• 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Tank level monitoring
#21
Thanks for pointing out these inductive sensors, It's nice to find such an economical way forward with tank monitoring. I thought I could get a small number of them going with the GPIO and Actions tabs in Openplotter but that hasn't really worked out and I'll have to see what I can do with your code.

They don't work on my aluminum tank of diesel, as we suspected. I have been wondering about a circuit to measure the resistance on the existing tank sending unit using the ADC chip already implemented. The catch is I think we need to disconnect the existing gauge momentarily to take the measurement. There must be a simple circuit that could do that periodically; once every 10 minutes would be more than sufficient. And I suppose on the software side we'd need to report the highest reading in 10 minutes instead of a realtime measure.
  Reply
#22
(10-14-2017, 01:17 AM)Saqqara Wrote: Thanks for pointing out these inductive sensors, It's nice to find such an economical way forward with tank monitoring. I thought I could get a small number of them going with the GPIO and Actions tabs in Openplotter but that hasn't really worked out and I'll have to see what I can do with your code.

They don't work on my aluminum tank of diesel, as we suspected. I have been wondering about a circuit to measure the resistance on the existing tank sending unit using the ADC chip already implemented. The catch is I think we need to disconnect the existing gauge momentarily to take the measurement. There must be a simple circuit that could do that periodically; once every 10 minutes would be more than sufficient. And I suppose on the software side we'd need to report the highest reading in 10 minutes instead of a realtime measure.

Nice that my work can be of benefit.

Have you tried to measure voltages etc on the terminals on the existing gauge ? They are normally a variable resistor and if so the voltage at the output might correspond more less linearly with filling. The analog input terminal is high input impedance so connecting this to the output should represent no problem. Then some calibration work need to be done.  Try it out with a multimeter first and figure out what's going on, then connecting it to OpenPlotter should be easy. Just think of isolation and common GND etc. 

Ole
  Reply
#23
(10-15-2017, 12:07 PM)olewsaa Wrote:
(10-14-2017, 01:17 AM)Saqqara Wrote: Thanks for pointing out these inductive sensors, It's nice to find such an economical way forward with tank monitoring. I thought I could get a small number of them going with the GPIO and Actions tabs in Openplotter but that hasn't really worked out and I'll have to see what I can do with your code.

They don't work on my aluminum tank of diesel, as we suspected. I have been wondering about a circuit to measure the resistance on the existing tank sending unit using the ADC chip already implemented. The catch is I think we need to disconnect the existing gauge momentarily to take the measurement. There must be a simple circuit that could do that periodically; once every 10 minutes would be more than sufficient. And I suppose on the software side we'd need to report the highest reading in 10 minutes instead of a realtime measure.

Nice that my work can be of benefit.

Have you tried to measure voltages etc on the terminals on the existing gauge ? They are normally a variable resistor and if so the voltage at the output might correspond more less linearly with filling. The analog input terminal is high input impedance so connecting this to the output should represent no problem. Then some calibration work need to be done.  Try it out with a multimeter first and figure out what's going on, then connecting it to OpenPlotter should be easy. Just think of isolation and common GND etc. 

Ole

Please report any success on the resistive sender front so I can shamelessly copy it !
I have been trying to accurately read a tank level/voltage using existing resistive fuel sender. 
In my case the old fuel gauge is broken so Ive setup a voltage divider using the 190 Ohm tank sender but my biggest problem seems to be getting a steady reference voltage to start with. 
Of course the boats 12v supply isn't very well regulated and even when I used a 10v regulator chip to supply the sender, that too seemed to suffer from variable loads. 
(My fuel, water and waste tanks are all metal so I actually have 3 resistive tank senders on this regulated supply and I guess that might be the cause)
All the best 
Steve
  Reply
#24
I voltage reference might help you out. 

Take a look at my testing breadboard at :
https://sites.google.com/site/olewsaa/ya...all-sensor
(the section : Using a 10 bit (low precision) MCP3008 analog to digital converter in differential mode and a WCS1800 35A current sensor.).

