[RCA] MSI B450 Pro VDH Max IC shorted itself

[dvader]

Hi Everyone,

I just wanted to quench my curiosity as to why a small IC in my MSI mobo burst out of nowhere. Long story short, I have a rendering PC (for my wife) and it works fine. One fine evening I turn it on and boom. My first thought was PSU (it's Corsair RM650x) but Corsair PSUs don't do that (at least not the ones just 3-4 years old). On close inspection, I found a small IC ( TPS25944L per boardview). Now this IC powers the ARGB header and I had nothing plugged into it. The motherboard should work without the IC though (will desolder it) and I will thoroughly test it after I install the new mobo. Any idea why would that happen?

Note: I checked a small SMD cap beside the small IC and it was shorted (don't know if that was before or after the fact).

For ref:

Thanks,
dvader

 

[nRiTeCh]

Usually argb wont affect unless you plugin some argb peripheral which might mess the voltage and cause further damage. As for the capacitor it is used to regulate voltage to the argb controller. In under warranty you shall get the mobo replaced and if oow then rendering already uses the system to its max so its best to get a new one.
You may try a few trusted pro pc component repair shop and they might do the job or msi sc.

 

[mokendkomer]

It would be great if it's under warranty, but I think op is taking more of an academic interest in the problem than looking for a solution as a consumer.

It really does seem odd that it would break when there's nothing plugged into the RGB header. My only thought is to check the bottom of the case to see if there's anything at all that may have come into contact with it somehow?
This is an extreme long shot but imagine a bug finding its way onto the board and bridging two points.

Please do share your findings once you desolder the chip and cap. I see the chip on sale for a few hundred in case the board is designed poorly enough to not work without it.

 

[dvader]

Usually argb wont affect unless you plugin some argb peripheral which might mess the voltage and cause further damage. As for the capacitor it is used to regulate voltage to the argb controller. In under warranty you shall get the mobo replaced and if oow then rendering already uses the system to its max so its best to get a new one.
You may try a few trusted pro pc component repair shop and they might do the job or msi sc.

I already ordered a new Mobo. No point in risking CPU and GPU. I am just baffled as to why it happened.

It would be great if it's under warranty, but I think op is taking more of an academic interest in the problem than looking for a solution as a consumer.

It really does seem odd that it would break when there's nothing plugged into the RGB header. My only thought is to check the bottom of the case to see if there's anything at all that may have come into contact with it somehow?
This is an extreme long shot but imagine a bug finding its way onto the board and bridging two points.

Please do share your findings once you desolder the chip and cap. I see the chip on sale for a few hundred in case the board is designed poorly enough to not work without it.

I thought about the bug and though it's possible, the bug would have blown up before the IC given the loud bang and big spark. The back of the mobo at this IC is just blank (in case it touched the back of the bcabinet somehow). Anyway if it works without the IC (most prob should) then I will use it as a server/backup board, one less IC to go wrong.

Also an SMD resistor (too small to even zoom) just between the IC and the shorted CAP is shorted too. again could be after the fact. I will just do an amputation surgery and remove shorted stuff (related to ARGB) and see if board boots. I have the board view file for the mobo so surgery should be easy. It's driving me a bit nuts at this point though.....

Please do share your findings once you desolder the chip and cap. I see the chip on sale for a few hundred in case the board is designed poorly enough to not work without it.

I just cleaned the burnt area with IPA and the blast took some of the solder mask around the chip away. So resoldering a new chip is now out of the question since the pads are gone. I could attempt adding some UV curing solder mask (don't have it on my bench though) but with the cost of extra equipment, I can get two mobos . or maybe this could be an excuse to buy some equipment.

 

[Emperor]

another story of MSI h/w issue, I think 3rd in last 2 months.

What brand board now you opted for?

 

[dvader]

another story of MSI h/w issue, I think 3rd in last 2 months.

What brand board now you opted for?

I have had a good experience with Asrock (I can call myself a fan at this point, have an RMA post here on TE too). Even bought their top-of-the-line (sort of) Taichi board. However, I think I should have a backup brand. Went with Gigabyte DS3H. Let's see how it goes.

