Weird throttle calibration problem

[Mikealpha2]

Hi,

 

I have a strange problem when calibrating my Throttletek 737 Throttle. The levers are not really synchronous, so I wanted to use the Sync function in FSUIPC. So I removed axis assignments in P3D and assigned in FSUIPC, and used  'send direct to FSUIPC calibration'. On the Calibration page (with no reverse zone) I just click Set for both Idle and Max position. Values from -16200 to 16200. Then I moved both levers synchronously and repeatedly pressed the Sync button. So far so good. Now if I move the levers (that's what I also see in the Sim) all is well for the Max position. The calibration page displays In at around 16200, Out at around 16200.

But at Idle it shows In -16200 and Out it shows zero (0). See attached screenshot. So the entire range from 0 to -16200 is missing or got lost.

What am i doing wrong ?

Mike

Edited November 18, 2017 by Mikealpha2

[Pete Dowson]

17 hours ago, Mikealpha2 said:

But at Idle it shows In -16200 and Out it shows zero (0). See attached screenshot. So the entire range from 0 to -16200 is missing or got lost.

Not 'lost'. Zero throttle = idle. Negative throttle is reverse thrust.  You have 'No reverse zone' set so you get the whole of the IN range, -16200 to +16200 for forward thrust, 0 (idle) to 16383 (full thrust).

Pete

 

 

[Mikealpha2]

But now if my hardware thrust levers are in idle position, the levers in the airplanes are at 50% forward thrust. That is not intended.

I just want the full range -16200 (levers at idle) to 16200 (max) to stay like that in the plane. My hardware levers at idle and levers in the plane at idle. 

How can I achieve that and still use the Sync functionality ?

Mike

[Pete Dowson]

11 minutes ago, Mikealpha2 said:

But now if my hardware thrust levers are in idle position, the levers in the airplanes are at 50% forward thrust. That is not intended.

What aircraft is this with? 

I could really do with seeing your FSUIPC INI file, where the settings are saved.

11 minutes ago, Mikealpha2 said:

I just want the full range -16200 (levers at idle) to 16200 (max) to stay like that in the plane.

Proper calibration should give you either 0 to 16383,  or -16384 to +16383 (depending on what is needed by the particular add-on aircraft and thus on the options you have selected in FSUIPC).  Normal "direct calibration' gets you 0 to 16383, because the values are written directly into the simulators engine values using Sim Vars as follows 

KEY_THROTTLE_SET	THROTTLE_SET	Set throttles exactly (0- 16383)	Shared Cockpit
KEY_AXIS_THROTTLE_SET	AXIS_THROTTLE_SET	Set throttles (0- 16383)	Shared Cockpit (Pilot only, transmitted to Co-pilot if in a helicopter, not-transmitted otherwise).
KEY_THROTTLE1_SET	THROTTLE1_SET	Set throttle 1 exactly (0 to 16383)	Shared Cockpit
KEY_THROTTLE2_SET	THROTTLE2_SET	Set throttle 2 exactly (0 to 16383)	Shared Cockpit
KEY_THROTTLE3_SET	THROTTLE3_SET	Set throttle 3 exactly (0 to 16383)	Shared Cockpit
KEY_THROTTLE4_SET	THROTTLE4_SET	Set throttle 4 exactly (0 to 16383)	Shared Cockpit

That is an extract from the P3D SimConnect SDK. Note the range for the values, 0 to 16383.

Your 16200 values are just the limits provided by your input levers. Full thrust is +16383, not 16200.

11 minutes ago, Mikealpha2 said:

How can I achieve that and still use the Sync functionality ?

Just using 'sync' shouldn't affect the range from min to max. If it is doing that I shall have to investigate as it would be an error. Please confirm that without sync you get what you want for the range.

