Modding the Toyota temperature gauge (no zener dead-zone!!)

The Toyota temperature gauge is designed to have a large dead zone in the middle, to find out how it does this and therefore how to 'fix' it, we need to understand how the gauge works:



There are 3 terminals on the back that screw into the dash, B+ (Ignition or Ign), Ground (Gnd) and T (from sensor). Only ground is connected to the gauge, B+ and T and just free terminals and if you disconnect the resistors and diode they connect to nothing.

There are ALSO 2 terminals on the needle side, these are the business terminals of 'meter power' and 'meter' position.

If you disconnect the diode the meter input now becomes disconnected, now if you then power up the meter the needle goes to the middle of the dial. This action is accomplished by two weak resistors inside the gauge that merely steer the needle to the middle in the absence of any input to the meter terminal. If you then connect (for instance) a 100 Ohm resistor to the meter input and with the other end touch ground the meter shoots up to the top needle stop, if you touch it to the meter power terminal it goes smoothly to the bottom or 'off'.

On the standard (original) gauge there is a 15 Ohm resistor from the B+ terminal to meter power, we can leave that there and ignore it. Then there is a 75 ohm resistor that goes from meter power back up to T, the dash connector where the sensor connects in on the head. This forms a voltage on the T pin that goes down with rising temperatures and up with lower temperatures.

Sensor values are:
1135 Ohm = cold
255 Ohm = just noticing on the gauge
100 Ohm = just right
19.5 Ohm = top of scale: very hot

Now the only thing on the standard gauge to connect the meter and swap if from its preferred centre position is that diode, forward it drops around 0.7 V, backward around 3 V (I think it's a 3V part, it's not that vital to know). This means that around the middle of the scale when the voltage on T rises to within 0.7 V of the middle the diode ceases to conduct, and when the voltage gets to +3V past the centre the diode again starts to conduct: this is what creates the large dead zone on the gauge, an unknown area of 3.7V. Two reversed parallel diodes would create a smaller dead zone, but I don't have time to analyse the zone position so I decided to eliminate the dead zone completely.

The key to my modification is that the 75 Ohm resistor and zener diode are entirely removed - snip them out carefully with small wire cutters, desoldering may damage the gauges.

In order to get the gauges to correspond to the same 'just noticing/just right/hot!!) positions I experimented and ended up with a slightly different topology. Connect a 220 Ohm resistor from meter power to meter. They are the two terminals on the needle side of the meter, and the 220 pulls the needle down to the bottom. Then where the diode was solder in put a 47 Ohm resistor. This is still between T and meter, and electrically adds a series constant 47 Ohm to the sensor resistance, the combined resistance then meets the 220 Ohm at the meter terminal and the resultant electrical tug-o-war will give a proper reading with no dead spot.

New resistors required
1) 220 Ohm  2W or higher rating, placed between the 2 lower face side terminals
2) 47 Ohm 2W or higher rating, placed where the old diode was (between T and meter).



My gauge is a 1996 Yazaki Meter one and the resistor values are based on characterisation of the old gauge, has not been tested on the van yet so there are no guarantees but when I fed in the same resistor values I'd seen on the original gauge it gave the correct indications on the new modified gauge so I'm confident it will be good, but I don't know quite where the needle will be pointing at 80C for instance, you'll have to get used to what is 'normal' for your own van.

You can see in the 2nd picture my fan of resistors I made with values of 255, 100, 47 and 19.5 ohms to simulate the sensor as I calibrated the gauge and selected the 220 and 47 Ohm resistors and the new topology (only one connection to the T terminal now).

I suppose at this point is is possible to add a warning lamp circuit but the gauge is quite clear to see and I didn't, but if anyone does please add to this thread!!

The Super Custom Limited's gauge is even worse! There's six green digital markers between C and operating temp, then a single marker above operating that's yellow, and three markers above it that I'm assuming are red. I'll probably be installing a separate coolant temp gauge and/or a pyrometer (exhaust gas temp) before summer hits.

