Best sensors for measuring fuel flow rate in vehicles

There are many ways to measure flow rate. Accurate flow rate measurement may require a positive displacement flow sensor, but they aren't usually cheap. Flow can be measured at the fuel injectors as well. Usually you can just tap into the OBD2 system to extract fuel data, because every car these days has fuel measurement already implemented. It is unlikely that a low-cost external measurement system will perform better. There also questions about safety that affect the answer. More information about your application is needed to give a better answer.

1) This device is to address short-fueling.(Yes, short-fuelling is a problem in many countries).
Not connecting with fuel pipe but fit around the tank's fuel cap so that when the vehicle is refueled, it notifies the actual volume amount going in the vehicle.

2) This is not the best sensor but to start with but can be used with fuels. e.g. similar plastic sensors for diesel. www.amazon.in/.../B093ZNNY84 . Planning to move to other sensors as we progress. If you know any better compact, cost efective,accurate sensors that can do the job, please do share.

3) Which is the simple, compact, cost effecient way to get refueling volume displayed to the users? On a seperate device or on a app or send a notification on SMS/Whatsapp? Please help share more in detail on devices available for each option & how it can be achieved?

"but fit around the tank's fuel cap".

Surely there's just the tiniest risk, maybe I'm blowing it out of proportion (but I don't think so) that this isn't the best idea, to put dodgy Amazon hardware or your DIY circuitry around the whole general fuel area?

What's the benefit anyway. Will you argue with the fuel station that this cheap project is more accurate than their dispensing systems which might cost a thousand times more? Where will that get you?

shabaz To make an informed decision & customer awareness. I repeat, yes short fueling is an issue in many countries. This is an open design research project & so progress. Not just with close minds.

As I said in your other thread - this flow meter is not very accurate. In reality, the accuracy is usually around 10-15%, even if such "manufacturers" claim 1% it will be under best conditions (i.e. calibrated against an external reference, at a fixed flow rate). Where the flow rate is not consistent, the result will be wrong by quite a bit - below the minimum flow rate, it won't register AT ALL or will do so inconsistently.

Fuel is tricky - its volume often changes quite a bit with temperature and proper fuel pumps compensate for change in density due to temperature. Such a sensor will not natively compensate.

You may be better off filling your own fixed-volume jerry cans of known volume and transferring that to your own tank. Or just fill a can and weigh it. It seems quite a bit safer and simpler ...

- Gough

As I said in your other thread - this flow meter is not very accurate. In reality, the accuracy is usually around 10-15%, even if such "manufacturers" claim 1% it will be under best conditions (i.e. calibrated against an external reference, at a fixed flow rate). Where the flow rate is not consistent, the result will be wrong by quite a bit - below the minimum flow rate, it won't register AT ALL or will do so inconsistently.

Fuel is tricky - its volume often changes quite a bit with temperature and proper fuel pumps compensate for change in density due to temperature. Such a sensor will not natively compensate.

You may be better off filling your own fixed-volume jerry cans of known volume and transferring that to your own tank. Or just fill a can and weigh it. It seems quite a bit safer and simpler ...

- Gough

Gough Lui  Your inputs on accuracy makes sense. Definitely accuracy & flow is to be addressed. Can using it with Temperature/density sensors help giving more accurate levels? Would love to discuss to address the concerns and get to a resolution. Thanks.

No one really wants to put the efforts to weigh the can each time refuelling :) 

boodle said:

Can using it with Temperature/density sensors help giving more accurate levels?

Likely not ... the reason purely boils down to the design of the sensor which is a windmill vane sitting in the flow. At certain rates, flow will just flow past the mill because the casing cannot be made to match very closely otherwise it could get stuck or wear out. This is the price of cheap sensors and why such sensors aren't used in actual practice where you want an accurate value of total amount as that is the time-integral of flow.

There are calorimetric flow sensors but they're not all that much more accurate either and need to be calibrated to the heat-flow properties of the fluid. Those often run $200+ for the sensor, but it has no moving parts, so the accuracy shouldn't drift over time as much as these mechanical sensors.

Calibration of fuel pumps as required by government where I am require the use of a "reference standard measure" - i.e. a cylinder of known calibrated volume. Notice they don't measure using a "flow sensor" because what you really care about is volume, not flow rate.

boodle said:

No one really wants to put the efforts to weigh the can each time refuelling :) 

Then just buy cans of a fixed size, fill them from empty to full. I mean, these are cheap and safe options which require no major work ... nor electricity, nor complex moving parts.

- Gough

@Gough, Accuracy +-1% is what I'm aiming.  To start with, I agree, I'm using low cost flow rate sensor but how true it is that the overall Turbine flow rate sensors have accuracy of more than +-10% as claimed by many above. If so, how come devices like this or this are commercially viable? All i'm trying to do is make the product cheaper & for end users which makes it a buyable product rather than fitting it in the trigger nozzle. 

I've already told you. I've used these sensors in various versions - brass, plastic, clear plastic ... they're all not very accurate especially in case of varying flow rates or flow rates that include rates below the minimum that the sensor is capable of. I used them in a number of applications including dispensing water for developing countries - we couldn't get it to accurately fill a 600mL bottle even with a consistent flow rate - even if you do calibrate it once and get it right, it doesn't take too long for it to drift away from the calibrated value due to changes in temperature, or movement of the sensor, or wear, or biofilms in the water, etc. Similarly, I have wanted for better, cheaper sensors for a long time ... but to no avail.

Devices like that aren't really viable ... I don't see any reputable company ever selling them. I don't see anyone who cares ever using them either. The fact they need to be calibrated first and only have 0.5% repeatability (i.e. discounting accuracy, the relative error is 0.5%) is pretty shocking to me.

You need to think carefully about the basics of metrology - as a rule, you can't use a clock of the same sort to check if a clock is accurate. You need something with more accuracy - usually 10x to 100x more accuracy to verify. Otherwise, all you're doing is measuring with very little certainty as the errors can go both ways for both devices in a nearly equal manner.

Why are you asking questions of people over and over when you could buy 20 of them and test them out yourself with your own code and a calibrated pumping system to see what the errors are like yourself? Do some of your own science - don't expect people to have all the answers you seek!

- Gough

But since you asked - this is probably the closest thing to what you're looking for that I can find on the open market with a remote display:

https://www.digiten.shop/products/digiten-lcd-flow-control-meter-g1-2-fuel-oil-gasoline-diesel-milk-gear-flow-counter

Claims a 0.5% accuracy just like your Chinese stuff because it's likely built around the same thing.

I have some of the 2%-rated Digiten stuff that is the same as the YF-201C /YF-S201 and similar Chinese flow sensors ... I measured the accuracy in a university lab to about 10-12% under a fixed flow rate, which is why i wouldn't trust the claimed accuracy regardless. That being said, you might get lucky ...

Otherwise, you'll just have to build your own as I previously explained.

- Gough