Another TX/T300 calibration problem thread...

Aug 20, 2017
17
8
36
Hey all,

Came to this forum searching for a solution to my problems. Seems like there's some very knowledgeable users on here and apparently a lot of the advice given has worked for many people.

Unfortunately I have continued to have issues and am now at a bit of a loss.

I bought the wheel super cheap, knowing it had issues. But reading these forums and having heard the symptoms from the previous owner, I felt confident that it was a cracked hall sensor mount or loose magnet.

It arrived and I powered it on and sure enough, it wouldn't calibrate, just twitched limply from left to right periodically. Hopeful that a recalibration using that PC software linked on the PCars forum might just cure it, I ran it and sure enough the wheel seemed fine following the procedure, appearing to complete it's power on auto calibration fine. I also ensure that it was on the latest firmware and all my drivers were up to date. I took it for a few laps of the Red Bull Ring in PCars2 but after no more than 3 laps, the FFB had diminished significantly.

I exited PCars2 and opened the TX Control Panel in Windows and tested with the test effects to rule out a PCars issue. The test effects were also considerably weaker, confirming it was the wheel. Another symptom was the rotation speed which was slower going left and faster going right. I confirmed this by moving the wheel by hand to the left and the centre spring fired it right back to centre quickly. Moving the wheel by hand to the right and letting go resulted in a limp and slow move left back to centre.

I unplugged the wheel and plugging back in resulted in the failure to auto calibrate and just twitching left and right every now and again.

I cracked open the case and inspected the hall sensor mount. The mount was not cracked at any point and was perfectly fine. The magnet was securely fixed to the motor shaft. I remade the solder joints on the hall sensor board where the wire attached as they didn't look fab and cleaned the magnet and hall sensor board with 99% IPA.

Leaving the cover off and putting the fan into forced cooling mode to rule out overheating, I reran the manual calibration tool and the wheel auto calibrated fine on next plug in.

I then repeated the test on Red Bull Ring but sure enough, the FFB went weak at the same point. Feeling the fins on the motor, it was barely warm. Same issues in the TX Control Panel test forces and same slow left rotation speed problems.

It might just be goosed but I'm hopeful somebody here may know of other avenues of investigation. Would be supremely grateful for any pointers.

If you made it this far then well done, and thanks for reading! :thumbsup:
 
  • Like
Reactions: toweleeie

asteroulis

250RPM
Oct 8, 2012
399
151
You've done all the logical steps known till now. Well done for the research.

I currently have one T300 with similar symptoms , so will keep you posted when I find the solution. Hopefully it's not something on the motherboard (even though I suspect it is).
 
Aug 20, 2017
17
8
36
Thank you for your response Asteroulis. I was wondering the same, though I've yet to delve deeper into the guts of this. I was putting it off in the hope that there may have been another known fix before I did.

I'm actually hopeful that it's the sensor mounted to the daughterboard, I'm quite comfortable with SMD repair after being forced to fix my GPU. Whilst the board is proprietary Thrustmaster, the chip is well marked so should be able to track them down. What do you think, a worthy avenue of investigation or is it likely to be fine, providing the right raw data, and more of a processing problem back on the motherboard?
 

asteroulis

250RPM
Oct 8, 2012
399
151
I'm 99% sure that it's not the hall sensor. You'd have positional issues which you don't. And I've tried that with replacing those parts. I have parts here. But if you're comfortable with SMD repair (I'm not) you could try and fix the mainboard.
 
Aug 20, 2017
17
8
36
OK so you're saying you've tried replacing the daughterboard mounted to the back of the motor with a known serviceable spare to no avail? If that's the case and our symptoms align then there's no point me going to the effort of replacing that.

If I had a known good base, I would be half splitting between the mechanical elements and the brains by swapping the motherboard. It would confirm or disprove your suspicions. Ideally I'd find another cheap U/S one with a different issue like PSU and make one good out of two bad, but they're few and far between it seems.
 

asteroulis

250RPM
Oct 8, 2012
399
151
Yes , I have been fixing T300's and TX's the last couple of years. I do have working ones and lot's of spares here. But sometimes it's a bit strange. One time I had two TX's with issues and swapping motherboards fixed both (!!!!!!) , lol . So really no recipe here. I think if I was handy with electronics I'd have more luck with motherboards, because what the heck if you replace the faulty component they should be fixed, right? I'll keep you posted.
 
Aug 20, 2017
17
8
36
Aha, that would explain your prevalence here and on other forums where these wheels are concerned! That is an odd one, I am genuinely intrigued now though.

I've decided to buy another 2nd hand set, reportedly working, unlike how this one was advertised. I will swap the motherboard purely out of curiosity to see if it cures the original one, and conversely, see if it induces the same fault in the 'new' one. It's no great shakes, I have a G920 to sell to cover the cost. Was hoping to repair the original one easily and cheaply and sell the G920 to be able to afford a better pedal set (it came with the 2 pedal trash set). Pedals will just have to go on hold though, never mind! Anyway, as soon as I do, I'll post back on that.

I do wish I knew much more about electronics. I'm handy at soldering and hot air rework to a degree, but if I knew the way components interacted with each other in more detail, I'm sure I too could repair things a whole lot more successfully. Sadly, unless it's a blatantly burned components or an obvious short, there's not much I can do either!
 
  • Like
Reactions: asteroulis
Aug 20, 2017
17
8
36
Back again. So finding a working TX at a palatable price was fruitless, however I was able to get a (supposedly :cautious:) lightly used T300 with the Ferarri GTE rim at a very good price which I was able to cover with the sale of my G920.

