Need replacement for Seasonic TX-850 with better OCP for 3090

Fortron/Source. At one point, they were also known as Sparkle.

They were great back in the Thunderbird/socket 462 era...Then largely moved back to mediocrity (even if it's just lack of marketing) once everything started going dual rail 12v.

Doubt you could say that they're bad, necessarily. But I haven't gotten any of their units since P4 either.
FSP usually works as an OEM for other companies like CWT in that they make some very good units like the Aurum series, but they'll also put out priced down models. They made a ton of straight up bad power supplies for years around the late P4 era, but their reputation is a lot better nowadays.
 
Today I installed the SeaSonic Prime TX-1000 to replace the older SeaSonic Prime 850 Ultra Titanium. So far, no accidental OCP (over current protection) triggers with the 3080Ti FTW3 Ultra. Handy that the cables were the same, so I could just swap PSUs without re-threading the entire case with a dozen cables.
 
I returned my TX-850 to Amazom and ended up buying a TX-1000 from BestBuy of all places. It arrived a bit tattered due to zero packing material, so BestBuy refunded me 10% and I was able to use my 10% off birthday coupon too. $262 shipped. :cool:

I'm curious what the build date is on the unit you received, if you don't mind sharing. It is the first four numbers after the R in the serial number. Mine is 2105, or May 2021.
 
I need advice on which PSU to keep. I have similar issues as described in this thread, so I want to ask for your opinion.

I currently have a 5950x, an Asus Strix 3090 OC on a Crosshair VIII Dark Hero, and a Seasonic Prime TX-850. I use this PC for work purposes as well, so I need a 100% stable system. I ran into some instabilities with the TX-850 three weeks ago shortly after I upgraded to Win 11. I have no idea if that had anything to do with it though, it might just be a coincidence.

Since then, I have had some of my games crashing my PC hard. No BSD, no soft crash, my PC didn't even reboot. Always a hard shutdown. Because of that, I ordered a Bequiet Dark Power Pro 1200W to be on the safe side. Now I don't know which PSU to keep.

Looking back at my process to figure this out, I should have thought about OCP a lot earlier. But, in the beginning, I assumed it had to do with either my CPU or GPU undervolt.
So I disabled PBO2 first, but that still shut down the PC in games. Only after I disabled the variable boost clock in BIOS, my PC did not crash anymore. But it was also severely crippled for my content-creation tasks.

Then AMD and Microsoft introduced their AMD CPU fixes, so I enabled boost clocks again. That re-introduced the shutdowns. It got a little better; one of the games did not crash anymore, two others, not as often. But the last one did way worse with crashes right when starting the game or shortly after.

Eventually, through reading this forum and Reddit, I realised it could be the overvoltage protection in the PSU. That made me dial down the GPUs max draw to 80%, which made everything stable again.

I then decided to try to re-enable PBO on my CPU after that again. Unfortunately, I did not save my original PBO values, so I spent the whole weekend achieving optimal settings again. After half a day of CPU tests, I raised my GPU's max power draw to 100% and had no more issues.

Long story short, the Prime TX's OCP does not kick in anymore, and I don't know why it all of a sudden does not... I even put my GPU to 123% max draw, which is the highest I can go in Afterburner. Even then, those games seem to be stable now. I ran the most problematic game and a CB23 (with 12 cores) at the same time. Total peak power of nearly 700W!

The Dark Power Pro never showed any weird behaviour in the few test I did with it. It always trucked along with solid, steady metrics.

I've read about a significant number of people who had issues with a Seasonic Prime TX-850 with various high-end CPUs and 3090 cards.

So now I am conflicted about which PSU to send back or sell respectively. Given that my first TX-850 came DOA and I had to RMA it, I am not very confident in keeping it. Yet, I would rather have some more opinions than my own before going ahead with it.

Please help me make the right decision!
 
I had the hard shutdown problem on an older TX-750 with a 3080ti. Was convinced it was an OCP thing as well after reading this thread and others.

Fortunately, I stumbled across the thread where Johnny Guru explains that it has nothing to do with OCP on the Prime models and is really a problem with high frequency noise (caused by 30 series GPUs) on the 12V sense line bugging shit out.

So I tried temporarily disconnecting that sense line and voila no more shutdowns. With further suggestion and ideas from that same thread, I tried putting a random ferrite bead around the now re-attached 12v sense line and that has so far worked out, no shutdowns since then.