 

[dvader]

Update:

I performed the surgery. I got the burnt chip/IC, one SMD resistor, and the capacitor out (the capacitor was shorted and the resistor had to be removed). The chip [blast] damaged the solder mask and copper lines were torn out. I just cut all copper lines around the chip in case they fused, since it's a multilayer board, the surgery didn't look good until after the final cleanup. I put white tape over the wound, as shown in the image below. I had to remove one resister because of the copper traces [being removed], (I checked the boardview it was just for ARGB).

Powered on the board and it worked fine. CPU, RAM, GPU you name it all works fine. Now here comes the interesting part, I got this board here at TE and since the beginning, its USB ports couldn't handle anything that needed more than 2-3 Watts of power. I mean they all worked but if I had a USB HUB connected to one of the ports and added an external HDD, a wifi dongle, etc, the devices would go out intermittently. This a known issue with some Ryzen boards and I just ignored it as MSI being finicky. Guess what, all my ports now work fine.

I believe that the capacitor or the IC was leaking current from the beginning and made my USB ports weak (since the ARGB IC was triggering some current/power limiting IC). Now that the IC(and)/Cap are gone it's all good. Do let me know if someone has a better explanation.

So in the end. The short actually solved my problem. I would rather have my USB ports than a useless 12v ARGB header.

Note: I cut and pulled out all copper traces JUST around the IC and CAP. when I say all, I mean ALL OF THEM. If the traces didn't blew up I would have loved to replace the IC though, since it's really cheap in my local market (INR 20-40 for a bunch of them) (It's actually used as an eFuse in digital circuits)

I have an Intel 4th gen NAS, I think it's time to upgrade

 

[guest_999]

Just fyi, if mobo has any signs of physical damage like the case here then good chances that rma request will be rejected & going by MSI after sales support reputation in India I say it has the most chances of rejecting such case.
@nRiTeCh @mokendkomer @Emperor

since the beginning, its USB ports couldn't handle anything that needed more than 2-3 Watts of power. I mean they all worked but if I had a USB HUB connected to one of the ports and added an external HDD, a wifi dongle, etc, the devices would go out intermittently.

A usb 3 port can supply at max 900mA while a usb 2 port can supply at max 500mA. If using a usb hub with usb power only hdd & that hdd is more than 2TB in size then using a powered usb hub is the recommended way.

 

[dvader]

Just fyi, if mobo has any signs of physical damage like the case here then good chances that rma request will be rejected & going by MSI after sales support reputation in India I say it has the most chances of rejecting such case.
@nRiTeCh @mokendkomer @Emperor

A usb 3 port can supply at max 900mA while a usb 2 port can supply at max 500mA. If using a usb hub with usb power only hdd & that hdd is more than 2TB in size then using a powered usb hub is the recommended way.

Not most chances....it's a definite no. If there is a burn mark RMA is rejected by default. F*** these bastards. How is that legal is beyond me but "m to aloo gobhi khane wala admi hu"

 

[guest_999]

How is that legal is beyond me

Indian customer protection laws are very weak & no party takes them seriously either to improve them.

 

[Emperor]

in 2020 I have got Intel original mobo Capacitor replaced from local mobile/TV repair technician, he was sceptical to replace Motherboard as feared that he might blow 10-15k board, but I assured him not to worry as board was 8 yrs. old, no warranty and due to lockdown can't go anywhere to repaired it, so he took chance and did his best (but with warning that it might not work) but the board worked for 10-15 days and RIP after that for ever.

 

[dvader]

in 2020 I have got Intel original mobo Capacitor replaced from local mobile/TV repair technician, he was sceptical to replace Motherboard as feared that he might blow 10-15k board, but I assured him not to worry as board was 8 yrs. old, no warranty and due to lockdown can't go anywhere to repaired it, so he took chance and did his best (but with warning that it might not work) but the board worked for 10-15 days and RIP after that for ever.

Let's see for how long this works, I am gonna stress this board like hell now. Just to see how much can I push it.

