Arduino Modular Engine Sensors Unit (~$50)

[haydent]

:)

So basically i recon i can put together an affordable Arduino (think matchbox computer) based engine sensor system for around $50.

Likely i would just put up the plans for anyone to copy, and sell completed packages for sale to those that are so inclined.

Being Arduino based it will be expandable, with any type and number of sensors available (e.g. oil, water, temp, g-force, body angle, anything, etc).


So what im thinking:

Sensors are connected to box under hood that has 'data connection' to display unit in cabin. (only 1 cable needing to be run through firewall)


Display unit in cabin could be choice of lcd display, segment led, led bar graph, led warning lights, etc all with audible warning settings

What do you think ? Any comments ?

[Homestar]

I like the idea - I could think of a few things you could do with it. For a start, a basic engine monitoring package - oil pressure, coolant temp, low coolant, etc. You could display the results, as well as set off a lamp or buzzer when a warning value is reached. I could think up a few fun things to go on top of this as well.

One of the things I eventually want to do is replace all the dash globes with LED's - all easy and good, but wouldn't it be better if you could dim them properly, and have them fade up and down using a PWM output from the processor? Rather than have the dimmer pot on the dash, have a button that fades the LED's up and down to the desired level when pushed...

Also, I'm sure we could find a use for the LOL (Lots of LED's) display...:)

Wiring wise, I personally would take care of all that myself, but I would be interested in getting the box of tricks all programmed up the way I wanted...

As I think of more, I'll post them...

Cheers - Gav

Edit - Haydent - did you just edit your post while I was typing this???:p

[Homestar]

Ooooh - Just had a thought - use the LOL shield to display any warning that the system was monitoring - how cool would that be???:p

[slug_burner]

Bring it on. Was thinking of a controller for a variable geometry turbo could be a good job for an Arduino.

For the display a 4.3" GPS touch screen with an AV input could do the trick as they come with a suction mount, ciggy lighter power via the USB connector. They run Windows CE so I don't know what other smart trickery you could get the thing to do. Last time I got one from the net (DX Deals Extreme) they were about $100 now they are down to $50ish.

[haydent]

i was even thinking wireless link between engine box and cabin readout for even less wires but that would be more work and something to consider after the basics are worked out.

this arduino stuff is soo cheap, you can get a nano for $11 that pretty much would do everything we are talking about

and the possibilities are many,

i got to thinking about all this for the fact that the engine already has all these sensors but not audible alarm, when you need to know right away if your oil pump bolt fell out and there is no pressure or you coolant hose just popped off...

ill be the first to admit i may not see a visual dash alarm straight away, due to road distractions

so i think ill start there (with the default sensors), but you can get thermocupler's cheap to put any where you want:
Bolt-On Washer Thermocouple Assemblies (http://www.omega.com/pptst/WT.html)

plus any other VDO or whatever brand sensor you can buy


i was also thinking bluetooth to your phone ! but that may not be so usefull for some of the chaps around here who havent gotten a grip on that technology yet ;)

i think making it as open and adaptable is the goal, as some people might just want a audible alarm and nothing else for that extra minimal look, while others might want to see everything at once.

arduino lcd | eBay (http://www.ebay.com.au/sch/?_nkw=arduino%20lcd&clk_rvr_id=355829927553&mfe=search)

arduino nano | eBay (http://www.ebay.com.au/sch/?_nkw=arduino%20nano&clk_rvr_id=355832810844&mfe=search)

[Homestar]

Yes, I can't believe how cheap all the stuff is.

One more thing - is there an RFID Shield that could be used to make some sort of ignition activation circuit - so if you are close enough with the correct RFID tag, the car could be started without a key like some of the new flash cars out there...?

[haydent]

for sure, this is the thing. we're not talking about rocket science. prob everything we want has been done, there are plenty of guides and equipment for Arduino RFID

[haydent]

ive ordered a nano to get things started :)

[alexturner]

Hey guys,

I'm actually in talks with some guys in commercializing a similar product. I'd love to contribute to something like this.

Here's a video on what I've been working on: YouTube - Broadcast Yourself. (http://m.youtube.com/'reload=3&rdm=m5u9car0#/watch?v=V85trMiIy_k)

I've also got it reading out Oziexplorer data

Alex

[isuzutoo-eh]

Interested... after seeing a Madman EMS the other day and looking at Arduino capability versus price. GPS location tracking would be very attractive too.

[banarcus]

I've been sussing the Arduino for signal emulation so one can do engine transplants into P38s and D2s and do away with the LR ECU altogether.

[jazzaD1]

I'm thinking of using arduino as an air suspension controller, I have the code, it is written in BASIC, can that be programmed to an arduino board?

[haydent]

I'm thinking of using arduino as an air suspension controller, I have the code, it is written in BASIC, can that be programmed to an arduino board?

[edit]

i think it can run basic: http://hackaday.com/2011/08/28/basic-programming-on-an-arduino/

[haydent]

I've been sussing the Arduino for signal emulation so one can do engine transplants into P38s and D2s and do away with the LR ECU altogether.

Wow, that's thinking out of the box literally ! I don't know much about ecu but pwm comes naturally for arduino. I got into our via rc multi copter flying.

[banarcus]

Wow, that's thinking out of the box literally ! I don't know much about ecu but pwm comes naturally for arduino. I got into our via rc multi copter flying.

PWM is what existing systems need as it is the medium that the LR ECU sends to hill descent control, temp gauge on the dash, for traction control in the SLABS ECU. Use inputs from a GM PCM with its native sensors and then convert them into LR friendly PWM signals.

I've got more research to do, I just need a oscillscope with data log to record during different conditions etc.

[haydent]

PWM is what existing systems need as it is the medium that the LR ECU sends to hill descent control, temp gauge on the dash, for traction control in the SLABS ECU. Use inputs from a GM PCM with its native sensors and then convert them into LR friendly PWM signals.

I've got more research to do, I just need a oscillscope with data log to record during different conditions etc.

sounds like the sort of thing plenty of people would be interested in. im sure youve read the data in rave about all this.

[haydent]

Hey guys,

I'm actually in talks with some guys in commercializing a similar product. I'd love to contribute to something like this.

Here's a video on what I've been working on: YouTube - Broadcast Yourself. (http://m.youtube.com/'reload=3&rdm=m5u9car0#/watch?v=V85trMiIy_k)

I've also got it reading out Oziexplorer data

Alex


very nice screen, where do you buy that ! and i can see you've put some work into it :)

what temperature sensors are/have you using from where ?

and are you interfacing with the cars computers at all ? or a layer on top of that.

i take it your getting your filter km's data from oziexplorer gps ?


i subscribed to your youtube channel :)

i think there's alot of people that would be interested in a system like your especially the amount of detail you are putting into it.

i want to start out small and simple, with just the basics, temp/coolant level and oil. and go from there.

talk more soon once i get my board and start programming.

[Homestar]

Hey guys,

I'm actually in talks with some guys in commercializing a similar product. I'd love to contribute to something like this.

Here's a video on what I've been working on: YouTube - Broadcast Yourself. (http://m.youtube.com/'reload=3&rdm=m5u9car0#/watch?v=V85trMiIy_k)

I've also got it reading out Oziexplorer data

Alex

Very nice Alex, keep us updated on this.

Cheers - Gav

[haydent]

It arrived today, well... i had no idea it was this small from the ebay photo's !

Here it is next to a AA battery, this thing will fit anywhere :D

http://www.aulro.com/afvb/attachment.php?attachmentid=48309&stc=1&d=1340621520

[loanrangie]

This looks right up my alley, pardon my ignorance what type of os for lack of a better word do/can these run ?

[EchiDna]

what kind of sensors can be attached to this thing? I've only seen thermistor types but to make it a goer, it really should have the ability to run upper/lower alarms at user defined set points for water temp, air temps from ambient to EGT, and items like oil pressure, boost gauge etc...

[stuee]

This looks right up my alley, pardon my ignorance what type of os for lack of a better word do/can these run ?

The arduinos have a boot loader that allows you to download custom code via usb. There is no os as such, you're writing code that directly interfaces the io and mcu resources.

Arduino has opened the world of micro electronics right over the last few years. When I last played with mcu's (both atmel and some Phillips/nxp only 4-5 years ago) you needed programmers which you could make yourself or buy off the shelf, and all io circuits generally needed to be built. I made use of some pre made boards with some idc headers which made life much easier but much more convoluted than todays stuff with simple plug in shields etc. What's even better about a lot of arduino boards is that they come with the isp header so you can still use your old programmers and code if you don't like the standard arduino interface.

My last project was a dual battery controller combined with a dual stage thermo fan controller configurable via a 2x16 lcd display. Was based on an atmega 128 ( http://www.futurlec.com.au/ATMega_Controller.jsp ) so had massive amounts of io to play with to expand in the future. It died when my mcu board blew up after sitting it on a metal screwdriver :eek: The board supplier was hopeless so got bored and packed it all up after that.

