Why I Built the BEAR Gauge After Ten Years at ICD TUNING


Why I Built the BEAR Gauge After Ten Years at ICD TUNING
By Ashwin Durai, The Madras Mechanic
I started servicing my own car because I was tired of taking it in, picking it up, and still having something to sort out. My thinking at the time was fairly simple. It’s my car. Let me try doing it myself.
Then I started working on cars for friends and family. That’s how ICD TUNING began in Chennai. I wanted the people who brought their cars to me to have the kind of ownership experience I’d wanted for myself.
Ten years later, I own a workshop full of diagnostic equipment, and I can still leave for a family trip with my scan tool sitting in the wrong car. That detail has a lot to do with why I wanted to build the BEAR Gauge.
What do you do when your mechanic is in another city?
At ICD, if a warning comes up, I can plug into the car, look at what it recorded and decide what to check. Put me on a highway with my family, a warning on the dash and my scanner at home, and my first question is the same as yours. How serious is this?
I’ve been in places where I didn’t know a technician or a workshop with the right equipment. Now imagine being the person who usually calls me. You ring from the side of the road. I ask what the car did. You tell me what you felt. Both of us would give quite a bit to see what the car saw in those few seconds.
For four or five years, I kept coming back to that conversation. What if the car could tell its driver what it had detected, in words they could understand? And when that driver called a mechanic, what if they could send the readings along with the story?
I knew what I wanted from the driver’s seat. I had very little idea who could actually build it.
The game developer who sent me an app a week later
I met Krishna through my brother Aakash. They knew each other through gaming. Krishna develops games and loves cars, which meant I could explain a problem the way I’d explain it on the workshop floor and he could picture what I meant.
The first thing we built together was a workshop app. Think about the perfectly reasonable phone call a customer makes to ask, “What’s happening with my car?” With a small team, somebody has to leave what they’re doing, find the technician, check where the job stands and call back. Multiply that by every car inside. We wanted the customer to be able to follow the progress and the team to have one place to update it. Krishna built that app first.
Then, during a casual conversation, I told him about the gauge. I described a driver who had never used a scan tool. If that driver had to search through menus and look up a fault code before the device was any use, we’d already made their job too hard. Could the app read the car through its diagnostic port, keep an eye on the available readings and explain a fault when it appeared?
Krishna’s answer was essentially, Anna, we can do that. About a week later he sent me a sample app. I was still thinking of this as a conversation we’d had. He was already telling me to open it.
And my first reaction, once I’d seen it work, was to tell him I’d forgotten to ask for something.
The most useful reading might be the one before the warning
Some of the hardest cars to diagnose are the ones that drive beautifully into the workshop. The owner felt a loss of power on a long highway drive. It happened under load, lasted a few seconds and went away. We can take that car out repeatedly and never quite recreate what happened to the person who drives it every day. Sometimes the eventual fix is simple. Getting the fault to show itself again is where the days go.
At ICD we already record data before and after tuning. Take boost, for example. I might want to see what boost the engine was requesting, what it actually made, and at what engine speed the two started to move apart. A single reading taken later in the workshop can’t show me that sequence. A log from the drive can.
That was my next question for Krishna. Could the gauge keep a running record of the engine parameters the car makes available? Could the software pick out the moment something started behaving unusually, even before it became obvious to the driver?
He put together a rough physical gauge with a display board, a GPS module, an accelerometer and a gyroscope. He got the diagnostic adapter and did the soldering himself. Then he showed me a recorded drive. The prototype could pick out an unusual section and suggest a few places to start checking.
That demo excited the mechanic in me. I know how long you can spend scrolling through a log looking for the few seconds that matter. Here was a piece of hardware Krishna had assembled, directing my attention to those seconds. I could open the trace, see the evidence and decide what to test.
The driver needs a different view of that same event. On the round display, you can follow readings such as coolant temperature, oil temperature or boost where the vehicle supports them. A reading that crosses its configured limit can take over the screen. The idea behind the AI-assisted explanation is to tell the driver what the car reported, why it deserves attention and when to involve a technician. That’s the information I’d want if my family were sitting beside me.
A drive record can be fair to the customer, too
After certain major repairs, we ask the owner to drive gently and bring the car back after an agreed distance so we can check it again. If something goes wrong in between, the first conversation can be uncomfortable for everybody. The workshop remembers its instructions. The owner remembers following them. Sometimes a component genuinely fails despite the owner doing everything we asked.
This is where the drive record became more interesting to me. How much time did the engine spend at full throttle? What temperature did it reach? When did the warning appear, and what happened afterwards? If the owner chooses to share that record, we can look at the sequence together. A driver who took care of the car deserves that evidence just as much as the person who did the repair.
We also explored what the driver could learn from a whole trip. The motion sensor can pick up hard acceleration and braking. Some cars provide signals such as indicator use. Krishna worked those ideas into a driving score. One hard stop tells you very little about a driver. A pattern across your regular journeys gives you something you might actually want to change.
