G940 bugs sorted out???

[Arrow]

[QUOTE=pupo162;220242

here you go, this is the issue I have on my throttle. throw all the video i only move the thortles, look how all the other axis just dance around. there is even a bit where i only move 1 throtle and the other has a spike too.


cheers[/QUOTE]

I just tested it, doesn't happen with my stick. I've heard that some units may have cables stuck in the throttle, so when you move it it creates the spikes.

[robtek]

As i have my 4th G940 now, i am shure thats the cable in the throttle unit!
As you might still have warranty -> get a new G940.
Apart from those Problems i love this stick, and the last one i've got even has a little stop in the middle of rudder and aileron trim.

[ACE-OF-ACES]

Originally Posted by Arrow

I just tested it, doesn't happen with my stick. I've heard that some units may have cables stuck in the throttle, so when you move it it creates the spikes.

Didn't happen.. or should I say I didn't notice it right away either.. But as so many noted at the Logitec forums as time goes by and as you use it more and more the symptom apears for all.. In that it takes some time for those so called stuck cables to come loose and wa-la you will get the noisy R1 and R2 and reversals.. Just give it time

[Artist]

Originally Posted by ACE-OF-ACES

In that it takes some time for those so called stuck cables to come loose and wa-la you will get the noisy R1 and R2 and reversals.. Just give it time

The Reversal Bug has IMHO nothing to do with the cable in the right throttle (trims, rudder have it, too). Rumors(!) have it, that the behaviour we know as Reversal Bug was implemented on purpose: Potientiometers wear out over time (1 million turns...) and being already used to this behaviour from the start, you wouldn't notice... That could be the reason they fixed it for stick x and y (Hall sensors!) but not for other axes. But as said: This is a rumour.

I am quite happy with having redirected throttle and rudder via a BU0836A (R1 and R2 to follow on my next day off, trims already external).

Artist

[ACE-OF-ACES]

Originally Posted by Artist

The Reversal Bug has IMHO nothing to do with the cable in the right throttle (trims, rudder have it, too). Rumors(!) have it, that the behaviour we know as Reversal Bug was implemented on purpose: Potientiometers wear out over time (1 million turns...) and being already used to this behaviour from the start, you wouldn't notice... That could be the reason they fixed it for stick x and y (Hall sensors!) but not for other axes. But as said: This is a rumour.

Agreed I don't think it is the cabling.. the cabling that was causing the R1 and R2 noise.. I am surprised to hear they used pots instead of encoders or hall sensors.. but that would explain why the BU0836 analog inputs worked with these sensors.. Leo does some filtering on the analog inputs so even if you had a 'spike' it would be filtered out.. Why Logi can't do the same in their firmware is beyond me in that it is 'standard' practice to do so with most analog inputs to A2Ds.. Not sure what or why they would impliment in firmware a reversal when using pots.. makes no since really.. Thus I would leave that in the rumor column.. More likly is it is a typical pot spike that spikes in voltage in a direction oposite of what your currently at and thus apears to be a reversal of sort

Originally Posted by Artist

I am quite happy with having redirected throttle and rudder via a BU0836A (R1 and R2 to follow on my next day off, trims already external).

Yeah you cant go wrong with Leo's stuff.. everyone uses his stuff.. Ill have to look but I seem to recal him having a board that would convert the hall sensor outputs to analog inputs on his board.. if so I would dump all the Logi electronics and go full Leo!

[Artist]

Originally Posted by ACE-OF-ACES

I am surprised to hear they used pots instead of encoders or hall sensors..

Price. Economy. Halls only for stick x and y (FFB )...

Originally Posted by ACE-OF-ACES

Leo does some filtering on the analog inputs so even if you had a 'spike' it would be filtered out..

He does? Didn't know. Didn't see - yet...

Originally Posted by ACE-OF-ACES

Why Logi can't do the same in their firmware is beyond me in that it is 'standard' practice to do so with most analog inputs to A2Ds..

Those spikes are way too big - and just imagine: We would not notice them until the cable really brakes - after warranty has expired...