Pete

 

[Mikealpha2]

Hi Pete,

first of all thanks for the insight. I think I have understood the problem now

it's the FSLabs A320. They have their own calibration tool to define the detend zones. They also have a range display. Without FSUIPC calibration, apparently with the standard -16384 to 16383 range, I see the range between -1.000 (idle) and 1.000 (max). Between this range the detend zones are defined.

Now if I use the FSUIPC calibration (I understood I have to to get the Sync functionality), I get a range of 0 to 16383, as you mentioned. Now in the FSLabs tool I see the range between 0 and 1.000. Consequently the levers in the plane are in the middle and not at idle.

I understand I can now redefine the detend zones, set the Idle range from now between -1.00 and -0.97 to 0.00 to 0.03.  I am just wondering, if I won't loose half of the resolution for that now. From -1.00 to 1.00 gives a Delta of 2.00. From 0 to 1.00 gives a delta of just 1.00.

So I think in that case the Addon would need -16384 to 16383 for the idle to max range.

I hope it is understandable with my "German" english.

Mike

[Pete Dowson]

1 hour ago, Mikealpha2 said:

I understand I can now redefine the detend zones, set the Idle range from now between -1.00 and -0.97 to 0.00 to 0.03.  I am just wondering, if I won't loose half of the resolution for that now. From -1.00 to 1.00 gives a Delta of 2.00. From 0 to 1.00 gives a delta of just 1.00.

Well, assuming that the aircraft responded properly to the forward thrust range of 0 to 16384, obviously your -16200 to +16200 "resolution" is still being spread over the whole range of 16384 units. So not less, more if anything.

However, in any case i think you'll find that the true resolution of the axis is well below the 15 bits used for 0-16383. Most throttle units hve 8 or 10 bit chips, giving a true resolution of either 256 or 1024 different values.

1 hour ago, Mikealpha2 said:

So I think in that case the Addon would need -16384 to 16383 for the idle to max range.

Possibly. It is a pity you didn't show me your INI file as I requested, as I could advise more accurately.  But here is an excerpt from the FSUIPC Advanced User guide (on one of the first few pages in fact). this may help:

 

UseAxisControlsForNRZ=No: This is a facility for the [JoystickCalibration] section(s) of the INI file, not [General].
It is a special option provided to try to cope with some different add-on practices (notably, in this case, the Wilco A320).
Normally, the 4-Throttles, 4-mixtures and 4-Prop pitch calibrations result in an output with either a range which includes
the reverse zone, or, if the "no reverse zone" option is checked, a range from 0 (idle) to 16383 (max). These are sent to
FS using the older "????n_SET" controls (THROTTLE1_SET, etc), since these are the ones providing the reverse zone
below zero.
If you set the [JoystickCalibration] INI parameter UseAxisControlsForNRZ to "Yes", then the NRZ (no reverse zone)
option for all three axis types will use the AXIS_????n_SET controls (e.g. AXIS_THROTTLE1_SET) instead, with a
range of -16363 (idle) to +1 6383 (Max). This will be Aircraft or Profile-specific if you set it in the appropriate calibration
section of the INI file.
 

Pete

 

[Mikealpha2]

Hi Pete,

ouch, sorry, I typed the previous message not from my FS PC, so I forgot the ini file. It's attached now.

Thanks again for the info, also regarding the resolution. So probably I should define the detends now and check how it handles during flight.

I'm a little bit lost after reading the manual excerpt you posted, but will try to understand it. That way, would it also then take the Sync values into account ?

 

Mike

FSUIPC5.ini

[Pete Dowson]

17 hours ago, Mikealpha2 said:

I'm a little bit lost after reading the manual excerpt you posted, but will try to understand it

Oh dear. I told you exactly what parameter to change!! Look, I'll repeat part of it, with the important bits for you in RED:

If you set the [JoystickCalibration] INI parameter UseAxisControlsForNRZ to "Yes", then the NRZ (no reverse zone)
option for all three axis types will use the AXIS_????n_SET controls (e.g. AXIS_THROTTLE1_SET) instead, with a
range of -16363 (idle) to +1 6383 (Max). This will be Aircraft or Profile-specific if you set it in the appropriate calibration
section of the INI file.

