Disco 2 Three Amigos refuse to die

This site contains affiliate links for which LandyZone may be compensated if you make a purchase.
NPG, Sierrafery, so lovely to hear all this. Don't understand a word, but this language has a lilt to it all of its own!!!!! (So glad there are guys like you amongst the members, where common-or-garden, shed mechanics like me would be completely stuffed by what seems like rocket science.)

Thanks Stanleysteamer. It wouldn't be such a nightmare if manufacturers and standards bodies agree for once to keep things simple and don't try to segment the market and obfuscate information in order to increase revenues. Just look at the phenomenal success story of the IBM PC which came out early 80s. It maintained more or less the same architecture, is based on open standards and you can still run software written 40 years ago on modern operating systems. But then, we cannot even agree on the shape of a 240V AC plug let alone.....

If DIY mechanics (myself included) think our cars are complex, wait when we're forced to use fully electric, self-driven vehicles. Maybe we'll just manage to change the wheels once in a while and even that I'm not so sure how easy it would be...probably you'll need special equipment to calibrate, align and so on :)
 
IBM PCs! Hmm!, I learned at school how to program a computer using paper tape, in Fortran. Lots of fun. Now I have absolutely no clue at all what is going on. I can still remember my dad's excitement coming home and telling me he'd seen a VDU! (He was an accountant at the dawn of computing.) I used to train staff on how to use Acorns for early Windows, Wordprocessing, Excel etc. Now things on cars in particular have just got beyond me. If I'd been an IT teacher instead of MFL I'd be better off.
 
So you'll probably be surprised when I say that the end-user diagnostics software communicates with the ELM327 chip inside the OBD scanner via a standard RS232 serial interface (UART)...a protocol that came out early 70s. Furthermore, the software talks to the ELM chip with Hayes modem AT commands (published early 80s) and does support the (in)famous ATZ command to reset the soft configuration.
 
So you'll probably be surprised when I say that the end-user diagnostics software communicates with the ELM327 chip inside the OBD scanner via a standard RS232 serial interface (UART)...a protocol that came out early 70s. Furthermore, the software talks to the ELM chip with Hayes modem AT commands (published early 80s) and does support the (in)famous ATZ command to reset the soft configuration.

You know I said I was an MFL teacher? Well, thanks for the info, but I simply don't understand it, it really is a new language to me!!! But I do appreciate the dates. My wife who used to be an international IT auditor, tells me that many systems banks use, especially in Germany, are really old fashioned no matter how fancy their initial appearance. At the end of the day, if security isn't compromised, if it is tried and above all TESTED, and it ain't bust, don't fix it! Wifey says most of the problems in the systems in banks and other huge companies are either due to users not following absolutely basic security protocols, (like she used to see managers whose password was "password") OR the systems hadn't been properly tested. As she said "that cost money". Another friend of mine, a computer programmer who worked on defence systems, resigned and never went back to it, because testing was insufficient and he could not honestly sign off on systems he didn't feel were really safe. Now we have, not one, but two big Boeings crashing due to software that had not been properly set up. Never mind the fact that the computer forced the the plane out of the pilots hands. It doesn't matter if you have three separate systems to fly the plane on, in case two of them go wrong, but if they are all running the same software....?!
Back in the days when I was an engineer, the first thing I was taught when designing circuits, (pneumatic control circuits), was to ensure that all systems would fail safe, And also there would be a big red button to hit if things went wrong. If Boeings had a big red button, it could be hit and control of the throttles, rudder, ailerons and tailerons, could be made manual again so the pilots could actually FLY THE PLANE. Might save the odd life? And they want us to have driverless cars and driven cars with speed control? Lord save us.
 
