[Csantucci]

Can you provide the download link for a loco that is affected by the problem?

[Laci1959]

My link
When you reach and keep the set speed. Good for deceleration and acceleration.
Sorry for copying it to the demo locomotive, but it seemed the fastest.

[Csantucci]

I have started testing. Can you confirm that the problem is only on the Tractive Force Curves and not on the Dynamic Brake Curves? I removed the Tractive Force Curves, modified the Dynamic Brake Curves to adapt to the values of the E464 and I got a good behaviour.

[Laci1959]

Csantucci, on 16 July 2021 - 09:12 AM, said:

I have started testing. Can you confirm that the problem is only on the Tractive Force Curves and not on the Dynamic Brake Curves? I removed the Tractive Force Curves, modified the Dynamic Brake Curves to adapt to the values of the E464 and I got a good behaviour.

Hello Carlo

Yes exactly.

Regards Laci 1959

[Weter]

@AlbertoSaurio65 This 1970-th era locomotive also had the speed-reference handle and traction/braking current-reference handle, in addition to wheel-style notchless throttle, but I don't know the way to model such cab controls now.
https://upload.wikimedia.org/wikipedia/commons/a/ad/Sr1_ohjaamo.jpg

[Csantucci]

Laci,
I don't understand your ORTSTractionCharacteristics table. The throttle has no influence in your table, which seems a bit strange to me: the force is the same from throttle at 0.1(10%) to throttle at 1 (100%).

Using a more proportional table, where the max force for every speed is reached at 90% of the throttle, I get a correct behaviour in my first test.

[Laci1959]

Hello.
I found it in a Czech vehicle. It was a little weird for me too. The speed controller has no fixed positions, it can be controlled continuously. What is the solution for a continuously controllable controller? Maybe ORTSMaxTractiveForceCurves?

Quote

ORTSMaxTractiveForceCurves (
0 (
0 0
160 0
)
1 (
0kmh 400kN
10kmh 400kN
20kmh 400kN
30kmh 400kN
40kmh 400kN
50kmh 400kN
55kmh 400kN
60kmh 370kN
70kmh 325kN
80kmh 280kN
90kmh 250kN
100kmh 225kN
110kmh 200kN
120kmh 190kN
130kmh 175kN
140kmh 155kN
150kmh 150kN
160kmh 145kN
)
)

This is the setting of a KISS motor train, in which the "Throttle" can also be controlled continuously. Anyway, the same phenomenon is the same.
Regards Laci 1959

[Laci1959]

Weter, on 16 July 2021 - 11:28 AM, said:

@AlbertoSaurio65 This 1970-th era locomotive also had the speed-reference handle and traction/braking current-reference handle, in addition to wheel-style notchless throttle, but I don't know the way to model such cab controls now.
https://upload.wikimedia.org/wikipedia/commons/a/ad/Sr1_ohjaamo.jpg

Hello.
It is very similar to the MÁV V63 Gigant locomotive. I haven’t dealt with this problem yet, but it will happen sooner or later.
Laci 1959

[steamer_ctn]

Laci1959, on 16 July 2021 - 01:25 PM, said:

I found it in a Czech vehicle. It was a little weird for me too. The speed controller has no fixed positions, it can be controlled continuously. What is the solution for a continuously controllable controller? Maybe ORTSMaxTractiveForceCurves?

0kmh 400kN
10kmh 400kN
20kmh 400kN
30kmh 400kN
40kmh 400kN
50kmh 400kN
55kmh 400kN................................................

The Tractive force curve shown here is using invalid values.

The values should be in metres/second, and Newtons.

This might be creating some of your problems depending upon how OR is actually interpreting these values.

[Csantucci]

Peter,
it's not so. If you look at the code for the InterpolatorDiesel2D or perform a test, you will see that for speed and force all units of measure generally accepted by OR can be used in these curves, and among them there are kmh and kN.

[Laci1959]

http://kepkezelo.com/images/wea2apihsfvo0eijqwl0.png
Hello
Compared to INVALID it keeps it quite accurate. See picture.
Regards Laci 1959

[Csantucci]

Laci1959, on 16 July 2021 - 01:25 PM, said:

Hello.
I found it in a Czech vehicle. It was a little weird for me too. The speed controller has no fixed positions, it can be controlled continuously. What is the solution for a continuously controllable controller? Maybe ORTSMaxTractiveForceCurves?

This is the setting of a KISS motor train, in which the "Throttle" can also be controlled continuously. Anyway, the same phenomenon is the same.
Regards Laci 1959

Laci,
I tried to use your above setting, with values divided by two (because the E464 has a max force of 200 kN) and with a load of 5 carriages, and the behaviour is satisfying.

[Laci1959]

http://kepkezelo.com/images/ril385g7hjyvkqjpd5r8.png
Hello.
This is a characteristic of a FLIRT multiple unit. Official publication for the training of engine drivers. The curve is straight in the first quarter of the speed range. There is only one curve for "all controller positions". There are no fixed jobs like ORTSTractionCharacteristics requires. I think the programmers have slipped over this.
Regards Laci 1959

PS.
So where could the mistake be? Should I make a video of it or have you seen the phenomenon?

[Laci1959]

Weter, on 16 July 2021 - 11:53 PM, said:

Wow, M(62)ашка! 3D cab?
I see very interesting table form!
I thought, the statements must be in pairs only, but here, everything looks very clear.

You must be in pairs when using ORTSMaxTractiveForceCurves. When using ORTSTractionCharacteristics in a table.
The table was made with Excell. I use the Save As menu item, where I select Formatted Text (space separated). The file is given the prn extension. I open it with Notepad and copy the spreadsheet and then paste it into the eng file.
I hope you understand.

[Weter]

Well, that's great, that ORTS accepts such form of data - this way, it would be very simple - to refer the real tractive characteristics for modeled locomotives.
And thank you, Laci for advice, how to make that easily.
As for Flirt.
I suppose, this is a modern stock with contact-less semiconductor (most likely, IGBT) -based, smooth variable traction control, handled by computer.
This way, driver only refers the desired result, but not controls motors directly.
It means, at factory, the optimal, averaged characteristics for traction curves was pre-programmed to PLC of the engine unit. (Programmable-logic computer)
So:
The flat part is intended for avoiding wheelslip and motors overload by current, at the moment of start or creep at minimal speeds.
Single curve is shown, maybe, because there is notchless algorithm here, instead of contractor-based controllers, where every notch had its own curve. Instead, at every moment, the most optimal voltage, current and frequency are computed for fitting driver's (or cruise-control) reference, and compensation variables are applied, taking in account the feedback signals from current and speed sensors.