I've used an Arduno Uno for many years to provide temperature measurements. This required a laptop to record data, so not particularly useful for driving around.

After playing with some cardboard box prototypes, I finally finished the engine compartment wiring and added a prototype display to where the radio usually goes. This enables real-time, accurate monitoring of one or multiple temperature sensors, one or more bright visual alert bars, a vacuum sensor, data-logging, and whatever else I want to add. Here's the plastic and electrical tape prototype. I need to 3D print a radio-face sized housing once I figure out which OLED display to use that best matches the stock interior.

Here are the teaser photos.

 

First some background. I picked up an Arduino shield that supports four channels of thermocouples. This worked great, I was able to reliably sense and record (through a serial port monitor on a laptop) four channels of temperature data. Here are plots that I've posted before that pretty conclusively show that using the L82-style dual-snorkel CAI provides the engine with much cooler air than an open element filter. The first plot shows some hard driving on a warm summer day with an open element filter. The second plot shows a similar driving profile on a similar day, but with the L82 dual-snorkel CAI. Despite loosing the coolant-temp channel, the carb intake air temperature is greatly reduced.

 

That experiment convinced me that the L82 dual-snorkel CAI is the best solution for a stock-ish L48 engine. I have them in both cars, and I have modified them to remove the bellows and heat stove input, meaning both snorkels get unimpeded cold air all the time.

In my 80, I noticed that there was a large gap at the sides of the radiator, where the air conditioning condensor once was. So I made sure the thermocouple on the upper radiator hose stayed attached this time, and I ran two passes on an identical driving loop within few minutes of each other. The loop featured some city driving, and then a blast down the highway, ending in a slow crawl through town. At no time did the coolant temperature rise above 180 degrees F, and my electric fans never turned on (195 F on, 185 F off, both switched together on separate relays).

Other than the carb intake air temp, the plots show that the air bypassing the radiator didn't really matter for this drive.

 

Here are the thermocouple wires, the condensor-free air gap, and the foam-filled radiator surround for the experiment in Post 2.

 

So I noticed that it didn't matter what I did. If my car was moving, it was NOT overheating, or even coming close. I have a 180 degree Milodon high-flow thermostat. So I did a longer driving, starting with a warm engine in a nearby town, and going over what passes for a mountain in Virginia. While going uphill, the temp was about 160. Coasting down the backside, the temp dropped to 150 (!).

 

This is early Fall in Virginia, not summer in Florida. But I was still surprised by the temperatures. The upper radiator hose thermocouple was under several layers of electrical tape, but perhaps cold air was still blowing on it. The other air temperature readings were interesting, but didn't tell me anything useful. I can't dynamically adjust the tune of the engine to take advantage of cold intake air. And the data collection process was cumbersome, requiring a laptop to collect data from the Arduino's USB port. It makes it very handy to plot in Spyder, but impossible to use while driving.

The thermocouple shield on the Arduino took over most of the Arduino's functionality. I could add another sensor, perhaps, but not display anything. I needed an actual coolant temperature sensor, and I needed a real-time display. Fortunately, I had these items, I just needed to write a new set of code, and wire everything up.

The teaser images in the first post show the result of all of that. I used a Wells SU109 2-wire ECT to take temperature measurements, and I added a Wells SU105 MAP sensor from a 1982 Corvette. I spliced the appropriate connectors to 18 gauge GXL-rated wire and ran it to an Arduino fitted with an OSEPP LCD 16x2 display shield.

All of the HVAC components are currently removed from the car. While I hope to have heat again someday, I removed the heater core bypass components, plugged the return port in the water pump, and used the ouput port in the intake manifold to hold the ECT. I needed a 3/8" NPT female to 1/2" NPT male adaptor to make the transition. Since the ECT is a two-wire system, there was no problem using pipe tape. The image shows the ECT, the plug in the water pump, and the temperature switch used to trigger the dual-Spal fans (2 wires for two relays).

 

Lots of charts/graphs/explanations to help me understand

I need to watch youtube Arduino to learn how to program my fans per DBL TRBL.

 

I am playing with this as well doorgunner. Teaching myself to code again has been fun.
Have merged DBL TRBL's Mazda PWM code with his test code for the stock sender using a C6 controller along with other snippets I have found on the Arduino sites.
Not finished yet but seems to be able to read the sender voltage ok so far and convert it to a temperature.