That looks a bit coarse, although quite cool!

When the engine is warm you can measure the resistance of your temperature sender to see what 'normal' is.
Then (probably later when it's cool) you can buy a potentiometer of the right type of value (buy a 2W one) and twiddle it to see how responsive your gauge actually is around the vital temperature.

So, how is this modification going? Does it still work well? Is your normal temperature now in the middle of the gauge or has it changed?

Thanks for sharing!

Hi!

It's all working well !!

But my cooling system maybe at a slightly different temperature than usual, because I replaced the 1KZ system with an external Landrover thermostat, to prevent engine temperature bouncing, as the side semi inlet thermostat can provoke.

With those resistor values it reads a little above the middle. So it could be tuned a little.

As I noted, my cooling system is different, but up a hill it moves about 1mm up, and downhill about 1mm down. When cold it's at the bottom of course, so I think my coolant system mod stabilised the engine temperature, and the gauge now tells me everything.

I did think about added a warning lamp for if it goes up high, but did not do that. It's a good idea though. In greece there are big hills and it gets very hot in summer, so I'd do the gauge anyway, and consider the coolant system mod later.. but actually the biggest risk with the old system is in cold weather, so you may not need it so much.

Hey @GPW did you ever trim those resistor values to bring it more central? I noticed that our old van's temp sat a little off centre as yours did with the values you used and I'm sometime going to check whether our Dyna / Hiace (whatever it actually is) has the same dead zone, if it does I'll mod it but it would be good to have it correctly central at normal temp.
I could measure the thermistor but the difficulty would be in knowing when the engine is actually at the correct temp since I don't have one of those fancy infra-red non-contact thermometers and don't know exactly what temp it should be anyway.

Hi!

Sorry, no I've left it as is for the moment !
The sensor could be measured in a saucepan with a thermometer, and the gauge run on a voltage equal to the usual van running voltage (14.2 - 14.4V) to test...

Turnigy do a cheap IR sensor for model planes (Hobbyking), but the ideal temperature.. well, it's a subject.
The actual best temperature for fuel economy is 'as hot as possible', so with a coolant mod and the Land Rover thermostat the temperature should be know, I think the one I fitted was 90C or thereabouts.

Is it possible to drive the van to a temperature you think is ideal, and then disconnect the sensor and measure the resistance? Then measure again at rest, cold, and use those two value to tune the gauge.

Also you need to select a 'too hot' resistance, I guessed, but perhaps heating the sensor to 110C or similar - and designating that as 'top of scale' would be interesting.

If we were to analyse the gauge more carefully (what voltage (or voltage ratio) corresponds to hot, ideal and cold) we could calculate the values needed.

It would be nice to have it calibrated, but it's a project in itself

Your suggested temps give me something to work on. I'm unsure if the sensor is just tapped into the metalwork or if removal will cause coolant to pour out which I don't want.
Anyway it's a project for when I get some enthusiasm and once all the other little things that have come to light recently have been fixed as best I can do or we can pay for.
Thanks for your thoughts.

I suspect it's in the coolant, most are: if you can buy a replacement sensor that might be good enough - they are supposed to behave the same, and you could always swap the new one in.

The coolant at the top of the engine, once pressure is released, is not going to come out fast - so swapping over is feasable: but leaving it open for a couple of hours is probably not ideal, unless you drain a little out first.

Several possibilities, but that issue isn't high on the list of priorities.
Today I think I have finally found the source of water entry into the rear entrance area, this has had me mystified for some time as it suddenly started happening but became persistent.
I need to replace the horn sometime as you could probably open the window and say 'beep' with as much effect as the stock one.
More rust removal is needed.
External drip mouldings need changing for longer ones to stop discharge right over the front doors.
The squeaking noise which sometimes comes from the engine needs to be located.
Battery charging systems need optimising.
I'm sure there's more, I just can't think of them right now!