March 16, 2025, 02:30:56 PM

Author Topic: Laminar's RX-8 EcoBoost swap thread  (Read 42896 times)

Offline Laminar

Re: Laminar's RX-8 EcoBoost swap thread
« Reply #240 on: November 18, 2024, 02:24:20 PM »
That all makes total sense. Here's something I'm not clear on - the absolute highest part of the engine bay is the coolant crossover from the rear of the engine to the front. I have a factory F-150 expansion tank. On the F-150, the expansion tank is the highest point in the system and the bleeds from the radiator and coolant crossover both go into the expansion tank.





Given my...unique...engine bay challenges, it will not be possible to have the expansion tank above the highest point in the engine. How much of a problem will this be?


This thread on LS1Tech has people arguing both ways - you can't fill a system higher than the level of the expansion tank because liquid will naturally find its level. But one guy claims you can fill directly into the radiator/hose and because the air exists in the expansion tank, coolant will only pressurize the air and not displace it.


My setup worked great on the stand, but I had the expansion tank sitting on top of everything. Maybe I should have tried dropping it down below the top of the engine to see what happened.


Someone set me straight.  :poke:

Offline digitalsolo

Re: Laminar's RX-8 EcoBoost swap thread
« Reply #241 on: November 18, 2024, 04:38:45 PM »
Is there nowhere you can stick a little air separator tank to bleed?

You can vacuum bleed with the radiator/bleed/etc. below the highest point.  It's just annoying.
Blake MF'ing McBride
1988 Mazda RX7 - Turbo LS1/T56/ProEFI/8.8/Not Slow...   sold.
1965 Mustang Coupe - TT Coyote, TR6060, modern brakes/suspension...
2007 Aston Martin V8 Vantage - Gen V LT4/TR6060, upper/lower pullies, headers, tune.
2021 Tesla Model 3 Performance - Stock...ish.

Offline Laminar

Re: Laminar's RX-8 EcoBoost swap thread
« Reply #242 on: November 18, 2024, 04:56:35 PM »
My coolant crossover is basically above the hoodline, and in the center of the bay where the hood is the highest:





Even if I put something back by the passenger side firewall it couldn't be above the level of that pipe.



Offline Laminar

Re: Laminar's RX-8 EcoBoost swap thread
« Reply #243 on: November 21, 2024, 11:29:44 AM »
One step forward, two steps back.


I went out last night to work on getting the fan secured to the radiator. For the top I drilled the shroud at the factory bolt locations, added a rubber grommet and washer, then screwed the shroud to the radiator.











And while working with the shroud I noticed the plastic at the bottom was broken. It may have come like that since I got it from a wrecked car in the junkyard, not sure. I whipped up some aluminum patches and JBWeld to glue everything back together. After applying the JBWeld and aluminum pieces, I went to clamp everything together.





Except when I installed the lower clamp, I accidentally caught the fan ring as I tightened it, which snapped off one of the fan blades.  :gruffy:


Looks like it's back to the junkyard to find another one of these fan assemblies.

Offline Laminar

Re: Laminar's RX-8 EcoBoost swap thread
« Reply #244 on: November 25, 2024, 06:56:16 AM »
We're so back.


Ran to the junkyard after work on Friday. The first three Fusions/MKZs I looked at ALL HAD THE FAN SWIPED. I don't know if people are finding out about these or what but I started getting concerned.


Luckily I still found a couple more and grabbed one.





Saturday was snow tire day, so I had to reassemble the front suspension, get the tires on, and clean out from under the car so I could roll it out of the way.










This is the first time I've had it on the ground with the new steering rack position.



On Sunday I went out and decided to trim the pinch weld back for AC compressor clearance.





So now the compressor will bolt up but it's juuuust touching the frame.








I can probably shorten the lower arms juuust a bit to give myself some frame clearance, but I can't lower it too much otherwise my belt will be too long.


With the compressor in place I threw the inlet piping in to see what that looked like. Looks like I can get 2" pipes up above the frame rails and then transition to something bigger.