In your INI file it's the 7th line in the section headed [JoystickCalibration]. All you needed to do, as it says, is change the two letters "No" into "Yes" on that one parameter!

You may need to re-calibrate afterwards, and reset the Sync points. So, best delete the two Throttle lines at the end, and the SyncSlope line. so you get fresh numbers from scratch.

Pete

 

[Mikealpha2]

Sorry, I am not that language perfect, occasionally need a bit longer to understand english manual text. Thanks for making it more clearer !

I was not aware I could store it aircraft specific. Good to learn how powerful FSUIPC is !

Anyway, works fine now !

[Mikealpha2]

Hi Pete,

 

bit too early celebration. Now with UseAxisControlsForNRZ=Yes I get the full range -16384 to 16383.  Just the Sync function leads to weird results.  I just do the same syncronous lever movement and Press Sync at various positions. When I do the same in the PMDG 777 and in the FSLabs A320 I get this :

PMDG 777:

SyncSlopeThrottle2=9/10,21/22,36/36,53/54,68/68,80/80,92/91,105/104,116/116,127/128

FSLabs A320 :

SyncSlopeThrottle2=12/23,34/23,54/23,76/23,90/52,105/83,127/128

Without FSUIPC calibration, moving the thrust levers simulaneously leads to thrust discrepancy around 1-3%. Now it is huge, maybe there is something interfering the Sync results in the A320 ?

The obvious solution, copy over the PMDG 777 line to the A320 block.

Anyway, I attach the fsuipc.ini

Mike

FSUIPC5.ini

[Pete Dowson]

1 hour ago, Mikealpha2 said:

When I do the same in the PMDG 777 and in the FSLabs A320 I get this :

PMDG 777:

SyncSlopeThrottle2=9/10,21/22,36/36,53/54,68/68,80/80,92/91,105/104,116/116,127/128

Does this work in the PMDG aircraft. most previous users have told me that they get serious diverse movements of the throttles if they calibrate in FSUIPC, never mind use sync, I thought PMDg read the axis values at a higher priority level than where FSUIPC feeds them to the sim engine, so the two diverging values cause conflict.

Generally, for PMDG aircraft (and some other advanced ones, possibly including the FSL ones),  advise only assigning to FS controls "Axis throttleN set" and not calibrating.  

But maybe that only applies to the way PMDG did things on P3Dv3, not P3Dv4.

1 hour ago, Mikealpha2 said:

Without FSUIPC calibration, moving the thrust levers simulaneously leads to thrust discrepancy around 1-3%. Now it is huge, maybe there is something interfering the Sync results in the A320 ?

Sounds like FSL do what PMDG used to do.   Is the throttle control erratic with FSUIPC calibration in any case?

Also, since the sync points are merely relating values coming in from your axes to each other, and scaling the changing values to achieve those points via the Throttle 1 calibration, the sync should be the same for ALL aircraft. After all, they ALL have Full thrust at +16383 and the Idle in the same place (0 or -16384 depending on which FS control method you set FSUIPC to use), so the throttle syncing should be identical too.

BTW the INI file tells me you are still on 5.103, a very early version of FSUIPC5. Could you update, please? The current version is 5.122.

It also shows me that you have a default calibration section and just the one Profile, for the A320. The default one is therefore being used for the 777, and that still has "UseAxiscontrolsForNRZ=No". It is rather different therefore.

And ... you have no yoke (aileron . elevator) assigned. Or rudder or toe brakes. Do you fly on keyboard only, or are you still using assignments in P3D? If you assign ANY joystick button or axis in FSUIPC you MUST disable controllers completely in P3D or you will get conflicts occurring. There's a check box in the P3D control assignments "other" section where you can disable controllers. You need to do that, or not use FSUIPC for assignment. You can still calibrate in FSUIPC even without assignment.

