Author Topic: Thrustmaster T150 racing wheel pcb faulty (no voltage to cpu) (Read 18287 times)

trez0r · « **on:** May 02, 2018, 08:01:12 pm »

Hi!

The wheel is not recognized by the pc and does not show any signs of life.

The origin USB 5V and PSU 25V are present but later the voltage "gets lost".

Pins 8 and 44 on the cpu (nxp lpc11u35) which are the VDD do not have any voltage on them.
Retracing back to coil which supposedly converts 24->12V - also have no voltage present.
Also half of the "big" caps at the start of the supply lane don't get any voltage.

By no voltage here I mean nothing much above 0V, there may be something like 0.1V or so, but I don't remember now - I'd have to re-measure to get exact numbers.

Short circuit somewhere? It's a multilayer pcb.
Hoping to get some pointers on finding out / fixing the problem. I don't have a strong background on electr(on)ic stuff, but I can solder and have a fairly good multimeter to work with.

A picture is worth a thousand words - see attachment

Thanks in advance!

LateLesley · « **Reply #1 on:** May 02, 2018, 10:10:50 pm »

First step, check the 0 ohm link resistors (all the ones marked FB17 - FB24), you can also check for power at each side of each one.

I've circled the bits to check in Red.

Also I'd maybe check the Large Diodes around the Caps, D8 and D9, the Diodes next to the coil D2 and D4, and the one at the bottom next to the 2 pin connector (I can't see what this one is labelled)

trez0r · « **Reply #2 on:** May 03, 2018, 07:55:50 am »

Hello!

First, thanks for chiming in!

I checked those resistors you circled and they all came up empty.

I suppose there's no other way to check the diode unless one end is desoldered?

I'll give it a shot later when I have some more time to tinker around.

Thanks!

EDIT:
I checked the diodes and they seem not too far out, please see attachment.
Or is 0.2V considered low for a typical silicon smd diode?

Actually I just looked up their datasheets:

D2/D8 - B260 (SCHOTTKY) datasheet max FVD 0.5-0.7V (https://www.diodes.com/assets/Datasheets/ds13004.pdf)
D4 - B120 (SCHOTTKY) datasheet max FVD 0.5-0.7V (https://www.diodes.com/assets/Datasheets/ds13002.pdf)
D9 - B560 (SCHOTTKY) datasheet max FVD 0.55-0.7V (https://www.diodes.com/assets/Datasheets/ds13012.pdf)

Forward voltage drop specified in chart of the datasheet is a bit higher than the measured even at lowest current at room temp. Not sure what to make of it.
Manual of my multimeter states 1mA is used for diode test. Maybe these readings are on point then at such low current.

LateLesley · « **Reply #3 on:** May 03, 2018, 02:43:23 pm »

0.2V does seem a little low, but I think since they are BIG Schottky diodes, they may be fine. Another way to check would be to apply power to the board, and just check the DC voltage at either side of the Diodes in relation to ground. You should see voltage at either side, with a small drop across it.

I'm having a bit of a hard time tracing the circuit, is there any tracks on the underside, as it does look like the board had a lot of via's in it, some of which I'd like to follow, to figure out how the circuit is set up? If you could get a better picture of the left half of the board top and bottom, it might help in tracing the circuit and finding out what's wrong.

I'm also trying to see what the markings are on the 6 pin IC under the inductor L1. I'm wondering if it's a PWM controller for the power, and maybe a feeder resisitor to power it is gone. Could be one of the SMD caps pulling things down too. It's hard to tell without knowing if there is anything under the board.

One wee trick i've seen, is to pour 99% isopropyl alcohol on the board with it powered, and watch for what gets hot, the alcohol will evaporate quicker on hot components. If it all seems ok, then chances are its maybe a feed resistor to the PWM circuitry for the DC-DC converter, or a shorted FET somewhere not switching. At this point, a lot of components are suspect, and we need to build a picture of the circuit, and follow the voltage, to see where it disappears. That narrows down where to look for the faulty part.

trez0r · « **Reply #4 on:** May 03, 2018, 03:14:41 pm »

Here's a slightly better pic of the board. And a pic showing both sides (on this I haven't yet soldered the diodes back on).
Actually there are more diodes on the underside, I should probably test those too.

EDIT:
Just checked them, they too appear to be fine (one way 0.185V FVD)

In fact I have a simple infrared camera (seek thermal), but the bad news is that I had already used it to watch the board when powered and didn't find any parts reaching excessive heat.

I can take a picture with it if you'd like.

EDIT:
That 6 pin IC under the inductor seems to be a controller similar to this https://www.diodes.com/assets/Datasheets/AP4313.pdf - the ground pin matches that of the diagram. There's no meaningful voltage on any other pin though.

trez0r · « **Reply #5 on:** May 04, 2018, 08:26:06 am »

Attaching a thermal image of a powered board. The t° delta is just 4 degrees, so basically none.

LateLesley · « **Reply #6 on:** May 04, 2018, 12:36:19 pm »

OK. This is gonna be difficult. It's a multi layer board. So we can't see some of the tracks. But we can trace some of the stuff.

Can you power up the board, and with a multimeter, set to voltage, Black lead on ground, Probe the following points, and give me the voltage reading?

either side of FB26, top left of the board, component side.

Reading of the middle and edge pin of Q13 (marked 2A next to the C48 label, top edge of the board)

The reading on D4, near the inductor L1. Right hand side should have power.

This is just testing the initial power rail.

Edit - also just noticed what looks like a damaged component trying to escape, just above the 4 pin connector (D5??) just down from the R53 marking on the board. It needs checking out.

trez0r · « **Reply #7 on:** May 04, 2018, 03:09:12 pm »

Here are the readings:

FB26 - 4.929V
D4 R/L - 4.919/4.944V
Q13 - 4.929V (M), 4.233V (L), 4.944V (R)

The suspect you mentioned is a diode - it appears to be in contact despite it being out of position. I added a blob of tin on that end to be sure. It tests out ok - 25.8/25.7V.