Sent from my Transformer Prime TF201 using Tapatalk 2

[loanrangie]

Thanks Stuee, i new it wouldnt use a normal os like windblows or linux , i guess language would've been a better word.
Sounds pretty cool what can be done, do i see a mini USB socket on Babicat's pic to use for programming ?

[stuee]

Yeh the one posted in this thread has a mini usb socket.

The model:
Arduino - ArduinoBoardNano (http://arduino.cc/en/Main/ArduinoBoardNano)

I gather the language is based off C++. Personally I prefer the more procedural languages like Basic and C as they better reflect the behaviour of the MCU (i.e process line by line), and they also compile into much smaller sets of code when loaded onto the MCU, which used to be important but modern chips have ample amounts of memory now days. Likewise those messing around with them before me would have preferred machine code.

I used to use the free version of this:
CodeVisionAVR (http://www.hpinfotech.ro/html/cvavr.htm)

Haydent would probably be aware of this if he's come from rc multi-rotors, but when I finish the 101 I want to make my own drone using the ardupilot (An arduino based platform).

Back on topic, you need to put some thought into:
- what type of sensors you will use. i.e standard automotive VDO temp and pressure or basic RTD's etc. My preference would be towards using standard automotive sensors as they have been designed for the harsh environments, are easily available if broken and are easy to mount.

- the power requirements for all the IO. i.e 5V from the arduino board or reg, or 12V from the car system. This will change how much wiring is required to go in and out of the unit to sensors. Alternatively, if using the 5V, you may need an additional regulator depending on how much power is needed. You will also need to look at heat dissipation as the standard 5V regulators dropping down from 13.5V-14V while running will burn a lot of heat. Consider a switch mode 8V supply to drop down before the Arduino.

- how accurate you want the sensors to be. If using the cars battery to power sensors you may want to have a reference voltage feeding into the Arduino to make measurement adjustments (don't feed 12V directly into the arduino).

- how will you interface the arduino to all the sensors, outputs, screens etc. Its easy enough to prototype it up on a breadboard, but I would try and make use of as many components or "shields" off the shelf if you plan to sell any units in the future. The biggest pain in the ass when I did my projects was manufacturing a pcb, then a couple of months later wanting to add or remove components and redoing it all. Limiting yourself to wahts available commercially has a lot of restrictions, but also drastically cuts down your testing and troubleshooting time.

- how will you communicate between the display and sensor unit (arduino)? If you drive a display directly from the arduino you will use up a lot of valuable IO and resources. I have previously used a standalone serial display that used basic serial comms between the MCU and display to present information. Only used two pins and was reasonably easy to program. The display had its own controller as such that did all the hard work. If you use a display driven directly off the arduino consider using two arduino's and interface them with serial, I2C, or even bluetooth or wireless if you want to get fancy.

A similar project to what your doing but from 2009:
multidisplay - Arduino based open source display for interesting measurand´s of a car. - Google Project Hosting (http://code.google.com/p/multidisplay/)
Their V2 looks impressive but still in development.

[haydent]

Not much. Just a basic boot loader that loops the scripts you write and put on it to work with it's input/output pins.

[haydent]

Didn't realise there was a second page on my tapatalk...

Thanks for the breakdown stu.

[haydent]

Also just to clarify the USB port is really only for programming via it as a virtual com port. It doesn't act in host or slave mode...

[haydent]

As far as pins go, im going to start with an lcd in 4bit mode which uses 6 digital pins. The nano has 11 "free", so that leaves 5 for buttons to control it. Then there are 8 analog input pins.

So that means upto 8 sensors.

That's enough for now.

Im also keen on later making a second sender arduino box that lives under the hood to either reduce the number of wires going through the firewall to ~2 for a serial com link, or none of a wireless link where used. Also it could increase the potential number of sensors if needed.

[haydent]

what kind of sensors can be attached to this thing? I've only seen thermistor types but to make it a goer, it really should have the ability to run upper/lower alarms at user defined set points for water temp, air temps from ambient to EGT, and items like oil pressure, boost gauge etc...

For sure I will be programming alarms, that's what it's all about for me.

[haydent]

- what type of sensors you will use. i.e standard automotive VDO temp and pressure or basic RTD's etc. My preference would be towards using standard automotive sensors as they have been designed for the harsh environments, are easily available if broken and are easy to mount.


What do you mean by RTD ?

[stuee]

What do you mean by RTD ?

Sorry, just referring to generic resistance sensor elements from places like RS or even Jaycar (technically a VDO temperature sensor is an RTD :whistling:). Although very cheap it would mean having to figure out a way to mount them.

Resistance thermometer - Wikipedia, the free encyclopedia

Thermocouples are a millivolt source, making them difficult/impossible to read with standard ADC on the atmega chips. They require external amplification to read, so while they are cheap to buy and replace, they require additional circuitry to run them, although you can buy small IC's that include all this for you. They are not as easy to use as RTD's which are very simple to set up in a simple voltage divider circuit. They are now typically used only for high temp measurements (i.e. EGT's) as RTD's are much more consistent for lower temps. So yes thermocouple fine for EGT, but I'd use RTD's for water/oil/transmission temp etc.

With any temperature sensor the key is the location. When I played with a thermocouple driven fan controller I first placed the thermocuple on a head bolt after reading how this had been done elsewhere. Big mistake. The temperatures fluctuated wildly depending on how hard I was pushing the motor and consequently the fan activation was sporadic. The coolant acts as a good damper for the measurement of temperature and is a good reflection of the average temperature of the engine. A simple 1/8npt vdo (RTD type) temp sensor in the top radiator hose provided a very stable output that was useful for controlling the fan. This was with a simple switch though. Technically speaking you could apply a digital filter to the signal from the head bolt mounted sensor in the arduino to average out the temps but to me this is a band-aid fix rather than implementing a proper solution.

Similarly, when measuring EGT's you want to know as best as possible the instantaneous EGT's so you can back of if your about to melt pistons etc. A thermocouple mounted on the outside of the exhaust will be filtered too much by the slow uptake and dissipation of heat from the exhaust pipes and average the temps. If you have a sharp rise in EGT's due to over boost or something then you may not notice it till too late.

Both the above are example of why I would stick to standard automotive sensors. The R&D has already gone into them, and its just a simple matter of buying/making adaptors.

Also, don't forget to save some digital pins for outputs for alarms and relays etc,

[Judo]

I'm very interested in this stuff guys! I've thought about putting a small laptop/netbook somewhere with a touch screen mounted on the dash and using it for GPS, maps, music, internet, etc.

But options like the raspberry pi and controllers like the arduino appear to be a much much better option.
The arduino for sensor monitoring and the raspberry pi for fatter apps like music and internet.

At the very least, I'd be interested in building one / buying one. (Almost certainly building - can't get enough of DIY!). Not sure what I can offer, but let me know if there is anything I can do to help.

Something that appealed to me greatly above is the idea of running everything through USB. I already have quite a number of power regulators, power leads, usb adapters and leads, audio leads and countless electrical items throughout the cab because they're all different voltages, signal types, plug types. Standardising these to mostly USB would be great!

[haydent]

Sorry, just referring to generic resistance sensor elements from places like RS or even Jaycar (technically a VDO temperature sensor is an RTD :whistling:). Although very cheap it would mean having to figure out a way to mount them.

Video Link: http://en.wikipedia.org/wiki/Resistance_thermometer (http://en.wikipedia.org/wiki/Resistance_thermometer)

Thermocouples are a millivolt source, making them difficult/impossible to read with standard ADC on the atmega chips. They require external amplification to read, so while they are cheap to buy and replace, they require additional circuitry to run them, although you can buy small IC's that include all this for you. They are not as easy to use as RTD's which are very simple to set up in a simple voltage divider circuit. They are now typically used only for high temp measurements (i.e. EGT's) as RTD's are much more consistent for lower temps. So yes thermocouple fine for EGT, but I'd use RTD's for water/oil/transmission temp etc.

With any temperature sensor the key is the location. When I played with a thermocouple driven fan controller I first placed the thermocuple on a head bolt after reading how this had been done elsewhere. Big mistake. The temperatures fluctuated wildly depending on how hard I was pushing the motor and consequently the fan activation was sporadic. The coolant acts as a good damper for the measurement of temperature and is a good reflection of the average temperature of the engine. A simple 1/8npt vdo (RTD type) temp sensor in the top radiator hose provided a very stable output that was useful for controlling the fan. This was with a simple switch though. Technically speaking you could apply a digital filter to the signal from the head bolt mounted sensor in the arduino to average out the temps but to me this is a band-aid fix rather than implementing a proper solution.

Similarly, when measuring EGT's you want to know as best as possible the instantaneous EGT's so you can back of if your about to melt pistons etc. A thermocouple mounted on the outside of the exhaust will be filtered too much by the slow uptake and dissipation of heat from the exhaust pipes and average the temps. If you have a sharp rise in EGT's due to over boost or something then you may not notice it till too late.

