forbiddenera

Yes that's what I am working off. If anyone else has any other schematics or info that would be great

Canadian cars don't use ALTC, I noticed our ECUs are missing q37. Adding it gave me the output, thanks to that schematic.
​​​
I picked up a 2nd flex fuel sensor form the JY today (first car in the yard this time..better luck than last time where I spent 3 hours one day and found 0 then did 3 hours of research and Vin checking of JY cars to find it the 2nd day).. so now I can install one in my car and have one to develop with.

I also grabbed some of the extra plugs I needed for my engine simulator.

I may consider including JY sensors with kits if people want but the sensor can be had for like 60 off rockauto anyway so not sure it'll be worth my time.

Once these microcontroller's arrive the kit will be put together. First ones should be ready in the next week or two. Again just providing them for conveniece, the code is open source and this is all based off off the shelf parts for now, unless I add a custom board which again is just for convenience for those who don't want to solder.


​​​​

forbiddenera

I would appreciate if anyone interested in ordering a kit could reply saying as much, so I can make sure I have parts enough for everyone. Of course again a reminder you can build it yourself and don't need to order anything from me.

As an aside, I am not sure about any rules pertaining to this but let me state that I am not really providing kits to make a profit - I am making all the information and code required for this open source so that anyone can build one, use it, improve on it, heck even sell it and profit themselves..but replies in this thread have mentioned people would be interested in sort of a premade kit so I'm happy to provide for convenience. Mods let me know if I need to do anything to stay cool.

Spaceturtle917

Would the screen even be needed? If you have Neptune you could have the voltage values converted to 0-100% values and the output into tuner view. At least I think you could but I have no experience with tuner view.

forbiddenera

You are 100% correct and that is why the screen is optional.

​​​​​

Spaceturtle917

I'm definetly interested though. I've been on ethanol with Neptune RTP and it's been a pain always having to fill up with ethanol and then the gas station changing the mixture slightly and having to tune.

Txdragon

that could be done with one of the Aux inputs on the board but, they're only 5V inputs and analog, if I recall correctly.

justYncredible

iTrader: (1)

i like this. i just got me a 3d printer so maybe, just maybe, i can design a box to place this thing. im a DIY person so i would attempt this

justYncredible

iTrader: (1)

screen for me, would be something i could mount under my stereo in my integra. like right under the tray near the lighter,.

forbiddenera

​​​​​​yes..Neptune, hondata, Honda tuning suite all support flex fuel with a 0-5v input

the issue is flex fuel sensors output a digital signal which must be converted to 0-5v, that's what this DIY is for. There are commercially available units but they're like $200 or so.

I grabbed a 2nd flex fuel sensor for testing. Interestingly it's like 3x the size as the fest one I got.



working on my ecu sim

microcontrollers haven't arrived yet.

if anyone wants to 3dprint an enclosure that would be cool.

justYncredible

iTrader: (1)

there might be stl files for the encoders already. let me know which one youre going to use and ill look it up on a few sites i use for files to print. i have a creality ender 3 3d printer.

forbiddenera

So I've been trying to decide how to proceed on this project for the last bit.

The way I see it, there's two audiences:
- those who don't mind soldering/programming an arduino
- those who don't mind wiring something into their car but want nothing to do with the above

The first audience, I don't think I really need to provide anything for. Take my code, upload it to an Arduino, change the pin output if needed depending on your Arduino. Wire in the sensor. 5v out to ECU through an RC filter.

In fact, I found out recently that one of the other developers of the HTS has already made a board that can take an Arduino and has the rc filter onboard. I don't see why this wouldn't work with my code if wanted. I believe he's asking $50 for the board or something, I have no idea whether that includes the microcontroller or not. If anyones interested I can try and get a link.

The second audience, I think I'm just going to make a board with a microcontroller on board. I suppose I will likely opensource the whole thing but will sell completed parts to those who want them.

So anyone wanting to build it themselves, the info is there, feel free to ask questions.