Pete

 

[Mikealpha2]

Hi Pete,

 

yes, works fine in the PMDG over the entire range. No erratic behaviour at all.

Now, in the FSLabs, if I remove the Sync line, it works over the entire range as well, of course with the levers out oy sync discrepancies. But no erratic movements. For me it looks like the sync function introduces the problem. If you look at the Sync values for the A320, for the first four values covering the range from idle to approx. middle position, the value remains at 23 for throttle2. From medium to full you see big increments to finally catch up with throttle 1 at max. position. Such as if Throttle2 would be still in the 0 to 16383 range, and Throttle 1 in -16383 to 16384 range.

Yes, so far I have only assigned throttle and tiller in FSUIPC. Both are disabled in P3D. For the other axes I have Brunner FFB equipment, they feed directly into P3D via their own driver. So aileron, elevator, etc. are disabled in P3D and not used in FSUIPC.

Ok, will update this evening.

 

[Pete Dowson]

1 minute ago, Mikealpha2 said:

For me it looks like the sync function introduces the problem. If you look at the Sync values for the A320, for the first four values covering the range from idle to approx. middle position, the value remains at 23 for throttle2. From medium to full you see big increments to finally catch up with throttle 1 at max. position. Such as if Throttle2 would be still in the 0 to 16383 range, and Throttle 1 in -16383 to 16384 range.

Well, FSUIPC is only recording the values your throttle is providing each time you press the sync button. I don't know why it needs to be different in the A320 compared to the 777 in any case, though I know Airbus systems are completely different to Boeing.

8 minutes ago, Mikealpha2 said:

Yes, so far I have only assigned throttle and tiller in FSUIPC. Both are disabled in P3D

Just unassigning in P3D, like FSX before it, is precarious. If P3D ever decides the controls are newly connected (maybe a USB hiccough, or a temporary disconnection beforehand) it will automatically re-assign.  This then causes conflicts and they puzzle users as they can't understand why it is suddenly different.

That is why I always tell users to disable the controllers completely, not simply try to de-assign (which then isn't necessary in any case).

Pete

 

[Mikealpha2]

1 hour ago, Pete Dowson said:

Well, FSUIPC is only recording the values your throttle is providing each time you press the sync button. I don't know why it needs to be different in the A320 compared to the 777 in any case

It really puzzles me. I have both levers in Idle position and sync reads 12/23.Just for testing, then I move only Throttle 2 up to the medium position and Sync still reads 12/23. Then above the medium position up to the max position it begins to increase up to 12/128.  If I do the same with Throttle1,  Sync values increase linear over the entire range.

Even with Airbus specialities in mind or FSLabs custom stuff, it still doesn't make sense.

I'll double check the throttles are deactivated, not only unassigned in P3D

Mike

 

[Pete Dowson]

5 minutes ago, Mikealpha2 said:

It really puzzles me. I have both levers in Idle position and sync reads 12/23.Just for testing, then I move only Throttle 2 up to the medium position and Sync still reads 12/23.

Well, what is the point of moving only one throttle? You are effectively telling FSUIPC to equate your meiddle Thr2 position to the idl Throttle1 position. "Sync" is trying to match the output values for both of the two Throttles. What do you expect it to do if you move only one?!

And what are the values actually arriving IN and going OUT? Where are you reading the 12/23 etc, only in the INI? Are you reloading the INI file all the time?

To view the IN values all the time, in real time, go to FSUIPC Logging tab, enable Axis logging, and Console Log, then run P3d in Windowed mode so you can see both it and the Log output both at the same time.

Really, without knowing how your throttle inputs are behaving I can't comment, apart from my comment that it is really nonsense to move only one throttle and trying to sync multiple positions on Thr2 with just one postion on Thr1.