Last edited:
So you'll probably be surprised when I say that the end-user diagnostics software communicates with the ELM327 chip inside the OBD scanner via a standard RS232 serial interface (UART)...
I'm not surprised that it communicates with it but i'd be very surprised if the retrieved data or codes are correct cos other OBDII scanners are comunicating with the ECU but most of the data is trunkated
 
You know I said I was an MFL teacher? Well, thanks for the info, but I simply don't understand it, it really is a new language to me!!! But I do appreciate the dates. My wife who used to be an international IT auditor, tells me that many systems banks use, especially in Germany, are really old fashioned no matter how fancy their initial appearance. At the end of the day, if security isn't compromised, if it is tried and above all TESTED, and it ain't bust, don't fix it! Wifey says most of the problems in the systems in banks and other huge companies are either due to users not following absolutely basic security protocols, (like she used to see managers whose password was "password") OR the systems hadn't been properly tested. As she said "that cost money". Another friend of mine, a computer programmer who worked on defence systems, resigned and never went back to it, because testing was insufficient and he could not honestly sign off on systems he didn't feel were really safe. Now we have, not one, but two big Boeings crashing due to software that had not been properly set up. Never mind the fact that the computer forced the the plane out of the pilots hands. It doesn't matter if you have three separate systems to fly the plane on, in case two of them go wrong, but if they are all running the same software....?!
Back in the days when I was an engineer, the first thing I was taught when designing circuits, (pneumatic control circuits), was to ensure that all systems would fail safe, And also there would be a big red button to hit if things went wrong. If Boeings had a big red button, it could be hit and control of the throttles, rudder, ailerons and tailerons, could be made manual again so the pilots could actually FLY THE PLANE. Might save the odd life? And they want us to have driverless cars and driven cars with speed control? Lord save us.

Agreed and there's so much one can say on the topic you've raised. Unfortunately I'd say, machines have entered the man-made eco-system of politics, finance and law where instead of them being there for the service of man, man has now become their slave. Talking about automation, what we have so far are machines that are either preprogrammed (i.e. procedural with fixed logic) or are capable of exhibiting some form of thinking with the use of an inference engine (AI-based). However, none of these machines can yet perform the creative problem solving that humans can do and therefore while human control could be assisted in many ways, it should never be relinquished in the execution of critical functions such as driving a car or flying a plane. There is another field that is highly controversial but emerging nonetheless - machines that are aware of themselves and their surroundings - very much like saying they have a consciousness. I'd say God help us all if this is where we're heading.

BTW, Alan Turing had put forward some really simple but ingenious ideas on how to tell humans and machines apart on the basis of their intelligence and thinking faculties, with the most well known being the Imitation Game.
 
I'm not surprised that it communicates with it but i'd be very surprised if the retrieved data or codes are correct cos other OBDII scanners are comunicating with the ECU but most of the data is trunkated

Surprisingly enough, the ELM chip can support lossless high throughputs on its RS232 interface, most notable at 57K6 and even going up to 115K2, with some limitations, especially wrt CAN. Some insertion delays are unavoidable for error checking and conversion of OBD data to ASCII but we're talking few μs. Most OBD standards include error checksums and also require checksums to be incorporated with the requests, which ELM does automatically. Some rounding does occur on floating point notation but the end-user software can make use of the chip's advanced facilities to calibrate parameter accuracy.
 
Surprisingly enough, the ELM chip can support lossless high throughputs on its RS232 interface, most notable at 57K6 and even going up to 115K2, with some limitations, especially wrt CAN. Some insertion delays are unavoidable for error checking and conversion of OBD data to ASCII but we're talking few μs. Most OBD standards include error checksums and also require checksums to be incorporated with the requests, which ELM does automatically. Some rounding does occur on floating point notation but the end-user software can make use of the chip's advanced facilities to calibrate parameter accuracy
i'm not gonna contradict you in your well argumented theory... what i can say for sure is that i plugged in ELM in my own Eu2 and and two other Eu3 D2s then compared the results with hawkeye(which is made by Omitec based on Textbook T4 dedicated LR software) and the results were different or some of them completely missing, i've got false fault codes about boost and air inlet temp which were not even stored and the only correct live data was about ECT and RPM all the others were shown but not accurate... i see that you know your stuff but IMO there is some hiddent code in the proper software made on purpose by the developers as to not be easy to emulate so better dont eat your mind by trying to read/diagnose Td5 with OBDII protocol
 
OK sierrafery, I understand where you're coming from. The reason I had mentioned ELM 327 in post #24 is to illustrate the benefits of protocol standardisation and the applicability of legacy protocols such as async RS232 and Hayes AT commands in modern technology.