Waiting for your input on next things to test. Thanks!

LateLesley · « **Reply #8 on:** May 04, 2018, 03:26:54 pm »

OK! So we do have a 5V rail. Now we just need to figure out where it going missing.

There's a resistor next to the label R25 (E62 on it, next to a Via) , can you check that there is 4.9V on the top pad of that, and can I get the part code off of U2? Just so I can grab the datasheet, it might help me figure out more of the circuit.

trez0r · « **Reply #9 on:** May 04, 2018, 04:43:19 pm »

U2 is lpc11u35f see here https://www.nxp.com/docs/en/data-sheet/LPC11U3X.pdf

R25 has 4.94V on pad closest to edge and 3V on the other.

LateLesley · « **Reply #10 on:** May 04, 2018, 08:32:37 pm »

ok.... I'm trying to trace the rest of it, and it's now getting too difficult to do with just pictures. pin 44 I suspect is the main power to the chip, but seems to be coming from an inner layer, just below the left side of C17, below the chip. it seems to have C17,C18,C22 and C21 on the powerline before the chip. Can you do an ohms reading across one of those caps?

Can you also check C13 and C16?

And can you clean the solder off the crystal pins, and resolder them? (the two solder joints next to SCLK on the underside of the board) They don't look too clever.

I don't know what tools you have, but you need a solder sucker or desoldering braid, clean off the old stuff, and resolder with fresh stuff.

It is getting hard to help now though, we're into trying to trace layers inside the board, and that's hard to do without having it to hand to probe about.

I'm beginning to think it might be a bad Via, not passing power through. Cos the power seems to get close to the chip, but not actually getting to it.

trez0r · « **Reply #11 on:** May 04, 2018, 09:23:57 pm »

C21/22 stop around 1.5 kOhm
C17/18/13/16 all stop around 20 kOhm

That sloppy soldering on crystal is my doing.

When I started this debugging ordeal, I replaced it with another same mhz unit just because someone said they had crystal fail on them in similar wheel.

LateLesley · « **Reply #12 on:** May 04, 2018, 10:27:33 pm »

ok.... so we don't have a shorted cap. I just just looked at the spec's of the chip, it seems to run of 3.3V - so that means there's going to be a regulator (probably a 5V - 3,3V LDO) to power the chip. Thats what we need to find next. I'm going to have a hunt around the board. might be tomorrow before I reply though, I'm kinda tired now.

Edit - OK, VR31 is on my radar, the 4 pin package top middle, Above the microcontroller U2. What's the marking on that component, as I can't see it clearly.

trez0r · « **Reply #13 on:** May 05, 2018, 07:00:19 am »

VR31 is "WR1 63".
I measured it's pins - lower left has 5V, right 0V, top left 0.015V, right 0.019V (top being closer to edge of the board).

About crystal - it appears to connect just fine - it's legs measure connectivity to chip pins 6 and 7 (XTALIN/XTALOUT).

Today I'm going offline till late afternoon, so I'll not be able to give any feedback till then. Again, really appreciate your help (and the debugging is exciting in its own way)!

LateLesley · « **Reply #14 on:** May 05, 2018, 05:42:59 pm »

Well, I'm afraid I'm kinda at the end of the road. I can't find the datasheet for VR31, I suspect it is an LDO, but I cannot confirm or get the pinout. It could be a transistor, there's a few different components use that type of package. If it is an LDO, it'll have an enable pin - which I suspect isn't getting powered due to the low readings, and I'm scared to inject 5V, as if we do it to the wrong pin, the microcontroller might get 5V and fry it.

You could try a continuity measurement between the bottom of C16, and the top 2 pins of VR31, that would confirm it's what supplies the microcontroller power. But even then, I can't trace where the enable signal is coming from, you'd need to probe about to find it. And that's assuming it IS an LDO, and not a transistor. I'm getting kinda stuck I'm afraid! Without the circuit diagram, or datasheets, it's difficult, unless it's a simple fault.

trez0r · « **Reply #15 on:** May 05, 2018, 07:45:06 pm »

What if we inject 3.3V? Could be easily doable from PC PSU.

Bottom right is ground for that 4pin IC. Top pins measure 6 and 10 MOhm to C16, so not really connecting.
EDIT:
Top pins are actually (L) USB Data (+) and (R) Data (-), lower left is USB 5V and lower right is ground.

trez0r · « **Reply #16 on:** May 06, 2018, 09:40:48 am »

Did some probing.
Almost all of the underside "lone" vias are connected to U2 power supply lanes (C21/C16/etc). Except the two + two marked in picture - those are connected in pairs one to unknown point and the other closer to 4pin power connector - to 26V input.
Do you think it would be helpful to mark all these vias on the front side? If so I'll do it later.

LateLesley · « **Reply #17 on:** May 06, 2018, 05:52:11 pm »

OK, so that kinda eliminates VR31. The microcontroller runs off 3v - 3.3v, and that must be getting generated somewhere on the board. So there must be a regulator circuit, an LDO, DC-DC, and Power Management chip of some sort doing it. What you need to do is probe the positive rail side of C16, and/or C21, and find where it is connected to. Hell, even just check continuity on both C16 to C21 to make sure they are on the same rail. If we can find the points where that rail goes, it could narrow down where it's being generated. I'd poke around that 6 pin device under the inductor, and around U1, and even U3, although it's a motor driver chip, so I don't think it'll be anything to do with it. But it's got to be being made somewhere! It's frustrating when they don't make it easy. :-)

trez0r · « **Reply #18 on:** May 06, 2018, 09:09:56 pm »

Ok, I'm pretty sure I marked them all. All the red dots connect. Which makes me think that inductor is actually supposed to convert to 3.3V if it had anything incoming (or am I wrong?).