Both the above are example of why I would stick to standard automotive sensors. The R&D has already gone into them, and its just a simple matter of buying/making adaptors.

Also, don't forget to save some digital pins for outputs for alarms and relays etc,

Thank you for you're valuable information and was thinking thermocouplers acted the way you describe RTD's as working...

Sent on mobile via Forum Runner

[haydent]

I'm very interested in this stuff guys! I've thought about putting a small laptop/netbook somewhere with a touch screen mounted on the dash and using it for GPS, maps, music, internet, etc.

But options like the raspberry pi and controllers like the arduino appear to be a much much better option.
The arduino for sensor monitoring and the raspberry pi for fatter apps like music and internet.

At the very least, I'd be interested in building one / buying one. (Almost certainly building - can't get enough of DIY!). Not sure what I can offer, but let me know if there is anything I can do to help.

Something that appealed to me greatly above is the idea of running everything through USB. I already have quite a number of power regulators, power leads, usb adapters and leads, audio leads and countless electrical items throughout the cab because they're all different voltages, signal types, plug types. Standardising these to mostly USB would be great!

Great, stay tuned. I'm waiting for some parts to arrive so I can get a prototype going, then you can help test and give feedback and suggestions on direction...

Sent on mobile via Forum Runner

[haydent]

So I checked out the arduino thermocoupler opamp Shields and the vdo RTD sensors.

The later would work out cheaper and easier to install and suitable for use.

Sent on mobile via Forum Runner

[Judo]

I was doing some reading last night on how this could be done on a raspberry pi and the biggest challenge looks like it would be the interfacing of the sensors to the PC via USB. USB is a fairly high level interface, but sensors are very much a raw signal.

There are USB interface devices for sale, but they are too expensive for this purpose. The best solution I found is a USB-Serial convertor, then read as though you have a rs232 serial interface on the PC. EG:

Page Title (http://www.web4robot.com/USBI2C.html)

My understanding is it would support up to 8 inputs analog or digital.
"Data Port pins may be individually selected to be digital input / output, analog input or PWM
output."

Thoughts? Anyone know of good ways to interface raw signals to a PC?

[stuee]

Thoughts? Anyone know of good ways to interface raw signals to a PC?

The best bet if you want to use the Raspberry PI is to interface it with someting like an Arduino and talk over usb using standard library's provided with the Arduino. Theres no reason that whatever Haydent comes up with can not be used (with some additional coding) to transfer the signals to PC using serial or USB.

Otherwise if you want a ready-made solution you will be looking at $100+. There are things like this about:
8 Relay 8 Analog USB IO Controller 12 V :: PC Based :: Controllers :: Welcome to Ocean Controls (http://www.oceancontrols.com.au/KTA-225.html)

A simple ASCII protocol allows control from Windows/Mac/Linux using either USB Virtual COM drivers or RS485.

[Judo]

The best bet if you want to use the Raspberry PI is to interface it with someting like an Arduino and talk over usb using standard library's provided with the Arduino. Theres no reason that whatever Haydent comes up with can not be used (with some additional coding) to transfer the signals to PC using serial or USB.

Otherwise if you want a ready-made solution you will be looking at $100+. There are things like this about:
8 Relay 8 Analog USB IO Controller 12 V :: PC Based :: Controllers :: Welcome to Ocean Controls (http://www.oceancontrols.com.au/KTA-225.html)
Yeah that's a good point. It seems interfacing will cost more than the arduino solution anyway - thus might as well just use Haydent's solution, or interface into that if additional functionality is required.

If a USB interface could be used, then it would also be possible to use a smartphone and app as the GUI... Siri can tell you when your radiator hose has burst. :angel:

[haydent]

If a USB interface could be used, then it would also be possible to use a smartphone and app as the GUI... Siri can tell you when your radiator hose has burst. :angel:

This could also be done via Bluetooth

Sent on mobile via Forum Runner

[stuee]

Yeah that's a good point. It seems interfacing will cost more than the arduino solution anyway - thus might as well just use Haydent's solution, or interface into that if additional functionality is required.

If a USB interface could be used, then it would also be possible to use a smartphone and app as the GUI... Siri can tell you when your radiator hose has burst. :angel:

Would be a lot of work getting Siri to tell you. Apple IOS and open source dont go together that well.

Android is a different kettle of fish. Google actively encourages development and easier to distribute open source software:
Accessory Development Kit 2012 Guide | Android Developers (http://developer.android.com/tools/adk/adk2.html)

One of the first kits released:
Google IO Android ADK (Arduino) - YouTube
A basic tablet interface:
Android Open Accessory DemoKit With Arduino Mega ADK Running On Motorola Xoom - YouTube

Having only stumbled across this stuff after looking at this project I think I'm going to order an Arduino Mega ADK to play around with on my Asus Android Tablet. So much cool stuff available now!!

[Judo]

I've had a (brief) go at dev on both Android and iphone and to be honest I didn't like Android that much. I was only playing around though - so perhaps the story would be different if I got really stuck into it.



Having only stumbled across this stuff after looking at this project I think I'm going to order an Arduino Mega ADK to play around with on my Asus Android Tablet. So much cool stuff available now!!

haha I am also thinking I'll order an Arduino after reading more and more on their website. These things are far too cool to not buy one. Not having a purpose for it yet is irrelevant right? :cool:

The question is, which one do I get... I'll be buying the SMARTGPU like Haydent as well. (Don't have an Android, so ADK is out for me anyway...)

[haydent]

I won't likely be developing any phone apps. To much work for limited use. And this really deserves a dedicated unit.

I got the nano, it's the cheapest arduino. Is recommend that if you want to easily copy my work. Though any arduino will work.

Sent on mobile via Forum Runner

[EchiDna]

how about the fusion brain from mp3car.com? US$60 and has shedloads of analog inputs...

[Judo]

how about the fusion brain from mp3car.com? US$60 and has shedloads of analog inputs...
It looks more or less the same as an arduino, but a faster processor.


Please note: This is a hobbyist grade product. As such, limited or no documentation is included with it and advanced technical skill may be required to use it.

This kind of scares me.... Without researching any further, the Arduino has quite a lot of documentation online and there are Arduino models available for about half the price. So although they are similar, I would always go the one with better online support.

Looks like a good site though - thanks.

[EchiDna]

It looks more or less the same as an arduino, but a faster processor.



This kind of scares me.... Without researching any further, the Arduino has quite a lot of documentation online and there are Arduino models available for about half the price. So although they are similar, I would always go the one with better online support.

Looks like a good site though - thanks.


suggest you read the forums, there is plenty of support info out there, but it doesn't come as a documentation pack with the hardware... more of a forewarning than anything else... I'm sure there are plenty of arduino owners who haven't got a clue about using one either, and yet they buy it.

[Dougal]

Can you make a stepper motor driver version so we can replace crappy factory instruments with ones that work? Thinking rev counter, fuel and temp.

I can do the hardware side, just need stuff to drive it.

[haydent]

how about the fusion brain from mp3car.com? US$60 and has shedloads of analog inputs...

does look impressive Mp3Car Custom Car Computers! (http://store.mp3car.com/SearchResults.asp?Search=fusion+brain) would def be good if you need that much, i dont though.


Can you make a stepper motor driver version so we can replace crappy factory instruments with ones that work? Thinking rev counter, fuel and temp.

I can do the hardware side, just need stuff to drive it.

prob a bit out of my scope at the moment sorry...

[haydent]

Can you make a stepper motor driver version so we can replace crappy factory instruments with ones that work? Thinking rev counter, fuel and temp.

I can do the hardware side, just need stuff to drive it.


just for your interest though:

Stepper Motor Driver Board ULN2003 for Arduino 5V 4-phase 5 line | eBay (http://www.ebay.com.au/itm/Stepper-Motor-Driver-Board-ULN2003-Arduino-5V-4-phase-5-line-/160772785983?pt=AU_B_I_Electrical_Test_Equipment&hash=item256ece073f)

[slug_burner]

Can you make a stepper motor driver version so we can replace crappy factory instruments with ones that work? Thinking rev counter, fuel and temp.

I can do the hardware side, just need stuff to drive it.

Is it pandering to old technology to have physical needles/pointers moving on a dial yet having to use stepper motors and controllers to achieve that outcome. You could just have displays that generate a graphical representation of the dials and needles simplifying the whole thing.

There is nothing like a good old mechanical dial even if it is driven by a digital circuit:cool:

[EchiDna]

if you plan on going THAT far with replacement instruments....

