Custom lettering on radio display??

[sbridgey]

There is no guide, certainly not a basic one!

I am a Software developer so I am sure I can figure it out. If you could just post the basic stuff you are using, that would be really helpful

 

[Chet]

Hardware doesn't require anything special, any arduino or similar would do.

I did have it running on an attiny (as in the boards above) but it's not working now. I started a new job this week so haven't had a second to look at it so I've no idea what's wrong.

 

[Dr HMS Derv Destroyer]

There is no guide, certainly not a basic one!

What ecu?

standard dci ecu. guessing i could take the signal
from the boost sensor alrrady on the car?

 

[Chet]

If it's not readable via obd you could add a seperate sensor

 

[sbridgey]

Am i right in thinking, you use a bluetooth OBD sender, connect to it using bluetooth on the Arduino, then run the data through a parser and display the output on the radio display?

 

[Chet]

I'm not using bluetooth but yeah to the rest!

 

[Hudsonkh]

Mine says kenwood on mine so u must be able to

 

[Andy_con]

how are things coming along whoever is working on this?

 

[Chet]

I still am but i have no time at the minute so no updates. I did find a mistake in the little boards so i think i might just sell them off as preprogrammed static displays for anyone that wants them. Still haven't managed to even get one working yet though...maybe at the weekend

 

[CRS-Mike]

I still am but i have no time at the minute so no updates. I did find a mistake in the little boards so i think i might just sell them off as preprogrammed static displays for anyone that wants them. Still haven't managed to even get one working yet though...maybe at the weekend

nice, il be after one. would that also work with an aftermarket radio /stalk adapter tho?

 

[Chet]

No these will only work if you're not using the standard stalk or radio

 

[CRS-Mike]

No these will only work if you're not using the standard stalk or radio

i see.... so i wonder if the connects4 adapter unit which is obviously forcibly displaying "renault" can itself be re-programmed to display different. hmmm

 

[Ezra-RS]

Does someone know which obd protocol the clio (172) uses? Is it ISO14230 (kw2000)?

 

[Chet]

Yes. If you look at my vid with the info on the display it shows the supported pids

 

[Chet]

i see.... so i wonder if the connects4 adapter unit which is obviously forcibly displaying "renault" can itself be re-programmed to display different. hmmm

Somebody said (somewhere) that one of the adapters could be preprogrammed but I haven't seen anything displaying anything exciting

 

[Deejuk]

Was checking this out yesterday as I bought my clio with a connect2 stalk adaptor in it and trying to work out which one it is... But if you have a CANBus later model and using the one for 05> you can update them!

http://www.connects2downloads.co.uk/c2/files/downloads/catalogues/SWC.pdf

 

[CRS-Mike]

not sure if i have the 004 or 005 one, will look later. so you can update them, but can you change the text in the update haha. had a look at that pdf, still pretty kool that you can do that tho

 

[Ezra-RS]

How's it going with the project, Chet?

 

[Chet]

Not much change. I'm hoping to talk to a supervisor in university about taking this on as my masters project which will mean time to spend on it and will also mean it would be well documented.

 

[Jono 69]

I am very interested in this too!! Hope there is some progress, seems a shame to lose out on all the work so far!

 

[Chet]

Ok, here's my intentions for this:

There will be a plug-in box that connects to the OBD port via cable and and also the radio/display. You will disconnect the display from the radio and plug it into this box, the box will connect back where you removed the plug.

With the OBD cable connected the box will be powered from it so it will not need connecting to external power. If the OBD cable is not connected the box will be able to source enough power via the radio connector to pass the radio signals to the display so it will work exactly as standard.

Assuming its connected to the OBD port the box will be activated via the stalk control by some command (probably the holding of a button for a few seconds) Once activated the display/stalk will then be under the control of the box. The radio will work as normal but would need to be controlled from the radio itself.

The box will display the available ECU parameters on the display, you will cycle through them using the stalk. There will also be a 0-60 timer and whatever else can be added - I will be open to suggestions.

If the OBD is connected and the box is not activated you will be able to connect your phone to it via bluetooth and use any of the phone apps to connect to the car. In testing I have seen a minimum of a 40% increase in connection speed via bluetooth compared to the ebay bluetooth obd modules.

​I will be looking for 3 testers to take one of the small boards I posted about before and fit it in the car, it will mean disconnecting the radio from the display temporarily while its installed and you will also need to wire it into a 12v supply. Don't volunteer unless you are comfortable doing this! I will be looking for some money to cover the costs (and to stop people running away with my board!) - something like £10. They will also need to be ok with posting the boards back to me for firmware updates if the need arises, I'm in ireland but postage shouldn't be too much from the UK. I have a number of other boards but I'm wanting to get a range of cars tested, if you have more than just a clio with a tuner list radio let me know. Testers will get a discount off the final units (if they want one) but I have no idea what they will cost, but they won't be cheap for a short run.