I salvaged the bead from a generic USB cable, sorry don't know any specifics about what composition or bandwidth it is. Still contemplating the RMA though, as this is supposedly fixed on all new models now.
 
I need advice on which PSU to keep. I have similar issues as described in this thread, so I want to ask for your opinion.

I currently have a 5950x, an Asus Strix 3090 OC on a Crosshair VIII Dark Hero, and a Seasonic Prime TX-850. I use this PC for work purposes as well, so I need a 100% stable system. I ran into some instabilities with the TX-850 three weeks ago shortly after I upgraded to Win 11. I have no idea if that had anything to do with it though, it might just be a coincidence.

Since then, I have had some of my games crashing my PC hard. No BSD, no soft crash, my PC didn't even reboot. Always a hard shutdown. Because of that, I ordered a Bequiet Dark Power Pro 1200W to be on the safe side. Now I don't know which PSU to keep.

Looking back at my process to figure this out, I should have thought about OCP a lot earlier. But, in the beginning, I assumed it had to do with either my CPU or GPU undervolt.
So I disabled PBO2 first, but that still shut down the PC in games. Only after I disabled the variable boost clock in BIOS, my PC did not crash anymore. But it was also severely crippled for my content-creation tasks.

Then AMD and Microsoft introduced their AMD CPU fixes, so I enabled boost clocks again. That re-introduced the shutdowns. It got a little better; one of the games did not crash anymore, two others, not as often. But the last one did way worse with crashes right when starting the game or shortly after.

Eventually, through reading this forum and Reddit, I realised it could be the overvoltage protection in the PSU. That made me dial down the GPUs max draw to 80%, which made everything stable again.

I then decided to try to re-enable PBO on my CPU after that again. Unfortunately, I did not save my original PBO values, so I spent the whole weekend achieving optimal settings again. After half a day of CPU tests, I raised my GPU's max power draw to 100% and had no more issues.

Long story short, the Prime TX's OCP does not kick in anymore, and I don't know why it all of a sudden does not... I even put my GPU to 123% max draw, which is the highest I can go in Afterburner. Even then, those games seem to be stable now. I ran the most problematic game and a CB23 (with 12 cores) at the same time. Total peak power of nearly 700W!

The Dark Power Pro never showed any weird behaviour in the few test I did with it. It always trucked along with solid, steady metrics.

I've read about a significant number of people who had issues with a Seasonic Prime TX-850 with various high-end CPUs and 3090 cards.

So now I am conflicted about which PSU to send back or sell respectively. Given that my first TX-850 came DOA and I had to RMA it, I am not very confident in keeping it. Yet, I would rather have some more opinions than my own before going ahead with it.

Please help me make the right decision!

I have a 5950x, Asus Strix 3090 OC on a Crosshair VIII Hero WiFi, and a Seasonic Prime TX-850 (that I got in 2019) and playing RDR2, I get hard shutdown, no BSOD, etc. I found I could replicate crash with Furmark and Prime95 running simultaneously (RealBench or Prime95 separately wouldn't work to replicate issue).
I recently upgraded from 2080ti, 3950x, 32GB Trident to 64GB Royal Elite, aw3480dw to CRG9, Win 10 to Win 11. I spent countless hours troubleshooting this issue.

My wife has a 5600x, Asus Strix 3080ti on a Crosshair VIII Dark hero, and a new Seasonic Prime TX-850. Did same testing on her machine and no hard shutdowns or crashes.

Going to RMA the PSU and report back if new one fixes the 12v Sense issue.
 
I have a 5950x, Asus Strix 3090 OC on a Crosshair VIII Hero WiFi, and a Seasonic Prime TX-850 (that I got in 2019) and playing RDR2, I get hard shutdown, no BSOD, etc. I found I could replicate crash with Furmark and Prime95 running simultaneously (RealBench or Prime95 separately wouldn't work to replicate issue).
I recently upgraded from 2080ti, 3950x, 32GB Trident to 64GB Royal Elite, aw3480dw to CRG9, Win 10 to Win 11. I spent countless hours troubleshooting this issue.

My wife has a 5600x, Asus Strix 3080ti on a Crosshair VIII Dark hero, and a new Seasonic Prime TX-850. Did same testing on her machine and no hard shutdowns or crashes.

