Overclock.net - An Overclocking Community - Reply to Topic
Thread: 2080 Ti - Working Shunt Mod 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)
03-27-2020 12:57 PM
Asmodian
Quote: Originally Posted by ThrashZone View Post
Hi,
I added just a little liquid metal so there is good contact two both shunts "and I do mean a little" and hot glued the extra shunt on the outside end just over the top and held the shunt down until the glue dried.
Works just fine and no soldering involved.
Want to remove it just use alcohol until it release the bond on the gpu board clean up a little and boom done no trace.
Liquid metal and solder are not safe together, the liquid metal will diffuse into the solder. This can mean nothing bad will happen, or the mod could stop working randomly, or that the original shunt desolders and the entire GPU stops working (probably unlikely with a very small amount). You also don't know what your real power limit is, though that isn't particularly important. The liquid metal is also likely to stain or leave a trace after being left in place for a long time. Soldering a resistor on top is quite easy and much less likely to change over time.

The only reason I can see for not wanting to solder is to abuse the RMA if it breaks.

Edit: Your hot glue and liquid metal is almost as likely to leave a residue as solder would. I can get a very clean desolder if I spend some time cleaning it up, and you actually flushed your warranty as soon as you started modding.
03-27-2020 12:15 PM
ThrashZone
Quote: Originally Posted by Imprezzion View Post
Doesn't that have a big risk of shorting the shunt and thus tripping the protection?
Hi,
Depends on how well it's done.
Beats the hell out of soldering and flushing a warranty down the drain.
03-27-2020 09:21 AM
Imprezzion
Quote: Originally Posted by ThrashZone View Post
Hi,
I added just a little liquid metal so there is good contact two both shunts "and I do mean a little" and hot glued the extra shunt on the outside end just over the top and held the shunt down until the glue dried.
Works just fine and no soldering involved.
Want to remove it just use alcohol until it release the bond on the gpu board clean up a little and boom done no trace.
Doesn't that have a big risk of shorting the shunt and thus tripping the protection?
03-27-2020 06:43 AM
ThrashZone Hi,
I added just a little liquid metal so there is good contact two both shunts "and I do mean a little" and hot glued the extra shunt on the outside end just over the top and held the shunt down until the glue dried.
Works just fine and no soldering involved.
Want to remove it just use alcohol until it release the bond on the gpu board clean up a little and boom done no trace.
03-20-2020 04:30 PM
Asmodian Worth it... hmm... No, it is not worth it. It would be very hard to notice the difference in any real world workload. But this is Overclock!

That said throttling is more complex than "it says above 1905 most of the time". Throttling happens on very short timescales, the MHz you see reported is simply the average over the sampling timescale. Also, to check if you are throttling or not you need to watch the Power Limit trigger 0/1 value in Afterburner or similar. You can see stable MHz and still get very short throttling due to hitting the power limit. I do notice the difference in benchmarks too, very slightly faster with a power limit high enough to never trigger.
03-20-2020 04:23 PM
JackCY
Quote: Originally Posted by Ben B View Post
Very interesting, never would have guessed a 12ohm would add that much power. With all things considered, what do you think would be the highest power outputs i could push the card without going over into serious problems and risking too much of its health? crazy to think you used a 8ohm resistor. is the gpu you have being LN2 cooled ?
It's elementary school math.

I didn't think there is a need for a spreadsheet on web. Everyone just makes the calculation themselves or blind trusts what ever from internet, I prefer the first and do the calculations, check power delivery and it's capabilities from specs etc.

1/(1/R1+1/R2) = result

That's all there is to it for parallel connected resistors.

You should not attempt to solder SMDs on expensive parts unless you have some experience with soldering small components already. And in that case this whole thread is common knowledge and nothing new that a parallel resistor will change how high the power is sensed. I'm rather surprised they did not go away from the resistor sensing already, especially for these higher currents on GPUs.
03-20-2020 04:09 PM
Imprezzion Now all I gotta consider is whether it's worth it. I ran some tests on a lower load game that doesn't max out the power limit and set it to 1.093v in the curve and it didn't even manage 2080Mhz. 2055 on 1.093 is a maybe. It ran for a good 10 minutes and then I went and played some Warzone but then again, is shunting it worth not being able to resell it easily and barely gaining 100Mhz IF that even proves to be remotely stable. I mean, I'm at 1950Mhz @ 1.043v now with just 310w BIOS and in any game I run it stays well above 1905 most of the time only occasionally dropping to 1935/1920. This non-A is quite terrible lol. At least it has Samsung memory that quite easily does 8000+Mhz.
03-20-2020 08:24 AM
Asmodian
Quote: Originally Posted by Imprezzion View Post
EDIT: I can't really get 0.007 Ohm here easily at the moment. 0.01 is easier to get. Would this work for example? Any wattage that they have to have?
https://www.conrad.nl/p/ralec-lr2512...1-stuks-454377
Those look good, the ones I used are only 1W. They dissipate very little power themselves.