 

[nRiTeCh]

Update:

I performed the surgery. I got the burnt chip/IC, one SMD resistor, and the capacitor out (the capacitor was shorted and the resistor had to be removed). The chip [blast] damaged the solder mask and copper lines were torn out. I just cut all copper lines around the chip in case they fused, since it's a multilayer board, the surgery didn't look good until after the final cleanup. I put white tape over the wound, as shown in the image below. I had to remove one resister because of the copper traces [being removed], (I checked the boardview it was just for ARGB).

Powered on the board and it worked fine. CPU, RAM, GPU you name it all works fine. Now here comes the interesting part, I got this board here at TE and since the beginning, its USB ports couldn't handle anything that needed more than 2-3 Watts of power. I mean they all worked but if I had a USB HUB connected to one of the ports and added an external HDD, a wifi dongle, etc, the devices would go out intermittently. This a known issue with some Ryzen boards and I just ignored it as MSI being finicky. Guess what, all my ports now work fine.

I believe that the capacitor or the IC was leaking current from the beginning and made my USB ports weak (since the ARGB IC was triggering some current/power limiting IC). Now that the IC(and)/Cap are gone it's all good. Do let me know if someone has a better explanation.

So in the end. The short actually solved my problem. I would rather have my USB ports than a useless 12v ARGB header.

Note: I cut and pulled out all copper traces JUST around the IC and CAP. when I say all, I mean ALL OF THEM. If the traces didn't blew up I would have loved to replace the IC though, since it's really cheap in my local market (INR 20-40 for a bunch of them) (It's actually used as an eFuse in digital circuits)

I have an Intel 4th gen NAS, I think it's time to upgrade

View attachment 206986

Good to know that this shorting/blasting actually sorted an old problem however, keep in mincd that the boards death is nearing as even a minor fault anywhere on the circuit/bus track is bound to keep reducing its life. I have gone through such pains and magics earlier so could say, rejoice as far as you can as one fine day the mobvo is gonna give up suddenly!
Unless you actually replace the surgically removed parts with the right ratings components it could give a new or an extending life to the mobo..

For now..rejoice and enjoy the surgery!

 

[Hachune]

The argb ic got power from 12v pin of the PSU while USB got power from 5V pin of PSU. Even if the 12v rail got shorted by the blown cap and leaking current, I wonder why it affects the 5v rail. Perhaps it has more to do with the PSU side rather than the motherboard. Out of curiosity, which resistor got shorted btw?

 

[variablevector]

Maybe a nitpick but ARGB works on 5V and uses 3 pins. The board in the photo has a 12V 4 pin RGB header. They're two different things and not interchangeable.

 

[dvader]

The argb ic got power from 12v pin of the PSU while USB got power from 5V pin of PSU. Even if the 12v rail got shorted by the blown cap and leaking current, I wonder why it affects the 5v rail. Perhaps it has more to do with the PSU side rather than the motherboard. Out of curiosity, which resistor got shorted btw?

yeah, most prob yes. My point was there would be a current control logic either on the mobo or on the PSU side and that is what was kicking in. The resitster was fine, it was just fused with the copper traces so had to take it out. I have attached a picture of the shorted cap if you are curious.

Maybe a nitpick but ARGB works on 5V and uses 3 pins. The board in the photo has a 12V 4 pin RGB header. They're two different things and not interchangeable.

Yup, two diff things.

Good to know that this shorting/blasting actually sorted an old problem however, keep in mincd that the boards death is nearing as even a minor fault anywhere on the circuit/bus track is bound to keep reducing its life. I have gone through such pains and magics earlier so could say, rejoice as far as you can as one fine day the mobvo is gonna give up suddenly!
Unless you actually replace the surgically removed parts with the right ratings components it could give a new or an extending life to the mobo..

For now..rejoice and enjoy the surgery!

I don't intend to use it for anything serious. I might use it in (one of my) secondary NAS if I feel like it but other than that it's just a bench motherboard for testing and troubleshooting. Replacing the parts is out of question (I would have loved to do that and even tried) cuz the copper traces were blown up and at least two layers fused together after the spark. I got my hands on the boardview files and realized the ARGB IC was not connected to anything else on the board, all it does is suck the power out of 12v Rail and power the 4 pins. Therefore the Appendix surgery.

Note:

You see the C170 capacitor that's what blew up