Going to RMA the PSU and report back if new one fixes the 12v Sense issue.
Next time you have no BSOD and just shut-offs you should immediately suspect the PSU.
 
I had the hard shutdown problem on an older TX-750 with a 3080ti. Was convinced it was an OCP thing as well after reading this thread and others.

Fortunately, I stumbled across the thread where Johnny Guru explains that it has nothing to do with OCP on the Prime models and is really a problem with high frequency noise (caused by 30 series GPUs) on the 12V sense line bugging shit out.

So I tried temporarily disconnecting that sense line and voila no more shutdowns. With further suggestion and ideas from that same thread, I tried putting a random ferrite bead around the now re-attached 12v sense line and that has so far worked out, no shutdowns since then.

I salvaged the bead from a generic USB cable, sorry don't know any specifics about what composition or bandwidth it is. Still contemplating the RMA though, as this is supposedly fixed on all new models now.
https://hardforum.com/threads/i-think-my-pc-diying-psu-or-shutdown.2016456/
 
Yeah so I had a surprising development yesterday. Well first, the latest TX-1000 + evga 3080 Ti FTW3 has been 100% fine since last Oct (so 7 months usage no hiccups). Yesterday I swiped the TX-1000 and 3080 Ti for a new build. I installed the original older 850W with a humble RTX 3050. Guess what, the PC again immediately black screened and reset with the offending game. (1) Perhaps even a 3050 can pulse enough power to confuse the older psu's ocp, or (2) it's actually the nVidia 3000 driver x86 code path making my Intel i9-9900KS pulse and tickling the older design psu. I will try swapping it with my SS 860W XP2 (platinum 860 instead of titanium 850) and see what happens out of curiosity ...
 
Hi all - this thread helped me resolve my issue so I figure I oughta post in case some other poor sap comes across this in the future. :)

I have a AMD 5950X rig on an ROG Crosshair VIII Hero (WI-FI) motherboard with the latest BIOS (4006, dated 03/10/2022) - power supply is a Seasonic PRIME TX-1000 (see the correlation). GPU is now an EVGA GeForce RTX 3090 Ti FTW3 ULTRA GAMING - which I installed (got it via step-up from my 3080 Ti) a few days ago and it ran fantastic. Fast forward to today I'm installing the latest NVIDIA driver (512.59 WHQL) - so I like to reboot my rigs after that to keep it clean. That's when all the trouble started...I was getting an error code 36 on the mobo with a white light (indicating GPU). Tried a bunch of voodoo but ended up swapping one of the PCI-E cables to another spot in the back of the TX-1000 and that solved the issue.
 
Two and a half years later this thread is still helping out. :LOL: My wife was playing Baldur's Gate 3 on my desktop (5900X + 3090 FTW3) and complained the PC was just randomly shutting down. Temps were fine. My TX-850 ran fine for years, but apparently BG3 gets the power draw up to 700W. Pretty sure the GPU was tripping OCP, since the PSU seems absolutely fine otherwise. I could probably just RMA the TX-850 to get one with more lenient OCP, but I got a TX-1000 to give myself a little more headroom and it's smooth sailing now.
 
I never had another OCP trip in two and a half years after I RMA’d my Seasonic Prime 850. It was a defective unit. If unit is just clicking off randomly, RMA.
 
The older Prime series (e.g. Prime Platinum 850, Prime gold 850, Focus Gold 850, etc) would probably trip, but not the rebranded OneSeasonic models, unless you were unlucky enough to get a TX-850 with an older OCP circuit and I'm not sure how that even happened.

Good intel, thanks for sharing... I currently have an old-ish (2018) Prime Ultra Titanium 850W and had quite a few issues with my old Vega 64. Since I upgraded to a 6900XT it seems to be fine...
 
Last edited:
My Seasonic-built CoolerMaster V1000 (circa 2012 or so) was very rarely tripping with my GTX 1080ti. About once every few months. I chalked that up to RAM/CPU overclock instability. Fast forward and I got a 3080ti, and it was tripping every time I got in game unless I undervolted the 3080ti. Swapped it out to a MSI MPG A850G and it's been good ever since.
 
I'm running a Corsair TX-850 almost 14 years now, that I got from RMA. It was still shrink wrapped when I received it, so I assume it was new. Funny enough the one I purchased didn't last 6 months. Despite the similar name, I get this is not the same but... I guess it happens.
 
Back
Top