Universal Instrument Panels (http://www.dakotadigital.com/index.cfm/page/ptype=results/Category_ID=69/home_id=59/mode=cat/cat69.htm)

[Judo]

I recently had a hire car that had both needle and digital reading for the speedo. The digits were way better to read IMO. Needles are so last decade. :P

I picture something like: each reading in a box/circle with a background colour that changes based on the reading. I.E. All reading normal would be a bunch of green boxes with digital reading printed inside each. The background colour slowly changes to orange, then red (using shading in between) to represent the status. This means for the most part you never even have to read the screen.. (Except when you want to see all the cool things you're tracking :cool: )

Another thought is, that most people (like me) will want a "screen off" feature for night driving. Ideally the screen would come on if any readings hit pre-defined levels.

[Homestar]

I recently had a hire car that had both needle and digital reading for the speedo. The digits were way better to read IMO. Needles are so last decade. :P


Have to agree - one of the things I actually like about the Dunny Door is that I can select a digital speedo display, which is the only one I look at. I never look at the tacho either, but that is because it is a company car...:D

[madtom]

I have just found this nice topic and made a quick read. Some ideas are wery nice. I have a "prototype" of arduino based (Mega2560) multi display for my TD5 Discovery (but it can be used in any car). TD5 is nice, as it has nearly all sensor already fitted (temperature, boost presure, ....), so I will add only oil presure sensor and EGT termocouple sensor. The most important think is an audible alarm. Because you can have 10 displays on the dash, but you will not recognize incoming problem, because you have to drive the car, not watch the numbers.
My idea is a bit different - one unit on the top of the dash with lot of wires to the sensors - only short wires, because my engine ECU is moved from engine into the cabin (to make it a bit more waterproof), so all signal are bellow the dash. And it is easier to make one unit, then making two units.
The most important think for automotive electronic is safety (using the right fuses, polyfuses or limiting resistor for all wires, .....) and electromagnetic compatability. There are very hard standards used for automotive. Also I don't want to disturb the function of engine or other parts of car. It is a good idea to at least read the standards and try to understand them. It can save a lot of problems during development and possible production.
I was also looking for ready made solutions, but didn't find any that fulfill all my needs and wishes. The EMS Madman was the closest one.
Some basic functions of my "prototype":
40x4 LCD (the biggest standard LCD - cheap and easy to buy)
measuring Main battery, Aux battery and managing of aux battery charging (I have a deep discharge one)
measuring Coolant temperature, Boost pressure, Exhaust gas temperature, Oil presure - all with two level alarms - yellow = warning (with short beep, I have time to find a place to stop the car), red = long beep, I have to stop inmediately
I'm also planning to add some speedo correction function, because I have fitted a bit bigger tires and the speedo is a bit pesimistic now.

[haydent]

sounds great madtom, pretty much just what i want. just for others interest this is the lcd setup/wiring ill be using to begin with. (maybe upgrading to larger one later)

Arduino - LiquidCrystal (http://arduino.cc/en/Tutorial/LiquidCrystal)

[haydent]

Download - Fritzing (http://fritzing.org/download/)

just found this great free prototyping software for making the images as seen in the link in last post:

https://www.aulro.com/afvb/images/imported/2012/07/1251.jpg

https://www.aulro.com/afvb/images/imported/2012/07/1252.jpg

[Judo]

sounds great madtom, pretty much just what i want. just for others interest this is the lcd setup/wiring ill be using to begin with. (maybe upgrading to larger one later)

Arduino - LiquidCrystal (http://arduino.cc/en/Tutorial/LiquidCrystal)
I'm still tossing up whether to get a Nano or a higher spec'ed one, but either way I plan on getting a SMARTGPU touch LCD to go with it. Being that the Nano has 8 analogue inputs, I'm leaning towards the Nano + touch LCD. Am I likely to run out of memory on a Nano? Or a non-issue in your opinion?

Re: your first post - are you still thinking of sharing the design and code so others can copy? Once I have the touch LCD, I will be attempting to code a touch based GUI. It'd be keen to interface it to your code if that works for both of us. Thoughts?

[isuzurover]

On a somewhat related note, which may interest some, I have been playing around recently with Sparkfun data loggers. They log to a micro-SD card.

We usually buy our stuff from Arduino & PICAXE :: Welcome to Ocean Controls (http://www.oceancontrols.com.au/Arduino-and-PICAXE)

They stock Arduino stuff as well. It is handy to be able to get everything in one place, and I have found the guys there good to deal with.

[Judo]

Looking into SMARTGPU more... appears I was dreaming when I thought it was around $60-$70.... ? They want $159 + $43 postage!!!!

But found this instead:

2.8 TFT Touch Shield for Arduino [2.8] ID: 376 - $59.00 : Adafruit Industries, Unique & fun DIY electronics and kits (http://www.adafruit.com/products/376)

$59 + $10 postage. :D :D :D

[haydent]

yeah, ill be making my code and designs open. i found this too:
multidisplay - Arduino based open source display for interesting measurand´s of a car. - Google Project Hosting (http://code.google.com/p/multidisplay/)
very similiar

judo-you may want to consider a better than nano if you have the space

i got my breadboard today and hookup wires and have started playing around.

[haydent]

its really great the detail rave goes into about some things:

oh and the voltage it operates at is 0-5 v perfect for arduino input

http://www.aulro.com/afvb/attachment.php?attachmentid=48739&stc=1&d=1341532993
http://www.aulro.com/afvb/attachment.php?attachmentid=48740&stc=1&d=1341532993

[haydent]

moved to last post

[Dougal]

Is it pandering to old technology to have physical needles/pointers moving on a dial yet having to use stepper motors and controllers to achieve that outcome. You could just have displays that generate a graphical representation of the dials and needles simplifying the whole thing.

There is nothing like a good old mechanical dial even if it is driven by a digital circuit:cool:

It's a great way to get classic looks with modern performance. I'm all for it.:)

It's the way that pretty much all cars have been running their instruments for at least a decade now. It's only recently that a few have one LCD display that looks like gauges (current RR etc).

[Judo]

Haydent, that information on the TD5 is great. Now I just need a TD5. :D (Give me a few years and I'll have SWMBO convinced..)

I'll be hooking up all my own sensors on my 300TDI. The only one I have at the moment is a EGT guage. I would like to get an engine head temp sensor, coolant temp sensor and battery voltage readings to test the Arduino. I figure they should be all relatively simply to aquire and hook up, but if anyone has suggestions on which sensors to buy I'm open to suggestions. (Haven't looked into it yet).

[Dougal]

just for your interest though:

Stepper Motor Driver Board ULN2003 for Arduino 5V 4-phase 5 line | eBay (http://www.ebay.com.au/itm/Stepper-Motor-Driver-Board-ULN2003-Arduino-5V-4-phase-5-line-/160772785983?pt=AU_B_I_Electrical_Test_Equipment&hash=item256ece073f)

Thanks for the info.
This one will drive one stepper. Is it possible to drive 3x instrument steppers from one arduino or does that necessitate 3 of everything?

[Judo]

Thanks for the info.
This one will drive one stepper. Is it possible to drive 3x instrument steppers from one arduino or does that necessitate 3 of everything?
My understanding (from a few days reading up on Arduino) is yes. The stepper board uses 4 pins on the Arduino board, so as long as you have an Arduino with at least 12 pins, you can use 3 steppers.

[isuzurover]

One of the guys who works for me has been pestering me to buy some Arduinos for the lab. However I think he mainly wants them for his Pajero and his house... He mentioned turning on his coffee machine when his alarm clock goes off so there is hot coffee waiting when he gets up in the morning... :D

However we do have applications where we need to control 240V devices... Does anyone have experience with any 240V relays and relay drivers?

[jazzaD1]

solid state relays work well for that purpose, can be driven by 3V DC or so, and will switch 240V, they are also silent and have no moving parts to wear out. A heatsink is required if switching relatively higher amperages, but it depends on the application

[haydent]

Haydent, that information on the TD5 is great. Now I just need a TD5. :D (Give me a few years and I'll have SWMBO convinced..)

I'll be hooking up all my own sensors on my 300TDI. The only one I have at the moment is a EGT guage. I would like to get an engine head temp sensor, coolant temp sensor and battery voltage readings to test the Arduino. I figure they should be all relatively simply to aquire and hook up, but if anyone has suggestions on which sensors to buy I'm open to suggestions. (Haven't looked into it yet).


I think you'll found most coolant sensors work the same way. Just need to check the voltage.

I bought a couple off head bolt k thermo couplers off Ebay that I don't need. I can send you one of you Pm me your address. Bear in mind you'll need a op amp circuit for these. Ill do one some day our buy one, but for now factory coolant temp sensor is enough for me with header tank float switch add backup. those and oil pressure are my initial goals.

Sent on mobile via Forum Runner

[haydent]

In regards to voltage if it's above 5v just need a simple voltage divider.

In regards to the older models, I'll be making one for my 92 v8 d1 too !

Sent on mobile via Forum Runner

[Dougal]

You can use glow-plugs as thermocouples for measuring head-temp. Apparently B-type thermocouple charts are close.