Offline digitalsolo

Re: Laminar's RX-8 EcoBoost swap thread
« Reply #245 on: November 25, 2024, 11:01:36 AM »
Progress!

I use that same fan controller on my Mustang.  I was cooking the "Chrysler" controllers from Dorman, that controller has been great.
Blake MF'ing McBride
1988 Mazda RX7 - Turbo LS1/T56/ProEFI/8.8/Not Slow...   sold.
1965 Mustang Coupe - TT Coyote, TR6060, modern brakes/suspension...
2007 Aston Martin V8 Vantage - Gen V LT4/TR6060, upper/lower pullies, headers, tune.
2021 Tesla Model 3 Performance - Stock...ish.

Offline Laminar

Re: Laminar's RX-8 EcoBoost swap thread
« Reply #246 on: December 02, 2024, 06:53:06 AM »
Most of my spare time over break was spent at the computer instead of in the garage. I'd really like to set up a triple a-pillar gauge pod to show some vital values that the factory gauges don't show, namely:

MAP
AFR1
AFR2
Fuel pressure
Oil pressure
VVT?
VSS?
IAT
CLT
Fan
AFR target
Ignition
Knock
mpg inst
mpg avg

CANbus gauges exist, but they're all $350+
https://www.scienceofspeed.com/gaugeart-can-gauge.html

https://drifthq.com/products/link-link-can-gauge-oled-52mm-101-0226?variant=43683736617147


I feel like I could do that for less. I ordered a 2.1" round display and an ESP32 board from Adafruit.
https://learn.adafruit.com/adafruit-qualia-esp32-s3-for-rgb666-displays

And also an automotive CANbus development board from AutosportLabs.
https://www.autosportlabs.com/product/esp32-can-x2-dual-can-bus-automotive-grade-development-board/

Progress over the weekend was getting them both set up, testing some display ideas, and then getting the two talking to each other over wi-fi.








I have the ASL board counting from 0-100 and back and sending that number over to the Adafruit board, which displays it on a series of gauges. The Adafruit board doesn't like refreshing the entire display more than about once per second, but if I keep my elements small I get 20+fps. I'm running CircuitPython right now, but word is the Arduino IDE has better performance, so I might wipe all of my progress and start over there.


I did get another package in the mail so I mocked up my intake concept - 2" pipes out of turbo inlets size up to 2.75" pipes, which will both go into the airbox.





Obviously everything gets shortened up to the proper lengths but hopefully this reduces any inlet restrictions due to pipe sizing.

Offline digitalsolo

Re: Laminar's RX-8 EcoBoost swap thread
« Reply #247 on: December 02, 2024, 09:07:10 AM »
That gauge is cool!  I’m planning on something similar integrated where a little status screen was in the Aston Martin cluster.   If you haven’t, check out “upir” on YouTube for display ideas too, he has some great guides.
Blake MF'ing McBride
1988 Mazda RX7 - Turbo LS1/T56/ProEFI/8.8/Not Slow...   sold.
1965 Mustang Coupe - TT Coyote, TR6060, modern brakes/suspension...
2007 Aston Martin V8 Vantage - Gen V LT4/TR6060, upper/lower pullies, headers, tune.
2021 Tesla Model 3 Performance - Stock...ish.

Offline Laminar

Re: Laminar's RX-8 EcoBoost swap thread
« Reply #248 on: December 04, 2024, 08:11:03 AM »
Put together my first gauge layout, set up a little bar graph for the MAP value and got the numbers to change colors based on high/low value settings.





I'm trying to decide if I want to do a-pillar or above the radio stack. The S2 has its own radio/HVAC display much higher than the S1 so the gauges would stick up a bit higher. The a-pillar has an airbag in it so anything mounted there becomes a projectile. Maybe I'll set a gauge in each location and see what feels best.