However, I'm almost sure that ELM 327 will not work properly on a TD5 (V8 perhaps) for reasons that you mention and also because LR seem to use proprietary codes and non-standard signalling for timing, initialisation and arbitration between electronic modules on these vehicles. I've been there myself and although my ELM device works 100% on more recent vehicles, the TD5 proved to be a challenge. Maybe another microprocessor such as the MC33290 would work as it's specifically designed for K-line protocols. But I agree - unless you're developing a product to market, it's not worth the trouble.

Cheers
 
OBDII should work fine on a D2 era V8 (US versions are certainly compliant and rather smart OBDII dashboard displays are available for them). Compliance for petrol engines pre-dates the requirement for diesels.
FWIW I have tried OBDII using ELM327 on D90 and D2 TD5s using TOAD software and got nowhere.
 
Agreed and there's so much one can say on the topic you've raised. Unfortunately I'd say, machines have entered the man-made eco-system of politics, finance and law where instead of them being there for the service of man, man has now become their slave. Talking about automation, what we have so far are machines that are either preprogrammed (i.e. procedural with fixed logic) or are capable of exhibiting some form of thinking with the use of an inference engine (AI-based). However, none of these machines can yet perform the creative problem solving that humans can do and therefore while human control could be assisted in many ways, it should never be relinquished in the execution of critical functions such as driving a car or flying a plane. There is another field that is highly controversial but emerging nonetheless - machines that are aware of themselves and their surroundings - very much like saying they have a consciousness. I'd say God help us all if this is where we're heading.

BTW, Alan Turing had put forward some really simple but ingenious ideas on how to tell humans and machines apart on the basis of their intelligence and thinking faculties, with the most well known being the Imitation Game.

The main problem with the aeroplane in question is that is has been designed, or misdesigned, with a fault which means that it has a tendency to stall, something to do with the positioning of the engines which rendered it inherently unstable, which, again, in these days of computers, we are supposed to think is "OK". The computer deals with this tendency, normally, but the pilots have to be aware of it, fully, and know how to work around it when they are panicking and too close to the ground. This inbuilt inherent instability is apparently OK as it makes it more fuel efficient. So all these people died cos its OK to risk lives in order to save a few gallons of aviation fuel. I no longer ever fly at all. Although I would love to be able to, as a private individual, but it costs so mch to learn how to become a pilot, my ATC days are long behind me! I agree entirely with your post. I do not want computers taking risks with my life!
 
So I finally got my hands on a better diagnostics tool, as it was cheap I also did the shuttle valve mod. The diagnostics point to the front left signal being low, which does appear to be the case as I can see that when driving the front left ABS sensor is reading a lower speed than the other three. I will be trying the "remove the o ring" fix that I've seen in a few posts on the same error code. But just in case this doesn't work I thought I'd ask here to see what the other possible causes might be?
 
Could just be needing a new sensor, or it could be that hub is worn. You might want to jack it up and check. A replacement hub comes with a new sensor already fitted. As always beware pattern parts; this is probably a time when paying extra for OEM may be worth it.
 
Could just be needing a new sensor, or it could be that hub is worn. You might want to jack it up and check. A replacement hub comes with a new sensor already fitted. As always beware pattern parts; this is probably a time when paying extra for OEM may be worth it.

Sorry, I forgot to mention that I've already replaced both front sensors. Okay so worst case I need to order a new hub, good to know. I'll investigate further, thanks
 
So I finally got my hands on a better diagnostics tool
What's that? name that tool... cos if it's ICarsoft you can forget about it and consider that it's not better nor worst .... it's just full of sh*t when it comes to D2... it's quite ''normal" with other LRs but not with the D2. Period
 
If it is any help I have the icarsoft i930 which is £96.00, I find it lots of help on the D2 V8, it has more than paid for itself.

On 3 amigos it showed a SVS fault, I changed the shuttle switches which "cured" it for a couple of months, the same fault and 3 amigos came back, I did the modification and it has been good for 18 months.

This mod can be done without removing the block and disturbing the hydraulics.
 
If it is any help I have the icarsoft i930 which is £96.00, I find it lots of help on the D2 V8, it has more than paid for itself.
The iCarsoft diagnostic testers are pretty good for petrol engines but are a bit sketchy for diesel engines.
I had the iCarsoft CR Plus and it had issues misreading the ECU on my wife's 2011 diesel BMW X5. I got rid of it and bought a Foxwell NT510 which worked better.
 
Back
Top