[haydent]

You can use glow-plugs as thermocouples for measuring head-temp. Apparently B-type thermocouple charts are close.
that's very interesting, do you have any links. ill keep googling...

[Dougal]

that's very interesting, do you have any links. ill keep googling...

No links, just some interesting old guy told me so I tried it myself. I was getting a range of 4-5mV with a multimeter onto one of my 24v 4BD1T glow-plugs at cruising . These aren't connected so I had no risk of blowing up any thermocouple driver.

You can find type B thermocouple charts online for reference.

[madtom]

Just one think - I have very bad experience with sofisticated LCDs for last 3 years. When the stock is empty, it is very hard to find some more (my job is development of electronic devices and small quantities manufacturing). For few last years the manufacturers don't make them for stock. This is the main reason, why I have choosed standard 40x4 LCD. Second reason is price. And touch screen in offroad car? Nice idea, but not, when you are moving off road. Also the needle type indicators - nice idea, but I prefare to look on the road, not on the instruments, when I'm driving. So an audible alarm is an option. And last but not least is the temperature range of the LCD - here in Central Europe I need -25°C in winter to +60°C in summer when the sun is shinning directly on the car.

[haydent]

heres my progress so far, i found a great menu library that handles the buttons and lcd with setting menu's, splash and home screens.

it support various lcd sizes:
https://github.com/brunialti/MENWIZ

so far i got it set up and showing the voltage on a input pin adjustable by a pot. if the voltage is below a certain level it makes pin 13 "high" and a led is illuminated.

eventually all alarm levels will be set via lcd menu.

ive also been getting some good tips from felow member here via email about a simple circuit to build to access eg coolant temperature sensor without effecting its signal

if anyone is interested in more details about my circuit or code so far please ask otherwise ill update again as i get more done.

http://www.aulro.com/afvb/attachment.php?attachmentid=49056&stc=1&d=1342264122

[Judo]

It's coming together nicely. I've just ordered:

1 x Arduino Mega 2560 R3 $65.00
1 x 2.8" TFT Touch Shield for Arduino - 2.8" $59.00

$12 postage. Not sure how long it will take, I ordered it from the USA. Will let you know once I have it and can start coding that touch screen!

[haydent]

sure. youll have to look for a different library for that sort of lcd than the one i use. it will make it alot easier rather than coding everything from scratch. checkout the playground section on the arduino website.

[c.h.i.e.f]

Ive been thinking lately about how to setup glow plugs to monitor combustion temps but only started scratching the surface...pre-determined resistance of the glow plug is a good start as most of us know..
B type thermocouples good for upto 1800*c and reading a mV output which we can cross reference onto a Btype chart eg(750*c = 2.782mV) I am yet to figure out a display or means of easily displaying a temp read out rather than a mV reading....

[stirlsilver]

Just some quick input from me. From the times when I was playing around with microprocessors to monitor sensors and control devices. I found one of the single biggest issues with this sort of system on a car is filtering out noise, especially on a petrol engine.

If you are planning on feeding a voltage signal to the analogue inputs, I think you will need to consider having some low pass filters in your circuitry to filter out ignition noise out of your signals. Or convert the voltage signal to a 4 to 20mA current signal at the sensor, run the signal cable to the location where you have the Arduino (inside a shielded box) and covert the signal back to a voltage. A current signal is much more immune to noise when compared to voltage so shielding and filtering isn't quite as important.

Also, another area which I tended to have issues with was getting a nice clean power supply to the microprocessor. The power supply on a car (again especially petrols) is very noisy and can cause microprocessors to behave strangely due to noise getting through the voltage regulators to the chip. So again, you may need to consider a nice filtered supply in your design to get the chip stable.

I really like the idea of the project, and I may implement something using an arduino on my project car also.

On a side note, for anyone that has an OBD enabled vehicle, you can consider simply hooking in a OBD to bluetooth adaptor (or OBD to USB ) and use an Iphone or Android device to run an app called Torque. I use it all the time in my car to monitor coolant temperature (along with other sensors) and also set some alarm trigger points on the sensors also should something go wrong.

[isuzurover]

Ive been thinking lately about how to setup glow plugs to monitor combustion temps but only started scratching the surface...pre-determined resistance of the glow plug is a good start as most of us know..
B type thermocouples good for upto 1800*c and reading a mV output which we can cross reference onto a Btype chart eg(750*c = 2.782mV) I am yet to figure out a display or means of easily displaying a temp read out rather than a mV reading....

You would need to calibrate using an oven and a known/reliable temperature guage.

[Judo]

BUMP! My Arduino mega + LCD touch screen arrived. :D Now to start playing around!

[haydent]

coincidentally i was just sitting here thinking about posting that i got a power supply cheap and small for it once its in the car. also available in 12v

im now waiting on a particular opamp type to use in the circuit

http://www.aulro.com/afvb/attachment.php?attachmentid=49758&stc=1&d=1344396278

3-Terminal 5V 1A Switching Voltage Regulator Power Supp | eBay (http://www.ebay.com.au/itm/3-Terminal-5V-1A-Switching-Voltage-Regulator-Power-Supp-/390327745637?clk_rvr_id=372876876511&mfe=sidebar)

[isuzurover]

Just some quick input from me. From the times when I was playing around with microprocessors to monitor sensors and control devices. I found one of the single biggest issues with this sort of system on a car is filtering out noise, especially on a petrol engine.

If you are planning on feeding a voltage signal to the analogue inputs, I think you will need to consider having some low pass filters in your circuitry to filter out ignition noise out of your signals. Or convert the voltage signal to a 4 to 20mA current signal at the sensor, run the signal cable to the location where you have the Arduino (inside a shielded box) and covert the signal back to a voltage. A current signal is much more immune to noise when compared to voltage so shielding and filtering isn't quite as important.

Also, another area which I tended to have issues with was getting a nice clean power supply to the microprocessor. The power supply on a car (again especially petrols) is very noisy and can cause microprocessors to behave strangely due to noise getting through the voltage regulators to the chip. So again, you may need to consider a nice filtered supply in your design to get the chip stable.

I really like the idea of the project, and I may implement something using an arduino on my project car also.

On a side note, for anyone that has an OBD enabled vehicle, you can consider simply hooking in a OBD to bluetooth adaptor (or OBD to USB ) and use an Iphone or Android device to run an app called Torque. I use it all the time in my car to monitor coolant temperature (along with other sensors) and also set some alarm trigger points on the sensors also should something go wrong.

We have had the same issues with the sparkfun datalogger based systems we have developed. All our sensors are pressure and temp, and not relying on or grounded to any of the vehicle or vehicle electrics. We thought that using 12V-5V and 24V-5V power supplies would remove the noise, but it was not the case. The noise meant that the datalogger would log 2x or 3x the max range of the sensor connected to that port - so completely useless - not just a bit of noise in the data. For our applications it was easiest just to switch to running the whole system from batteries instead...

[Judo]

Took me a while to get started, but I've been playing around over the last week. Started with the example "paint" app and changed a few things here and there to better understand the code.

http://imageshack.us/a/img255/8281/arduinoaulro.jpg

Then I started drafting up some screen ideas. Initial thoughts are each sensor value can be "Green" "Amber" or "Red" status which would be the background for that cell. Starting with just 4 sensors, but can split the screen into however many needed. Haven't figured out what the menu options will be available yet but I'm thinking the black dot in the middle could be the menu button.

The colours are a bit "fluro" for my liking - I'm sure I can make them nicer - but it's a starting point. ;)

http://imageshack.us/a/img442/6461/arduinoscreentemplate1.jpg

Now I need a sensor or something to connect so I can start reading/displaying real data.




I bought a couple off head bolt k thermo couplers off Ebay that I don't need. I can send you one of you Pm me your address. Bear in mind you'll need a op amp circuit for these.



im now waiting on a particular opamp type to use in the circuit


Did you find an op amp? Any links for these sensors/op amps you can send me?

Re: the head bolt coupler, do you know the bolt size needed?

[Judo]

I'm thinking something like this for head temp:

Thermocouple Sensors K type 18mm id Washer for Cylinder Head Temperature (CHT) | eBay (http://www.ebay.com.au/itm/Thermocouple-Sensors-K-type-18mm-id-Washer-Cylinder-Head-Temperature-CHT-/251005573587?pt=AU_Car_Parts_Accessories&hash=item3a711919d3#ht_1932wt_907)

Just need to find the bolt size first...

[haydent]

I'm thinking something like this for head temp:

Thermocouple Sensors K type 18mm id Washer for Cylinder Head Temperature (CHT) | eBay (http://www.ebay.com.au/itm/Thermocouple-Sensors-K-type-18mm-id-Washer-Cylinder-Head-Temperature-CHT-/251005573587?pt=AU_Car_Parts_Accessories&hash=item3a711919d3#ht_1932wt_907)

Just need to find the bolt size first...
thats the ones i have, you can have one if you want, just pm me your address.

im still awaiting on the op amp, it seems to have been lost in post, so i have to reorder.

as for thermocouplers' you will have to build a little circuit to go with them to amplifiy the signal.