For those looking for a cheap controller you can wire in, stay tuned and I'll be releasing something you can either build yourself or buy from me.

328p dip chips are cheap and easy to solder for anyone wanting to build their own, so I think I will keep the whole thing thru-hole for easy assembly.

Working on a board layout now.

forbiddenera

DigiKey f'd up and sent the products to my old address..
They are resending..

tempted to hop on my bike and go look at my old place see if I can find it.

blew my engine so can't drive over

forbiddenera

The microcontrollers are here.

I'll put together the first few sample boards as soon as I can; who wants 'em?

In the middle of swapping my engine though, spun a bearing.

forbiddenera

haven't posted there in a while..don't have any D besides the one I was born with right now. Lol.

anyone interested in getting a prototype version, send me a PM for details.

I still need crystals. Was gonna go pick some.up but spun a bearing. Might order some from digikey but the spare engine is ready just need a new clutch.

quickhondaz

you have me thinking I really want one of the prototypes, but haven’t made up my mind on whether I want to pull the trigger on e85 just yet.
if I did decide to go ahead with e85 and did get the plug and play kit from you, what else will I need to do to my ecu? It is currently on Neptune with a Demon board. I was told by HRTUNING that I would need the necessary output/converter for Neptune to read it properly. I also have a tunerview that I will be using so I don’t think I’ll need a screen for it. And after reading all of the above info, I already have the flex fuel sensor saved in my cart on Amazon.

forbiddenera

This device would be the 'necessary converter'.

In order to get fuel temp on the tunerview you will need to use an extra input for that, but only if you want to see that data.

I've heard of issues with TV and demon, though I'm not sure if that's just the app or the hw device too. Basically the demon outputs at 921.6kbaud where most ecus output at ~9600 or ~38400.

Depending on your ECU and what you may or may not already have hooked up, you may need to modify one of your ECUs analog inputs to accept 0-5v as the inputs on the Demon are for logging only. You may also need to add a pin to your ecu plugs to be able to wire it in.

​​​​​Besides that you will just need a tune after that. Neptune supports flex fuel. Also, Honda Tuning Suite (the successor of eCtune) supports it and the Demon as well. I may try and see about using extra inputs on an emu (demon/tamer) for corrections but this may not be possible (without custom fw on emu) and if it is may still have reliability concerns but could be a nice feature, but that will be down the road.

My personal car should be up and running in the next week or so and I'll be doing the fuel pump which will make me comfortable running more than e10

​
​​​​

Chrisfrom1986

iTrader: (1)

I'd like one just to have for the future. I don't have any knowledge of arduinos or programming unfortunately.

forbiddenera

Once the prototypes are tested and validated than I will be making a ready to go version that simply needs to be wired into the car and selling it through my business. I don't want to break any rules here so I'm not really trying to promote that although anyone can PM me regarding it.

I'm how to get prototypes finished within the next 2 weeks and have them tested over the month to follow.

There are 2 maybe 3 spots open for.testers. You will.receive a discounted prototype ready to wire in the car as well as a discounted or potentially free version of the final product.

​​​​​Testers need not have any programming or arduino knowledge but it can't hurt. You will need to be able to.wire the unit into the car and tune the vehicle or have someone do this for you. I do need to set an expectation that the testers will actually test and not.leave it sit for 2 years while they finish their project as then everyone else will be waiting for testing to finish.
​​​​​​
the code and everything else will still be open source and available to anyone wanting to do this themselves.

Once my clutch arrives from.tdot performance (not.impressed - 2 business days and no response or tracking number, their email auto reply.says at LEAST a week to get an email.reply.from them due to pandemic which IMHO is bull) I will be putting my new motor in along with the flex fuel sensor and my controller. I can then test myself.

I am also.awaiting an engine sim from.Blundar to assist in testing..should be here shortly but he's been busy.
​​​​​

forbiddenera

So I've been kind of lazy on getting crystals; then I realized that the chips I was planning on using have a built-in clock; less parts.

