The Madras Mechanic

Hi, I’m Ashwin Durai aka The Madras Mechanic.
I’m the founder and CEO of ICD TUNING, a car enthusiast, builder, and someone who believes in keeping things real.

[The Madras Mechanic @ashwindurai_]

I think people are missing something very important.

I’m a workshop owner. I tune cars for a living.

Do you realise how badly I have FAILED as a businessman during this E20 panic?

This was my moment. 😭

I should’ve made a reel with scary music, red flashing text and a fuel pump on the floor.

“⚠️ E20 IS SILENTLY KILLING YOUR ENGINE ⚠️”

Then sell you my revolutionary E20 GUARD ULTRA PRO MAX™ additive.

₹2,999 per bottle.

“Developed after 17 years of research.”

Research location: my office toilet.

Then launch an E10 TO E20 CONVERSION TUNE™.

“Your ECU doesn’t understand E20, bro.”

₹25,000.

Laptop connected.

Progress bar.

Three aggressive keyboard clicks.

Congratulations. Your car is now E20 compatible. Yayyy!

Customer comes in with a misfire?

E20.

Don’t check the vacuum leak.

Don’t smoke-test the intake.

Don’t look at fuel trims.

Just stare seriously at the customer and say:

“Sir… ethanol.”

Change the fuel pump.

Clean the injectors.

Replace the fuel lines.

₹87,450 please.

Actual problem?

₹500 vacuum leak.

But unfortunately, I’m not intelligent enough to build a business like this.

Not cunning enough either.

My parents raised me for a different world. Terrible parenting, really.

I probably could’ve made far more money selling E20 fear than the imaginary money the government apparently pays me to “promote E20.” 😂

And this is EXACTLY what pisses me off.

People with GENUINE vehicle problems are walking into workshops already convinced E20 destroyed their car.

Misfire? E20.

Fuel pump? E20.

Injector? E20.

Bad mileage? E20.

Bike parked for eight months with contaminated fuel and a rusted tank?

E20 BROOOOOO.

When you make one thing the villain for EVERY failure, nobody investigates the actual cause anymore.

I DO NOT support how the government rolled out E20.

I DO NOT support the lack of clarity.

I DO NOT support treating consumers like beta testers without proper communication.

I’ve said this repeatedly.

But apparently nuance is a paid government feature.

The funniest part?

When it comes to ethanol, vehicles may actually be the LEAST interesting part of the argument.

There are bigger questions.

Water usage.

Agriculture.

Crop allocation.

Economics.

Lifecycle emissions.

Policy.

Infrastructure.

Those arguments actually have weight.

But no.

“MY 14-YEAR-OLD BIKE DIDN’T START THIS MORNING. ETHANOL DESTROYED INDIA.”

Question E20.

Question the government.

Question the rollout.

But occasionally…

maybe…

just for fun…

question the guy selling you the cure after convincing you that you’re dying.

#e20 #saveindia

[The Madras Mechanic @ashwindurai_]

People really don't understand how engineering tolerance works.
A manufacturer designing a machine expected to survive 10–15 years does not engineer every component to operate exactly on a razor-thin line.
Because the real world is not a laboratory.

The engineer knows the car will be driven by a 70-year-old grandfather and a 20-year-old enthusiast. It will sit in traffic at 45°C, climb a mountain fully loaded, cold-start after weeks of sitting, be driven gently, be driven aggressively, get perfect maintenance, get delayed maintenance and sometimes be filled with fuel of inconsistent quality.

Engineering has to account for variation.

That's why there are design margins, control tolerances, material margins, safety factors, adaptation strategies and validation boundaries.

And if there is one country that has spent decades unknowingly conducting real-world experiments on engineering tolerance, it's India. 😂

Five people and luggage in a hatchback designed to comfortably carry four? Done.

Family vacation? Seven people, twelve bags, one pressure cooker and half the house inside the car. The suspension is sitting on its bump stops, but somehow we're still climbing Ooty. 😂

Washing machine rated for 7 kg? We don't weigh clothes. We push the door until it closes.

