Searching \ for '[EE] Driving an automotive gauge' in subject line. ()
Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=driving+automotive
Search entire site for: 'Driving an automotive gauge'.

Exact match. Not showing close matches.
PICList Thread
'[EE] Driving an automotive gauge'
2012\05\11@175602 by

Hello.

I am building my racing simulator and I want to use a real fuel level and
water temperature gauges.
I would like your opinion about  which way to drive these gauges is the
best. Sorry if I say somethig silly but my knowledge is limited.

So the gauges have 3 wires: 12V, ground and the signal.

http://dl.dropbox.com/u/3544050/gauge1.GIF

the signal connects to the sender that converts the fuel level or water
temperature in a variable resistance. the resistance varies from few Ohm
until 20-25 Ohm.

my doubts regard how to generate the variable resistance.
I have three options in mind:

1) using a digital pot. the smallest pot I can find is 1KOhm so resolution
could be a problem. I think I can solve putting a resistor in parallel.
2) I read I can use a mosfet as a variable resistor.I have never done that
but I will investigate it
3) I have replaced the variable resistor with a mosfet and applied a pwm
signal to the gate. it works but the gauge because noisy, it makes a sound
like BUZZZ...
also on the drain I have voltage spikes up to 120V. I suppose because there
is something inductive in the gauge...

Does anybody have suggestions or consideration to make?

Thank you
Cristian
If the "gauges" are D'Arsonval moving-coil movements there is
indeed something inductive inside
Do the math for putting a 1K digital pot in parallel with a 27 ohm resistor.  I don't trust my quick try which says the pot should vary from 77 to 338 ohms, which would not be a resolution problem, would it?
Joe W

On 5/11/2012 5:55 PM, Cristiano Cesaretto wrote:
{Quote hidden}

> Cristiano
On Fri, 11 May 2012 22:55:59 +0100, Cristiano Cesaretto
<cristianocesaretto.it> wrote:
>I am building my racing simulator and I want to use a real fuel level and
>water temperature gauges.
>I would like your opinion about  which way to drive these gauges is the
>best.

Those gauges often use a bimetalic strip that moves the pointer by
bending from the heat from the variable current flowing through it.
Check if that is what you have!  In my opinion you should not try to
match the resistance of the sender, but instead drive it with a PWM
switching signal, perhaps with some fixed resistance in series
I've tinkered with these gauges before.  Though I can't remember much of the specific data on them, I'd approach it this way...

Are you sure the range is a few ohms to 25ohms?  That's really low, unless they're "active"/electronic gauges (vs. very common electrical, or newer stepper-motor-needle gauges).  Common values for fuel-level are 0-90 ohm, 0-180, 240-33, etc, though much older cars can stray from this.  Temp senders usually range from a few hundred ohms to a few k-ohms.  What brand are these gauges, or what car are they from?

You might want to measure to be sure.  Put resistors of different values on the sensor line, and figure out what resistance gets you at either end of the scale.

Measure the current with these two resistors in place.  Now, you can determine if a digital pot can supply the required current.

I'd expect pwm to cause the buzzing, but are you filtering the pwm?  Fuel level and water temp are slow functions, so you should be able to add a significant filter.

Cheers,
-Neil.

On 5/11/2012 5:55 PM, Cristiano Cesaretto wrote:
{Quote hidden}

> Cristiano
I suspect putting two resistors this much apart together in parallel would result in a significant linearity issue, and a resulting resolution issue at one end of the range.

That 77 should be below 27 I bet though.

Cheers,
-Neil.

On 5/11/2012 7:00 PM, Joe Wronski wrote:
> Do the math for putting a 1K digital pot in parallel with a 27 ohm
> resistor.  I don't trust my quick try which says the pot should vary
> from 77 to 338 ohms, which would not be a resolution problem, would it?
> Joe W
>
If the vehicle is new enough to have OBDII (~1996 and newer), the shop manual emissions book should have all the info, to be able to test the sensor and the sending units.

On 5/11/2012 9:41 PM, Neil wrote:
{Quote hidden}

>
I mis-read the "few ohms" part of the resistance range and thought the range was 20-25.
Joe W

On 5/11/2012 9:44 PM, Neil wrote:
{Quote hidden}

>>
I've actually designed a product for someone previously for this application - I can give you advice but not schematics etc.

The gauges can have up to 200ma (often 10v regulated through a 50r resistor) through the "signal" wire when shorted - you won't find a digital pot that can handle this.

PWM will work to an extent for mechanical bimetallic gauges but not the newer digitally controlled gauges.

You want to use a constant current source - easy to do with a mosfet or NPN transistor, a current sense resistor, a fairly fast opamp configured as a comparator and a reasonably accurate voltage reference.

Using the above will drive most gauges and you can use the digital pot as the input for the comparator so choose the current flowing through the mosfet. You could also use PWM and a filter for the comparator input - the gauges tend to be so slow to respond that a lot of filtering isn't necessary.

If you Google "electronic current source" all will become a lot clearer.

Dom

{Original Message removed}

{Quote hidden}

This is very probably a simple air-core gauge, most modern car instruments use these as they are cheap and robust. If so then internally there will be two fixed coils, wound at 90 degrees to each other, and a cylindrical magnet in the center attached to the spindle.  By applying two currents in quadratrure to the windings the pointer can be made to move through a full 360 degrees, and this is how they are used for speedometers and tachometers etc..  For fuel and temperature gauges where they only need to move through ~90 degrees the drive is much simpler; usually the two coils are connected in series and the resistive sensor shunts current from the center tap of the two coils.  The gauge (or instrument cluster) may provide some additional biasing resistors to scale and linearise the movement (some coolant temperature gauges use a zener to crudely linearise the reasonse of the thermistor used in the sensor) .  This beautifully simple design automatically compensates for change!
s in the system voltage.

In any case you do not need to drive the gauge with a resistance, you can simply apply a suitable voltage to the sensor connection on the gauge.  With the gauge connected normally, simply measure the range of voltages you see on this pin for the expected range of sensor resistances, then you simply need to make a circuit to provide the same range of voltages.  The voltage source obviously must be able to source and/or sink sufficient current.  Most importantly, if the gauge is powered from an unregulated supply (i.e. car battery) then your voltage source must be referenced to that voltage.  If not then any change in the supply voltage will be mirrored in the gauge reading.

Cheers

Mike

=======================================================================
This e-mail is intended for the person it is addressed to only. The
information contained in it may be confidential and/or protected by
law. If you are not the intended recipient of this message, you must
not make any use of this information, or copy or show it to any
person. Please contact us immediately to tell us that you have
received this e-mail, and return the original to us. Any use,
forwarding, printing or copying of this message is strictly prohibited.
No part of this message can be considered a request for goods or
services.
=======================================================================

More... (looser matching)
- Last day of these posts
- In 2012 , 2013 only
- Today
- New search...