Just need to pick out and order some kind of connector and 5v power. I have 3D printing access now as well as a lot of other things.

Who's interested in testing the first prototypes? I know I've kind of asked before, but, I'd like to ideally get immediate testing from someone who's ready to use one.

forbiddenera

Ported the code to the new chip tonight along with testing.

Datalogging directly from the device is possible; outputs 4 fields right now: ethanol content %, fuel temperature, sensor HZ, sensor Duty.

Ethanol & Fuel temp are exposed as analog outputs.

There will be some configurable settings on the system, not sure if all will be exposed to the user:
- minimum output voltage (0.5v default)
- maximum output voltage (4.5v default)
- default ethanol value (0% default) (what gets sent if no sensor or bad data)
- default temperature value (15c default) (what gets sent if no sensor or bad data)
- averaging depth / time (how long / how many samples to average for, protects against sensor blips and huge/fast changes, for example, if the device misses some data or gets a bogus value, it won't instantly send that to the ECU, causing your tune to jump)
- max deviation / time (protects against values changing too fast)
- eeprom startup function (save last good value to eeprom occasionally and use this on startup)
- eeprom startup function time
- perhaps definable warnings for the oled screen

The defaults should be fine for most but I plan on making a web-app that can change the settings (I'm also working on a web app that can datalog ECUs directly from the web browser over serial or usb on pc/mac/linux/android without plugins or downloads). You will likely need an FTDI cable/breakout/hulog type device for this, as it would increase the cost to include a built in USB connection. Same with direct datalogging, though you'll of course see it in your ECU's normal datalogging if you've hooked it up to (an) analog input(s) there.

Testing with the internal clock does seem to work but I think it will be more stable with an external clock, so I'm going to order some crystals and test, if there's a big improvement in stability than I will use a crystal.

I've begun building the first prototype PCB. I still have to solve power input from the car; you could power it off the ECU's 5v out (sensor out) but I don't trust people to do that / not connect it to 12v. A linear regulator won't do so well from 14-15v down to 5v, so will have to use a buck converter.

Still need to finalize what I'll use for connectors; the prototype(s) may just be hardwired. Screw terminals is an option perhaps. The OLED I will use a DIP-4P connector, this is already the connector on the screens for the prototypes and allows an easy cable option that should work at least 6ft so you can mount the screen wherever.

I'm stress testing this current one overnight right now.

forbiddenera

Todo:
- stress test with real gm sensors & stress test on real ecu hardware
- xtal stability test, 16mhz test
- power source
- finalize proto/alph board layout
- also design beta/final pcb
- configuration code
- connectors for sensor & ecu
- design & 3d print case for board
- design & 3d print case for screen
- put source on git

My goal is to make this module at least much better/more reliable/better code than the "other" Arduino-based DIY flex fuel thingys; already my code should be more reliable, the original code wasn't fully using the timers to their advantage and temperature wasn't really of concern, nor reliability: none of the code I've seen has any averaging stack or anything, so a if the system suddenly got a bad value, it could send that to the ECU; not good if you're full boost on E10 and suddenly the ecu thinks you're on E85 and adds 10 degrees for a few cycles then gets pulled right back as the device catches the next good value.

Currently compiles and runs on the 328p, with the gigantic OLED font library, it's using 50% program space and 90% of ram, abusing all 3 timers (t0 used for both pwm out and timing, t1 pwm in, t2 pwm out).

Can't wait to get a few of these out to people, test it and make the final design!

forbiddenera

PM'd you as you're the only one who's replied so far who seems to be 'ready' to use one now. Let me know if you want to test.

TunerThings

I'd definitely be interested in testing a kit out!!!

detroitm2

@forbiddenera Can you share your code on this? I'm working on a similar project and would love to base some of my work from it.

Thanks.

forbiddenera

Hi, get in touch with me and I'll see what I can help with. forbidden era#2332 on discord.