TDC1000: Liquid level sensing through the bottom of a metal tank

Hello Louiza,

Thank you for your question.

The issue is that metal is not a good medium for ultrasonic energy. So the energy that is transmitted through the metal tank is attenuated significantly and if the metal is thick then it is heavily attenuated. In some instances you may need to raise the excitation voltage of the signal being transmitted in order for it to still have enough SPL to go through the tank transmit through the water and return through the tank once again. Typically if you are using a soft metal like stainless steel, the results will be better.

Some things to check: Ensure that you are using a contact transducer you want to make good mechanical contact with the tank so a contact transducer is needed. Also check that the transducer is mounted properly to the tank (No gaps in between transducer and tank). Make sure that the back of the transducer is sealed/dampened. You may also try to use different frequency transducers to see if they are a better acoustic match with the metal tank.

Best,

Isaac

Hi Isaac, 

Thank you for your response. 

Sorry, I should have mentioned that we have already validated the sensor and mounting method using the BOOSTXL-TUSS4470 evaluation board. We have been getting some good results out of it, with both plastic and metal tanks. 

Another thing to note is that I based the design on the TDC1000-GASEVM kit. Is there anything else you would recommend to us to try? Either hardware or firmware changes?

Thank you. 

Hey Louiza,

That is good to hear that you have had success using the BOOSTXL-TUSS4470.

At what voltage are you transmitting for metal tanks using the TUSS4470 and the TDC1000? If you are using the same voltage/frequency on the TDC1000 then the issue could be somewhere else, if you could provide some scope captures of what the transmission of the metal tank looks like it might give us a better idea of what is going on.

Best,

Isaac

Hey Isaac, 

Yeah, of course. In the screenshots below you can see the successful plastic tank measurement and next, the measurement with the exact same setting on the metal tank - the cursor is placed where  we were expecting to see the stop pulse. 

With both devices (the evaluation kit and custom board) we were transmitting at 30V and divided the clock down to 1MHz, as that's what the sensor is specified to. 

I've also added a screenshot from the BOOSTXL-TUSS4470 GUI, which shows distinct peaks. These correlate to the correct ToF. 

Do you think it could be an implementation issue? And if so, do you offer any schematic reviews outside the forum?

Thank you. 

Hello Louiza,

From the screenshots it looks like were not getting a return echo from the metal tank. Are you using the same 1MHz transducer and setup from the reading you provided on the TUSS4470 setup?

It is possible it could be an implementation issue but I highly doubt it if you are getting a reading on the plastic tank. I would be glad to review the schematic for you, I will send you a friend request here so that you may provide the schematics for review through private message.

Best,

Isaac

Hey Isaac, 

Yes, the transducer was the same one on both set-ups. 

Perfect, thank you - I received the request so I'll send the files there. I think we can now close this thread. :) 

Kind Regards, 
Louiza