LateLesley · « **Reply #19 on:** May 07, 2018, 04:48:32 pm »

No, I do not think you are wrong. I think we are on the right track, I'm suspecting that 6 pin package. I think the circuit is similar to this :-

The problem is, I can't find any datasheet for that chip. (It is marked BN0U 15? )

In fact, I think the circuit is more like this one -

So, upshot is, I think that's the faulty component, and we either need to find that part, or rig something similar into place. I think what you need to do now, is try and figure out the pinout of that package. Find out what pin is Vin, and what voltage it is, we pretty much know pin 6 goes to the inductor, we need to find the enable and feedback pins. one will have a resisitor to Vin, feedback will have resistors to V out and ground. Probably high value ones.

trez0r · « **Reply #20 on:** May 07, 2018, 07:10:14 pm »

It's either BNOU or BN0U (yes, underscore below last symbol) 15.
I couldn't find anything on it in web as well.

I probed that IC and it seems it doesn't get any volts on any leg, so I guess the problem might be somewhere earlier. Also could it be that we are looking at it from wrong direction - as in we expect volts from it to inductor, but in reality it's the other way around (if that makes sense)?
Anyway please see the picture, I mapped the close proximity to that IC (green is ground).

LateLesley · « **Reply #21 on:** May 07, 2018, 08:03:48 pm »

NO, I do think we are looking at it right, because the inductor is connected to the powerline for the microcontroller which is 3.3V, and we already know the other two lines are 4.9V and 25V - so if they landed on the inductor, it would fry the MCU.

The fact that you can't find power on the 6 pin IS a problem though, I would have expected to find the 25V there, if not the 4.9V. This could mean it is earlier in the circuit that the problem lies. So now we need to find which pin is the voltage in, and trace that circuit back. I'm guessing since it's at the bottom on the board, it's meant to come off the 25V line. Now if I remember right too, there were two caps with no power too, wasn't there? Can you probe out if either of them connect to any pin on the 6 pin converter? (I suspect pin 5.) In fact, can you find out where pin 5 connects to? I can see it on R42 (which I suspect is the enable resistor) and C28, but there is a Via, so it goes elsewhere. If we can find where, it'll help.

trez0r · « **Reply #22 on:** May 07, 2018, 08:35:29 pm »

You remember right. The closest I could find these 2 caps connect to is the driver chip of the motor (diagonal pin to the orientation dot).

I updated the pic with your request on pin5. It in fact is leading to one of those 2 pairs of "lone" vias I mentioned in earlier post.

LateLesley · « **Reply #23 on:** May 07, 2018, 11:19:45 pm »

where does the top left of Q10 go? And are you sure about that purple line to the big cap? cos I think it's top edge is ground, and may be something to do with the fault. I'm wondering of Q10 is shorted. That motor driver chips coming on the radar too, I'm wondering if it has something to do with generating the rail. It might be worth poking about its two top left pins, see where those Via's lead it to.

trez0r · « **Reply #24 on:** May 08, 2018, 07:09:21 am »

Q10 checks out and does not short. Top of Q10 goes back to 5V actually (it's more like 4.3V near Q10 though). You are right about ground being at top of that cap, I updated the pic (does it still seem fishy?). Those resistors on purple line are ~80/100kOhm. Driver's chip's two top left pins go to the diode cluster on the backside (which I think play role in operating the motor).

trez0r · « **Reply #25 on:** May 08, 2018, 08:47:50 am »

Here's clipped front and back side (flipped horizontally). Opening these in two separate tabs and then switching between them should help with identifying the vias between both sides.

LateLesley · « **Reply #26 on:** May 08, 2018, 03:29:04 pm »

OK, that was helpful. Q10 seems to track back to one leg of XFO1 via R40 and R41, its all taking us back up to that area with XFO1 and VR1, which all seem hooked up to that connector J_USB. It's starting to look like it looks for the USB connection before it pulls up that rail. so the next step is going to be hooking up the J_USB connector, and whatever is associated with it, and see if we get the rail powered. I suspect it looks for 5V from a USB connector, and if it doesn't see it, shuts off the rail. Whats the markings on that XFO1 component? I think we need to identify it too.

trez0r · « **Reply #27 on:** May 08, 2018, 06:07:27 pm »

XFO1 is unmarked but it is some kind of mini-coil - both sides show windings.

I always connect both USB and 220v-to-PSU when powering up the board. Unless usb is connected the psu won't even switch on - it has some triac/opto switchboard inside. I probably should've told this earlier.

LateLesley · « **Reply #28 on:** May 08, 2018, 11:08:45 pm »

It may have been helpful! still, I think we may have circled around the same components in the end.

I think I now need you to do some voltage readings with the power on - All round XFO1, R40 and R41, Q10 near the R100 resistor. And Q9 for the hell of it. I'm thinking one of them might be faulty. (Q9, Q10)

trez0r · « **Reply #29 on:** May 09, 2018, 06:48:37 am »

I updated the pic. All colors having voltage have it specified. Rest are dark.

LateLesley · « **Reply #30 on:** May 10, 2018, 12:03:54 am »

Is there anything around the motor driver IC at all? It's starting to look like it provides the powerline for the 3.3V rail converter, so now we need to figure out where it gets it's power from. Time to check that, and the 3 big components at the bottom of the board, TVS1 diode, the big resistor 472 printed on it (guessing it would be 4.7K), and the other component (Diode??) I think these power the motor chip.

Rasz · « **Reply #31 on:** May 10, 2018, 11:08:58 am »

Quote from: LateLesley on May 05, 2018, 05:42:59 pm

Well, I'm afraid I'm kinda at the end of the road. I can't find the datasheet for VR31, I suspect it is an LDO,

>VR31 is "WR1 63".