If someone has custom requirements for something they would like controlled via the stalk and display I will be able to help with that.

This will be an open source project, once finished and documented everything will be made available so you can make your own.

Any thoughts?

​

 

[aliengravy]

Hi Chet....

I am a recently graduated Electronics Engineer myself. Now a self-employed freelance engineer for various companies...

I've been following you since the start with this project as I have always wanted to do something similar myself. Unfortunately I just haven't had the time to get out there and start from scratch like you have done, and I think highly of you for that. Unfortunately I have too many personal projects of my own including a full EFI system for automotive use (my final year project, and now a commercial project) to do this.

I think its a great idea as a university project to do, but I think there is a commercial application for this, all be it a slightly limited market, I wouldn't knock it. Because of this, as much as I would love to know the protocol ect... I think you should probably keep your cards close to your chest with this one. Although I think it is very kind of you to say it would be open source, I think you should use this as a university project, and then market it after! (Make sure you ok this with the uni first.. they like to keep the IP of your final year project if you don't state your intentions....)

Having said all of this, I would be very interested to test, and willing to support you with your project in anyway I can.

 

[Chet]

Thanks for the feedback. Fair play for going freelance - are you specialising in any area?

I have still to speak to my university about it but I'll be making sure what i do remains mine. I might market it for a while if the interest is there. As part of my plan the first run would be 10 or 20 which would be part of my project so that would be before anything is released. Its a tough one really, I would like to make some money as after going back to college I've been broke for a long time but i do believe in sharing info too so i don't know what to do!

 

[The Fatty]

you wont make any money by making it open source. If it was me (I also have my own company and make bespoke software/devices and have a BSc & MRes in Computing), I would be keeping I very closed source at least until the potential business value has dropped very low, then then I would only be releasing the methodology, not the source code, circuits etc.

 

[Chet]

It's hard to gauge the size of the market though, I can't see these costing much under £200 to be honest - the connectors alone are mad money. Saying that, i paid more for my old fastchip XRD and i was happy to, that was much simpler too.

 

[The Fatty]

FLOL where the heck did you pull that figure from?! Use pre-built cables (about £10) and RJ45 jacks. PCB & components £20. Sell them for £60 without cable, £70 with cable.

 

[Chet]

You're crazy! When it's designed i will gladly get you to make them for me - i'll give you the £60 for them

​

 

[The Fatty]

You have to be kidding me?! An ELM327 chip is only £10. MCU + PCB + small parts will be no more than £10 (unless you have opted to use some stupidly powerful and unneeded MCU?!) development time is probably a couple of days to perfect, but split across multiple unit sales. Unless there's a part you are using i'm missing?!

I don't imagine anyone will buy them for £200, you will be hard pressed to sell it for £60 imo.

 

[Tunst]

Id be willing to test one for you

 

[Chet]

You have to be kidding me?! An ELM327 chip is only £10. MCU + PCB + small parts will be no more than £10 (unless you have opted to use some stupidly powerful and unneeded MCU?!) development time is probably a couple of days to perfect, but split across multiple unit sales. Unless there's a part you are using i'm missing?!

I don't imagine anyone will buy them for £200, you will be hard pressed to sell it for £60 imo.

A genuine ELM327 is closer to 20 but I'm not using that anyway as they're quite slow. I'm building this the way I want it, not down to a price. Retaining and replacing the existing radio comms costs more than the OBD <-> display hardware. Ensuring the power supply works in an automotive environment is much more than sticking a regulator in. Bluetooth again is a significant expense. The ISO connectors are next to impossible to source and stupidly expensive. It all adds up, and quickly! There is no way you could get a finished unit out for £60 on this small scale.

 

[The Fatty]

Bluetooth?! for what use? I think I may have missed a major jump in the point of this project! I thought you were just displaying current readings on the existing screen cycling every few seconds? - edit: just read your long post at the top of this page, I now see you want to make it Bluetooth enabled for use with apps like Torque.

what's your reason for not using a stabilised regulator protected with a zener? Others seem to get away with that alright. I have never built one for in car use but I cant see why it would be a problem?


I thought you had already sussed the in-car communication bus and interfaced it with your MCU?

The ELM327 chip is not slow btw and can be had for under £10 if you import from china.