Will post it when I am done.

 

Thanks for the reminder! Here's @DblTrbl's excellent thread. Hopefully this thread will end up being helpful to someone, too.
https://www.corvetteforum.com/forums...oft-start.html

Please post code, links to other projects, anything that can help! I wanted to get the backstory down, but the most useful thing I've done is to figure out how to use the 2-wire ECT with an Arduino. I need to write that up and post it.

 

I used a 2-wire Wells SU109, which was used in various GM vehicles from 1980 to 2014. It cross-references to a Delphi 12146312. Here's a link to the datasheet.

https://pe-ltd.com/assets/coolant_temp.pdf

I pulled numbers from the datasheet, centered at 180 degrees F, and used this website to get the β model coefficients.
Thermistor Calculator

This gave me β, T_0,and R_0, which I define in my Arduino code:

#define SU109_ECT 4117.31f,298.15f,2896.96f // B,T0,R0

I could probably tweak the model a bit, or use the Steinhart-Hart model and melt my Arduino, but this seems to work fine for a car that rarely sees more than 190 F.

I used this code as an example
https://playground.arduino.cc/ComponentLib/Thermistor2/

And wrote my own version of the function.

Example Function call:

ect1degF = Temperature(1, 3, SU109_ECT, 330.0f); //ECT1 is on analog pin 1

And the function

// Temperature function inputs
// 1. AnalogInputNumber - analog input to read from
// 2. OuputUnit - output in celsius, kelvin or fahrenheit
// 3. Thermistor B parameter - calculated
// 4. Manufacturer T0 parameter - calculated at 25 C (but in Kelvin)
// 5. Manufacturer R0 parameter - calculated (ohms)
// 6. Your balance resistor resistance in ohms

float Temperature(int AnalogInputNumber, int OutputUnit, float B, float T0, float R0, float R_Balance)
{
float R, T;

R = R_Balance * (1024.0f / float(analogRead(AnalogInputNumber)) - 1);

T = 1.0f / (1.0f / T0 + (1.0f / B) * log(R / R0));

// Default OutputUnit = Kelvin
// OutputUnit = 2, Celsius
// OutputUnit = 3, Farenheit

if (OutputUnit == 2) {
T -= 273.15f;
}
if (OutputUnit == 3) {
T = 9.0f * (T - 273.15f) / 5.0f + 32.0f;
}
return T;
}

 

I used a 330 Ω reference resistor, since that is close to what the ECT's resistance is at 180 F.

In this image, I unplugged the ECT, and have about a 125 Ω equivalent resistor plugged into the ECT connector, which the Arduino interprets as 234 F.



It doesn't come across well in this photo, but I am using an Adafruit Neopixel 8 element stick to digitally display temperature. I can pick the color and intensity (and could even change the brightness if the headlights are on or off, or from a brighness sensor on the dash). As the temperatures go above where my fans kick-in, the colors get more red, and much brighter. At full-brightness, an over-temp situation looks like a shift-light. You won't miss it, as you might with a gauge. I am also monitoring vacuum (here, the car is not running), but the Arduino could easily measure oil pressure, and have a similar bright display.

 

Very nice. Once you get the hang of it you'll be like a kid in a candy store, always wanting more.

Let us know which Oled you're leaning towards. I went with a 7" Nextion display cause it saved me lots of work and it's a touchscreen. Down side is it's not easy to read in direct sun.

 

Thanks!

I've reached the limit a few times with Arduino Unos, so I'm glad I'm not trying to use them for fuel injection or ignition timing. Rather than an all-in-one solution, I'll likely use separate ones for different major sets of instruments.

For example, it was all one of these Arduinos could do to calculate display real-time GPS speed. (My dining room table was moving at zero mph). The other one used most of its pins just talking to the LCD display, but perhaps it could handle a sensor or two. I don't know about its daylight visibility, though, hence the idea of using OLEDs.

 

I understand the using up the pins problem with some displays. Some displays use tons of pins while others use a serial approach that only uses 2 pins.

I like the idea of having separate processors do separate tasks, but I'm having a hard time understanding how you maxed out an Uno with just a gps and a display. Maybe the libraries are taking lots of resources or some other reasons.
The mega will give you more pins and more memory. But the mega speed is pretty much the same as an Uno. With careful design and execution, you would be amazed at how much one board can do.