PRTR5V0U2X USB protection and clamping, nothing to do with power

U4 (6 pin bno?U) pin 5 - where is it connected? does it have 26V?
and check c28 for short

at this point I would just remove U4 and inject 3.3v from external supply

trez0r · « **Reply #32 on:** May 10, 2018, 08:25:53 pm »

Hi!

U4 pin5 goes to purple line.
No pin of U4 has meaningful voltage.
C28 is not shorted.

Resistor 472 is indeed 4.6K

I've updated the pic with voltages at several more locations.

LateLesley · « **Reply #33 on:** May 10, 2018, 08:44:06 pm »

take that big component with 0.14V on one end off the board, then try and power up again. Make sure not to connect the motor. I'm thinking it might be a protection diode gone short.

Maybe even see if it gets hot with the power on, before you pull it.

trez0r · « **Reply #34 on:** May 10, 2018, 08:51:36 pm »

It doesn't get hot. Btw I had it previously unsoldered on one end and it checked out ok - it was the X diode on that paper slip few posts back.

EDIT: just desoldered it and powered up the board - no noticeable change in behaviour.

I traced down one of the quiet electrolytic caps, see updated pic. ~~Can't find the other one though.~~ Found the other too - it is connected to the R100 (0ohm) resistor on the backside.

LateLesley · « **Reply #35 on:** May 10, 2018, 10:34:18 pm »

ummm, next thing i'd try then is taking out the TVS1 diode. Something around there has got to be holding the power down. After that, I'd be thinking U3, the motor driver is faulty

Rasz · « **Reply #36 on:** May 10, 2018, 11:14:55 pm »

purple line means R45, "104" being 100k resistors, is shorted
measure U4 pin 5 resistance to ground
and find everything connected to this pin (purple) on the pcb, this is supply for 3.3 switching regulator, and its missing
I would even start with removing C28 (has a dark line across one joint, either dirt or crack)

trez0r · « **Reply #37 on:** May 11, 2018, 01:20:34 pm »

Here's a new pic.

I've marked the pins on both U3 and U4. For U3 I used it's data sheet and for U4 I used this http://www.ti.com/lit/ds/symlink/lmr12010.pdf which seems close enough by the looks of it.
Dashed lines are connected through vias.

U4 pin5 to the ground is 200K - it can be observed that it has those 2x R104 on the way to gnd.
And those R104 are not shorted, the old pic was a bit misleading.
I repeat that C28 is not shorted as well.

TVS1 diode is fine - albeit the new pic shows 8v on both sides, in reality there was a voltage drop which I omitted.
I don't think driver chip U3 is a suspect yet. It gets supply of 25V and produces two nearly identical outputs of ~8v. It's vref is empty (this is the 3.3V lane) and it's inputs are also empty (because the main cpu is not powered, because the 3.3V lane is dead).

To be fair I fail to understand the logic of the board - how would U4 ever get any Vin unless there are some other connected points which I don't seem to be finding anywhere else on the board.
Q10 as soon as the USB 5V is connected, slashes the U3pin5 resistance-to-gnd in half - see green doted arrow in pic. Q9 - I fail to see the point of it at all - ground on the gate (when Q10 switches on), ground on the source too. ~~What's the use? What do I not know?~~ Oh, ok, it seems to disable U4 (by supplying gnd to EN), when the USB is not connected, so the 3.3V line stays dark.
Anyway it seems there really has to be another place from where the purple/U4 supply line gets voltage. Will try to find it again.
Any hints where to look at?

LateLesley · « **Reply #38 on:** May 11, 2018, 01:59:48 pm »

Can you do me a favour, and meter out C27, and C44. Is C44 cracked right across?? I cant tell if it's cracked or got debris on it. Same with C27, looks like it might be cracked top left corner, but it could just be a fibre laying on top.

trez0r · « **Reply #39 on:** May 11, 2018, 02:15:17 pm »

I can only assure that they do not short. Visually under magnifying glass they look fine too. Sometimes I wish I had some better camera than just phone..
Also I went over both sides of the board and can't find anything else that connects to pin5 of U4..

Rasz · « **Reply #40 on:** May 11, 2018, 04:50:05 pm »

inject few volts into purple (with low current limit) and start going over everything with multimeter until you get a reading
or remove 4 elements connected to it and measure resistance to ground again, if zero then might be as weird as internal track broken

either way I wouldnt have this much patience, would just solder $1 ebay 3.3V DC-DC converter in place of U4 long time ago

LateLesley · « **Reply #41 on:** May 11, 2018, 05:13:19 pm »

Right, I think I've almost got to grips with it. There's something wrong with the diode cluster underneath the board, and the resistor underneath should have 8V on either side of it. I think... Can you measure all the connection points of the diodes and resistor underneath? and also feel for heat? (might need to leave the power on a few mins). I suspect that resistor underneath will warm up, and possibly one of the diodes. D11??

If they do, I'd like to see the board with all the diodes underneath off. Do you have a pic of them from before you soldered them?

trez0r · « **Reply #42 on:** May 11, 2018, 07:21:00 pm »

Something strange is going on.
What previously on U3 was 24V and 8V now is 18V and 6V. And these keep oscillating slowly +/- few volts in a matter of several minutes. What is going on?
Anyways the resistor and diodes do not get warm even a bit.

Here's a pic of V measurements.

trez0r · « **Reply #43 on:** May 11, 2018, 07:30:33 pm »

Rasz: How would you go about installing a separate dc-dc converter? Would cutting off trace from pin6 of U4 to inductor be enough to disregard it? Would you install the module before or after inductor? Would you take VIN from USB?
Would this one work?