It's getting shipped today so intrigued to know what commonality there is between the T300 and TX? Obviously they will have differences with one being for Xbox and the other being for PS3/4, but I'm wondering if those differences are isolated to just the motherboard, the switch boards on the front and the chassis itself. If the rest is the same it would make sense from a cost to produce standpoint, and I could swap the common guts from the known good T300 (provided it is good) to the TX and find out if the motherboard is the problem.

If it is, I can then keep my eye out for another TX with different issues and make one good one to sell on, or keep the TX for the common parts to use as spares in the T300 in the future.
 
Last edited:
Aug 20, 2017
17
8
36
Of course, I completely forgot about the difference in rotation angle - damn. Oh well, provided all is well with this new T300, I probably won't risk taking it apart until it breaks.

Back on the TX though, I tore it down this morning before work and couldn't make out anything obviously wrong with the motherboard which was disappointing. As mentioned in previous post, I'll still keep my eyes peeled for another broken TX but for now I think I will have to park this.

I would still be keen to learn if you find a solution to your own problematic TX and if I get another I will update here with my findings. I'd just like to take this opportunity to say thank you @asteroulis for leaning in on the thread and offering your advice, it really has been much appreciated :thumbsup: hopefully I can repay the favor if I get to the bottom of it.
 
Aug 20, 2017
17
8
36
Strange development... Maybe in a good way but I'd say it's still inconclusive.

As I mentioned in a post above, I completely tore down the TX to see if there was anything obviously wrong but there wasn't. What I didn't mention was that the JST connector for the hall sensor was ever so slightly not square in the socket on the motherboard on turning it over but it was so slight that I disregarded it. This point might have a bearing.

So in my lunch hour I cleaned all the connectors with IPA, brushed the motherboard down with it and proceeded to reassemble. I was ensuring that all connectors were fully fitted and square and then hot gluing them to make sure they stayed that way.

I had also taken the liberty of getting a replacement hall sensor mount 3D printed by a local guy using a file on Thingiverse before this broken TX even arrived. As I mentioned in the OP, I expected this to be the cause. I never bothered with it before as the original was completely fine and to be honest, the dimensions of this 3D printed one were a bit off, I think the scale in one of the axis was wrong and it was taller than the original. But I thought why not try it, I was interested to see the relationship between the sensor and the magnet at a different distance anyway.

Upon rebuild, I reinstalled the firmware and reran the manual calibration software. Up to this point all seemed as before. However checking in control panel showed the centre to be ever so slightly off by ~1°. I corrected this using X, Y and MODE and launch PCars2.

I completed a 3 lap race of RBR without the FFB going weak and the centre value held. I exited PCars2, disconnected the wheel and reconnected and it auto calibrated again but again was slightly off. This is a vast improvement over the feeble twitching that was being displayed before.

I have noted though, that it is still slower returning to centre moving right to left than it is moving left to right, but this isn't apparent in game.

I decided to "stress test" it by completing 4 laps of the Nordschleife since this is both long, twisty and a scanned track in this game so it is asking a lot of the wheel. It didn't skip a beat.

I'm by nature, a cautious person, so I'm reluctant to claim that this is fixed yet. I still want to test it for longer so hopefully tomorrow evening I can get a few hours non stop in to see how it fairs. And if it holds up, I still want to try and find out what it was, my shotgun approach to repair with this goes completely against logic and my OCD won't allow it! Maybe I'll start easy and put the original hall sensor mount back in, take it from there.
 
Aug 20, 2017
17
8
36
I was right to be pessimistic, the problem hasn't been resolved.

The strange thing is, I've replaced nothing but the sensor mount. All I did was take it apart, clean it and put it back together and whilst the FFB still weakens and it is still slow moving left, it's not nearly as pronounced as before and it's still entirely playable. I spent a few hours on PCars2 last night, mostly because it is now my only wheel until the T300 arrives, but also because it was actually ok. The forces were exactly as I would have expected, just not particularly strong after the first few minutes, like having the strength set to 50%.

Still a shame, more testing to do when the T300 turns up and still looking for another broken TX. Will continue to update.
 

asteroulis

250RPM
Oct 8, 2012
399
151
It's rated at 24V/2.5A, usually measured around 29~31V with the voltmeter. With a 19V psu you'll get less power for sure but it will work. If you have one spare it's ok. But you can order a 24V with 3 or more Amp rating for cheap and it will be a better choice imho.
Anyway, haven't touched my base yet so cannot comment further. When I finally do (unknown when) I'll come back to you.
 
Aug 20, 2017
17
8
36
I've measured the PSU at idle at 32.2V but as I understand it, the PSU in these is unregulated so that's in the right ballpark whilst it's not under load and my multimeter is by no means top of the line or necessarily calibrated!

On a positive note, the T300 has turned up and it appears to be fine on preliminary inspection and a quick power on - will give it a more comprehensive run through either tonight or tomorrow.
 
  • Like
Reactions: super_gt
Aug 20, 2017
17
8
36
Found another broken TX - listing says it smells of china smoke - hopefully I can get it for a song and I can make one good out of two bad... If the seller ever responds to me!

T300 is fabulous btw - now I know what the TX should feel like :D shame about the pedals!
 
  • Like
Reactions: toweleeie

toweleeie

75RPM
Oct 31, 2009
79
17
did anyone get to the bottom of this problem?
same issue on my TX base to a T

ok, so i took it apart after reading a similar post on ISR, and it is the glue on the core thats gone
can any one recommend decent glue for the job? or replacement motors?

thanks

MM
 
Last edited:
Aug 20, 2017
17
8
36
If you need to buy a replacement motor, they're cheap on AliExpress but they don't come with the magnetic top cap attached to the core anyway. So one way or another, you'll be doing some gluing. I would say any epoxy would be fine, it doesn't turn super fast and there won't be any load on the magnetic top cap beyond it's own small amount of mass.
 
  • Like
Reactions: toweleeie