How will you modify the ECU voltage to the gauge for proper read out. As I see it from a EE perspective, the voltage does not change until there is a wide coolant swing into the danger zone. That voltage is controlled from the ECU per the RAVE schematics to the gauge. Just curious how you would mod the readout without re-flashing the ECU. If there is no voltage/resistance swing from the ECU it won't matter what you do to the gauge itself.

 

IMHO - even when the gauge is made to read out in smooth analog, a change of 3 or 4 degrees might not appear except as a cat whisker of movement. But it will show on the Ultra Gauge. There is room for experimentation, the sensor has two thermistors in the same housing, only one is used in the D2. From the LR fuel injection guide:

NTC type sensor
• Sensor contains two elements, only one is used on Discovery, on Range Rover one is also used for the instrument temperature gauge.
• Output = Approx. 4.9v at -50°C (-58°F) to 0.75v at 130°C (266°F). Approx. 1.8v at 70°C (158°F)
• ECM fault default value = dependant on software map up to 60°C (140°F), after which Fuel System Inputs 75 defaults to 85°C (185°F)

 

The position of the dial can be controlled in two ways: by varying the voltage (analog control) or by pulsing a single voltage (digital control). The ECU uses the pulsing method (pulse width modulation or PWM).

My design does not attempt to modify the ECU as the internal design is unknown (to me) and it is too expensive to chance breaking it. I would "hack" into the wiring harness by cutting the one wire running to the indicator and installing a connector. My device would replace the ECU's pulses by sending it's own pulse train to the gauge.

In terms of where to get the actual information about temp, that comes from simply monitoring the actual temp by asking the for the data via the OBDII port.

 

Yes you are right: a digital readout will be able to show a tiny change in temp. because it is digital in nature.

My question is- do you really want to watch changes in individual degrees? I don't- I'm too busy driving to care about such small changes. I don't care if the temp this precise moment is 184 or 186, it's just not useful information. I just care if it is in a normal range (and of course a knowledgeable owner should know what that normal range is) and I want to know if it is generally holding steady, rising or falling.

For decades studies have shown during an instrument scan, the typical person can't process a mass of numbers but they can easily process positions of dials. This is why even with video screen dashboards on cars and aircraft, manufacturers most often still use the dial gauge format to convey information. For me, a dial with a small numeric digital inset would be the best.

Personally I would prefer if LR had silk screened actual temps on the gauge, but unfortunately they didn't...

For me, cutting holes in the dash and filling them with LED readouts and buttons, or otherwise attaching new gadgets just clutters things up. So I would only do so if I couldn't come up with a way to use the current equipment to fill my needs. That's just my preference.

BTW: in a case where I would like to know the actual temp (such as in an a serious over-temp situation) the dial gauge can be used to communicate that info and visually call your attention the condition. It's been done before by lots of companies, and it involves ticking the value with the dial. It has been long known that the human eye is quite adept at perceiving even tiny movements with peripheral vision so motion is a very effective way to capture the driver's attention. Whereas you would be much slower to pick up this condition via an alphanumeric screen.

 

Forgot to respond this this part-
not planning on hacking into the sensors at all. Not necessary.

 

Same here. That's why my preference for gauges is 270 degree sweep analog. You can clock them to the normal reading then ever really have to mentally process the data they are displaying. They become basically a Yes/No indicator. If it's 'Yes', then no need to spend more time thinking about it. If it's 'No' then you can look at it longer and decide if you need to worry about it, or is a normal 'No', like lower oil pressure idling at a traffic light.

 

OK. But I'll bet a pail of used differential fluid that the cost for an average owner to construct an OBDII data collection device, and convert that to a signal to drive the gauge, and get all the wiring and parts working as planned might take longer than plugging in the Ultra Gauge and having the info and a lot more to look at. I mentioned the second sensor because it offers a path that would allow analog connection (using second sensor and perhaps a few resistors) while leaving the ECU to monitor the original sensor and continue to drive various decisions in the ECU.

But either way, being able to see coolant temp in real fashion is an improvement over the slow roasting over hickory coals afforded by the Rover gauges.

 

UltraGauge OBDII Scan tool & Information Center
Done.

 

Agreed. Long sweep dial better than short.

 

Agreed, fastest method would be to use any of several OBDII solutions.

Obviously the first version of my solution will take longer for me to develop than using OBDII or installing a second gauge system.

After it is designed and running, installing my system will take probably 30 minutes,,, if you were to buy as a kit.

Now if someone wanted build my system from scratch. You are right,, many would not be able to make it work given any amount of documentation and time. I'm talking about using a micro-controller to do the work. It costs less than $10 but most would not know what to do with it, how to program it, etc.

Yes- you could re-purpose the current gauge, isolate it from the ECM, and drive it directly with $5 worth of inexpensive passive electronics with input from a separate or unused sending unit. Me personally? I would rather make my old analog gauge, "smarter".

FYI- I would be VERY careful trying to tie into the current sending unit in any way as the parasitic effects of the this may unpredictably impact the readings to the ECM,, and you REALLY don't want to start sending confusing cooling water temp info to the ECM as the engine operation and reliability downside is enormous.

Cause let's face it,,, these trucks are unreliable enough aren't they?