Rasz · « **Reply #44 on:** May 11, 2018, 08:13:01 pm »

remove U4 with hotair, bodge regulator with flying wires, hotglue somewhere
one you linked is linear, but should be ok

but first check with lab supply, either remove U4 (or coil) and inject 3.3V, or if you are super double sure there is no voltage on purple and no short to ground then inject 5V there, either way set current limit to 100-200mA and go from there

trez0r · « **Reply #45 on:** May 11, 2018, 08:34:20 pm »

If only I had lab supply.
The best I can do is 3.3V or 5V from a PC psu and use a 6-12ohm or 10-20ohm resistor to limit the current to 500-250mA.

LateLesley · « **Reply #46 on:** May 12, 2018, 12:10:10 am »

ok, so that slow oscillating voltage goes somewhere we can't trace. it'll come out the Via's on the top side, middle top of U3, next to C27, and to an Isolated Via right under the R of R42, I can't see if it hops off anywhere. Anyway, it sent me wandering around the board for other clues - What's going on with R13, over near the crystal? Looks like the side nearest the Via isn't soldered too well. And the diagonally opposite side of the board near the power connector, see that Diode D5 I questioned way earlier? next to the 4 pin connector? Look close with a magnifier, one side isn't soldered, and it looks toasted. You need to replace that. And R5 needs the side nearest the 4 pin resoldered. I'd remove the plastic of the 4 pin again to do it, to save toasting it any more. Then we may move on to the diodes underneath. I need to figure out their job first, but they may be doing some voltage clamping of some sort.

Rasz · « **Reply #47 on:** May 12, 2018, 04:13:29 pm »

Quote from: trez0r on May 11, 2018, 08:34:20 pm

If only I had lab supply.
The best I can do is 3.3V or 5V from a PC psu and use a 6-12ohm or 10-20ohm resistor to limit the current to 500-250mA.

injecting 3.3V should be mostly safe after lifting U4, injecting 5V might go bad without precautions, for example U4 can be bad, pass 5V to microcontroller and fry it

kabacis · « **Reply #48 on:** May 16, 2018, 08:24:12 pm »

Decided to draw on whiteboard the part around DC-DC converter. It more or less matches what everyone wrote before.

trez0r · « **Reply #49 on:** May 28, 2018, 11:52:22 am »

Sorry for getting back so late, had some priorities to attend to!

Kabacis is actually my work colleague and he conveniently had recently acquired a lab power supply..

We had some fun with it today and I think we done goofed now.

We injected voltage on "purple line" which is the U4 input and observed the voltage on the output after the coil which is the 3.3v line.
Around 10V the output came alive with around 1V on it.
Around 20V the output had ~1.5V which is close to the lower bound of lpc11u35f cpu operating voltage (1.8v-3.6v)
We decided to slowly approach the 25V which was probably the intended input voltage for the U4 to produce a working 3.3V line (1.8-3.6V really).
At around 24V the output was just under 2V and then something happened and the voltage was brought down while the amperes spiked to around 0.5A.
And that was it. Now the 3.3V line is short to the ground.
As a last resort we cut off the output trace of U4-to-coil and tried injecting some amperes to 3.3V line directly. All we got in return was CPU getting hotter the higher we went. Not sure why but it appears to be fried now.

Anyway thanks all for your time, It was fun (at least for me).

LateLesley · « **Reply #50 on:** May 28, 2018, 04:03:54 pm »

Sorry it turned out bad. To be fair, there was always the chance the micro-controller was already toasted, and that was why it wouldn't lift the powerline. And I hate the modern way that it's now technically junk, all because you can't get a new chip with the programming in it.

Anyway, it was a fun investigation at least.

trez0r · « **Reply #51 on:** May 28, 2018, 07:19:09 pm »

I wholeheartedly agree.

Before I started hacking this piece I wrote several emails to TM support. The best they could do is paid RMA. That means sending the whole wheel set to France, paying for shipping there and back, paying for the work hours and parts. All in all an unattractive package which certainly does not make sense financially for a used wheel of this caliber.
They absolutely declined selling/sending only the motherboard, which is a shame because there are precedents in other forums where they have agreed on such move after a bugging them enough. And I really tried within staying polite..

Anyway, thank you again Les!

Rasz · « **Reply #52 on:** May 29, 2018, 12:17:43 am »

Quote from: trez0r on May 28, 2018, 11:52:22 am

We injected voltage on "purple line" which is the U4 input and observed the voltage on the output after the coil which is the 3.3v line.
Around 10V the output came alive with around 1V on it.

as I said earlier: "if you are super double sure there is no voltage on purple and no short to ground then inject 5V there, either way set current limit to 100-200mA and go from there"
what made you inject >5 V??? "go from there" doesnt mean keep injecting more voltage

, it means continue diagnosing. There is no situation possible where 3.3 DC-DC converter doesnt work at 5V, but starts working at 10V (other than it being faulty)

Quote from: trez0r on May 28, 2018, 11:52:22 am

Around 20V the output

and you kept on going

Quote from: trez0r on May 28, 2018, 11:52:22 am

At around 24V the output was just under 2V and then something happened and the voltage was brought down while the amperes spiked to around 0.5A.
And that was it. Now the 3.3V line is short to the ground.

just as I wrote: "injecting 5V might go bad without precautions, for example U4 can be bad, pass 5V to microcontroller and fry it"

tldr either:
-you blew U4 and it passed 24V to the MCU
-or there was always short to ground on 3.3V rail

Quote from: trez0r on May 28, 2018, 11:52:22 am

As a last resort we cut off the output trace of U4-to-coil and tried injecting some amperes to 3.3V line directly. All we got in return was CPU getting hotter the higher we went. Not sure why but it appears to be fried now.

That was supposed to be the _first_ thing to try

At least it was a learning experience for you.

trez0r · « **Reply #53 on:** May 29, 2018, 08:11:39 am »

Ok, yesterdays post may have been worded too lazy, sorry for that.
We weren't as reckless as it may have sounded from it.

