[DerAlte]

Originally Posted by Sutripta

QNX / AIX? Or some other RTOS? ...

Some other, usually home-grown not-commercially-recognizable ones.
* License cost matters, since the first concern is saving pennies
* Commercial dependence on another vendor is a killer (though the manufacturer depends on vendors completely for parts, for all other stuff all vendors will produce identical parts from the manufacturer's drawing, so changing vendors is easy. What does one do when every RTOS vendor has his own idea what an RTOS should be? A single IP contention case on the vendor, and you have to shut your car production. Who wants that risk?)
* The best of the commercial RTOSs are too wasteful of resources for the purpose. Standardization is required only outside the ECUs - e.g. inter-ECU comm, diagnostics etc. - and there is no UI to speak of, so that is not a concern. Nowadays, customers (vehicle manufacturers; for example BMW and Daimler have their own algorithms for controlling crankshaft angular acceleration limiting) participate in software development, but they anyway have to learn the rules and environment if they want do so.

Originally Posted by Jaggu

... All depends on the cost that the manufacturer strikes up with the automobile company ...
GB talking to ECU needs to be trained? ...

Ekdum correct hai, ji!

And no, Txn ECU doesn't need to be trained. The protocol and the parameter values are worked out *before* the ECU is designed/produced.

Originally Posted by vina

I understood that the transmission and the engine communicate with each other (though I guess, when they both communicate to the ECU, since ECU is the master for both they effectively communicate to each other anyway).

What I didn't understand was what would they communicate and how does it help?

There are 2 separate ECUs - the engine ECU that controls injection etc., and the transmission ECU that controls up/down shifting. No 'master ECU' - risky, costly, unnecessary.

The simplest communication is Txn ECU telling the Engine ECU to reduce engine RPM for a short time to enable smooth gear shift.

[nitrogary]

i loved this thread guys!
spike, it would be a good idea to put together all the questions and take them to Kannan in calibration. he ll give some good insight. points which need more explaination in my view are:
1. zero fuel calibration/zero fuel quantity
2. fuel balance control
3. injector bank balancing
4. injector drift
5. injector code (which is etched on the head of the injector) and how this affects the dynamic corrections in the fuelling when you enter the code in the software.

[SPIKE ARRESTOR]

^^ @Nitrogary, who better than you; who has exposure to one of the strictest emission norms for Diesel engines today? The T2BIN5. Do share your thoughts as well.

Spike

[vina]

Originally Posted by DerAlte

...
There are 2 separate ECUs - the engine ECU that controls injection etc., and the transmission ECU that controls up/down shifting. No 'master ECU' - risky, costly, unnecessary.

The simplest communication is Txn ECU telling the Engine ECU to reduce engine RPM for a short time to enable smooth gear shift.

Thanks

Is there are other way the two can work with each other?

Do the engine control maps depend on the state of the transmission as well (i.e. wheel rpm, gear engaged etc.)?

[headers]

Originally Posted by nitrogary

i loved this thread guys!
spike, it would be a good idea to put together all the questions and take them to Kannan in calibration. he ll give some good insight. points which need more explaination in my view are:
1. zero fuel calibration/zero fuel quantity
2. fuel balance control
3. injector bank balancing
4. injector drift
5. injector code (which is etched on the head of the injector) and how this affects the dynamic corrections in the fuelling when you enter the code in the software.

Originally Posted by SPIKE ARRESTOR

^^ @Nitrogary, who better than you; who has exposure to one of the strictest emission norms for Diesel engines today? The T2BIN5. Do share your thoughts as well.

Spike

Hey guys, it seems both of you are working for the same company. Good to have you on board Mr. Nitrogary.

BTW, why do you think numbers are etched on the head of the injector? What does it signify?

[Sutripta]

Originally Posted by DerAlte

Some other, usually home-grown not-commercially-recognizable ones.
......

Thanks. Samurai did reply

Originally Posted by Samurai

It was a real-time yamos, native to Bell Labs. Had commandline interface like Unix, but behaved liked pSOS. pSOS, now that's something I haven't heard in a long time.

And I would think it was related to System 7 rather than cars.

Interestingly Vijay said that they use Infineon chips (understandable) but using a GCC toolchain.

Regards
Sutripta

[Sutripta]

Originally Posted by star_aqua

The tone wheel can be divided into segments. segment speeds can be measured very accurately by specifying start angle and stop angle.

What is the resolution of the crank position sensor? Signal:- Is it quadrature + index or single stream +index? How accurate is the index (TDC).

Overrun is where we are sure injectors are completely shut.

Sure, but during overrun, crank is rotated by something else (the car). Will that not override everything else.

Code is converted into tester parameters by using a simple excel tool. i haven't checked the logic behind. lemme decode that excel sheet first.

Great. Further questions after this.

A new transmission on an used engine will definitely have torque mismatch which will affect the shifting quality.

What do you mean by torque mismatch?
Take the simple case of standalone ECU and TCU. Does it mean that TCU will change shift points depending on what it 'reads' from the engine. Once again, it will be interesting going through the algorithm.

Regards
Sutripta

[DerAlte]

Originally Posted by vina

... Is there are other way the two can work with each other?
Do the engine control maps depend on the state of the transmission as well (i.e. wheel rpm, gear engaged etc.)?

1. CAN Bus is the most modern method. Earlier there were dedicated I/O connecting one ECU to another. BUT that method needs more pins on the ECU connector.

If you meant on which other functions they can collaborate, there are a few like Hill Climb Assist, etc. in vehicles with AT

2. They are catered to (not Gear engaged, wheel RPM though), but not via maps - instead via auxiliary modifiers. For example, clutch depressed causes (in some ECUs) a slightly higher idle anticipating sudden onset of torque, for example while stopping or starting on a slope (vehicles with MT).