A truck rated for 10 tonnes? Somewhere in India, somebody has looked at that number and interpreted it as a suggestion.

Fuel gauge reaches reserve? “Bro, another 40 km will come.”

Oil change at 10,000 km? “Sir, 17,000 only. I was busy.”

Tyre pressure? Checked when the tyre visibly starts looking depressed.

We have abused the crap out of engineering tolerance in almost every aspect of our lives, and engineering has quietly saved us more times than we realise.

But suddenly, with E20, people imagine that an E10-compatible vehicle has a microscopic switch inside the fuel system.
10.0% ethanol: 😊
10.1% ethanol: 💀

That's not how engineering works.

Now, don't twist this.
Tolerance does not mean infinite tolerance.
A 10-tonne-rated truck carrying 15 tonnes and surviving does not make it a 15-tonne truck. A washing machine surviving repeated overload does not mean overloading is recommended. And an E10-certified vehicle tolerating E15 or E20 does not magically make it officially E20-compatible.
Certification, recommendation and actual failure threshold are three different things.
This is the distinction people keep missing.

The manufacturer gives you an approved operating boundary. Engineering tolerance gives the system some ability to survive real-world deviation from that boundary.
How much deviation? That depends on the component, material, calibration, operating conditions and duration of exposure.

But the idea that every machine instantly fails the moment you cross its declared specification is simply not how robust engineering works.
Engineering tolerance is probably the most abused feature in India. We have trusted it unknowingly for decades.
E20 didn't suddenly delete it.

[The Madras Mechanic @ashwindurai_]

@iamskg69 @jsensarma @Mr__Quickdraw I don’t support the fact that that e20 is the villain of every failure! I support the science!

[The Madras Mechanic @ashwindurai_]

@prithwiray @jsensarma @Mr__Quickdraw Yes true, but i have to add this, unlike you and me who goes WOT at any given opportunity, most of these cars have never seen 100% throttle in their lives! Pump duty was never the problem. Even if it was, it just moved the timeline a little early not a catastrophe!

[The Madras Mechanic @ashwindurai_]

@prithwiray @jsensarma @Mr__Quickdraw But e20 makes the overall octane rating 97 right. So the bump in DC isn’t as much to alter trims!

[The Madras Mechanic @ashwindurai_]

@prithwiray @jsensarma @Mr__Quickdraw Agreed! People claiming brand new cars failing because of 10% additional duty cycle is absurd! And most of them have never seen full load!

[The Madras Mechanic @ashwindurai_]

And you still haven’t understood my point. I never said octane affects the fuel pump. The octane example was about engineering tolerance.

Your car says 95 RON minimum. Put 91 RON in it and, in many modern applications, it doesn’t instantly knock itself to death. The ECU has knock detection, ignition correction and operating tolerances that may allow the vehicle to continue running, albeit potentially with reduced performance or other compromises.

So the number printed by the manufacturer is not automatically a physical cliff where the system dies one unit beyond it.

That was my point.

Yet when the same principle is discussed with E10 and E20, suddenly people assume an E10-labelled fuel system sees 20% ethanol and immediately “barfs”.

I’m not saying octane and ethanol compatibility are the same engineering problem. They aren’t. I’m using one to explain the concept of engineering tolerance.

If engineering margins and control strategies can allow a 95 RON-rated vehicle to operate on lower-octane fuel without immediately knocking itself to death, why is it apparently impossible to even consider that an E10-labelled fuel system may also have material and operational tolerance beyond its certified blend?

Especially when comparative E10/E20 pump durability research exists, and my 10+ years of workshop experience dealing with these systems every day does not support diagnosing E20 from a picture of a failed pump.

E10 is a declared compatibility limit. It is not proof that E20 is the failure threshold.

That’s the point you keep arguing around.

[The Madras Mechanic @ashwindurai_]

Here’s my favourite E20 irony.

E20 may have EXPOSED your poor fuel storage…

And then taken the blame for it. 😂

“Sir, we found water in the fuel. E20 absorbs water.”