There surely was not a short circuit on purple or 3V lines as it have been measured over and over again.
We started at 0V and near 0A. We incremented volts in small steps, and were especially careful around 3/5/25V bounds. Multimeter was attached to 3V line and constantly measuring the voltage as we were injecting on purple. 3V or 5V on purple didn't yield much on the output; We figured the U4 is most likely is a ~25/3V converter of sorts.
10V was were it crossed the 1V on output. At no point did the voltage on 3V line exceed 2V. We didn't touch amps at all, they were extremely low / not much more than maybe 10mA. At around 24V the voltage suddenly dropped to 0.5V and the amps went to 500mA.

I guess if we really did blow the U4 we wouldn't have noticed the voltage spike on output anyway. I will measure if input of U4 is shorted to output now.

Rasz · « **Reply #54 on:** May 30, 2018, 01:19:29 pm »

you wouldnt see the spike on multimeter due to slow update rate

trez0r · « **Reply #55 on:** May 31, 2018, 08:11:32 am »

I measured U4 and in/out on it is indeed now shorted. Rasz's guess of what went down was spot on it seems.

Rasz · « **Reply #56 on:** June 01, 2018, 06:22:37 am »

well shit, now I feel its my fault for even mentioning to you possibility of injecting anything before U4

LateLesley · « **Reply #57 on:** June 01, 2018, 09:24:17 am »

I wouldn't feel bad, It was a valid diagnostic move, there should have been voltage on that purple line.

1) It was faulty to begin with.
2) There's a good chance the Microcontroller was already toasted, and thats why it wouldn't lift the line.
3) It was at a point where it needed voltage injected to see what was going on.

I'd be interested to know the status of the diodes underneath the board. I wanted them removed, cos I had a suspicion they were pulling things down. My last suggestion was to remove them to see how the circuit was configured.

But what is done is done. We can't win them all, but we can learn things from it. The board might be toast, but we have gained knowledge. It's taken me a long time to get used to the idea that you can't fix everything. I think its in our DNA as repairers to try and rescue everything, but it just can't be done. So don't feel bad. Go rescue the next challenge.

james_s · « **Reply #58 on:** June 02, 2018, 07:19:11 pm »

Keep an eye on ebay, maybe you can find a working unit in poor cosmetic condition from which you can salvage the PCB.

trez0r · « **Reply #59 on:** June 05, 2018, 06:32:03 pm »

Rasz, don't be silly, you were 100% spot on with your advice many posts earlier.

I can only blame myself (not that I do so much, as Les well put) for assuming that the u4 was a 10x step-down converter. To be fair it looked/measured much like it until it let out the magic smoke.

Some background. Turns out I'm a bit of a car-sim fan and wanted to upgrade from my old logitech driving force gt I've owned for some time. Meanwhile I thought I could save some money and also satisfy my cravings to tinker which I get from time to time. So I did some thrift shopping few months ago on ebay and in fact this wheel is just one out of three which I didn't manage to salvage. The other two were a step higher models - TM T300 with faulty pedals (a dirty pot which was just fine after cleaning) and TM TX with a much more sophisticated problem - the electric brushless motor magnetic core had come loose from the shaft and was messing up the calibration after a short while of usage. This sure was a bit harder to crack.

Anyway thanks everyone for input and opinion. I'll be back soon enough with the next problem to troubleshoot.

Czary · « **Reply #60 on:** February 16, 2019, 06:12:07 am »

Hi, greetings from Poland.
Couple of days ago I've bought TM T150 wheel. Problem with mine is that it doesn't see any input (no wheel turn, no pedal input). The good point is, that it reports itself in Windows and I was able to do a succesful firmware upgrade.

I suspect faulty PSU- there is 4 pin DC output connector but I only got 25VDC on 2 pins while other 2 doesn't report any voltage. Is that correct ?

Best regards, Czarek.

LateLesley · « **Reply #61 on:** February 17, 2019, 02:57:38 am »

Hi Czarek,

This was from a while ago, but if you look at the picture in the first post in this thread, you will see that trez0r helpfully marked voltages on the 4 pin connector. I believe you should have 25V and also 5V, the 5V possibly being derived from the USB connection, so you may have to test with it plugged into USB. I'd also check for a short on that 5V line, obviously with it unplugged and powered off.

Best of luck

Kind regards, Lesley

Czary · « **Reply #62 on:** February 17, 2019, 11:34:18 am »

Hi Lesley,
thank you very much for quick reply.
As trez0r posted earlier there is triac/opto switchboard inside the PSU. To switch the triac on we need 5VDC, but my PCB only generates 2.7-1.73VDC

So far I've checked all FB's, voltages and ohms mentioned in the previous posts :

FB26: 2.7V
Q13: 2.7V(M), 5.05V(L), 2.7V(R)
D4: 5.05V(R)

C13,16,17,18: 24.6k
C21,22: 1k

Is that correct for Q13 ?
trez0r measurements was different for this component.

Also I've noticed that USB voltage always shows 5.05VDC while other points (FB26 or Q13(M)) showed 2.7VDC just to drop to 1.73VDC, after about one hour of PCB being powered by USB

LateLesley · « **Reply #63 on:** February 18, 2019, 03:15:29 am »

Hi Czarek,

To be honest I'm not sure, I dunno if it's maybe an in-line FET for a power rail, or something different. One check might be, does it get hot? If it does, it''s not necessarily dead, but something could be pulling the line down. I'd just try checking all the caps for shorts to start with, it is surprisingly common for them to fail. It took me ages to realise caps fail quite often.

infortec · « **Reply #64 on:** August 23, 2020, 09:48:22 am »

HELLO.
SOMEONE CAN GIVE ME THE REFERENCE OR THE VALUES OF THE XFO1 COMPONENT

LateLesley · « **Reply #65 on:** August 23, 2020, 03:16:41 pm »

