Honeycomb yoke dead zone concern

[paulpenney 3]

48 minutes ago, ark said:

I find it disappointing that the Honeycomb yoke appears to suffer the same internal controller A/D software issue as the Saitek and if I were to buy this 'brand new' yoke I might have to modify it with an 'extra' A/D board just like I did for the Saitek. 

Al

Yeah, like seriously - how much more can a higher resolution converter be? The answer is probably not that much. 

[ark 1,295]

1 hour ago, paulpenney said:

Yeah, like seriously - how much more can a higher resolution converter be? The answer is probably not that much. 

Hard to know for sure, but I don't think it is necessarily an A/D resolution issue. It could be an issue with how the yoke's internal controller chip firmware processes the A/D values near the axis center position.

Al

[ark 1,295]

2 hours ago, MarkDH said:

The switches are SPDT ON-ON switches.

OK, I see. Thanks.

Al

 

[oemlegoem 29]

2 hours ago, MarkDH said:

The switches are SPDT ON-ON switches. The top left switch (labelled ALT) is wired as buttons 12 and 13. If you watch it in Windows Control Panel, button 12 is on when 13 is off, and vice versa. Hence there isn't any problem with these switches in FSUIPC.

Strange. For a cost optimised design, it would be more logical to use SPST switches = less wiring in a cramped space, and generate the "missing" switch through software in the controller.

Edited November 16, 2019 by oemlegoem

[DaveCT2003 1,820]

4 hours ago, paulpenney said:

The problem overall is caused by the software in the onboard joystick controller

Okay, you're actually talking about FIRMWARE, not software. Not semantics, there is a big difference.

The firmware is already on my list to speak with Honeycomb about next week.

 

[paulpenney 3]

47 minutes ago, DaveCT2003 said:

Okay, you're actually talking about FIRMWARE, not software. Not semantics, there is a big difference.

The firmware is already on my list to speak with Honeycomb about next week.

 

Correct its the firmware. The firmware's raw output has a large dead zone. Hopefully they can release an update as it appears the microcontroller is programmable via the USB. 

[DaveCT2003 1,820]

4 hours ago, paulpenney said:

Correct its the firmware. The firmware's raw output has a large dead zone. Hopefully they can release an update as it appears the microcontroller is programmable via the USB. 

Hey Paul,

Any chance you can make a short video to demonstrate the deadzone issue you're seeing, that shows what I'm typed below?  If you need some help setting up the recording please let me know, I'd be very happy to help.

1. Start with showing your AXIS assignments,

2. Then show your AXIS Calibration Settings,

3. The  show the following all at the same time while you move the Yoke along the X and Y axis very slowly:

           a. Aircraft Yoke

           b. Windows USB Game Controller Calibration Screen

           c. JoyTester Display

 

This will help me to see the differences between your Yoke and mine, and give me some good data to show and discuss with Honeycomb this coming week.

 

To create the video I recommend using OBS Studio, but if you have something different that you're used to then that's certainly fine.

 

Many thanks in advance!

 

Edited November 17, 2019 by DaveCT2003

[ark 1,295]

6 hours ago, DaveCT2003 said:

The firmware is already on my list to speak with Honeycomb about next week.

I wonder if some yokes already have updated or different firmware versions? If so, that might explain why the severity of the dead zone issue varies between users. If, as has been reported, there are no mechanical issues and the firmware is the same for all, then I would expect the 'direct' USB yoke output would show the same thing for all users.

Al

Edited November 17, 2019 by ark

[ark 1,295]

18 hours ago, paulpenney said:

Hopefully they can release an update as it appears the microcontroller is programmable via the USB.

Interesting - where did you find the info on the programmability of the microcontroller firmware via USB?

Al

Edited November 17, 2019 by ark

[MarkDH 234]

 

[oemlegoem 29]

Mark: I appreciate very much the high professional content of your video. More of that! 

[paulpenney 3]

18 hours ago, DaveCT2003 said:

Hey Paul,

Any chance you can make a short video to demonstrate the deadzone issue you're seeing, that shows what I'm typed below?  If you need some help setting up the recording please let me know, I'd be very happy to help.

1. Start with showing your AXIS assignments,

2. Then show your AXIS Calibration Settings,

3. The  show the following all at the same time while you move the Yoke along the X and Y axis very slowly:

           a. Aircraft Yoke

           b. Windows USB Game Controller Calibration Screen

           c. JoyTester Display

 

This will help me to see the differences between your Yoke and mine, and give me some good data to show and discuss with Honeycomb this coming week.

 

To create the video I recommend using OBS Studio, but if you have something different that you're used to then that's certainly fine.

 

Many thanks in advance!

 

Dave,

Sorry to say but I have already modified my Yoke to alleviate the issue. I created a guide for others that want to do so in the second video below. In the first video below the issue is clearly demonstrated and I did confirm it in both the windows calibration screen and using JoyTester before the modification. The result of the modification is absolutely zero deadzone and great performance now. 

 

[DaveCT2003 1,820]

56 minutes ago, paulpenney said:

Dave,

Sorry to say but I have already modified my Yoke to alleviate the issue. I created a guide for others that want to do so in the second video below. In the first video below the issue is clearly demonstrated and I did confirm it in both the windows calibration screen and using JoyTester before the modification. The result of the modification is absolutely zero deadzone and great performance now. 

 

Thanks my friend!

 

[ark 1,295]

It will be interesting to see what Homeycomb's response to all this is. Hopefully it will be positive and result in the Honeycomb yoke reaching its full potential.  I see two issues:

1. The A/D resolution. Currently based on 8 bits which provides 256 different values for each axis. I'm not optimistic this can be changed with firmware, but clearly depends on the type of controller being used. While the resolution is not great, it probably is 'good enough' and would likely not be a major point of interest if there were no other concerns.

2. The dead zone around the center point of each axis. This seems to me to be the more critical issue and fortunately, likely more amenable to a firmware fix then the A/D resolution.  I suspect creating a dead zone around an axis center point, such as by ignoring a couple of values on each side of the center point, is one way to help ensure reliable 'return to center' when the yoke is released.  The downside, of course, is a dead zone, and the lower the A/D resolution the greater the impact of ignoring a value, or in effect, equating it to the true 'center value'.

As a side note, for those motivated to implement a dead zone fix but are unfamiliar with the Arduino, you can use an A/D board like the Leo Bodnar 12 Bit Joystick Controller being used here www.youtube.com/watch?v=YEqkz2JxoFo  to eliminate the dead zone in a Saitek yoke (did this, works great). The advantage of this approach is no firmware programming is required. The disadvantage of this approach is no firmware programming is required -- and thus you can't make changes if so desired at a later time.

Al

Edited November 18, 2019 by ark

[paulpenney 3]

10 hours ago, ark said:

Interesting - where did you find the info on the programmability of the microcontroller firmware via USB?

Al

I plugged it into the arduino IDE and the AVRISP spit out some information on the chip, but I couldn’t write to to it. Probably a matter of having the right board setup