Yep, drag racing is very humbling. I can think of some things you can improve. If you are doing a wet burnout, you need to burn 2nd not 1st to get enough temp in the 4 seconds allowed. Obviously, the wing is probably robbing a few mph there. A full tank is what I always run, extra weight in the trunk will help too, don't go with 1/2 a tank unless you want to spin. Video looked pretty fun.

 

I'll take whatever advice you have haha. I actually watched your video on burnouts and realized why you mentioned using 2nd shortly after I pegged redline in 1st. I assume more wheel speed also means more heat.

Interesting point on the full tank. I figured less weight would be better but I hadn't considered weight over the rear axle. I probably need to launch 1-2000 RPM higher to actually get the tires spinning.

Firebird has 100 MPH and 110 MPH stickers depending on the speeds you hit, so the wing will probably come off next time, because I like stickers.

Let me know if you ever find yourself up north.

 

That's pretty typical of a Miata build, except for SofH of course. I have 330hp dyno'd HP, I've been 3-4 times, and was super happy to hit 12s. 12.971 to be exact. The rest is tires, gearing, suspension, and a lotta practice. Those first three I'm not too keen on changing just for 11 seconds of racing, so I'm happy where I ended up.

 

108mph is def solid 12s territory if you can figure out the rest of the setup and driving. Road race setup miatas are pretty bad for drag racing though, honestly not bad!

 

I'll hopefully go another couple times this season and break into the 12's. If nothing else I want that 110 MPH sticker.

Pretty happy with some progress I made tonight. A while back I grabbed a cheap (around $20) ESP32 board with a display attached to it. The idea was to make a gauge similar to the CanChecked or Link Can gauges. I pulled some libraries, wrote a few classes and such, but after wiring in a buck converter and a CAN transceiver I never had much success. I think it was the classic mentality of "do all the prototyping and development, and then test the final product".

So today I ripped things apart and got them back on the breadboard, simplified the code and plugged into one of the CAN headers on the Link. Sure enough, with a bit of tinkering I was able to get some values popping up on the screen.





My older code, which was just spoofing sensor values, was much cleaner. I also had some red, yellow and green LED's set up to indicate warnings and errors. Regardless, I'm pretty happy to have a value on the display. Assuming everything is working correctly, I was able to pull IAT, CLT and Oil temp tonight. I wasn't able to get fuel level.

At this point I should be able to slowly add my existing code back in and figure out where the issues are. Eventually, I'd like to 3d print an enclosure for this and mount it on the steering column cover. With the ESP32, CAN transceiver, buck converter and a 3D printed housing, I think this will end up costing around $50. Not bad when a CANChecked is around $300?

ETA - I already am able to monitor most of this using CAN output on the Link to a wireless OBD2 dongle and my phone as a display, but I'd rather have a solution that's integrated into the car and doesn't require my phone being mounted to the dash and loading up an app. Even with a good suction cup mount, I worry about it falling off at an inopportune moment.

 

Encouraging.

It'll need a housing, more permanent wiring, and I might throw those LEDs back on, but it's actually reading CLT data from the ECU (I believe IAT as well). There's a small button on the board that I have setup to change which sensor it reads from and clear peak values. This is also drawing power from the car's CAN port, so no laptop or external power is required.

I just realized the peak value isn't storing properly, but that's likely just a quick update to the code. At some point I'll also rearrange the display values to make them as visible as possible. The actual value read is a bit off as well, but I need to look into what the ECU sends out (F vs C for example). It's in the ballpark though and the value did start to tick up when the car was running for a minute.

 

Late to the party but glad to see the fans might be a kick in the right direction for your CLTs.

The drag racing actually looks fun as hell haha. I haven't flat shifted once since putting the turbo on my car but you're making me want to after watching that video and seeing your log lol.

I was always under the impression that not only the camber of a road race Miata, but also the lack of any anti-squat provisions that make these cars less than great at launching? Either way, would be fun to see you hit low-13's and/or bag that 110mph trap speed on the next one. In for more drag content for sure haha.

 

Still plenty of issues to work out, but progress is progress.

 

Everyone always thinks these cars can't launch, you maybe on to something with the alignment though. However, here are my 2 quickest time slips, I let off 9mph in Eureka too. On crappy continentals.

 

When I was sending data to an OLED screen it seemed like it caused a delay in receiving CAN messages. Many of the newer microcontroller boards have two processors, so I would use the first processor listen to CAN messages and the second processor to drive the display. I like the Adafruit RP2040 CAN feathers. I am using the Ada RP2040 CAN feather to drive shift lights on my Ecotec Miata. I also used to run Microsquirt, so I created an Microsquirt emulator to play with my telemetry system offline. Some stuff on Github if you want to look at it.