I can't, as I've never had a hands on look at this board, but considering it looks like it connects to the USB connector, I'm gonna guess it's possibly a common mode choke for the USB, part of the ESD protection. So I don't know if i'm sending you down a wrong path, but it might be worth having a look at components in this search -

https://www.google.com/search?q=smd+inductor+common+mode+usb+0805&hl=en&source=lnms&tbm=isch&sa=X&ved=2ahUKEwjEz5PyzbHrAhXKesAKHXg_Bx8Q_AUoAXoECDUQAw&biw=1345&bih=612

Good luck in identifying the part.

nonkapo · « **Reply #66 on:** January 30, 2021, 04:12:14 pm »

Good evening all,
Another dead Thrustmaster T150, however with slightly different symptoms. Let me share my experience so far. Tinkering with the device i try a lot of advises with no luck. The wheel gets recognized and i was even able to do firmware update. However, the self centering and buttons do not work at all. Only the led make you think there is some life in it:-). So i was relaxed at least the power supply, processor and interface are ok and finally decide to open it. Thanks to the nice high resolution photo of the pcb on this site i was able to identify one fried chip.Expecting it to be sort of pwm controller/driver but it was so badly burnt you can not get much from the letters identification. So thanks to your photo i get its A4950 pwm driver. However i'm still curious what might be the reason for the chip to fail as fixing it is easy job but burning it again is not my cup of tea;-). Note it has bunch of protections inbuilt so should be not easy victim. Suspecting issue with the motor off cause. Some other ideas will be really appreciated.
Thanks in advance.

Rasz · « **Reply #67 on:** January 31, 2021, 01:12:01 am »

Isnt that the chip driving the motor? Your ff motor might be fried, shorted, tired, etc, or maybe back emf suppression diode died? hmmm looking at the datasheet none of that should be possible, it has
▫   Motor short protection
▫   Motor lead short to ground protection
▫   Motor lead short to battery protection

and build in back emf diodes. Since its whole 60 cents I would just check supply voltage, replace it and see what happens.

nonkapo · « **Reply #68 on:** January 31, 2021, 10:52:23 am »

Thanks,yeah that's the plan. However until the part arrive from Farnell i must wait a bit. Cant get it on the local market.
The chip is visually fried kind of blown. I remove it yesterday. Thinking to try starting the motor with 20A ESC from a RC model and see how it behave. Also curious which way it might affect the buttons so they wont work. Some kind of power supply issue? Any schematics available or only back tracing can be done?

nonkapo · « **Reply #69 on:** February 06, 2021, 11:41:15 am »

Just curious if the motor brush less or not? There are three cables coming out but I'm still not confident.

Edit: Without chip installed there are 25 volts on the VBB pins. It should be able to withstand 40V so considering it normal. Should i solder the chip what do you think?
Test points around the processor are measuring 3.3volts.

Rasz · « **Reply #70 on:** February 06, 2021, 04:04:37 pm »

A4950 is meant to control brushed motors, so 2 poles and shield
is the big low value resistor ok? is motor ok? i would just go for it. I assume you bought 2 chips just in case

nonkapo · « **Reply #71 on:** February 06, 2021, 07:23:33 pm »

Thanks for the answer. Yes i order two but only one arrive;-). Meantime before i read your answer i look at the motor and the brushes are visible. I did not have suitable power supply to check the motor but the low value power resistors have some kind of bulge on them. Both measure 7 Ohm. Its late now so i will look for psu in my shack tomorrow and update the results here.
Edit: Motor is tested on 12 volts. Spins both directions and no obvious damage. Soldering the chip and making cross sign:-)))

nonkapo · « **Reply #72 on:** February 07, 2021, 01:51:17 pm »

After the hard but successful re-soldering the wheel is back in working order. Will post the photo with voltages on all testing points for future reference. Thanks to all involved again.

nonkapo · « **Reply #73 on:** February 13, 2021, 07:28:36 pm »

Before i manage with the photos, another question comes. I get a bit disappointed to figure out Thrustmaster T80 pedals do not work with T150. Trying to figure the pin out so far failed. I take the t80 pedals apart and noticed the cable have only 4 wires connected, wherever t150 as per what i see in you tube they are 6. So despite both types use rj12 offset connector and are plug compatible, the wiring is different and there might be even potentiometers value difference. Lets see what my future research will show.

nonkapo · « **Reply #74 on:** March 06, 2021, 11:35:58 am »

Here the overview of the problem with photos.
After connecting, the unit is recognized by the computer, but no buttons are working neither self calibration is happening. The power light is on after USB connection as it should be. Firmware update is attempted and completed successfully but no change in the status.
Follow up disassembling and visual inspection of the PCB which reveal cracked/melted PWM IC A4950. De soldering and soldering of the replacement completed and the wheel is back to life. Some photos of the PCB showing the multi meter measured voltages.
Note at those points(accept power connector) measurement must be done with oscilloscope to make sense, so I've done multi meter one just for reference.
Note also the power connector voltages are available only after USB is connected to the PC(both 5V and 25V!!!).
Measurements are done at TP2, TP1, SCLOCK(clock signal?!?), SDATA(data signal?!?!), power connector.

nonkapo · « **Reply #75 on:** March 06, 2021, 11:39:02 am »

Here the rest. At all Sclk, Sdata, TP1 and TP2 3.3volts approximately are present.

culebron · « **Reply #76 on:** May 18, 2021, 01:48:49 pm »

Hi,
seeing other people finding this thread and replying about their issues, I'll post my one too.

Thrustmaster T150RS. One day it stayed turned on (but idle) for 6 hours, then I started driving with it, and force feedback turned off. (I suppose, the motor was heated but still within limits, but after it started working again, it heated above a threshold and something failed.) No driver reinstalls or reboots helped.

Currently, the device turns on and rotates the steering wheel from lock to lock both ways, absolutely normally. It gets detected and work fine in games, except for force feedback.