[mike_ie]

However we do have applications where we need to control 240V devices... Does anyone have experience with any 240V relays and relay drivers?

Yeah, I've used various types over the years in different applications - CEBus, etc. A good source for them in Australia is security alarm system wholesalers - they are used to drive home lighting, security lighting etc when the alarm is tripped. Alarm systems run on 12V, so the relays are specifically designed for just the application you describe.

[mike_ie]

Can you make a stepper motor driver version so we can replace crappy factory instruments with ones that work? Thinking rev counter, fuel and temp.

I can do the hardware side, just need stuff to drive it.

I worked on something like this quite a few years back, using an 8051 microcontroller, but the premise is much the same. It's actually reasonably straightforward - either read the resistance of the analogue senders or pop in digital senders, and start generating a square wave for each gauge. It has other benefits too, such as being able to smooth out signals from the fuel sender, matching different gauges to the stock fuel sender, so on and so forth.

In my opinion, a combination of stepper motor driven gauges and an LCD display would be the ideal. Gauges for normal usage, plus the LCD displaying additional information, logging data, flashing bright red when tolerances are exceeded etc. I'd imagine that you'd have to integrate a multiplexer into the design to accommodate all of the digital inputs and outputs, but again that's very straightforward.

I need to get me one of these kits, ASAP :)

[alexturner]

Ahhh very cool,

I'll have to upload the code to the unit I've built :) anyone had any luck with the obd2 integration?

Alex

Sent from my GT-I9300 using Tapatalk 2

[Dougal]

I worked on something like this quite a few years back, using an 8051 microcontroller, but the premise is much the same. It's actually reasonably straightforward - either read the resistance of the analogue senders or pop in digital senders, and start generating a square wave for each gauge. It has other benefits too, such as being able to smooth out signals from the fuel sender, matching different gauges to the stock fuel sender, so on and so forth.

In my opinion, a combination of stepper motor driven gauges and an LCD display would be the ideal. Gauges for normal usage, plus the LCD displaying additional information, logging data, flashing bright red when tolerances are exceeded etc. I'd imagine that you'd have to integrate a multiplexer into the design to accommodate all of the digital inputs and outputs, but again that's very straightforward.

I need to get me one of these kits, ASAP :)

Thanks Mike, I might call on you for ideas later on. Basically I want to drive needles to keep the factory look. No other reason.

[mike_ie]

Thanks Mike, I might call on you for ideas later on. Basically I want to drive needles to keep the factory look. No other reason.

No worries. Had you any particular gauges in mind? Wouldn't mind taking a look....

[Dougal]

No worries. Had you any particular gauges in mind? Wouldn't mind taking a look....

Yes, the tacho, temperature and fuel needles in an 85 RRC cluster. Which are the three that are never better than a rough indication (sometimes on strike) of what is actually happening.

It's now diesel so the tacho could be alternator fed or from the 0-3v converter box I conjured up a few years back to make it work. Fuel level and water temp senders are variable resistance to earth, I'm not sure what the ohm ranges are right now or even what they are supposed to be.:angel:

[Judo]

Yes, the tacho, temperature and fuel needles in an 85 RRC cluster. Which are the three that are never better than a rough indication (sometimes on strike) of what is actually happening.

It's now diesel so the tacho could be alternator fed or from the 0-3v converter box I conjured up a few years back to make it work. Fuel level and water temp senders are variable resistance to earth, I'm not sure what the ohm ranges are right now or even what they are supposed to be.:angel:

Speaking of fuel gauge, on the weekend I knew I was getting close to empty based on the trip metre. 90 litre tank in my Disco. Fuel gauge showed just under half, no where near the red zone. I filled up at the servo and the pump didn't stop until 88.5 litres. :D Based on the mileage I did on that tank, I had 12kms left before empty.

[matbor]

Ahhh very cool,

I'll have to upload the code to the unit I've built :) anyone had any luck with the obd2 integration?

Alex

Sent from my GT-I9300 using Tapatalk 2

some things i have come across in the past that might help, no idea though...
Antipasto Hardware Blog: Introducing the LCDash GT: Hacking Your Engine with Open Source Hardware (http://antipastohw.blogspot.com/2010/07/introducing-lcdash-gt-hacking-your.html)

and...

http://code.google.com/p/opengauge/

[haydent]

I finally made some time to sit down with this for a bit more past 2 days. Here's a photo of it so far, ready for testing in car. Currently just monitors Coolant Temp, with Alarm and Max reading. Will add oil pressure sensor support next, then make a pcb boxed prototype.

http://www.aulro.com/afvb/attachment.php?attachmentid=65626&stc=1&d=1379239579

[FANTOM P38]

Nothing like a bit of progress! Have you considered Auto Trans Temp?
very handy for towing!

[Kevin B]

Watching with interest..

[steveG]

Good work!

I've got an Arduino board here and a few one-wire temp sensors - just need to make time to play with it.

Steve

[haydent]

Nothing like a bit of progress! Have you considered Auto Trans Temp?
very handy for towing!

sure, im trying to make something as open ended us possible so its easy to adapt to as many sensors and size screen as desired.

[jonesybj]

Something I've been working on also. For my rear air suspension to show pressure in each bag going to add bit more to it though :)

Ride height gauges - YouTube

[Judo]

Something I've been working on also. For my rear air suspension to show pressure in each bag going to add bit more to it though :)

Ride height gauges - YouTube (http://www.youtube.com/watch?v=qhem0H95to0)
It's too dark to see much - what screen is that?

BTW, my Arduino project was derailed when I bought a 2nd Landy. :angel: Hopefully it will work its way back up the project list again...

[dkaos]

Looks good so far, hell I would be happy with that for now! Although the other stuff you have mentioned does sound great. Hope you can get it finished in time for Xmas :)

Cheers
Clint

[haydent]

I just finished moving from breadboard to 'veroboard' prototype stage. I found testing it in the car with a breadboard was too unreliable due to vibrations & movement, this fixes that. So i can now 'install' it & start using/testing it full time.

http://www.aulro.com/afvb/attachment.php?attachmentid=67632&stc=1&d=1382774729http://www.aulro.com/afvb/attachment.php?attachmentid=67633&stc=1&d=1382775053http://www.aulro.com/afvb/attachment.php?attachmentid=67634&stc=1&d=1382775053http://www.aulro.com/afvb/attachment.php?attachmentid=67635&stc=1&d=1382775053

[haydent]

A couple more photo's, this time of the plug cables i made that go inline with for example the ECT sensor or Oil pressure switch, this way you are not cutting any loom wires. The plug type are standard fuel injector type plugs but the male variant was a little hard to find.

http://www.aulro.com/afvb/attachment.php?attachmentid=67754&stc=1&d=1383004571http://www.aulro.com/afvb/attachment.php?attachmentid=67755&stc=1&d=1383004697

[Xenon149]

does look impressive Mp3Car Custom Car Computers! (http://store.mp3car.com/SearchResults.asp?Search=fusion+brain) would def be good if you need that much, i dont though.



prob a bit out of my scope at the moment sorry...


This site stocks some pretty good stepper motors.
Stepper Motor: Bipolar, 200 Steps/Rev, 20×30mm, 3.9V, 0.6 A/Phase | Australian Robotics (http://www.australianrobotics.com.au/products/stepper-motor-bipolar-200-steps-rev-20%C3%9730mm-39v-06-a-phase)

[pickstock]

didnt even realise this was here, now to spend some time going though it.

Im just getting into arduino
do you have a project blog or anything? id love to do this and any other projects around

[haydent]

I don't but was thinking of creating something to put it all together. Glad to get some interest. The project has been delayed in testing since my last post as the zf reco'd transmission in it only lasted 11 months and the tc died . So it's been getting replaced under its warranty. ... In told zf don't even do their own tc's , contracting it to someone more specialized. ..

[pickstock]

would be awesome if you could, i started putting one together but 17 pages was a little much.
If theres anything i can help with gimme a yell.

Ill send you a pm about it all mate

Makes sense though, we have a contractor to come in and do a lot of our mechanical work. Im able to do it but the risk is better off being with him.

[haydent]

Put 17 pages together for what?

[haydent]

I have a website I'll likely put it on. ..

[pickstock]

i meant go though this thread and pull the bits of information like what you used and sensors etc

[haydent]

Ok yeah. There's actually very little technical info just photos. I'll write it up.

[pickstock]

how you going with this mate?

[haydent]

funny, i was just working on it: https://github.com/hayden-t/ArEngineMon

adding photos and a readme, about to add the source

[pickstock]

YOU SIR ARE A GOD!!!

