Throttle - Index rotary jitter

SOLVED! solution here: https://forums.eagle.ru/showpost.php?p=3725345&postcount=17

/***********************************************************************/

The index-finger rotary axis on my throttle has jittering problems, I've noticed that issue immediately but lately it's gotten a bit worse. I've tinkered with the software a bit but so far no luck. Has anyone found a way to fix this issue?

The ideal solution would be an interpolation of the readings from the potentiometer, to cut the input spikes. Deadzones and curves don't really work.

I don't know if this would help, but you can increase the smoothing for that axis. See if it's better, worse or the same.

I think there's no fix for this problem since this axis uses a non hall sensor, only the two throttle axis has. Looks stupid but this rotary with a conventional pot was the reason that I've asked for a refund, I knew it that soon or later I would have trouble with this axis so I just kept the order for the T-50 base and I don't regret it.

Maybe for a next model Virpil changes these pot for some hall sensor, there's no reason for such a great product to use those cheap pots anymore.

It's not a big deal per se since I use it for the zoom. It's just that sometimes it starts moving by itself, shaking the view as if I were entering VRS. Scary stuff sometimes.

@Seb: I've already maxed the smoothing but it doens't really seem to help.

Oh well, I'm afraid I had to write support again..

Just a clarification for my understanding.

Does the axis jitter in windows game controller and in the VPC setup software, or does it just jitter in DCS?

It behaves in the same way wherever I try.

I have been one of the first to receive the throttle; I recorded three months ago and the jitter is clearly visible at 2'32".

I remember that at least a couple other people had the same issue (I don't remember if I met them here, on reddit, or somewhere else though) so I was wondering if someone has fixed it.

I see.

To better monitor the situation, use VKB JoyTester.

Have you calibrated the throttle using Windows game controllers?

The index-finger rotary axis on my throttle has jittering problems, I've noticed that issue immediately but lately it's gotten a bit worse. I've tinkered with the software a bit but so far no luck. Has anyone found a way to fix this issue?

 

 

The ideal solution would be an interpolation of the readings from the potentiometer, to cut the input spikes. Deadzones and curves don't really work.

I have had the same issue from day one. Only notice it say when you are partly zoomed and left in that position so not sure if its an issue with the sensor or the software

UPDATE: I just emailed Virpil support, lets see what they say about the issue

Have you calibrated the throttle using Windows game controllers?

No. Windows' calibration tool is a big no-go for Virpil stuff. I have used it more than one year ago when I got the T-50 but later I had to fix the curves with DxTweak. There's no point nowadays.

@mithandra: nice, keep me posted mate :)

No. Windows' calibration tool is a big no-go for Virpil stuff

That’s why I asked. :)

I have had the same issue from day one. Only notice it say when you are partly zoomed and left in that position so not sure if its an issue with the sensor or the software

 

UPDATE: I just emailed Virpil support, lets see what they say about the issue

Same here, from day one the index rotary wheel has had quite a lot of jitter. It seems to be quite stable at the min/max/center position, but in between those positions the rotatory start to jitter quite badly making it pretty much useless. Ruins the premium feeling a bit for me.

I already tried maxing out the smoothing settings in the Virpil configuration but... :joystick:

"Good" to see that I'm not the only one. I'm quite sure they can smooth the pots reading via firmware. Let's see if @mithandra gets an answer from the support.

I logged a support ticket with Virpil, they told me that this issue was resolved in 2 firware updates ago. I downloaded and flashed the virpil throttle and stick (stick because why not have the latest version)

I still have the judder at certain points, however overall its better than it was

Cheers Specter

I followed your input and updated the firmware. Something has changed but slightly, the jitter is still there. I will play with the settings a bit when I have some spare time and see what happens (for instance, I haven't checked if the axis settings have been overwritten by the firmware update).

It's no matter of software, firmware etc etc, it's mechanical, it's a common potentiometer like Saiteks and CH Products. They must have a better quality, but they're not hall effect sensors so there will be spikes/jitters no matter what you guys do, unfortunately.

It's sad because it's a well place rotary, you could use for many precision inputs like radar antenna, prop pitch and many other good stuff on DCS. Hope in the future that Virpil change this pot for a hall sensor pot or something more precisely.

By Tapatalk + Samsung S9

Anyone that's having some judder on this wheel axis - load up the VPC Software, click "Step 1. Load Device Profile" on the VPC MongoosT-50 Throttle. Double click on the axis that is giving you juddering - we've added a dynamic deadzone/filter "Dynamic (1/10)%".

This is set as tenths of a percentage and will filter out these small judders without impacting the accuracy of the axis. Try setting it to "5", click "Save" at the lower right and then click "Step 3. Save Device Profile".

