Hyundai Coupe/Tuscani (Siemens ECU)

[alawadhi3000]

Thanks for the consideration.

I tried the instant fuel consumption and it seems that its a bit low.

When I'm going 110KPH (3000rpm @ 6th gear) with A/C it says that I'm doing 5L-5.5L per 100KM which is too low, My car is highway rated at 7.9L/100KM (in the UK) and 9.8L/100KM (in the US).

I'm calculating the MPG using the MAF method, is there any way that I can increase its accuracy? I filled the tank before a few hours and gonna see how much Hobdrive calculation differs from the real world.

[cail]

When I'm going 110KPH (3000rpm @ 6th gear) with A/C it says that I'm doing 5L-5.5L per 100KM which is too low, My car is highway rated at 7.9L/100KM (in the UK) and 9.8L/100KM (in the US).

How that could be rated differently? Different measure rules?

My car is 8L/100KM rated, but driving on cruise control on a straight road with constant speed 60KPH (4th gear, ~2600rpm) gives me 4.5L/100Km ;-]]

I mean the MPG rate is not a plain rate - it takes some averages (speedups and slowdowns). To compare and make sure hobdrive is precise, you have to compare real gallons which you spent on a full tank.

MAF method is basically can't be calibrated. The only value you can change - is AFR - but this is not quite correct to change it.

Most probably your protocol gives you an injector timings - if you find it, you then may try using it, since normally injector based calculations are in better precision than MAF.

[alawadhi3000]

Well I will see how much the Hobdrive calculation is off the actual figure once I run this tank out (should take about a week).

About the injectors PIDs how many bytes they usually take? I have six consecutive bytes that are always the same, since my car is a V6 it must have something to do with the cylinders/injectors.

The values are B0 B0 B0 B0 B0 B0 when the car is off and vary when the car is running, I think it goes down the harder I accelerate.

BTW I'll PM you the XML with the values that I'm %100 sure about so that you can add them in the next beta.

[cail]

V6 engines normally have injector pulses for two banks separately, and also a joined (probably average) injector pulse width. These normally takes two bytes.

You better build the charts in xls file I gave you - from it it'll be possible to findout injector pids just by curve analysis.

thanks for your help on supporting this ECU.

[cail]

I've PMed you hobdrive beta 1.1.38 build with preliminary support for MIL reading and clearing for any KWP protocol.
I've also included your ECU into the list (and it now in separate file). So you can edit your own file, instead of other's ecu (dont forget to change this in vehicle settings).

Not sure if you have any simple way of triggering a MIL (to test) - most simple way is to detach some sensor you know. Another case is to remove power from one of the sparks and run engine for couple of seconds.

[alawadhi3000]

Thanks, I'll test the MIL and work on the Excel sheet later today.

[alawadhi3000]

I restarted the CarPC today after installing the WIndows updates and boom Hobdrive data was lost.

It even asked me to enter the vehicle settings and info (like when you configure it for the first time).

The fuel tank was nearly empty and now I have to start again lol.

[cail]

Probably your config files were corrupted for some reason. Have some plans to introduce backups to improve stability.

BTW, whats with the MIL codes? Have you had a chance to test it?

[alawadhi3000]

Sorry, been busy for the past few days and didn't have the chance to test.

I'll test it today for sure and report back, I'll do the excel sheet also.

[alawadhi3000]

MIL works great. (only disconnected one sensor).

I'm having a hard time figuring out which byte is the RPM, I've done a test to narrow it down but still no luck.

I stopped on the side of the road and started Hobdrive then accelerated in the first gear until redline.

Redline is ~6750RPM @ 57KPH.

I've attached the Excel sheet for this run, column "U" is the speed, the first green row is when the car started accelerating and the second one is when I hit the redline.

[cail]

Thanks for MIL. But be aware that MIL descriptions will probably not match.
I'll checkout if there any standard KWP MIL code descriptions.

Do you remember what MIL was raised and on what sensor detach?

Regarding the RPM, I think it is in columns V and W. Try it with

<wordle-21>1</wordle-21>
(if I'm not mistaken with offset)


I also see that offsets 13 and 14 are a single word
<wordle-13>1</wordle-13>
But I can't clarify what it means. It could be an InjectorPulseWidth (it is similar to MAF in shape)

[alawadhi3000]

I disconnected the Intake Air Temperature sensor and it gave me P0113 which is correct.
Description was fine also but there were 3 lines total (2 Russians? and 1 English), I took a picture of it and its attached.