Yes.

Congratulations.

You discovered ethanol chemistry.

Now let’s finish the chapter.

Ethanol has an affinity for water.

And because ethanol can interact with both water and the hydrocarbon fuel mixture, ethanol-blended petrol can hold a certain amount of water in the fuel phase before phase separation occurs.

Here’s the inconvenient part:

MORE ETHANOL CAN MEAN GREATER WATER TOLERANCE BEFORE PHASE SEPARATION.

E0?

Water and petrol are basically that divorced couple who refuse to sit at the same table.

Introduce enough water and it separates.

Ethanol walks in and says:

“Fine. I’ll talk to both of you.”

E10 can tolerate more water in the fuel phase than straight petrol.

E20 can generally tolerate MORE than E10 before phase separation, under comparable conditions.

So when people say:

“E20 attracted water and contaminated my fuel!”

My next question is:

WHERE DID THE WATER COME FROM?

Did E20 manufacture H₂O inside your tank?

Did two ethanol molecules get bored at night and give birth to a water molecule?

NO.

The water had to ENTER the fuel system.

Poor underground tank management.

Water ingress.

Condensation over storage cycles.

Contaminated transport or handling.

Poorly maintained storage infrastructure.

Or fuel sitting under unsuitable storage conditions.

Ethanol didn’t create the water.

It interacted with water that was ALREADY INTRODUCED INTO THE SYSTEM.

And here’s where things get interesting.

Ethanol also has a cleaning/solvency effect.

Years of deposits, gums and contamination sitting happily inside old storage infrastructure?

Ethanol-blended fuel can disturb and mobilise some of that material.

Filters clog.

Sediment appears.

Contamination gets noticed.

And everyone turns around and says:

“LOOK WHAT E20 CREATED!”

Created?

Or exposed?

Imagine not cleaning your house for 15 years.

Then someone switches on the lights.

“THIS LIGHT HAS FILLED MY HOUSE WITH DUST.”

No, brother.

THE DUST WAS ALREADY THERE.

YOU CAN JUST SEE IT NOW.

Can ethanol-blended fuel create storage and handling challenges?

YES.

Can excessive water cause phase separation?

YES.

Can phase-separated or contaminated fuel damage a fuel system?

ABSOLUTELY.

But if water entered your fuel…

The engineering question is:

HOW DID THE WATER GET THERE?

WHERE WAS THE FUEL STORED?

HOW WAS IT TRANSPORTED?

WAS THE STORAGE TANK MAINTAINED?

WAS WATER BOTTOMING CHECKED?

WAS THE FUEL SAMPLE TESTED?

Instead, we skipped the entire fuel supply chain and arrested E20 at the crime scene.

E20 didn’t get a lawyer.

Didn’t get a trial.

Didn’t even get CCTV footage checked.

CAUSE OF DEATH:

ETHANOL. 👍

Maybe E20 didn’t create every storage problem.

Maybe…

It exposed how badly some fuel was being stored and handled.

And unfortunately for E20…

THE MESSENGER ARRIVED JUST IN TIME TO TAKE THE BLAME FOR THE MESS.

[The Madras Mechanic @ashwindurai_]

Now apparently every low-pressure fuel pump failure is also caused by E20.

Fantastic.

A low-pressure fuel pump has one fairly unglamorous job:

MOVE FUEL FROM THE TANK, THROUGH THE LOW-PRESSURE CIRCUIT, TOWARDS THE HIGH-PRESSURE SYSTEM.

Typically at a few bar of pressure, depending on the vehicle.

It is not sitting inside the tank doing chemical analysis and screaming:

“OH MY GOD 20% ETHANOL 😭”

Yes, fuel chemistry matters.

Yes, pump materials, lubrication, seals, electrical design and flow demand matter.

But a failed low-pressure fuel pump is NOT a forensic report that says:

CAUSE OF DEATH: E20.

And then comes my favourite diagnosis:

“Sir, contaminated fuel entered. E20 destroyed the pump.”