Adding a 0.01 Ohm on top would give you a +66.67% power limit.

Edit: At 500W draw, with a 0.00333 Ohm total resistance (0.01 Ohm on top), both resistors together would dissipate 1.667W and most of that goes through the original one.
03-19-2020 04:20 PM
Imprezzion
Quote: Originally Posted by Asmodian View Post
Mine is water cooled with a EK full cover block and a crazy pump and 8x140 of radiator (with the 7900X in the same loop).

I think doubling the power limit is fine when water cooled (temps <60°C). I wouldn't run furmark though. No idea on actual risk, I don't have anything but anecdotes that make it seem pretty safe if temps are kept low. The card still throttles for temperature too so anecdotally more power is probably not that dangerous, only upping the load on the power supply. Not sure how happy a card would be with no power limit but throttling due to temperature all the time. It would probably have a short life but also likely similar to running stock power and throttling due to temps all the time, only with temperature issues being much more likely.



Personally I would use 7 mOhm resistors. You can always lower the limit below 100% with Afterburner if you need to and with a 7 mOhm mod you can explore where your limit really is. It also prevents throttling for shorter spikes, as long as your game does not run at 100% power all the time. Even 12 mOhm is well above the limit of the heatsink, assuming you keep the same BIOS, and once you go above the watt capability of the cooling system temps go up fast. How are your temps now?
It's sitting in the mid to high 50's in Modern Warfare Warzone which maxes the current power limit pretty well between 114-124% constantly. It's at 1950Mhz @ 1.043v with 8000Mhz memory (memory clock impacts power quite a lot.

If I loop Superposition 1080p Extreme which just slams 124% power constantly it flatlines at 62c.

My fan setup is quite Frankenstein. I got a Accelero IV on it with 2 Cooler Master MF140 RGB fans mounted to it and a whole load of Enzotech copper VRAM and Alphacool copper VRM heatsinks.

The fans PWM control wires and RPM wires are connected to the card, the 12v and - to my motherboard to not blow up the cards fan controller.

It sits at about 44-47% fanspeed which is around 850-900RPM. The fans are capable of 1550RPM full throttle and that drops temps like a rock but gets loud. Max comfortable noise level is around 60% @ 1200 RPM. That is low 50c territory. However, I don't know when I'm going to simply overwhelm the heatsink and at which point fanspeed won't matter anymore. I pushed 420w on a 1080 Ti with XOC unlimited BIOS on this heatsink and 420w constant load saw 70-75c on 1200 RPM.

EDIT: I can't really get 0.007 Ohm here easily at the moment. 0.01 is easier to get. Would this work for example? Any wattage that they have to have?

https://www.conrad.nl/p/ralec-lr2512...1-stuks-454377
03-19-2020 04:08 PM
Asmodian
Quote: Originally Posted by Ben B View Post
Very interesting, never would have guessed a 12ohm would add that much power. With all things considered, what do you think would be the highest power outputs i could push the card without going over into serious problems and risking too much of its health? crazy to think you used a 8ohm resistor. is the gpu you have being LN2 cooled ?
Mine is water cooled with a EK full cover block and a crazy pump and 8x140 of radiator (with the 7900X in the same loop).

I think doubling the power limit is fine when water cooled (temps <60°C). I wouldn't run furmark though. No idea on actual risk, I don't have anything but anecdotes that make it seem pretty safe if temps are kept low. The card still throttles for temperature too so anecdotally more power is probably not that dangerous, only upping the load on the power supply. Not sure how happy a card would be with no power limit but throttling due to temperature all the time. It would probably have a short life but also likely similar to running stock power and throttling due to temps all the time, only with temperature issues being much more likely.

Quote: Originally Posted by Imprezzion View Post
What resistance would you guys recommend? 7 mOhm, 10mOhm or even a bit higher like 12 or 15?
Personally I would use 7 mOhm resistors. You can always lower the limit below 100% with Afterburner if you need to and with a 7 mOhm mod you can explore where your limit really is. It also prevents throttling for shorter spikes, as long as your game does not run at 100% power all the time. Even 12 mOhm is well above the limit of the heatsink, assuming you keep the same BIOS, and once you go above the watt capability of the cooling system temps go up fast. How are your temps now?
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