[nobbyclrk]

Nice project. :)

If it starts getting easy and you want to improve the graphics to WOW! Check these displays out. 4D Systems | Turning Technology into Art (http://www.4dsystems.com.au/product/3/24/Arduino_Display_Modules_and_Shields/4Display_Shield_128/)

Touch screens too so you can put the buttons on the screen. The company is from Penrith NSW.

[haydent]

thanks, one may consider that type of screen if the need arrises. it really depends how much you want to display on screen.

[mzs]

Here is an OBDII Arduino kit: OBD-II Adapter for Arduino | ArduinoDev.com (http://arduinodev.com/hardware/obd-kit/#library)

This is suitable for statndard OBD protocols, but you can freely program the device on demand.

Maybe it's possible to discover the TD5 non-standard protocol too (protocol, frequency, baud rate etc.), and write a special program to

It should be possible because i have Autocom CDP/ Delphi diagnostic unit and this can connect to the TD5 ecu, can read/clear fault codes, and monitor many parameters on the fly.

[haydent]

definately something ive thought about but something i wont bother with.

what your talking about is essentialy its a nanocom

[mzs]

definately something ive thought about but something i wont bother with.

what your talking about is essentialy its a nanocom

Let's say, but for the seventh of the Nanocom price.

[haydent]

yeah but its a fair amount of work for a select model, whereas what ive bade could be put on any vehicle, and does stuff you cant get via nanocom, though you could add it via aux inputs, eg coolent level

[mzs]

I agree. I think your project is great, and the ecu independent and extended sensor reading is also a very good point.
I only wanted to inform you that your hardware and software can easily extended with an OBD reader using only a serial port of the Arduino to get information from the ECU.

[haydent]

I agree. I think your project is great, and the ecu independent and extended sensor reading is also a very good point.
I only wanted to inform you that your hardware and software can easily extended with an OBD reader using only a serial port of the Arduino to get information from the ECU.

normally yes, but i believe the td5 has a propriety diagnostic protocol no ?

[haydent]

here's a couple of photos of a common ebay float switch that i shortened by cutting a bit out and re glueing & "plastic welding" the end back on.

this makes it short enough to fit in the end of a td5 header tank :)

[Fraz]

That's right, the TD5 doesn't conform to OBD standards

[haydent]

just finished installing 'new' modded header tank, and the float switch works. how long it will last i will find out.

i couldnt bring the engine up to temp and pressure as i didnt realise on the header tank i just installed there is a extra drain outlet that i need to block up first...

so just a reminder all the source code is still available here: with photos and description
https://github.com/hayden-t/ArEngineMon

some day i hope to design a pcb and housing to finish it off.

[stobbie]

Great work. I want something like this for my Land Rovers, but it would also be great for my boat to mesure EGT and boost.
Haven't done anything with Arduino yet, so I will have to start studying first I guess.
Any plans for EGT and boost readouts?

[haydent]

arduino is very easy and affordable. order a 'nano' or two off ebaey, thats what i use for this as its enough.

EGT, Boost and any other sensor would be possible too, not that i have immediate plans for them it would be easy enough done though

[Fraz]

I'm working on an expanded version of this. Currently running an Arduino Due as a display driver to output a 256 colour VGA signal to a 7inch colour monitor left over from an old carpc project. As the VGA system is so CPU intensive, I'm using anther arduino uno to measure everything and send it all to the display driver via serial.

There isn't much limit as to what you can measure - it's getting the maths right that's the tricky part

[haydent]

Just finished fitting one of these common oil pressure senders along with a 1/8" 27 NPT T piece to the engine: all up only about $40 off ebay australia !

now i just have to code support for it.

the sender is 0 - 10Bar translating to 3 - 160 R (ohms)

http://www.aulro.com/afvb/attachment.php?attachmentid=81800&stc=1&d=1407464741

[3lud13]

Started reading this the other day very good ideas here and something I am interested in. Just spent the last 20minutes trying to find it again to continue reading so now with a comment its much easier to find.

[haydent]

thanks ! unfortunately i cant edit he opening post so its hard to present an upto date summary of the project state, i can understand people not wanting to read through 13+ pages for a few relative posts...

check out my github page too.

[d2dave]

thanks ! unfortunately i cant edit he opening post so its hard to present an upto date summary of the project state, i can understand people not wanting to read through 13+ pages for a few relative posts...

check out my github page too.

So do you have a unit available that I can buy?

[haydent]

not yet, but i could make one up or maybe make some instructions/schematic

would you go more for diy or off the shelf ?

[d2dave]

DYI if it is not hard. My definition of hard is when programing is involved.

If it is just a matter of following instructions and some soldering I would love to DIY.

I recently came across a site by a bloke called David Pye from the UK. He was making kits for lowering SLS in D2's and was happy to share the info for DIY.

He built the kit using Arduino but after communicating with him I found out that an eprom(I think that was what it was)had to be programed.

See posts 31 to 34 here.

http://www.aulro.com/afvb/discovery-2/43361-sls-remote-4.html

[haydent]

great, as it will be easier at this stage of development for me to provide a schematic. would that be ok? or would you need a pcb by itself ? or pcb with components ?

its arduino based and you just load my code onto it which is 1 click via usb

currently i use it with the factory coolant temperature sensor, factory oil pressure switch, and float switch added to coolant tank. im just testing an oil pressure sender too.

theoretically it would work with any sensor. (apart from k-type thermocouples as they put out such small voltage that needs amplification that i have not got in this circuit, though it could be added)

[d2dave]

The stuff is cheap so not really much to lose if I stuff it.

I would love to have a crack so long as you don't mind helping.

[haydent]

so how do you want to do it ?

[haydent]

Updated screen layout
Temperature & Oil reading/maximum/alarm

http://www.aulro.com/afvb/attachment.php?attachmentid=82433&stc=1&d=1408526917

[3lud13]

Yes, I can't believe how cheap all the stuff is.
One more thing - is there an RFID Shield that could be used to make some sort of ignition activation circuit - so if you are close enough with the correct RFID tag, the car could be started without a key like some of the new flash cars out there...?

Im currently thinking the RFID would be better used as a sort of immobiliser just keep it on a keychain or something I guess but not let your car start unless its within range (not sure if you could code multiple RFID chips if you had multiple sets of keys but couldn't see why not


I picture something like: each reading in a box/circle with a background colour that changes based on the reading. I.E. All reading normal would be a bunch of green boxes with digital reading printed inside each. The background colour slowly changes to orange, then red (using shading in between) to represent the status. This means for the most part you never even have to read the screen.. (Except when you want to see all the cool things you're tracking :cool: )
Love the idea of changing led colour depending on temps


It's a great way to get classic looks with modern performance. I'm all for it.:)

It's the way that pretty much all cars have been running their instruments for at least a decade now. It's only recently that a few have one LCD display that looks like gauges (current RR etc).
LCD display that looks like gauges would be quite cool and if I have my way would be what I prefer.



currently i use it with the factory coolant temperature sensor, factory oil pressure switch, and float switch added to coolant tank. im just testing an oil pressure sender too.
theoretically it would work with any sensor. (apart from k-type thermocouples as they put out such small voltage that needs amplification that i have not got in this circuit, though it could be added)
Im interested in the float switch you added to your coolant tank I believe you have a td5 however I have a 300tdi wondering if something similiar could be hooked up with the 300tdi cooling system
Also a huge thanks for kicking my brain into this train of thought
haydents website so I can find it easier later (https://github.com/hayden-t/ArEngineMon/tree/master/images)
Currently I am thinking this could be a great project for the dash of my car and engine too I guess. Im thinking arduino could monitor all my engine pressures, temps, fluid levels for anything my engine requires at a fraction of the cost of buying actual gauges.

[haydent]

Sure. You monitor and display any sensor. There's a white float switch that I would try first in my pictures. It is smaller and easier to fit. It will likely be even easier in your header as it is not compartmentalized ?

[splatters]

If you are still interested im looking at your source code so i can re-start this little project of yours, PM me for a phone number and we can chat about arduino, for the benefit of the group

[3lud13]

I'm still interested though no longer for a disco I upgraded to a perentie so dash space is at a premium I wanted to purchase one of the 5 channel temp sensors from the shop on this site but when I got around to it they had sold out so currently might have to dig this information up again and work on a temp sensor arrangement to keep an eye on things.

Sent from my SM-G900I using AULRO mobile app

[steveG]

Doesn't look like Hayden has been around here for a while, but was wondering if anyone is currently playing around with Arduino gauge projects.

After some cold nights and a recent thread on the 1Gauge (http://www.aulro.com/afvb/communications-car-audio-electronics/238987-1gauge-some-questions.html) I decided to dig out my dusty box of Arduino bits (mainly a Arduino Uno, 16x2 LCD shield, and some DS18b10 one-wire temperature sensors).