If you do end up dividing the tasks among multiple boards, then I suggest using CANBUS to share the information.

 

That is some interesting work with the Arduino there Bikespace and DblTrbl!

I like the use of the Arduino for measuring and showing the active temperatures! Your intake air is cool and cooler is better!. Is there any way to help get ambient air into the intake, directly? That is what I like so much about the L88 Hoods that GM designed and built back in the 1960's. I have wanted to generate that type of data you are showing but with my L88 Intake setup. I want to see the air temperature as it enters the throttle body on my High Compression C3. Getting the coldest air possible was why they designed it the way they did as it is critical when racing a Big Hot engine.

I have been using a pair of thermo-couples on my Radiator IN and OUT to see just how efficient my BeCool Radiator really is using the EVAN's NPG in it. As soon as I am done with the Radiator I will focus on the temperature of the incoming combustion air. I will then move on to get the fuel as cool as I can before introducing it to the engine.

Being a Private Pilot and being aware of the "engine monitoring devices" they use in the aviation industry I have always wanted something like one of those watching over the Corvette's engine all the time. They are monitoring the vacuum, the temperatures, operating temperature and individual exhaust temperatures. I have headers now but am contemplating going back to the factory exhaust manifolds as they were a kind of a header in 1968 on the 427's. I will drill it out for a thermo-couple and have it Ceramic Coated with the Higher than Normal temperature Ceramic coatings. My Headman headers are over 25 years old and showing their age a bit.

The Holley Sniper System uses CANBUS and it is really a great way for things to communicate. I am sure that you know far more about it than I do but it is very handy and easy to setup. I am still learning....

My Son is a Computer Software Code writer and he absolutely LOVES the Arduinos and other PLC's. He writes software for them and has dozens doing little tasks for him all over his house. He can't understand why I don't use them as much, it happens to have something to do with I was taught COBOL and FORTRAN. These newer programming languages are much better and easier to use (IF they make sense to you).

Thanks for Educating Us!
Chris

 

Holley's CANBUS is proprietary. You can't connect with it unless you know their secret sauce. They want you to buy their interfaces.

You're dating yourself with the COBOL and FORTRAN. I'm one of those too, LOL. It will take your son 1 hour to get you started so that you can use a $3 arduino to log all your temperatures and pressures and control your fan(s).

 

I've played with CANBUS a bit. For what I want to do, dedicated small-controllers work fine, and don't need to communicate. I could have one that does temperatures and controls the fans (though my experiments show that I do not need to control the fans any better than a switch is already doing). Another one for HVAC that reads temperatures, and controls an electric valve on the heater core line, and a PWM fan controller.

I'll have to revisit the GPS issue again. I had thought that the bottleneck was actually the Arduino doing math. The GPS receiver can operate at 10 Hz, but connected to the Arduino I was lucky to get 1 Hz. Perhaps I have something wrong in my code, or it's actually the receiver that's the problem, or I'll just switch to a Raspberry Pi for that application. Having a 3-axis accelerometer connected would be pretty nice, if I can get back to a 10 Hz update rate.

I would love to see the data from your radiator. I don't know how easy it would be to repeat an experiment using 50/50 ethylene glycol/distilled water mix, but I would really like to see if there is a measurable difference.

As for the CAI question, in both cars I installed the L82 system as designed, and only modified the dual-snorkel assembly to remove restrictions, and block off extra ports. I did NOT seal the gaps where the plastic pieces join to go around the (nonexistant) AC compressor, for example, but that would certainly help. As it is, the system gets cold air from the front of the car, but it is above the radiator. It has a long, uninsulated, and slightly leaky path to follow to get to the carb. You can clearly see in the data when I had to stop at a stoplilght. The intake air temp creeps up, as does the engine compartment temp.

Using a better-sealed system, an insulated system, or a system like @pauldana sells, might help, as would plumbing the intake down low behind the grilles in the front bumper. A friend of mine totaled his C5 when he drove through a puddle with a similar CAI setup, so I won't be doing that.

If your son is an Arduino wiz already, give him a list of requirements, and you'll quickly end up with a system 10-times better than anything I can write.