Pete

 

 

[Mikealpha2]

Hi Pete,

>> Well, what is the point of moving only one throttle?

just for testing purpose and to make the behaviour more clear. After each sync click I look in the ini-file again

Throttle 1 and 2 at idle. Sync stores 12/23. Now I move only throttle 2 and would expect to see the second value increase. It doesn't, up to the medium position I still get 12/23.  During all that IN and Out values change symmetrical.

So four Sync clicks while moving the throttle from idle to medium gets four times 12/23.

If, again for testing purpose, I keep throttle 2 at idle and move  throttle 1 up to medium position, I get 12/23, 34/23, 54/23, 76/23

It is somewhat beyond me why Throttle 1 should behave differently then Throttle2. If without Sync they increase thrust simultaneously ?

Anyway, will check the Axis logging as you describe it.

 

[Pete Dowson]

46 minutes ago, Mikealpha2 said:

It is somewhat beyond me why Throttle 1 should behave differently then Throttle2. If without Sync they increase thrust simultaneously ?

The numbers in the sync are proportions (128 = 100%). I am puzzled at the 23 not changing, but that is why I need to see the logged axis values. And with the log i can see if I can reproduce the problem here.

Pete

 

[Mikealpha2]

I have enabled Axis logging, and Console window, but I do not see any values appearing in the console window when I move the throttles. Do I have to set anything additionally ?

Anyway, have updated to 5.122 now and started again. So I press SET for both throttle1 and 2 in the Idle and Max position. I can observe IN and OUT changing more or less simultaneusly over my entire range -16213 to 16317. Looks good. Now with both throttle 1 and 2 in Idle Position I press sync. This time it writes 0/129 in the file, not 12/23. From then on OUT is in disorder, doesn't make sense to store anymore Sync values.

 

Just an update. With any other planes the Axis logging works, I see the values in the Console Window. But not with the A320.

[Pete Dowson]

32 minutes ago, Mikealpha2 said:

I have enabled Axis logging, and Console window, but I do not see any values appearing in the console window when I move the throttles. Do I have to set anything additionally ?

Is this whilst in the Calibration? Sorry, i think the general axis logging only takes place when in normal flight mode. I really at this stage only wanted to see the value of each axis being sent to FS as you moved the two levers together from idle to max using the Sync line as it is at present.

Then maybe a separate log with the same but with the Sync line removed or "commented out" (precede it with a ;) so i can see the good or bad effects of the sync that you say you get.

There is a debugging option i can get you to enable which logs a lot more stuff -- including raw input values -- even whilst in Calibration, but that log will be hige. You'd need to ZIP it up and either attach it or send it to me by Email.

However, I'm unlikely to be able to do much more for you now until Thursday, so if you can manage okay without the sync line for now, maybe we can continue then?

I note that earlier you said this:

6 hours ago, Mikealpha2 said:

Without FSUIPC calibration, moving the thrust levers simulaneously leads to thrust discrepancy around 1-3%.

In view of such a small discrepancy I do wonder why you want to use sync. Real throttle levers in real aircraft are rarely level for the same thrust setting. They vary for several reasons -- age, connection streccting/linkage differences, minor difference in hydraulic pressures in the different systems, age and wear and performance of engines, fuel flow differences, etc. Take a look at some real life cockpit videos.

Pete

 

[Pete Dowson]

Addendum to last message: 

Until I can get to check this, could you try changing "UseAxisContrlsForNRZ=No" again on the FSL. I know this gives you only 50% to 100% thrust movement, but I'd like to double-check that it is the difference this parameter makes to the Sync Action. If so i will know exactly where to look. the code in FSUIPC is actually quite complex there, having been developed and added to quite a bit over the years.

Make a safe copy of your FSUIPC5.INI file first, so you don't lose any work you've already done. You can then quickly restore it afterwards.

Pete

 

[Mikealpha2]

Quote

Is this whilst in the Calibration? Sorry, i think the general axis logging only takes place when in normal flight mode.

