Overclock.net - An Overclocking Community - Reply to Topic

Thread: Titan X (Pascal) Shunt Mod Done With Resistors Reply to Thread
Title:
Message:

Register Now

In order to be able to post messages on the Overclock.net - An Overclocking Community forums, you must first register.
Please enter your desired user name, your email address and other required details in the form below.
User Name:
If you do not want to register, fill this field only and the name will be used as user name for your post.
Password
Please enter a password for your user account. Note that passwords are case-sensitive.
Password:
Confirm Password:
Email Address
Please enter a valid email address for yourself.
Email Address:

Log-in


  Additional Options
Miscellaneous Options

  Topic Review (Newest First)
10-21-2019 10:48 PM
Asmodian
Quote: Originally Posted by chibi View Post
How feasible is it to run the shunt mod on the stock air cooler? Blower style for the Titan Xp.
Honestly I wouldn't do it myself. The VRM cooling would worry me and they already get pretty hot with the blower coolers. The power/heat for performance trade off due to this mod is not great, Nvidia can save a surprising amount of power/heat without hurting performance too much

If you don't go crazy with the overclock, or especially voltage, it would be fine; no fires or anything. Just not really worth it.
10-20-2019 08:01 PM
chibi How feasible is it to run the shunt mod on the stock air cooler? Blower style for the Titan Xp. I have my blocks disassembled right now and not sure I want to go back to watercooling for the time being. Getting tired of the yearly maintenance lol.
10-20-2019 01:18 PM
Asmodian
Quote: Originally Posted by HyperMatrix View Post
Certainly possible. But how would the liquid metal work its way upwards to the solder connecting the resistor to the board?
This is the well known issue with using liquid metal. It is exactly the reason not to use liquid metal. The gallium is soluble in the solder, it will wick its way through it. Gravity is not enough to prevent it, it is like how water will climb up cloth or paper if you dip a corner in it. It takes a long time with gallium and solder but it will make it there eventually, assuming there is enough gallium to 'wet' all the solder.

It definitely wasn't heat, if the resistance was high enough to get hot the GPU would think it was using a crazy amount of power and instantly throttle. This mod works by lowering the resistance, a lower resistance causes the shunt area to generate less heat for the same amount of power and only the same amount of heat at the new modded power limit if the GPU was hitting it 100% of the time.
10-17-2019 05:24 PM
HyperMatrix
Quote: Originally Posted by chibi View Post
I think what you've experienced is more in line with the liquid metal eating away at the solder joints and not heat.
Certainly possible. But how would the liquid metal work its way upwards to the solder connecting the resistor to the board?
10-17-2019 01:46 PM
chibi
Quote: Originally Posted by HyperMatrix View Post
Just a heads up on this mod. Personally I went the liquid metal route from day 1. Not sure if the same problem could pop up with 2 resisters stacked, but what I found was that due to the increased heat over the resistor, the solder would actually get hot enough to melt.

Snip.

I think what you've experienced is more in line with the liquid metal eating away at the solder joints and not heat.
10-16-2019 06:56 PM
HyperMatrix Just a heads up on this mod. Personally I went the liquid metal route from day 1. Not sure if the same problem could pop up with 2 resisters stacked, but what I found was that due to the increased heat over the resistor, the solder would actually get hot enough to melt. After about a year, the resistor had fallen off the board on one card. This actually was a huge problem because I didn't realize what had happened. My computer shut off, and wouldn't come back on. Bought a new motherboard because I was getting mobo error codes. Didn't work. Then I noticed the resistor had come off. So I resoldered it, and my computer still wouldn't turn on. Fortunately I had a second computer at that time so I was able to test that the video card was working properly now that the resistor had been reinstalled. However, in the process, it had burned my CPU. No idea how. But just a recap:

- Did liquid metal resistor mod
- Got so hot the resistor actually fell off the board after a year
- Killed CPU
- Resoldered resistor
- All working fine now

So if you get a scare and something happens, your video card may not necessarily be dead. Check your mod. On my second video card, the resistor had shifted, but not fallen off yet. It can be very hard to solder the resistor back on the board. On one of my cards, I had to run 30 awg wire from other contact points on the board that were connected to each end of the original resistor location, and attached those wires to a new resistor.

I think I ended up using a healthier dose of solder this time around, along with supporting it with 100 degrees celsius rated glue sticks, which I could take off in the future if needed.

So based on that experience, my recommendation would be to not stack the 2 resistors on top of each other, but to run a wire from the original resistor to the new one, mounted a bit farther away.

Cheers.
08-10-2019 12:15 AM
Asmodian
Quote: Originally Posted by chibi View Post
When stacking the 2x 5M0 resistors, do I still need to utilize MSI Afterburner to raise the power limit?

Similar question, if I order the 3M0 and do a replacement, does that still need the power limit raised in AB? I would prefer not to have third party software installed if the PL issue can be resolved with the hardware fix since no bios flash is available. TXp card by the way.

Thanks!
No in both cases, this is a hardware mod so AB or similar are not required for it to work.

Once it is done the card will always think it is using less power than it really is, we are actually modifying what it uses to determin how much power it is drawing so with everything in the drivers/software/BIOS at stock the real power limit will actually be 2x stock (when stacking another 5 milliohm resistor, using a single 3 milliohm resistor would raise the power limit by 1.667x).
08-09-2019 11:14 PM
ThrashZone Hi,
Default clocks/ boost only throttling would be caused by heat not power demand...
Monitor memory and core temps possibly the water block isn't mounted properly
I myself used the wrong screws/ longer ones instead of the shorter ones and contact was crap and so were temps.
08-09-2019 08:06 PM
Zfast4y0u
Quote: Originally Posted by chibi View Post
My bad, I was not being specific enough. I don't want to OC my card as it's fast enough for the games I'm playing. I just don't want it to power limit under load. Yes, my GPU is watercooled with Heatkiller block and 2x 360mm rads.
if u dont overclock ur card, u have more then enough headroom not to hit power limit ever. in games that is. if u undervolt ur card as well ( i assume ur gpu clocks till 2000mhz ), its going to be even better.

dont try to fix if it aint broken imo. i have 1080ti and max what i saw power draw was like 90-100% max with heavy overclock (in games), and i have headroom till 120%...
08-09-2019 04:38 PM
chibi
Quote: Originally Posted by ThrashZone View Post
Hi,
Not sure how you'd oc the gpu without msi af or similar :/
But yes you'd still need to raise/ unlock voltage control and raise it that is what oc'ing is

It's just the cards firmware won't be able to tell and trip safety features.
Water cooling is recommended.

My bad, I was not being specific enough. I don't want to OC my card as it's fast enough for the games I'm playing. I just don't want it to power limit under load. Yes, my GPU is watercooled with Heatkiller block and 2x 360mm rads.
This thread has more than 10 replies. Click here to review the whole thread.

Posting Rules  
You may post new threads
You may post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off