Okay.

What was the contamination?

How much water was present?

Was the fuel sampled?

Was phase separation confirmed?

Was the tank inspected for sediment or corrosion?

Was low-side fuel pressure logged?

Was pump current draw measured?

Was the filter inspected?

Was the pump controller checked?

No?

Excellent.

E20 IT IS.

Here’s another small inconvenience called ENGINEERING.

If enough water or badly contaminated fuel enters the system to seriously disrupt combustion, the engine has a problem much earlier.

Because unfortunately…

WATER DOESN’T BURN.

Very inconsiderate of it.

When enough water reaches the combustion chamber, it contributes essentially nothing to the chemical energy release required to produce torque.

Misfires begin.

Combustion becomes unstable.

Rail-pressure or mixture-related faults may appear depending on the system.

The engine loses power.

And with severe contamination, it may stall or refuse to run.

On systems equipped with water detection, you may also get a water-in-fuel warning.

In other words, the car can become VERY unhappy before your mechanic finishes his speech about ethanol melting the entire fuel system.

Can water damage pumps and fuel systems?

YES.

Can ethanol compatibility matter to fuel-system components?

YES.

Can a low-pressure fuel pump fail?

OBVIOUSLY.

But if your entire failure analysis is:

PUMP FAILED + E20 EXISTS = E20 DESTROYED PUMP

Congratulations.

You haven’t diagnosed a car.

You’ve solved a WhatsApp forward.

SHOW ME THE FAILED COMPONENT.

SHOW ME THE FUEL SAMPLE.

SHOW ME THE CONTAMINATION.

SHOW ME THE PRESSURE DATA.

SHOW ME THE CURRENT DRAW.

SHOW ME THE ACTUAL FAILURE MODE.

Until then…

“E20 destroyed the fuel pump” is a conclusion.

NOT A DIAGNOSIS.

[The Madras Mechanic @ashwindurai_]

I answered an engineering question about the fuel. Now answer mine: what is the evidence that E20 caused the pump failure?
I’ve replaced hundreds of failed fuel pumps over 10+ years, long before E20 panic.
A failed pump proves the pump failed. It doesn’t prove E20 caused it. Show the failure mechanism.
Otherwise, you’re not diagnosing. You’re assuming.

[The Madras Mechanic @ashwindurai_]

People are posting pictures of their fuel caps saying:

“E5 / E10 ONLY. SEE! PROOF! E20 WILL DESTROY MY ENGINE!”

Interesting.

Look at the same fuel cap.

It also says:

95 RON MINIMUM.

Now be honest.

How many times did you fill 91 RON fuel because that was what was available?

Did the engine explode?

Did the fuel pump melt?

Did the injectors resign?

Did the piston come out through the bonnet holding a resignation letter?

No.

The ECU detected combustion behaviour, adapted within its control strategy and, depending on the engine and conditions, you may have lost performance or efficiency.

A specification on a fuel cap is important.

But the moment you see “E10”, you suddenly interpret the label as:

E10 = PERFECTLY SAFE

E10.1 = ENGINE DEATH

That’s not how engineering works.

Compatibility is not a magical cliff where the car is perfectly healthy at 9.9% ethanol and mechanically disintegrates at 10.1%.

There are calibration limits.

Material limits.

Fuel-system limits.

Operating conditions.

Engineering tolerances.

And TIME.

Yes, E20 can have implications in a vehicle not certified for E20.

I’ve never denied that.

But posting a fuel-cap photo is not failure analysis.

And “MY CAP SAYS E10” is not proof that E20 caused your misfire, fuel-pump failure, injector problem, divorce or male-pattern baldness.

Read the label.

Respect the specification.

But please…

STOP USING A FUEL CAP AS A POST-MORTEM REPORT.

[The Madras Mechanic @ashwindurai_]

@jsensarma @Mr__Quickdraw Octane rating controls knock. Higher octane better knock resistance. That doesn’t mean if you use less octane fuel the engine explodes. The ecu adapts and runs accordingly. Same goes with e10-20. The ecu adapts and runs accordingly.