I opened the lid and checked if there were any noteable defects, but don't see any. I suspect the thermal resistor to be the reason. Currently, it shows 43-45 kOhm, and when heated it drops to 33 kOhm. Comparing to datasheets of other thermal resistors, it seems way off the scale.

I tried a fix, but it did not help: I connected a usual resistor instead: 43K, then 20K, 11K, 5K, 3K, 1K, 510, 325, 100, 30 Ohm, and even shorted the 2 pins. (Every time I would disconnect the wheel and power it down.) None of that seemed to work. Only once with 325 Ohm the FFB would show signs of life, but I could not reproduce that.

Question: what resistance does this thermal resistor have on a working model?

culebron · « **Reply #77 on:** May 18, 2021, 01:55:55 pm »

As an afterthought: I measured the voltage between VCC and GND pins connecting to the resistor, and it was 0.16 V. Could it be that the thermistor lowered resistance so much that something in the net that's before it failed?

MC87 · « **Reply #78 on:** May 21, 2021, 08:27:00 am »

Hi Culebron,

I have the same issue although with my thrustmaster tmx slightly different board.
I measured everything could not find anything until i saw your post about the thermal resistor which would be obvious

.

When i plug off the thermal resistor connector from the motor to the board the force feedback works again in the test tab on the thrustmaster software.
I guess i need a new resistor then, i am tempted to just play with it plugged off though since the motor is fine and cool.
Does anybody know which resistor is used here?

FYI the measured voltage on my pins to the resistor is 3,3V so you definitely have something wrong in the supply and nothing in the resistor side will make it work for you.

Anyway thanks for pointing out the obvious:D

culebron · « **Reply #79 on:** May 22, 2021, 01:32:58 pm »

Hi, MC87,
that's curious information. If no resistor (= infinite resistance) makes it work, then it makes sense why my one is not working.

If my resistor is ok, then 43kOhm @ 25° should work. These seem to have the required property:
https://www.aliexpress.com/item/1005001495663005.html

this one has 33kOhm @ 25, but I'm not sure this is high enough.
https://www.newark.com/epcos/b57891m0333k000/thermistor-ntc-radial-leaded/dp/63W2797

In theory, you could test it with usual resistors and find the threshold at which it turns on.

Can I ask you to help with cross-checking the part on the board? If you have a working one, finding a failed part would be easy, if we chat over Zoom, for instance.

MC87 · « **Reply #80 on:** May 25, 2021, 07:15:49 am »

Quote from: culebron on May 22, 2021, 01:32:58 pm

Hi, MC87,
that's curious information. If no resistor (= infinite resistance) makes it work, then it makes sense why my one is not working.

If my resistor is ok, then 43kOhm @ 25° should work. These seem to have the required property:
https://www.aliexpress.com/item/1005001495663005.html

this one has 33kOhm @ 25, but I'm not sure this is high enough.
https://www.newark.com/epcos/b57891m0333k000/thermistor-ntc-radial-leaded/dp/63W2797

In theory, you could test it with usual resistors and find the threshold at which it turns on.

Can I ask you to help with cross-checking the part on the board? If you have a working one, finding a failed part would be easy, if we chat over Zoom, for instance.

Maybe i Can help you yes.
But as i mentioned i have a thrustmaster TMX and i saw from the pictures on this topic that i have a slightly different board and i dont even understand my board fully.
Can you take a picture of where you think the 0.16v starts from so i can measure or see if i even have that component on my board, you could use the pictures in this topic as a reference for the 3.3v needed? you must have some sort of short somewhere.
The 0.16V is definitely a problem for you.
It would be nice if someone has a working resistor so we have the specs to order a new proper one. My resistor was shorted having 0.3ohm.

AndreiLC · « **Reply #81 on:** March 02, 2023, 07:59:57 am »

Resurrecting this old topic, maybe someone can help me

Recently bought an Thrustmaster T150 with faulty 2 pedal set.

The problem can be seen here: https://youtu.be/xLhe_Hb6ZRs . It sometimes occurs when the brake pedal is depressed, and applying gas. Or sometimes it happens on gas only.

What i tried:

The the cable was kincked and some whires where unportected, so i shortend the whole cable past the damaged section.
Disassambeld and cleaned out the two pots, the problem still persisted, so bought new pots and assembeld them with new wires directly to the board, bypassing the old pedals cable and the internal cable from the motherboard to the pedal set connector. But the problem still presists.

Any suggestions ?

Right now i am considering modding the pedals with arduino and plug them dirrectly into the pc.

Many thanks.

james_s · « **Reply #82 on:** March 03, 2023, 11:28:29 pm »

Have you measured the voltage on the pots in operation? Best place to do that is at the other end of the cable where they plug in. Typically they will have one end connected to a voltage source and the other to ground then the signal comes off the wiper. It's possible that the voltage source is noisy or sagging. It could also be driver issue, this is a quite old peripheral isn't it?

nonkapo · « **Reply #83 on:** April 04, 2023, 06:43:28 pm »

Dear, do not consider the pedals are faulty. I've recently seen for sale T150 wheel with T80 pedals. They are not compatible. So check first what pedals you have. You might be lucky if i'm right. The other thing is there is update from 2019 or 2020 which changes the way pedals are recognized. Check if you have that update and see if it helps. Let us know how far you went.

nonkapo · « **Reply #84 on:** April 06, 2023, 02:33:10 pm »

Another faulty T150 here. Now the patient have different symptoms. Initially it was not firing up. After disassembly and check i figure out solder shorted pins. After removing the drop of solder the wheel start the initialization turns but after that it gets disconnected. Feels like some sort of protection circuit switch it off. The fan keep trying to turn though. Any ideas?? I'm thinking about the power supply fault but it may be anything else connected to power regulation. Any ideas? Thanks.


EEVblog Main Site	EEVblog on Youtube	EEVblog on Twitter	EEVblog on Facebook	EEVblog on Odysee