A feature for anyone who’s ever abandoned a fuel log
I once downloaded an app to track my fuel expenses. I think I entered about ten fills. Then I forgot. I still wanted to know how much I was spending. I just didn’t want another bit of homework every time I stopped for petrol.
So we worked backwards from the fill-up. On a compatible car, the fuel-level reading rises. The app knows the fuel type and can use the city, date and local fuel price to estimate the cost. Add the distance covered and the owner has a useful expense history, even if they’re exactly as enthusiastic about manual entries as I am.
Planning a family trip brought up a different set of questions. Give me two routes and I’ll often take the slightly longer one with the better road. Around 150 or 200 kilometres in, somebody is going to want coffee or a clean restroom. And if we’re stopping for fuel, I’d like to know whether there’s a well-reviewed pump within the range my car actually gets.
Those are the details I gave Krishna for the trip planner. Route options that take road quality into account. Fuel stops based on the car’s average consumption. Places to take a break, informed by what other travellers say about the facilities. Speed-limit and camera alerts alongside the route. I’d been opening several apps to plan one drive. Getting those decisions into one place appealed to the person in me who has to get the family packed and out of the house.
Then I gave Krishna a Dragy, a VBOX and a RaceBox
This is the part where the racer in me got carried away. We were already putting GPS in the gauge. Naturally I wanted to know whether we could use it to time acceleration and braking on a closed course. A 0–100 km/h run, a 100–200 pull, a quarter mile. I gave Krishna the timing devices I knew so he could compare the results.
I expected him to fit the gauge and show me what its GPS could do. He came back with the phone.
He’d been working with the phone’s own location data and comparing it with the reference devices. What impressed me in those early comparisons was how closely the readings tracked, and then how much detail he’d built around the run. I wanted to see where the car gained speed, what the ambient temperature was and what the density altitude looked like that day. He understood why I’d ask for all three. That first demonstration is still one of my favourite parts of the project.
We call the timing feature Launch Control. The idea is for an enthusiast to be able to use it as a paid option on a suitably mounted phone, even before buying a gauge.
We’ve also been working on a leaderboard and a community where people can share runs and the interesting cars they spot. The gauge’s vehicle-health functions have a much wider audience. Somebody buying it for a family car should be able to pay for the features they’ll use. Our plan includes health monitoring, drive recording and fuel tracking with the gauge, with timing and the trip planner with road alerts as optional extras. We’re still working out the payment model for the services that carry recurring data costs.
Getting the everyday details right matters just as much. A Volkswagen and a Hyundai can expose different information. Our plan is to configure the gauge for the car’s vehicle identification number, or VIN, and support moving it to another compatible car through a paid software update. The readings we can show depend on what that particular vehicle provides.
What I want to happen when somebody calls for help
For all my excitement about timing a quarter mile, the feature I keep coming back to is the garage network.
Picture a driver away from Chennai with a car that starts but has a fault they’re worried about. Roadside assistance sends someone out with a battery and jump leads.
The driver already knew the engine would start. The technician has made a trip and still needs to work out what to bring back. That’s a frustrating afternoon for two people who were trying to solve the same problem.
What we’re planning is a network of garages across India with the technicians and equipment to handle the cars referred to them. The app would show nearby options. The owner could choose a garage and send it a unique access code. With that permission, the technician could review the available diagnostic data before setting out. Maybe the car needs recovery. Maybe the technician can take the right tools on the first visit. Either way, that first phone call becomes a much better conversation.
My partners asked a fair question. What happens if we give owners enough information to sort out some problems themselves? My answer was simple. If they can sort it out, let them. I’ve had cars sit in my workshop for weeks while we chased an intermittent fault. Imagine four cars occupying four bays for ten days. Now imagine being able to identify the checks sooner, order the right part and give even one of those owners their car back days earlier. I know which workshop I’d rather run.
There’s one more place I’d like this journey to go. We see people buy used cars, particularly European cars, and discover only afterwards how much work the previous owner put off. It bothers me to be the person who has to give them that news. We’re still figuring out how service records and vehicle history could help a buyer ask better questions before money changes hands. I’d love for somebody who looked after their car for years to be able to show the next owner what that care was worth.
Ten years ago I could do something about one car, so I picked up the tools and serviced it. Today I get just as much satisfaction from the idea of a technician I’ve never met, in a town I’ve never visited, opening a drive log with the owner’s permission and saying, I can see what happened. Here’s what I’d like to check first. If something we build here in Chennai can make that conversation possible, that’s a tenth anniversary worth celebrating.
The BEAR Gauge is taking interest registrations at an early-bird price of ₹14,999 + GST. You can check compatibility for your car and register on the BEAR Optimizations page.
Yours truly,
The Madras Mechanic
Your BS Filter for Car Myths.




Comments