Maps in the Engine ECU only contain normal operating profile (normal = without any other dynamic influence).

Originally Posted by Sutripta

... Interestingly Vijay said that they use Infineon chips (understandable) but using a GCC toolchain. ...

Where?

Infineon, Motorola, and a couple of others (including a PowerPC based SoC from IBM/Motorola). Almost all of them are SoCs or microcontrollers (as they were called earlier), i.e. processor, some RAM/ROM, I/Os etc. all on one chip. We should see some multi-core stuff coming up in the next few years.

GCC, yes, definitely - 0 cost, infinite tailor-ability, rugged and robust, well known interfaces into debugging, etc. etc. One doesn't need to pay for neither flab nor idiosyncrasies of a tool vendor. If you want a GUI interface for 'modern' programmers, give them Eclipse on top of GCC.

[nitrogary]

Originally Posted by headers

Hey guys, it seems both of you are working for the same company. Good to have you on board Mr. Nitrogary.

BTW, why do you think numbers are etched on the head of the injector? What does it signify?

thanks headers, spike

when the injectors are manufactured, is suppose they are tested for their flow characteristics, solenoid response time, flow rates and other performance parameters. the variation from nominal value is captured in the code. the injection maps are based for the nominal injector behavior. the deviation in the actual performance is countered by the ECU when we enter the injector code.
for example: say, the injector is supposed to send 10 units of fuel. but in reality it is sending 10.5 units and we know this during the manufacturing testing. so by entering the code, the ECU knows that this injector is sending 10.5 units instead of the desired 10 and suitable corrections can be made.

sutripta: the least count of the crankshaft sensor is about 3 degrees at full speed.

star, correct me if i am wrong. i suppose a tone wheel is situated on the cam shaft and not the crank. i am not sure what the crankshaft wheel is called. i used to call it missing tooth wheel. maybe i missed some comments in between where this is mentioned.

[headers]

Originally Posted by nitrogary

when the injectors are manufactured, is suppose they are tested for their flow characteristics, solenoid response time, flow rates and other performance parameters. the variation from nominal value is captured in the code. the injection maps are based for the nominal injector behavior. the deviation in the actual performance is countered by the ECU when we enter the injector code.

Thanks NitroG a r y : I understand the stratification, but it leads to one more question - Does it mean that the injector using 10.5 instead of 10 is less efficient?,

Or does it all come under the tolerance bands of injector manufacturing?

Or is it just that the ECU is told to send 10.5 instead of 10 for that particular injector to ensure the pressure is constant?

Sorry for the silly questions!!

[DerAlte]

Originally Posted by nitrogary

... tone wheel is situated on the cam shaft and not the crank. i am not sure what the crankshaft wheel is called. i used to call it missing tooth wheel. ...

What's a 'tone wheel'?

There is nothing on the cam shaft (too little space) usually; sometimes the distributor shaft in petrol engines is the host to the disc. The 'missing tooth' wheel on the crankshaft is what gives the crankshaft angle. And the missing tooth signifies TDC of cylinder 1, or the main index of 1 rotation. Not sure whether it is the leading edge or the trailing edge. The ECU sees this as an event for further processing.

@Sutripta saar, there is no encoder there. The pulse train from the sensor goes directly to the ECU. One cannot calculate acceleration produced by each cylinder if using an encoder, na? That's needed to figure out if one cylinder is not firing, as well as being associated with selective cylinder shutdown (engines with 6/8/12 cylinders).

I think only Honda uses, or used to put, an optical encoder wheel in the distrib. All others use a simple Hall-effect one today at the crankwheel, I think.

[Sutripta]

^^^^
Daadaa,
Tone wheel is the generic name given to any encoding ring/ reluctor ring/ slotted disk etc which can be used to determine progress of shaft as it rotates. Even applies to wheels (for ABS)

The 4 stroke cycle is 720 degrees, so just knowing position of crank will not do, position of cam has to be known. But from cam, crank position can be known, so that is what is normally done, though it can go by any fancy name.

However, when measured from the cam, TDC cannot be determined with any degree of accuracy. And going through available literature (as is available to laymen like me), for a modern diesel, correct timing of the pilot pulse is extremely important (within 2 degree) . I don't think this precision will come from determining TDC from cam.

Quadrature output should not be necessary except for marine diesels, but you never know with the (we are not happy till its complex) Germans.

It will be nice to hear from the professionals, of whom quite a few are taking an active interest in this thread.

Originally Posted by DerAlte

@Sutripta saar, there is no encoder there. The pulse train from the sensor goes directly to the ECU. One cannot calculate acceleration produced by each cylinder if using an encoder, na?

?!?!

Regards
Sutripta

[Sutripta]

Originally Posted by nitrogary

when the injectors are manufactured, is suppose they are tested for their flow characteristics, solenoid response time, flow rates and other performance parameters. the variation from nominal value is captured in the code.

Exactly my thoughts. But to go from philosophy to engineering, one needs some concrete data. So what is the production spread of these values, how are those compensated for, how do those vary with time/ age/ usage, and how is this drift detected.

Regards
Sutripta

[SPIKE ARRESTOR]

Dada, AFAIK the primary reference is from the Flywheel / TDC sensor, the Cam phase sensor is more for Synchronization purpose.

Spike

[Sutripta]

^^^
Yes. Whichever way you look at it (and implement/ engineer it), you have to know where in the 720 deg cycle you are. (Think of wasted injection!). So you HAVE to know the cam position also. So cam has to indexed.

Regards
Sutripta