 

Thanks for the offer! I refactored most of the code today and ended up with a much better data rate. I also have the ESP requesting data for all sensors at the same time now, so the values for each sensor are updated as the data comes in. Honestly these little boards are pretty awesome for $20.

I think it's going to be plenty fast for my purposes. I mostly want to be able to keep an eye on coolant temps without needing to have my phone out and bluetooth'd to the OBD dongle.

I think I'm sending requests around every 50-100 MS right now and things are keeping up? I think it could go faster in theory, but I probably won't push it until I've got a pretty decent prototype in the car.

 

Here is some of what I did on github: https://github.com/rb92673/racecar
Maybe it will be helpful.

 

Thanks for sharing. Now that I think about it, I do probably need a button box for something

 

I need to refine the model, but progress is progress. I embedded some small magnet to hold the front cover on, but it's not secure enough for me to trust it for anything other than just driving around town. I mainly wanted to try that approach because I'd recently gotten the magnets, but I think they'll be recovered and used in another project.

I'll slap some tape on it for now and hopefully get it in the car tonight for a test drive.

 

Sorry for spamming all the updates but...

 

Maybe I'll just stick to drag racing... Shook the dust off this weekend out at the local "track". Fun to get back out and see everyone and give the car a shakedown. There weren't any real issues all day from a couple weird cuts like you'll see in that video. I think one was a false triggering of flatshift and the other was a weird lean spot or fuel starve? I didn't see any protections triggering on the second one so I'm not 100% sure what it was.

The camera setup is coming along. I think I'll pull the mic out of the engine bay next time and move it to the sun visor mount I printed. I think that'll get plenty of turbo noise without all the clicking and clacking from all the solenoids and injectors. The downside is that you'll have to listen to my commentary from inside the car, but I guess everyone will have to make do.

The OBD gauge worked great, better than I expected. The faceplate stayed secured with the magnets, which was one of my main concerns. So nice to be able to glance down after a run and see current and peak temps. I could already monitor live temps with my phone, but it's nice to be able to glance down without any extra steps/interfaces, just a little OLED with a number on it. Coolant temps peaked somewhere around 96*? I ended up changing the duty cycle ramp after the event to be a bit more aggressive. I figure if I'm above ~95 C (203 F) then the fans should be full on tilt. Regardless though, for not spraying the rad down at all or running the fans at 100% between and during runs, it seems like a step in the right direction. It was only like 22C (72 F) and sunny today, so not worse case, but I'm still optimistic, especially if I seal up all the holes in the shroud.

 

Ripping dude! Car sounds and looks great, and that video was sweet. I thought the audio was pretty damn good there.

 

I'm happy the fans seem to be working for you.
They really do pull an assload of air.

And for the audio... I get more from hearing the tires than I do the engine (at least in driving critique). Almost anywhere you can hear the "load" of the engine (ie, if you're in it, or partial, or lifted, etc) but with good tires the engine noise can overwhelm the tire chatter before the squall of beyond the limit.
That said, I can't really think of a location where a mic would be better to get the tire noise (especially the front) that isn't either in the bay, or stuck off the side/front of the car.

 

I needed to destress yesterday so I ran the car up the local ski hill. I did see my warning light pop on, which meant I hit 95 C, so I threw more duty cycle on the fan, and at a lower temp. I'll probably test that out today or this weekend with another cruise when it's warmer out. I also tested and found that the fan will run up to 94% or 96% duty cycle before it stops, so I bumped the max in my table from 85% to 90%.

I also added some tape and weather stripping to the top of the fan shroud to seal it up. There was a pretty sizable gap there. I figured I'd see if the fans were going to keep up without it, but at this point I'm done with experimenting and just want the car to stay cool.




I ended up finding some offcuts of this bubble top weather stripping in one of my bins. Clipped onto the fan shroud pretty easily and seems like it'll do a good job.




I've been driving myself insane trying to improve the housing for this CAN gauge. I swear I've modeled and scrapped 5-10 different designs at this point. Fusion does some things that I can't comprehend at times. It's a powerful tool, but man can it be frustrating. I finally got something I like better than the original design. Not perfect, but good enough for the time being. The lid is a bit more secure and it doesn't block as much of the speedo and warning lights.

also added Ethanol % and TPS %, fixed a bug that was causing a reset when clicking through the different sensor and added the ability to choose which sensor to display by default. Honestly, after having coolant temp displaying constantly it felt weird not being able to see where it was at when I was cruising around yesterday.

 

glad to see im not the only one using cut up softtop seals for sealing my rad 😅, big fan of the gauge.