Offline Laminar

Re: Laminar's RX-8 EcoBoost swap thread
« Reply #249 on: December 06, 2024, 08:53:07 AM »
Made my second screen layout. Since I'm going multi-fuel I'm forcing myself to get used to Lambda values instead of AFR. Also the ECU's map and fuel calculations are all natively in kPa. I could display psi but again I'm trying to force myself to get a good sense of kPa. Why make things easy?



Offline digitalsolo

Re: Laminar's RX-8 EcoBoost swap thread
« Reply #250 on: December 06, 2024, 10:20:45 AM »
Awesome progress.   I'm perpetually fighting to use lambda mentally.  I've been in kpa land forever, but fight to deal with absolute or atmo reference in that.   Too much mental math...

Are you still working in Python or did you go C++ (Arduino)?   I'd be interested in your code to cheat off of when I start making my gauge.  :D
Blake MF'ing McBride
1988 Mazda RX7 - Turbo LS1/T56/ProEFI/8.8/Not Slow...   sold.
1965 Mustang Coupe - TT Coyote, TR6060, modern brakes/suspension...
2007 Aston Martin V8 Vantage - Gen V LT4/TR6060, upper/lower pullies, headers, tune.
2021 Tesla Model 3 Performance - Stock...ish.

Offline Laminar

Re: Laminar's RX-8 EcoBoost swap thread
« Reply #251 on: December 06, 2024, 10:59:11 AM »
Still in CircuitPython. I think the arc-style gauges that required refreshing half of the entire display hurt my fps. Sticking with bar gauges that only affect a narrow band of pixels helps, and it seems that updating text values is more efficient. I just put in an order this morning for two more screen/ESP32 combos. The next big hurdle is going to be setting up the ECU on my test stand and getting the ASL ESP32 to decode its CANbus. The Link has a few styles of dash outputs but I'm not sure any of them will have everything I want - the normal stuff is easy, but I want to see flex %, per-cylinder knock retard, afr correction per bank, avg/instantaneous mpg, VVT angles, VSS sensors, etc. I can set up custom CAN outputs, but between the factory RX8 integration, gauges, ABS, plus my CAN switchboard, I'm running out of CAN channels. I'm sure I can make something work.


I'm saving all of my code iterations in a Google Docs...maybe I should actually use my GitHub account to track this stuff properly.

Offline Laminar

Re: Laminar's RX-8 EcoBoost swap thread
« Reply #252 on: December 07, 2024, 01:20:26 PM »
Here's the Github. Nothing's functional at this time, it's all just made up calcs to simulate to the screen.


https://github.com/MikaelVroom/ESP-Trigauges


Third screen:



I'll read the knock retard for all six cylinders and pick the largest retard value to display. Four big numbers plus the moving instantaneous mpg gauge is pushing the limit of what I can refresh in a timely manner. If I let it sit for a while, it inevitably fault due to an invalid buffer. There are commands to flush the buffer, which I'll need to work in periodically to prevent the buffer overrun.

Offline Laminar

Re: Laminar's RX-8 EcoBoost swap thread
« Reply #253 on: December 09, 2024, 08:04:37 AM »
Started on the CANbus. Got as far as getting the ECU powered from a 12V wall wart and hooking the CAN wires between the ECU and ESP. I set up a basic CAN receive routing and I'm actually getting messages!








Now I have to actually learn the protocol, how to format messages, and how to interpret them.

Offline digitalsolo

Re: Laminar's RX-8 EcoBoost swap thread
« Reply #254 on: December 09, 2024, 01:01:20 PM »
Are these just broadcast messages from OBDII protocol?  There's lots of docs around how to read those if so.   My code has some processing of acquired data from CAN for similar purposes:  https://github.com/analoghan/AstonMartinVH_DashCAN
Blake MF'ing McBride
1988 Mazda RX7 - Turbo LS1/T56/ProEFI/8.8/Not Slow...   sold.
1965 Mustang Coupe - TT Coyote, TR6060, modern brakes/suspension...
2007 Aston Martin V8 Vantage - Gen V LT4/TR6060, upper/lower pullies, headers, tune.
2021 Tesla Model 3 Performance - Stock...ish.