Both ! But that's only in the A320, axis logging in flight mode works in all other planes including the PMDG 777. Seems FSLabs has suppressed the logging ?

 

Quote

I'm unlikely to be able to do much more for you now until Thursday, so if you can manage okay without the sync line for now, maybe we can continue then?

that´s fine, I also don't have that much time until the weekend. For now I have just copied the 777 sync line to the A320, that gives the best results so far.

There are some threads in the FSLabs forum currently left at "FSUIPC calibration' doesn't work". Besides my setup, would be also nice to post a solution/instruction there

 

Quote

In view of such a small discrepancy I do wonder why you want to use sync. Real throttle levers in real aircraft are rarely level for the same thrust setting.

maybe 1-3% was a bit optimistic, I looked at it again, over the range it's also around 5-6% occasionally. But thanks for info, I didn't know that, always thought real throttles are perfectly synchronous.

Anyway, the idea came up after seeing a 'levers not in sync' message in the A320 display. Apparently one lever was in the detend already, the other not. Unfortunately my throttle doesn't have physical detends, so I thought I try the Sync function. Of course one can widen the detend zones in the A320 config, but I didn't want to do that too much.

And the Sync function works REALLY well, in the PMDG 777 it's great !

 

Quote

Until I can get to check this, could you try changing "UseAxisContrlsForNRZ=No" again on the FSL.

Ok, will do that !

 

[Pete Dowson]

1 minute ago, Mikealpha2 said:

Both ! But that's only in the A320, axis logging in flight mode works in all other planes including the PMDG 777. Seems FSLabs has suppressed the logging ?

That's rather odd. I suspect you'd get the logging if you used the FS controls in assignment instead of "direct to calibration". Seems that FSL take the values before SimConnect can give them to me.

As I say, I do have some debugging log options to get them before that, but let's leave that for now.

3 minutes ago, Mikealpha2 said:

For now I have just copied the 777 sync line to the A320, that gives the best results so far.

Interesting. That implies the "UseAxisControlsForNRZ" option is only adversely affecting the vlues input to the syncing operation.

These are all good clues. I don't have the FSL aircraft, but I might be able to set something up to do the same sort of thing.

5 minutes ago, Mikealpha2 said:

There are some threads in the FSLabs forum currently left at "FSUIPC calibration' doesn't work".

Well, as I said earlier, i generally advise not to use "direct to" nor "calibration" for aircraft like all the PMDG boeings and other similarly advanced aircraft, which do tend to be designed to read devices and do their own best effort with them in any case. This is especially true of throttles, far less so for yokes, rudders and ailerons where there's a lot of personal taste to how the handling feels.

Odd that there are few if any other posts here about the FSL Airbus.

Pete

 

[Pete Dowson]

On 11/21/2017 at 4:46 PM, Pete Dowson said:

Interesting. That implies the "UseAxisControlsForNRZ" option is only adversely affecting the vlues input to the syncing operation.

I've managed to do some tests here, now, and, indeed, it is the "UseAxisControlsForNRZ" option which is messing up the sync point making facility. It is using the -16384 to +16383 range for the throttle to be sync'd, but the 0-16383 range for the master throttle.

This is actually a really good catch, so thank you! It's an original bug, dating right back to when the sync option was implemented! Amazing that it has never been reported before!

And you found a good work-around for your case -- copy over the sync positions from a "UseAxisControlsForNRZ=No" calibration. The other way it to calibrate and sync first with that option set No, then change it afterwards and "reload" the calibrations.

The fix internally, which I will implement, is basically that -- ignore the "Yes" setting whilst recording sync points. It will be in the next interim updates (4.972a and 5.122a) which I'll post in the Download Links subforum in a few days.

Pete

 

 

[Mikealpha2]

excellent, thanks for info. Glad it could be sorted out !

A plane needing that full range and a Sync user, probably a rare combination :-). I'll post some instructions in the FSLabs forum after the new versions are out.

Thanks and keep up the great work !

Mike