The water tank level indicator display has broken in my van. There are four resistor-based sensors in the tank and I'd like to replace the display with an lcd or led type designed for batteries. Powering from the 12V nominal lead-acid or LiFePo4 battery (it needs to work for either) in the van I could replace the resistors on the tank as required for the display.

Assumed circuit is: Battery > Regulator > Tank Sensors > Level Indicator > Ground

The voltage of the battery will vary. How is:

1. power level regulated so the indicator measures tank resistors rather than battery voltage
2. required tank resistors calculated

Would a lead-acid level indicator with a more even voltage versus capacity curve or a Lithium one?

 

Hello there welcome to AAC! Year,Make, Model of your van please.
.

 

These types of sensor never work for long, the forum I'm on (motorhomefun) has many posts about the problems all vans seem to have with them. Our van had 2 stainless rods going down into the tank and a moving coil meter in the control panel, after a couple of years one of the rods (anode?) got a coating on it even though the gauge was momentary action.

I fitted a float switch at the 25% water level to light a red LED on the control panel and it's worked perfectly ever since.

 

Hey, thanks for that.

It's more of a non-descript stand-alone trailer. 12V 100Ah AGM battery charged by portable solar panel likely to be changed over to LiFePo4 with a MPPT controller. The faulty 4 led display is powered from the battery for a 60L plastic potable tank with readily accessible resistors (part A7552) on the sensors like the following:



Hmm, better assume circuit is: Battery > Regulator > Level Indicator> Tank Sensors > Ground

There is no power regulator or level indicator and open circuit is empty. I'm happy to pull the sensor screws out of the tank and clean them as required, been working for 10 years until recently.

 

I've checked the voltage steps on the 12V Lead Acid led level indicator in the first post (100% to zero in 25% steps) with 0A setting on power supply: 13.2, 12.5, 11.9, 11.2, 0

 

This type of system depends on the conductivity of the water to pass a small current between sensors covered by water, which turns on a transistor, which turns on an indicator light to indicate empty, ¼, ½, ¾, or full level.


I recently saw a circuit posted here using a ULN2003 in place of individual transistors.

 

I recently saw a circuit posted here using a ULN2003 in place of individual transistors.

https://forum.allaboutcircuits.com/threads/uln2004-water-level-indicator.100327/

I want to use an existing module, preferably the led one. Hopefully, I just need to change the existing resistors and add a power regulator chip.

 

The tank level indicators in GMC Motorhomes (made from 1973 – 1978....yes, the is a cult following!) fail because the transistors fail or the incandescent bulbs burn out. I haven't got to fixing ours in the 1976 Royale we bought almost a year ago, as it not working is a minor inconvenience compared to some of the problems we've had.


We knew there was a design issue with the alternator charging circuit that could lead to bad things if the alternator failed in a particular way. Somebody way back when determined they could save the cost of a resistor in the charging circuit by using about 5 feet of nichrome wire instead. This 5 feet of toaster element wire leads from the fuse block on the right side of the dash to the instrument cluster on the left. To keep things tidy, it's bound with about 20 other wires leading from the fuse block under the dash.


When the internal voltage regulator in the alternator failed, the 90+ volts the alternator was putting out through the toaster wire did exactly what you'd expect. Since we knew about the issue, we quickly knew what was happening. The wiring harness was just fused into a molten blob of plastic instead of catching on fire! Of course the blob was in the almost unreachable part of the dash.


We were fortunate that there wasn't a fire and that we had an accurate wiring diagram – about 5 of the wires in the harness were orange. It took a while, but we got everything spliced back together with only 2 mixed up connections.

 

These types of sensor never work for long

No, it's only a momentary display that runs for 5 seconds when the monitor button is pressed. That seems to avoid all the corrosion problems with running a current through water and wear-and-tear on a physical gauge.

Anyway, I'm surprised no real implementation advice has been given. It seems a buck/boost converter is needed to provide a stable power supply allowing the sensor voltage to be measured instead of the battery. A potentiometer should allow the resistance to be set through trial and error to allow say 10-20mA through.

Any suggestions for a simple power supply through the described batteries would be appreciated.

One issue is the lack of sensitivity by using a meter designed for a battery:

Voltage	%max	Voltage - 10.5V	%max
13.2	100%	2.7	100%
12.5	95%	2.0	74%
11.9	90%	1.4	52%
11.2	85%	0.7	26%

Could something like a diode be used to adjust battery voltage and avoid the need for a power supply?

 

I think this would work better with a 2S 8.4V voltage indicator and a zener diode power supply.

Assume:

1. power supply drops voltage from 14.6V to 8.6V
2. V0 -V4 are 5.6, 7.3, 7.0, 7.7, 8.4 V respectively
3. allow 4m of AWG26 cable and 0.5m 200 Ω·m water (ie both negligible)
4. required current returned V0-V4 is 2/3, 3/4, 5/6, 11/12 and 1 * 8.6V