is registering about 125 C.

I suspected the voltage stabiliser, which is a new solid state unit, but that is giving out bang on 10v which is what the gauge requires, so I am at a loss unless the gauge is faulty, but I cannot understand why 3 different temp senders with different resistances give the same end result.

Ralph

Edited February 4, 2021 by Ralph Whitaker

[Lebro]

I guess you need a sensor from a vitesse ?

Bob

[RobH]

You only tested the senders at room temperature. They probably have different response curves, which is why they differ in response rate during warm-up, but all end up at about the same resistance when hot.  You could try testing them in boiling water?

That new gauge will have been calibrated to suit a particular sender which is probably different to the one you are trying.  Is that gage supposed to be powered from 10V in its original application ? 

Calibrate the gauge and sender as a pair - there are instruction on how to do that on a recent thread. 

Edited February 4, 2021 by RobH

[DesNoble]

Am I being a bit simple here. Surely if all the gauges read high then it indicates that your thermostat is duff. Have you tried replacing that and seeing what the temperature reading is.

Des.

[Lebro]

I think there is only one gauge, but several sensors have been tried.

Bob.

[Ralph Whitaker]

As the gauge is one of the "new" Smiths gauges it doesn`t come with a particular application, though it would have been useful had there been some indication of which sender I am supposed to use. I guess I will email the seller and see what they advise. As Bob says, one gauge, many senders, although as they are all supposed to do the same job , other that differences in the fitting (thread, length, connector etc) one would expect them to be similar internally. I have a Vitesse one (GTR1080, but it is different where it screws into the thermostat housing, it seals on a fibre washer under the head unlike the TR one that seals in a taper inside the housing.

Anyway, will keep at it, perhaps I should move the question to the 4/4a forum who have the electric gauges as standard and may have come across this before.

Ralph

Edited February 4, 2021 by Ralph Whitaker

[RobH]

A little bit of research on the web finds this:

"The correct temp sensor is the Smiths TT6811-01, which has an 1/8" x 27 pipe thread. The resistance of the sensor varies with temperature as indicated below. All plus or minus perhaps 10%."

25°C - 820 Ohms,  50°C - 306 Ohms, 70°C - 170 Ohms, 80°C - 142 Ohms,   90°C - 106 Ohms, 100°C - 79 Ohms, 110°C - 53 Ohms. 

 

https://www.competitionsupplies.com/smiths-oil-water-temp-sender-1-8-x-27npt-to-suit-bt2240-temp-gauges

below are typical curves for some different senders. You can see that there is considerable variation at the cold end but they are fairly close together at 100C.  

Edited February 4, 2021 by RobH

[Lebro]

Could you try the vitesse sender in a bowl of hot water just to see if it reads correctly with your gauge. At least you would know that the gauge is OK.

Bob.

[Ralph Whitaker]

6 minutes ago, Lebro said:

Could you try the vitesse sender in a bowl of hot water just to see if it reads correctly with your gauge. At least you would know that the gauge is OK.

Bob.

I could, but as the gauge is not from the vitesse it wouldn`t achieve much. Rob H has found a link to a site that sells the gauges and the correct sender. As it is a different sender to the ones I have tried before I have ordered one and will see if that is better. Thanks anyway.

Ralph

[ijonsson]

I went through all of this and ended up adjusting the position of the needle in the gauge, after a tip here on the forum. My gauge has two holes on the rear where you, with some care taken, can adjust the needle position. This after measuring with another brand of gauge and with IR thermometer in parallel to be absolutely sure that the Smiths gauge was reading way to high.

[RobH]

Procedure is in the second post in this thread but the modern gauge may be different.

Edited February 4, 2021 by RobH

[Ralph Whitaker]

I received my TT6811-01 transmitter today, now I will have to spend tomorrow making an adapter if I have the right taps and dies. Pity they don`t make them in the right size to fit the TR range.

Ralph

[Lebro]

Why not test in hot water first in case it's no better than the others ?

Bob.

[Ralph Whitaker]

1 hour ago, Lebro said:

Why not test in hot water first in case it's no better than the others ?

Bob.

That`s not a bad idea. I have emailed to see if they do the same range sender but in the more common size used on Triumphs, Minis, Mgs etc. but I can check it first to see if its likely to work.

Ralph

[Ralph Whitaker]

I emailed Competition Supplies and it would seem that this temperature sender (TT6811-01) is the only one suitable for the current range of new Smiths gauges, and fit has to be achieved by means of adapters which they stock.

I was able to order an 1/8NPT to 5/8UNF adapter specifically made for applications such as the TR where replacing a capiliary type gauge, and has the tapered seat to mate with the taper in the thermostat housing.

My only concern is that the new sender is half the size of the original and by the time a reducer bush is added into the equation the bulb will barely protrude into the water flow. I don`t know whether it will make any difference to the operation, I will have to wait and see.

This TT6811-01 sender is the standard fit for, among others, Lotus, Jensen, & Reliant Scimitar, but numerically I would have thought that these are vastly outnumbered by the surviving Minis, Mgs, and Triumphs which all use a different size transmitter, but I suppose if you only want to make one that will cover everything you make the smallest which can then be made to fit others with adapters.

Ralph

[Ian Vincent]

There might be a very slight lag in the temperature reading accurately but ultimately I wouldn't expect it to make any difference.  The coolant (in the head) will all end up at the same temperature as will the thermostat cover.

At least if you have an adaptor you will minimise the risk of the sender becoming welded into place although I suppose that isn't a big issue if you have an electric connection, because it is easy to unplug.

Rgds Ian

[Ralph Whitaker]

OK, the adapter arrived, shorter than I was expecting, and comes with a Dowty washer to make a seal against the thermostat housing. I think it has been deliberately kept short because as you can see from the photo the transmitter barely protrudes. The sender in this pic is only screwed in hand tight and may go another turn with a spanner but not much more as it is a taper thread. Will not get chance to fit it today as I am in the middle of re-building a gearbox for my Renault Trafic van, and in any case the TR is jacked way up at the front at the minute so I can fit the inner wing splash plates, but hopefully next weekend I will be able to fit and try the temp sender.

Ralph

[Ralph Whitaker]

OK, fitted the new sender yesterday and I am happy to report that the gauge now reads correctly at around 95-100C when fan switches on. (Fan switch is an 86C item in the return from radiator to pump), and given that the drop in temp through a radiator is usually in the region of 10 degrees I am pretty happy that is close enough.

Thanks for all the help.

Ralph.