Regarding RPM you're absolutely right, it works perfectly thanks, I do have a question, what is the difference between <word> and <wordle>? They both take two bytes but whats the relation between the two bytes?

Regarding the injectors I think there's a high possibility that you are right also, I tested it and
<wordle-13>0.02</wordle-13> and it gives me reasonable consumption while cruising but very low consumption when Idle (Like 0.7L/hr) with A/C on, I'm reading lots of info about injectors and their fuel economy equations at the moment.

Also, Columns P,Q and R are related to the MAF, if I disconnect the MAF sensor they turn to 0, Columns S and T are not related to MAF but they are always/nearly the same as Columns Q and R (Q=S & R=T), any idea about that?

[cail]

Thanks for the info.

word takes bytes in big endian order (like obd2 does),
wordle is little endian (first low byte, next high byte of a word) - like on most KWP protocols.

With injector you should also know that it should be calibrated with injector pulse width in vehicle settings.

In your case you seems to calibrate it via 0.02 coefficient
You may try to setup injector based MPG on idle to your normal idle consumption, and then see how real ride MPG shows up.
It may be the case that it differs from MAF. The reason is in fuel trim corrections, which could be applied to MAF values by lambda sensor.

On my vehicle, Injector base consumption is in general higher than MAF-calculated. But at the same time it is more precise (thats sad, but truth is more important

PQR could be a MAF derived sensors, f.e. MAP.

[alawadhi3000]

I see.

My car has the instant L/100KM meter but I removed it to install the Double DIN PC monitor, I'll see if I can fit it somewhere else to compare the real instant consumption vs Hobdrive injector based calculation but I doubt it'll fit somewhere else without heavy modding.

The interesting thing is that the OEM gauge uses fuel injection and speed to calculate, so that should be the most accurate way.



Any tips on finding other sensors? the remaining sensors are:-
Fuel system status (Closed, open).
Short term fuel.
Long term fuel.
Fuel pressure (Not important).
Cylinder Timing (Not important).
Throttle Position.
Oxygen Sensors.

There are only two oxygen sensors on my car (B1S1 and B1S2), and I think that the second sensor is stuck at 0.98V because when I replaced the ECU with an OBD-II compatible ECU it showed that B1S1 was changing a lot (working normally) but the B1S2 was stuck at 0.98V.

Thanks a lot.

[cail]

> Fuel system status (Closed, open).
Just reproduce engine brake (shiftdown after speedup). FSS should change a bit during this time.
Also Injector normally shows Zero value during this condition.

> Short term fuel. Long term fuel.
These should float around zero (with 128 value as zero base). Ltft is more stable

> Throttle Position.
Thats the easiest one, no? Just press the pedal

> Oxygen Sensors.
You may try find it and detach (it is often on outtake somewere near catalitic converter)

[alawadhi3000]

Lol, I know that I need to press the gas pedal but doing so will change like 50 values.
I found the throttle sensor and disconnected it, and got the Throttle Position Sensor memory location.

Regarding the Oxygen Sensor I'm afraid thats too much hassle, my car is lowered and I'm a tall guy so I can't get down to disconnect the sensors, not to mention its veeeeeeeery hot out there.

I'll try the rest tomorrow.

Thanks.

[cail]

;-]
Forgot to tell you can press gas pedal with active ignition but with engine not running.
This would not change that much of the parameters.

[alawadhi3000]

Oh thats why I didn't get the sensor without disconnecting it, its because I assumed that it'll be 00 if the car was off, I didn't think that the sensor still send signals even when the car if off.

Any chance of implementing Power and Torque in the next versions? I'm not requesting but more like wondering if its considered or not.

If there's a change-log thread with whats you're working on for the next version that would be cool.

BTW congratulation of the win today.

[cail]

Yep, had plans to add power/torque calculations, will give you a notice when these'll be available in beta.

for change-log, you probably can take a look at hobdrive.userecho.com - but there are no system or internal planned features listed (which are in internal tracking system and not avaialable).

thanks for the congrats

[alawadhi3000]

I'm back.

I have more questions.

1) How can I implement this equation in the XML:-
(A-128) * 0.78125
My understanding is if I put the value as 0.78125 and the offset as -128 the program will multiply first and then subtract.

2) I found the fuel system status PID
Value is 0 if open loop, 1 if closed loop (using O2 sensors).
How can I put that also.

3) If my injector size is 190cc and my car is a V6 (has 6 injectors), what will be the fuel coefficient value? I heard that there is an equation for that but I couldn't find it.

4) Any tips on locating the Catalyst temp? I've seen the shop manual and I definitely have one.

Thanks.