These voltage references are very cheap and quite accurate.  190 ohms are fairly lowso you'll need a resistor ladder to suit your analog input range. 


Ole
  Reply
#25
(10-16-2017, 05:04 AM)affinite Wrote: Please report any success on the resistive sender front so I can shamelessly copy it !
I have been trying to accurately read a tank level/voltage using existing resistive fuel sender. 
In my case the old fuel gauge is broken so Ive setup a voltage divider using the 190 Ohm tank sender but my biggest problem seems to be getting a steady reference voltage to start with. 
Of course the boats 12v supply isn't very well regulated and even when I used a 10v regulator chip to supply the sender, that too seemed to suffer from variable loads. 
(My fuel, water and waste tanks are all metal so I actually have 3 resistive tank senders on this regulated supply and I guess that might be the cause)
All the best 
Steve
I built one using a plastic stick (the long bottom part of a plastic clothes hanger), and bare copper wire wrapped around it in four places. The bottom is the reference, the other three tell me low, medium and high level. It all sticks down into the water, mounted to the screw-top of my water tank. I used an Arduino Pro Mini and four LEDs. The arduino senses the connection through the water at the various levels. Very simple, and works great. I don't need to know more about my water level than low, medium and high.

Very simple, but I like simple solutions! If you wish to pass that information to OP, I'm sure it wouldn't be difficult to add some code to pass the data via firmata.
  Reply
#26
Regarding the single transducer ultrasonic sensor, I finally had a chance to test the one i purchased from Banggood.

The good news: it works (no surprise there). I set up a simple test rig with an Arduino Mega and using sample code, it immediately started reporting fairly accurate distances from the sensor.
The bad news: Range is limited, both long and short. Short range was no closer than 20cm. Long range was difficult to really measure, but it appears to be around 4 meters.

I think olewsaa's suggestion that two devices be used is a good one. From what I read the single sensor needs a certain amount of time to settle before the receive signal is reliable, hence the close range limit. In my case, no good for tank level monitoring.

I think in my case I'm going to go back to my contact sensor design for my forward tank. It doesn't give me much resolution, but all I really need to know is approximately what the tank level is.

As far as using the ultrasonic sensor as a parking sensor, I suppose you could do it, but not to 40'. I would guess for longer distances you would need considerably more power to send out the pulse, and more sensitivity to receive it. Then there's the problem of spreading of the beam and actually detecting what you want. I also think it would be cool to know how close I am to my pontoon, but there are other priorities right now!
  Reply
#27
Has anyone ever seen any specifications for the accuracy of external capacitive (foil) sensors ?
My current tank monitor uses the external capacitive setup and it seems less than accurate, possibly because there is only one trim resistor.
Measuring of my black water tank seems to be all over the place and I would LOVE to have something more accurate than the current 4 LED display.
In any case, I would like accuracy to be within maybe plus or minus 2-4% for the integration with OpenPlotter to make any sense.
Measuring at a particular level (inductive ?) seems pretty useless to me except maybe as a warning level (with a great big red flashing alarm light in the loo) before overflowing the black tank or similar.
  Reply
#28
Has anyone experienced with fuel flow meters as the alternative to the always troublesome tank level sensors? Maybe not as the alternative but the addition to fuel management.
Ultrasound tank level measure then being just a semaphore for thresholds but flow meter to calculate remaining fuel based on calculation (similar to what we nowadays do in the fuel section of the logbook - needs obviously resetting with every fillup)

What sensors have you used? and what accuracy can be expected?
  Reply
#29
Are you talking about the Hall Effect sensors that are out there? It might be interesting to look at.

Resetting could be done with a float switch or other level sensor in the tank (appropriately damped to eliminate issues with sloshing).
  Reply


Forum Jump:


Users browsing this thread: 2 Guest(s)