Originally Posted by ACE-OF-ACES

Not sure what or why they would impliment in firmware a reversal when using pots.. makes no since really.. Thus I would leave that in the rumor column.. More likly is it is a typical pot spike that spikes in voltage in a direction oposite of what your currently at and thus apears to be a reversal of sort

Maybe... When I fix R1 and R2 I could test a high quality poti on the logitech controller... and report here

Originally Posted by ACE-OF-ACES

Yeah you cant go wrong with Leo's stuff.. everyone uses his stuff.. Ill have to look but I seem to recal him having a board that would convert the hall sensor outputs to analog inputs on his board.. if so I would dump all the Logi electronics and go full Leo!

Just went to his website: All but his main page report 404. Obviously he's reconstructing... IIRC *all* his board BU0836, A , and X support the A1302KUA-T Hall sensors directly. I've just ordered some. We'll see.

Artist

[ACE-OF-ACES]

Originally Posted by Artist

IIRC *all* his board BU0836, A , and X support the A1302KUA-T Hall sensors directly. I've just ordered some. We'll see.

I thought I read something on it too.. but.. I seem to recall HALL sensors putting out a pulse like an encoder.. Thus you need something to read/count/monitor the pulses.. Where as BU0836 boards only have analog inputs.. I had a few email exchanges with Leo and actually got the data sheet on the PIC controller he uses on his boards.. I can not remember if Leo added the filtering or if it was a standard feature on the PIC controller he uses.. I got all my notes on it at work and I am at home today so if needed I could look it up on Monday

Saw that his site was down..

http://www.leobodnar.com/

Hope he didn't go out of business!! In that we had planes on building about 25 more of the force sensing sticks

EDIT I looked up that A1302KUA-T.. it is a IC package that has an analog out so yeah that would work as an input to the BU0836.. But I think you will still need to read/count/monitor the pulses? Do you know if this is that the HALL sensor used in the G940?

[Chivas]

I actually thought that I had finally found a joystick to replace my MSFF2, after the new drivers, and firmware fixed the reversal bug. After flying with the G940 for a month or so I got tired of trying to fine tune the centering tension and forcefeedback, and reinstalled my MSFF2 joystick, Cougar Throttle, and Saitek Pro Rudders. I was immediately surprised how much smoother and more accurate my old system is than the G940. The G940 electrical gears are really course, and the software to control the center tension and forcefeedback are also alittle flaky.

I'd try the new hydraulic joystick that is being developed now, but it looks like crap.

[Artist]

ACE-OF-ACES,

Originally Posted by ACE-OF-ACES

I can not remember if Leo added the filtering or if it was a standard feature on the PIC controller he uses.. I got all my notes on it at work and I am at home today so if needed I could look it up on Monday

That would be nice...

Originally Posted by ACE-OF-ACES

EDIT I looked up that A1302KUA-T.. it is a IC package that has an analog out so yeah that would work as an input to the BU0836.. But I think you will still need to read/count/monitor the pulses? Do you know if this is that the HALL sensor used in the G940?

About the G940 I do not know...

I checked Leo's site again (it's still disabled) and found a comment of him in some other forum (mentioned in an earlier poste here): His employee left him, he's on a project, so he won't take orders he's not able to process speedely enough...

Anyway: In *his* forum he once gave instructions for the A1302KUA-T:

The connections are:

BU0836 GND === Hall sensor "-"
BU0836 +5V === Hall sensor "+"
BU0836 INPUT === Hall sensor "O" (output)

Just make sure you follow Hall sensor pinout carefully. It's a semiconductor device and could be damaged if not connected as designed.

and the two illustrating pictures from that post

Artist

[F19_Klunk]

thanx guys for all your input. Even though I would love this set, it seems to me that there is a risk I have to wrestle with the gear, or with Log support... considering the cost, I think I'll pass...

Actually I have no clue where to go to now.. The Hog and the X65 is way too expensive as I won't get any peds.... hmm I have to think hard about it.