4 replies to this topic

[Sciosophy]

Hi everyone,

 

I’m working on a small DIY project and I’m trying to figure out the most reliable way to measure the ethanol content of gasoline using a very small sample (ideally just a few drops).

 

Context:

 

In automotive tuning it’s common to run ethanol blends like E40–E50, but the ethanol content of pump “E85” can vary a lot depending on season (for example E60–E80). Because of that, people often measure the ethanol content before mixing fuels.

 

The common manual method uses a water separation test in a graduated tube, but I’m interested in building a small electronic handheld tester that could determine ethanol percentage from just a drop of fuel.

 

I’ve read that possible measurement principles could be:

 

• dielectric constant / capacitive sensing

 

• impedance or conductivity measurements

 

• density or refractive index

 

• optical or IR methods

 

My goal would be something like:

 

• handheld device

 

• a few drops of fuel as sample

 

• ethanol range roughly 0–85%

 

• accuracy within maybe ±2–3%

 

I’m curious from a chemistry or instrumentation perspective:

 

Which physical property would give the most stable measurement for ethanol in gasoline?

 

Would dielectric constant measurement be reliable enough given the additives in gasoline?

 

Are there known compact sensor approaches used in industry for ethanol/gasoline mixtures?

 

Any pointers to measurement techniques or sensors used for fuel analysis would be greatly appreciated.

 

Thanks!

Edited by Sciosophy, 03 March 2026 - 10:26 PM.

[Pilesar]

Perhaps some ideas from here: https://copilot.micr...dp2VTeJtp9UWcUn

[breizh]

Hi,

• Water Extraction Method (Field Test): A common, simple method involving adding a known amount of water to a fuel sample in a graduated cylinder. The ethanol binds with the water and separates from the gasoline; the increase in volume of the water layer indicates the percentage of ethanol.

This may be sufficient enough!

Breizh

[katmar]

Disclaimer - I have no experience with measuring ethanol in gasoline but I am familiar with measuring ethanol water mixtures in the beverage industry.

Your biggest problem is going to be in establishing a base line for the gasoline as the range of chemical species that can make up gasoline is very wide. From my limited experience with fuels, the "ethanol" that is used can also contain other alcohols which have similar fuel values but for which the other properties that you are considering can vary as well. This probably means that whichever method you finally decide to use, you are going to have to calibrate your instrument by making up a range of samples and measuring them individually to create a calibration curve for you particular circumstances.

In the beverage industry the most frequently used methods used (outside of specialist laboratories) are density and refractive index. The cheapest way of measuring density is with a hydrometer but this needs a similar sized sample to the water separation test which you are trying to avoid because of the sample volume. There are electronic density meters that vary in price from a few hundred dollars to tens of thousands of dollars. These meters use only a few millilitres of sample.

The densities of ethanol and gasoline do not differ as widely as ethanol and water, so you may have some accuracy issues with using density.

Refractive index might be better from an accuracy perspective, and certainly it requires much smaller samples - typically a few drops. Refractometers vary in price from very cheap for hand help analogue instruments, through to a few hundred dollars for portable electronic units, and right through to the tens of thousands of dollars for lab grade units. Here again you will need to generate your own calibration curves.

[MrShorty]

I think density as a property to get at this will be a non-starter, because you will need to know the density of the starting gasoline feedstock before the ethanol is added. Since this can vary, it will be difficult to impossible to determine ethanol concentration based solely on density.

While I didn't find a lot of technical information on the internet, a quick search suggests that the "flex fuel" sensors in your car are using an electrical property of some sort to measure EtOH content in real time. I'm wondering if an automotive sensor would be a good starting place for a project like this, since an automotive sensor must be able to reliably get EtOH content based solely on the properties of the fuel in the tank (and stored calibration curves).