[The Madras Mechanic @ashwindurai_]

@sajaniaf @AudiIN @AudiOfficial The cap also says 95ron minimum. Were you using 95 ron before e20 rollout? And did the engine implode when you filled regular fuel which was 91ron? Genuinely curious.

[The Madras Mechanic @ashwindurai_]

@Mr__Quickdraw If evidence is presented and someone still chooses outrage over understanding it, yes. Car enthusiasm isn't an exemption from stupidity.

[The Madras Mechanic @ashwindurai_]

My cars have been equipped with ethanol content sensors since 2021. I have personally logged E10–E20 blends across different parts of the country, even back in 2022.

Yet the stories of “massive mileage drops” suddenly exploded only after E20 became national news.

Here's the problem.

Stoichiometry doesn't watch the news.

If ethanol content changes, the required air–fuel ratio changes immediately. The lower volumetric energy content of a higher ethanol blend exists from the moment you burn that fuel. Lambda feedback and fuel trims respond accordingly.

That effect doesn't wait for a government announcement or an Instagram reel to become active.

Wear and tear? Sure. That's a long-term discussion.

Material degradation, corrosion, seals, hoses and component ageing can take time to show symptoms. I'm not dismissing that discussion.

But stoichiometry is not a long-term phenomenon.

If E10–E20 was already entering our tanks and causing the kind of massive mileage losses now being claimed, the intrinsic fuel-consumption penalty associated with that ethanol content existed the moment the fuel was burned.

Stoichiometry doesn't wait for the news. Chemistry doesn't wait for a government announcement. And your ECU certainly doesn't read Instagram.

Now, before someone conveniently twists this into “he is defending the government”:

No. I am not.

I am not defending the way E20 was rolled out. I am not defending inadequate consumer communication. And I am certainly not supporting the idea of changing the fuel landscape while leaving millions of vehicle owners confused about what is going into their tanks.

My point is simple: many of the claims being made today are wildly exaggerated, both about mileage and vehicle failures.

Fuel pump failed? E20.

Injector failed? E20.

A 15-year-old hose cracked? E20.

Check-engine light? E20.

Poor mileage in bumper-to-bumper traffic? Obviously E20.

At this point, if your power window stops working, someone will probably ask you which petrol pump you filled from. I am seeing people diagnosing ECU failure on Instagram comments and very well linking it to E20.

A vehicle failing while running E20 is not automatically evidence that E20 caused the failure.

Correlation is not causation.

And another major problem is that many people are confusing contamination with compatibility.

Water ingress, dirty storage tanks, sediment, poor fuel handling and compromised fuel are contamination problems. You cannot show me contaminated fuel, a rusted tank or a failed fuel pump and automatically conclude:

“See? Ethanol destroyed the car.”

You first have to establish what was actually in the fuel and identify the failure mechanism.

And compatibility itself is widely misunderstood.

Compatibility is not a magical engineering cliff where an E20-certified car survives and an E10-certified car explodes the moment ethanol content reaches 10.1%.

A manufacturer's compatibility declaration is a certification and liability boundary built around engineering validation.

The actual vehicle is designed with material, fuel-system, control-system and calibration tolerances. Whether those tolerances are sufficient for a particular ethanol concentration, exposure period and operating condition is an engineering question.

It is not something you establish with one un verified failed fuel pump and an angry Instagram reel.

Question the rollout. Question the government. Question the policy. Question the lack of communication. I have absolutely no problem with that.

But if we're going to question the engineering, bring data.

Because panic is not evidence. Correlation is not causation. And stoichiometry still doesn't read the news.

[The Madras Mechanic @ashwindurai_]

@amitkilhor @MercedesBenz I thought engineering was the problem. Turns out English isn't doing too well either. This country really needs saving from the government and people like you.

[The Madras Mechanic @ashwindurai_]

@NsNitinsinha That's exactly the question we should be asking!