Please report back!

It's no matter of software, firmware etc etc, it's mechanical, it's a common potentiometer like Saiteks and CH Products. They must have a better quality, but they're not hall effect sensors so there will be spikes/jitters no matter what you guys do, unfortunately.

It's sad because it's a well place rotary, you could use for many precision inputs like radar antenna, prop pitch and many other good stuff on DCS. Hope in the future that Virpil change this pot for a hall sensor pot or something more precisely.

 

By Tapatalk + Samsung S9

Have you worked with potentiometers and resistors on a PIC? I guess no. Via firmware you can do almost whatever you want, in this case interpolate the values read by the input pins in order to smooth the HID output.

In fact, here comes the cavalry! I'm afraid I won't be able to test the setting for a while due to the holidays. If I manage I will update you, @Cyph3r.

Have you worked with potentiometers and resistors on a PIC? I guess no. Via firmware you can do almost whatever you want, in this case interpolate the values read by the input pins in order to smooth the HID output.

 

 

 

 

 

In fact, here comes the cavalry! I'm afraid I won't be able to test the setting for a while due to the holidays. If I manage I will update you, @Cyph3r.

I'm not cursing Virpl man, I'm very happy with my T-50 base, but IMO it's sad that they not used a better pot for this axis.

Anyway, good luck messing around with the software...

By Tapatalk + Samsung S9

why dont you just buy a hall sensor and mod it? or

replace the potentiometer

Anyone that's having some judder on this wheel axis - load up the VPC Software, click "Step 1. Load Device Profile" on the VPC MongoosT-50 Throttle. Double click on the axis that is giving you juddering - we've added a dynamic deadzone/filter "Dynamic (1/10)%".

 

This is set as tenths of a percentage and will filter out these small judders without impacting the accuracy of the axis. Try setting it to "5", click "Save" at the lower right and then click "Step 3. Save Device Profile".

 

Please report back!

Hot damn! I just did that and after quick test it truly seems to work. There is still a bit of judder at 89-91% at setting 5, but I increased the percentage setting to 6 and that seems to eliminate the pot jitter/juddering completely. So far so good, thanks for the tip! :thumbup:

Although in future you guys might want to consider using a better pots. I would expect something like this from Saitek/Logitech...:music_whistling:

Hot damn! I just did that and after quick test it truly seems to work. There is still a bit of judder at 89-91% at setting 5, but I increased the percentage setting to 6 and that seems to eliminate the pot jitter/juddering completely. So far so good, thanks for the tip! :thumbup:

 

Although in future you guys might want to consider using a better pots. I would expect something like this from Saitek/Logitech...:music_whistling:

Great news!

Well - typically pots will have this sort of filter built into the firmware from the start, so you'd never notice. Our software and firmware is always evolving though so you get to see these progressions take place!

If I'm not wrong, Sokol's used this hall sensor to upgrade my HOTAS Cougar throttle few years ago http://www.potentiometers.com/Series6120.cfm

I don't know if this axis uses a pot with this same dimensions but there's a lot of options in the market to get rid of from conventional pots. Althoug the price should increase with a new upgraded throttle I would definitely go for the Virpil throttle without thinking twice.

Again, I am not belittling Virpil and its products, they are a great company and as a customer I am just putting my opinion for more awesome products, with better components. :thumbup:

I have set "Dynamic (1/10)%" to 5%. I use the axis as zoom, so I don't really case about having a perfectly correct read; I'd rather have it stable.

I'm going to try it ASAP and come back with a feedback. Thanks Cyph3r :)

Hoo really? Let's time say for itself. Let's see if this beloved pot will resist for more than two years with constant use.

I'm not cursing Virpl man, I'm very happy with my T-50 base, but IMO it's sad that they not used a better pot for this axis.

Anyway, good luck messing around with the software...

 

By Tapatalk + Samsung S9

It's not a matter of pots (not entrirely). I can show you the values read by the analog pins of the same ATmega32U4 I use on my control boxes and those values are usually unstable (many reasons, from the lack of precision bit-wise to electronical inferences). As I said, you always have to interpolate those values in order to provide a stable output.

Acutally, if you look at the values read by the hall sensors, they too are "unstable" and it's not because of Virpil (of course quality and eletrotechnical design matters); it's totally normal. The fact that you don't notice on other products is simply because they already interpolate the numbers from the factory.

why dont you just buy a hall sensor and mod it? or

replace the potentiometer

It's an option but not worth the effort in my case. Pots are not a bad solution per se, issues come on the long-term (I bought my CH stuff 15-16 years ago and they still work fine).

Alright, I finally managed to test for longer than 5 minutes and I must say it's definitely better. Thanks Cyph3r!