I honestly don't see anyone buying it for £200 for a Mk2 Clio when they could have it on there phone for £5 via one of these
http://www.ebay.co.uk/itm/V1-5-ELM3...iagnostic_Tools_Equipment&hash=item485503e8bf - that's just my opinion though, you might get lucky

 

[aliengravy]

Yup you got to pull your cost way down! I can build a full Bluetooth compatible ECU for under £50!
And voltage regulation isn't that bad in the automotive environment now days... Stuffs got a lot cleaner!

http://www.obddiag.net/allpro.html $6!

I specialise in automotive electronics, engine calibration and control systems, and end of line test equipment. All involve PCB design and build, c, C#, java and php....

 

[Chet]

Yep, bluetooth for torque as its awesome. When it became my intention to take it as my masters project it all needed to be over engineered (no getting away with it) and as i said, this is something I'm making for myself firstly so costs wasn't really a consideration. Comms is sorted, having it be able to intercept the display from the radio is more complicated and only half tested so far.

​Those 327's from china are counterfeits. I actually spoke to the author of Torque about my OBD implementation and he was the one that said how slow and unreliable they are.

£200 may have been an over estimate, I honestly haven't sat down and fully planned it out at all. The biggest expense by far is the case (which isn't picked) and connectors. I don't want to get bogged down in this now anyway as aside from this fully functioning prototype that works in my head it's a way off Now that I think about it I imagine the project would look better if it was approached from the point of view of a commercial product rather than a one off.

BTW my XR25PC USB cables sold for up to £75 when the instructions for making them were on my site.

 

[Chet]

Yup you got to pull your cost way down! I can build a full Bluetooth compatible ECU for under £50!
And voltage regulation isn't that bad in the automotive environment now days... Stuffs got a lot cleaner!

http://www.obddiag.net/allpro.html $6!

I specialise in automotive electronics, engine calibration and control systems, and end of line test equipment. All involve PCB design and build, c, C#, java and php....

Is that £50 for just PCB and components? I can see no way of meeting a cost like that for a finished product. BTW my £200 wasn't a cost to make, it was a selling price based on the usual rule of 3's and then bumped down.

That thing is $25?

 

[The Fatty]

​Those 327's from china are counterfeits. I actually spoke to the author of Torque about my OBD implementation and he was the one that said how slow and unreliable they are.

As I understood it, the already built 327 Bluetooth devices only function in the slow mode due to the poor implementation. The 327 chip used in them is capable of high speed data transfer without errors. I have one of the cheap 327 BT devices, and even in the slow speed they are find for reading 1 node at a time. They do get slow if your try and read 3+. Also if you used a 327 chip (v1.3a), you should be able to talk to the radio display and stalk controls using only the ODB port (provide Renault have wired up the CAN bus wires on the ODB port) as they do CAN as well from what I understand.

 

[Chet]

With my cheapo 327 i can get up to 5 pid/sec. I'm looking at getting 8 but that all needs more testing.

The CAN is wired to the port on my 172 but it seems like a very basic implementation, its separate from the tuner list display. The update lists use CAN but again i think this is isolated from the rest of the car.

 

[The Fatty]

http://www.obddiag.net/allpro.html $6!

That's a very interesting kit, they are using a standard USB enabled PIC MCU and loading the ELM327 code onto it; does that mean the ELM327 instruction set is publicly available then? Because that would mean you could implement everything from just 1 MCU. When I played around with it, I used a PIC18F4550 to talk to he ELM327.

 

[Chet]

That's a very interesting kit, they are using a standard USB enabled PIC MCU and loading the ELM327 code onto it; does that mean the ELM327 instruction set is publicly available then? Because that would mean you could implement everything from just 1 MCU. When I played around with it, I used a PIC18F4550 to talk to he ELM327.

The ELM327 is just a rebranded PIC. If you have a look in your cheapo chinese ELM adapter you'll see its most likely a PIC

 

[The Fatty]

With my cheapo 327 i can get up to 5 pid/sec. I'm looking at getting 8 but that all needs more testing.

do you need to read all 8 at once though? Surely you can only display 1 at a time? OR is that for the BT implementation?

The CAN is wired to the port on my 172 but it seems like a very basic implementation, its separate from the tuner list display. The update lists use CAN but again i think this is isolated from the rest of the car.

interesting. I suppose you tried the single wire implementation on pin 1, CAN1 on pins 6/14 and CAN2 on 3/11? (I am assuming is wired as a SAE J1962?)

 

[The Fatty]

The ELM327 is just a rebranded PIC. If you have a look in your cheapo chinese ELM adapter you'll see its most likely a PIC

yes I knew that, but I wasn't aware the code was public...