Gunfighter Mk.II base loose X-axis connector

[gerodinis]

I have a Gunfighter Mk. II w/MCG Pro which I hadn't used for a while and while trying it again, it had a severe problem with the X axis. It started centered alright, then jumped around as I moved the stick making flight impossible.
After a successful firmware update and calibration and the problem persisting, I decided to open the base to check the wiring.

As I found out, the white connector on the X-axis board in front of the MARS sensor is loose. Meaning, it just can't be firmly secured in position like the other connectors. Tested for continuity between the cable end that goes to the Y-axis board and the pins behind the connector on the X-axis board with a multimeter and found there is a problem when the cable is moved around. The connector is also not firmly secured on the board and can be moved by hand - but the pins remain steady. So it's not a bad solder.

For the time being I secured the cable with electrical tape on the X-axis board, so that there is no play affecting the connector. But obviously this is not a good solution and will not last long.

What can I do? I think it's a waste to order a whole new replacement board just for a plastic piece. What do you suggest as a permanent solution?

As a recommendation to VKB, I think if the cable was just half a centimeter longer, there would be no problem.

[redribbon]

Yes it will happen again!
That white part of the connector, loose one just push it till it get on it's place but x axis spiking will occur every few months so you'll need to disconnect and reconnect cable/wiring and it will work again.
I guess it is design flaw and longer wiring could solve it.
Maybe ypu can prolong it by yourself or try to buy new male connector and build new cable by yourself since VKB seems doesn't care much about it.
What i'm afraid is after many reconnections connector and it's pins will worn out and circuit board pins (soldered ones) could be damaged too.

[rtrski]

I got a new gunfighter base along with a Kosmosima grip recently and noticed that they had significantly improved the quality of both these cables and of the connections with some positive reinforcement pieces. This is compared to a first-generation gunfighter mk ii base from December 17.

I asked if retrofit parts were available for purchase and was told no. However I have purchased my own good silicone insulated wire and the right size connectors to attempt replacement of the cable myself one of these days. If I notice it behaves much more smoothly and doesn't require periodic reseating I will update this thread.

[gerodinis]

Thanks for your answers.
I'm thinking I could try a drop of epoxy to secure the cable on the PCB (not on the connector itself!) so that it stays in place.

[redribbon]

I got a new gunfighter base along with a Kosmosima grip recently and noticed that they had significantly improved the quality of both these cables and of the connections with some positive reinforcement pieces. This is compared to a first-generation gunfighter mk ii base from December 17.

I asked if retrofit parts were available for purchase and was told no. However I have purchased my own good silicone insulated wire and the right size connectors to attempt replacement of the cable myself one of these days. If I notice it behaves much more smoothly and doesn't require periodic reseating I will update this thread.

hi can you give us info on connector and wiring size and model?
It would be good if VKB stuff provide us with such information if they are unable to provide us upgrade over this flaw.

[AeroGator]

It would be good if VKB stuff provide us with such information if they are unable to provide us upgrade over this flaw.

My fault I haven't seen this thread before.
There will be something done about this.

[Roke]

That’s great to hear.

Being able to take pre-emptive action to prevent the issue from happening again would be very welcome.

[rtrski]

hi can you give us info on connector and wiring size and model?
It would be good if VKB stuff provide us with such information if they are unable to provide us upgrade over this flaw.

Sorry for not replying quickly.

Yes I think I have the connector information at home and can post again later. As for the size of wire etc that I'll be using, let me actually do it and prove it works before I give you values that might not be correct.

[rtrski]

Yes I think I have the connector information at home and can post again later. As for the size of wire etc that I'll be using, let me actually do it and prove it works before I give you values that might not be correct.

OK, replying to myself. I finally built it a couple days after that last post and have given it a little time. I don't believe that I have caused any new problems, but I'm also not 100% sure I've cured all problems. Details to follow.
Please note: all this pertains to the VERY FIRST lots of Gunfighter Mk II (B) gimbals that were made available for sale in Dec 2017, in my case purchased along with the MCG-Pro (B) with Cyrillic markings and shipped to the US. I have had no such equivalent behavior from a newer GF Mk II B that came with a LH Kosmo Premium just this spring.

I'll try to embed an Imgur Gallery here but it may not work:
https://imgur.com/a/dLHOy1X#LPgiSpM
The photos didn't upload in order and while I can view them in original image name order seems like to others they always show in order uploaded....sorry. I tried to put titles and comments to make them clearer.

ISSUE: The cable that goes from the gimbal X-axis sensor board to the Y-axis sensor board (which also accepts cabling from the stick shaft itself, and then out to the black box connector) appears to be the primary culprit of occasional intermittent behavior. In my case after not using the stick for a little while (like during the work week) I have suffered the occasional phantom press of one of the 3 4-way-hat "center push" buttons [usually the one labeled "Master" at bottom center of the MCG-Pro back face, but sometimes either the index finger side or thumbrest hat positions too], plus some glitchiness in the way the X-axis value gets sensed...it might center just fine but then try to travel to -X without stick actuation.