If progress stays ahead of the enthusiasm decline my plan is to put together a data logger for the county with

A few DS1820 temp sensors for head temp
RPM
K type thermocouple inputs for EGT and turbo compressor temp (since they'll exceed the 127degC limit of the DS1820's)
Speed
Battery voltage (just because its easy!!)
SD card for logging
Display of the various values
Visual and audible alarms


So far I have working setup for 3x DS18b20's and a functioning frequency meter with optocoupler isolated input to take "W" wire output from the alternator (already have the W wire installed and currently drives a normal VDO tacho).

Does the County need such a logger?
No
But that's not a good enough reason not to play around and make one ;)

Steve

[Judo]

My enthusiasm curve turned out to be inversely proportional to time.

I look forward to your updates tho! :D

[steveG]

My enthusiasm curve turned out to be inversely proportional to time.

I look forward to your updates tho! :D

Time spent, available or just elapsed?? ;)

Seems Hayden is still doing Arduino stuff, but no longer able to post in tech sections http://www.aulro.com/afvb/showthread.php?p=2574629

Best I start my own thread rather than continue to hijack this one...

Steve

[BennehBoy]

Bit of a thread Resurrection here, but I've got a micro-controller based gauge system in development...

It started out life on Arduino Nano, but I rapidly ran out of GPIO & memory so it's now moved onto STM32 (32bit 76mhz vs 8 bit 16mhz).

All the code is here -> GitHub - BennehBoy/LRDuino: In car multi gauge system - Arduino, 0.96" SPI OLED, MAX31856 (https://github.com/BennehBoy/LRDuino)

I've got one of my early Uno based prototypes in my TD5 Disco:

https://www.aulro.com/afvb/images/imported/2017/02/66.jpg

It turns out that 8 of those OLED screens fit perfectly in a CB cutout so I've upsized the bench test system. It now has a digital inclinometer with animated vehicle images, and it also will display readings from an ELM327 type OBD dongle via Bluetooth.

https://www.aulro.com/afvb/images/imported/2017/02/67.jpg

The screens are not blurry at all, my phone cam just doesn't like taking pics of them.

I'm currently porting Luca72's (the guy who wrote td5 map editor) td5 opencom to the same microcontroller, it provides a cut down version of nanocom. Datalogging to SD card, fuelling info, fault reading & clearing.

You can see it running in instrument mode here, it's hooked up to an Arduino Uno which is running as a basic td5 emulator. I'm using a logic analyser to sniff the serial connection between the two, it makes life a lot simpler.
https://www.aulro.com/afvb/images/imported/2017/02/68.jpg

The idea is that I will then integrate this into my multi gauge system.

[VladTepes]

Just saw this thread and am very interested in these type of projects, always looking for something like this at an affordable price for the ute, as well as possible projects for the bike 'just because'.

Oh and I have a Defender so just in regards to the above pic, I don't need that 102 mile per hour bit.... [bigrolf]

[BennehBoy]

Oh and I have a Defender so just in regards to the above pic, I don't need that 102 mile per hour bit.... [bigrolf]

Yeah it's a bit optimistic even for my D2 - big tyre's and lot's of offroad toys :)

[vanaxel]

Great project. Im start to look at arduino to get information from the ECU. This info is very good.

[steveG]

Nice work BennehBoy - those individual OLED's look great.

Steve

[BennehBoy]

The porting of the opencom code has been a fourfold task...

1) I had to make it work on STM32 (Maple Mini) which is a 32bit architecture, this meant removing a lot of PROGMEM code. (ardunio uses PROGMEM code to move static definitions into program memory rather than RAM, STM32 does this automatically)
2) I had to uplift many areas of code to the current Arduino IDE way of doing things. The original code was written on 1.0x and the IDE is now at 1.8.1.
3) I had to swap the code from using an LCD display library to using an OLED library (because I did not have an LCD to hand).
4) I had to physically rewire everything and change the code to use the different pinouts.

All of that was fairly straightforward.

However, in the process I found that one of the submenu's would not work, this was the ECU menu, which contains some of the interesting stuff. I don't know if I've broken this, or it's a combination in the way the IDE and/or STM32 work.

So, I ordered an LCD Display. I then used a very old version of the IDE to compile the original code onto an Arduino Mega, and the menu's work. I needed to check this because there are areas in the code that are obviously incomplete and I did not know whether the menu was non functional due to my changes, or if it was broken anyway.

Why am I telling you this?

Because I'm going to refactor all of the interesting parts from Luca's code into my gauge project using a different menu system.

So, next steps are to:
1) get an alternate menu system working on STM32 -> GitHub - neu-rah/ArduinoMenu: Arduino generic menu/interactivity system (https://github.com/neu-rah/ArduinoMenu) (at present it breaks due to a difference in the compiler standards, but this ought to be a simple fix).
2) Merge Luca's comms code into my LRDuino project
3) Uplift all the hardware from breadboard onto vero/prototyping board

Lot's to go at and never enough time :)

[BennehBoy]

My project now has a menu system and can talk to an emulated TD5 Ecu.

I've got an MSB ECU here with topside switch failure that I'll be bench testing against once the econoseal connector for the black plug arrives in the post.

Assuming all goes well there, I've got a little bit of K-line sniffing to do to get my code handling fault reading/clearing, then it's time to build out on stripboard for the car.

Here's a photo (although caught the displays mid refresh) of the system reading a few PID's from the emulator (running on an Arduino Uno with LCD screen):

https://www.aulro.com/afvb/images/imported/2017/03/530.jpg

[BennehBoy]

Datalogging to micro SD card now working.

I'm building a cut down version for a friend, he just wants one display with a button to rotate around 4 analogue sensors, Boost, EGT, Fuel Pressure, Oil Pressure.

Parts come to about £20 GBP, Sensors if bought from Ali Express cost about £40 all in. I priced the same from a commercial provider and sensors alone come to £180 (for no better quality I might add).

I'm still waiting on some parts to arrive so that I can get the ECU hooked up on the bench.

[AK83]

Datalogging to micro SD card now working.

....

Sounds good!

How hard is it to set it up to a clock of some type, recording date and time.

Makes it a lot easier to keep a track of the log data in the long run.

This is the only annoying aspect of the 1Gauge I fitted to the D1.(I'm not worried about ECU interface and so on).

what I have to do now with my logs:
I download the (many multiples of)log files saved to it's SD card.
I sift through the very short files. They may only be a few mins or so, and I'm not that pedantic. Anything less than 10Kb I delete.
The remaining files are sorted by file name, and collated into one large file(command copy *.csv <filename>.csv) this produces one large csv file for the month.

Thanks to a member here for the tip .. I then edit and adjust the time values column to continue on for each logging session. So if log0001 has a time value of 200 sec max, then log0002 continues from 200 and beyond.
If this isn't done, it's impossible to see the entire log in one session as a graph, as it loops back to the previous logs start time.

I adjust the time value on the X-scale of my log viewing software so that it displays the time in a readable format .. ie. 3 days 12 hrs 31sec ... not 269012 sec(which is meaningless to a human! :D)

Having a clock interface that can keep track of time would be handy(for the log data).


.. anyhow, just a suggestion.

[BennehBoy]

Sounds good!

How hard is it to set it up to a clock of some type, recording date and time.

It's fairly straightforward, a real time clock (RTC) module needs to be included in the solution, I had considered doing this and/or hooking up to GPS which also provides time. The output could then be combined with a mapping service like Google to show performance information at any given geographical location.


Makes it a lot easier to keep a track of the log data in the long run.

This is the only annoying aspect of the 1Gauge I fitted to the D1.(I'm not worried about ECU interface and so on).

what I have to do now with my logs:
I download the (many multiples of)log files saved to it's SD card.
I sift through the very short files. They may only be a few mins or so, and I'm not that pedantic. Anything less than 10Kb I delete.
The remaining files are sorted by file name, and collated into one large file(command copy *.csv <filename>.csv) this produces one large csv file for the month.

Thanks to a member here for the tip .. I then edit and adjust the time values column to continue on for each logging session. So if log0001 has a time value of 200 sec max, then log0002 continues from 200 and beyond.
If this isn't done, it's impossible to see the entire log in one session as a graph, as it loops back to the previous logs start time.

I adjust the time value on the X-scale of my log viewing software so that it displays the time in a readable format .. ie. 3 days 12 hrs 31sec ... not 269012 sec(which is meaningless to a human! :D)

Having a clock interface that can keep track of time would be handy(for the log data).


.. anyhow, just a suggestion.

Alternatively the data could be exposed via bluetooth and an Android app used (it looks fairly simple to build an android app to do this, the device's GPS could then be used).

[BennehBoy]

BTW, now that this project is drawing to a close I'm looking about for what I might do next...

A 4HP24 controller to replace the one that ships with the D2 (and as a compushift alternative) is on the cards...

There's a superb aftermarket automatic controller available for the Merc 722.6 gearbox, it's a great bit of inspiration toward building something halfway decent for LR users. See here -> Home (https://ofgear.dk/)