In most cases manipulating the stick through all positions (doing circles clockwise and counter clockwise, or "zig-zagging" by going max to min X while slowly panning from max to min Y) followed by a calibration will eliminate the problem for quite some time. Occasionally I have resorted to remove the gimbal from the housing, unseating the cable, reseating the cable and kind of "retraining" the wiring path to fix. So after a couple years I finally decided to try my own replacement to see if it would help. I believe I have observed (see gallery link and pics above for further details) that the general 'stiffness' of the PVC-jacketed wire, plus the routing and length, appears to both allow the cable to be crowded against the inner wall of the gimbal under operation, as well as get 'tugged' off of the X-axis board end. As the connector and plug at that end may themselves together kind of slide down the header pins this may be the source of the intermittent connectivity.

ATTEMPTED FIX: My replacement uses silicone-jacketed wire. I should've used something on par with what was already present but all I had was 24 gauge, which I'd judge to be at least 2 increments bigger than what was there (28??). I'd not necessarily recommend anyone attempting to do what I have done here without downsizing. I've got the connector endpoints to make another 49 of these assuming I don't goof up pins as I'm attempting crimps, so I can take another stab later.

Connectors: the connector ends (and this information was provided directly to me by VKB headquarters, after being forwarded their way by rrhodhe from the US VKBcontrollers side) are JST Mini-micro GH 1.25mm 4-conductor. The female socket pins crimp to the wiring, and the male header pins are on the boards. Without any formal endorsement intended I found them off eBay:

https://www.ebay.com/itm/100-Sets-Mini-Micro-JST-GH1-25-1-25mm-2-3-4-5-6-Pin-Connector-Plug-Female-Male/273191492901?ssPageName=STRK%3AMEBIDX%3AIT&var=572461579242&_trksid=p2057872.m2749.l2649

In terms of length, after removing the old cable (photos first to make sure I knew how the 4 wires travelled from end to end) I stretched it on a sheet of paper and marked its length to the backside of both plastic parts, and cut my new cable about 1.5 to 2 cm longer. After stripping and crimping the pins I've still got about 1 - 1.25 cm [~1/2 inch] more length than the original.

INITIAL VERDICT: I was a bit worried I was introducing problems with my overly-large wire and the chance for stray single Cu strand bridging between the 4 conductors, but so far I have not seen that. After replacing the original cable with my own (keeping the original of course as backup) and re-calibrating, I saw no intermittence in 2-3 play sessions. But then after 12 days or so of no play, and a Windows Update (probably not related) I launched VKB Joy Tester and ran the stick thru the range and did see it glitch the brake axis and "button 18" (same back lower center Master hat mentioned before) just *once*....then it went fully quiet and acted nicely. I even tried yanking the stick hard in multiple directions and let it just oscillate back to center (I keep dampening almost off and use weak springs, because I'm more of a space simmer). No further misbehavior.

So I believe my slightly longer more flexible run may have corrected the crowding and tugging issue, but there may be just enough of a slight oxidation build-up after lack of use that might weaken the female-pin-socket to male-pin-header contact points somewhat, and that has to be burned back thru by jacking with it a bit...or I may in time need to open it all up and unseat/reseat that connection again because of the way the connector plastic and socket plastic move together vs. one plastic end being fixed to the board more securely.

Last point, from looking at pictures of my much newer Gunfighter Mk II (B) base that I got with my Kosmosima this year, there's been some improvement to the positive attach of the 3-wire cable up to the stick and of the 4-wire cable from the Y axis sensor board down to the Black box connector board, but no obvious visible change to the X- to Y- connection endpoints that I'm talking about here, aside from the cabling itself feeling like more flexible, higher quality insulated wire. So this issue may only really exist for the first generation folk like me.

I hope this helps someone. FYI crimping those tiny <bleep!>ing pins on to the wires neatly is a royal pain, but might just be harder for me because of my previously mentioned use of too-large a gauge of wire. Strongly suggest you have some sort of "third hand" grip and magnifying glass to assist the assembly work if you try this yourself, and do NOT over-strip the wire ends, strip just barely enough to crimp down on. (Don't overinsert the wire into the crimped pin either, as you need the socket end of the pin open to slide over the header male pins to make a mate.)

[Roke]

Thanks for taking the time to write this up rtski...

I hope that has solved the issue for you. Will be interested to know if you stay glitch-free over the coming months!

Cheers

Roke

[PhoenixCNE]

Yes, big thanks for being so thorough. I have the exact problem you describe with my gen 1 gunfighter and have always found it to be quite curious how the phantom button presses problem "fixes" itself after you give the stick a good run through all of its range of motion. It has been the 1 quirk so far with my gunfighter but has been so easily manageable that I haven't given it much more than raised eyebrows.

[rtrski]

Yeah, even with my new cable - same thing after several days of non-use. I open JoyTester and wiggle a little, see a few glitches, actuate the brake lever 2-3 times and then run the stick around full range of motion until I see the button presses stop. Good to go thereafter.

So my cable replacement may have been a waste of time (then again previous to it I might've needed to go as far as disconnect/reconnect/recalibrate to get glitchies to stop, and (so far) I have not had to again, so jury still out.

Really strange, but quite survivable.