Rampage III Gene fullcover waterblock?

[tiborrr]

MrToad, I've just checked everything and there is enough 0.5mm thermal pad encased with the water block to fit all power regulation chips. You need to trim the thermal pad in smaller pieces and cover individual PWM chips if necessary. Just follow the manual and you should be OK. I would really suggest you to remove the gray spongy shim, mentioned in previous posts and do a 'dry run' mount. By 'dry run' I mean install the block as per instruction manual, remove it and inspect the contact. When everything is OK, re-mount the waterblock the same way as before. If there are problems with the contact inspect the mobo/block assembly and troubleshoot the source of bad contact. Take a picture of the possibly problematic spot afterwards!

EDIT: Install the tubing on the water block prior to the final installation of the FC waterblock on the motherboard! There is a chance you have chipped off the SMD elements on the X58 organic packaging in previous attempts.

[MrToad]

MrToad, I've just checked everything and there is enough 0.5mm thermal pad encased with the water block to fit all power regulation chips. You need to trim the thermal pad in smaller pieces and cover individual PWM chips if necessary. Just follow the manual and you should be OK. I would really suggest you to remove the gray spongy shim, mentioned in previous posts and do a 'dry run' mount. By 'dry run' I mean install the block as per instruction manual, remove it and inspect the contact. When everything is OK, re-mount the waterblock the same way as before. If there are problems with the contact inspect the mobo/block assembly and troubleshoot the source of bad contact. Take a picture of the possibly problematic spot afterwards!

EDIT: Install the tubing on the water block prior to the final installation of the FC waterblock on the motherboard! There is a chance you have chipped off the SMD elements on the X58 organic packaging in previous attempts.

I'm "starting fresh" again tonight (brand new mobo and block) and I've worked out an easy way to put the tubing on the block BEFORE mounting it, so no excessive pressure (or torsion/traction) is applied to it after is seated.

I will indeed mount it WITHOUT the "spongy shim"

I still can't see how that caused so many problems, however is the only major difference between my trail of unsuccessful mounts and hardware failures and other people's successes...

Maybe if that's the source of the issues the very first mount on each motherboard caused the damage and then it didn't matter how many attempts or changes I did afterwards, with or without the shim, the damage was already done so the outcome was always the same...

Either way, thanks to everyone for your suggestions (and to Gregor for replying to my ticket)

Wish me good luck

[MrToad]

No joy...

Haven't fried the motherboard this time, but the first mount, doing everything "by the book" resulted in the same symptoms as the other two.

Unless I leave the block almost loose, all I hear is a "click" on the PSU and then powers itself off (for the record I've tried with three different PSUs, which boot the board with the stock cooling fine).

I'll give it another shot tomorrow, now I'm too tired and I know I'll make a stupid mistake and kill something...

[MrToad]

Apologies for stealing your pic Psygnosi, didn't have any of my board available at work...

Did some last late night experiment and this is what I found out. I removed all screws from the block, removed the block itself and then gently seated it in place.

Then I applied force downwards using my fingers on different points. The green circles mark where applying downwards force doesn't prevent the motherboard from powering on, the red circles mark where applying force does stop it.

I checked the thermal pads for anomalies but there seems to be even (and not excessive) pressure in all three of them (the grooves caused by the mosfets on the stock thermal pads are at least 0.3 mm deeper than the ones on the block pads).

Honestly, shy of sending both the board and the block to EK I don't know what else to do... I'm running out of ideas

Is not like is the first time I do this, this is the fifth motherboard model I fit waterblocks on... Except the very first one (a nvidia 680 Abit board) where I crushed the mosfets with the block (novice problems, I tried to use thermal paste instead of thermal pads on them... ) I haven't experienced any other issues, but the feeling of inadequacy is really setting in now...

[Tackleberry]

Did you use backplate on mosfets?

[MrToad]

Did you use backplate on mosfets?

Yep, I did. And no third party pads either. Cut the included 0.5mm pad as instructed.

I also verified contact there too. No shorts, no uncovered mosfets.

[Vetalar]

try to isolate block surroundings (SMDs, chokes etc.) with tape or smth. check MOBO bending while tightening block screws. and what conditions (case or open stand)? maybe bad earthing of system could cause that symptoms...

[MrToad]

try to isolate block surroundings (SMDs, chokes etc.) with tape or smth.

Done. Used ESD safe kapton tape to cover everything in the surrounding area that doesn't get cooled.

check MOBO bending while tightening block screws.

I can see slight pcb warping on the top left corner of the motherboard as soon as I do anything but leave the bolts retaining the mosfet part of the block anything but completely loose.

and what conditions (case or open stand)? maybe bad earthing of system could cause that symptoms...

Both inside the case fully assembled (first few attempts with first motherboard) and "open air" on an ESD safe workstation.

If I have the will to face it during the weekend, which is unlikely really, I might try 0.5 mm thermal tape on the mosfets, starting from the ones adjacent to the NB and then the ones next to the CPU depending on bending/contact...

Thanks for reading and posting the suggestions tho... After so many unsuccessful attempts is not unusual to slip into a "can't see the wood for the trees" state...

[Vetalar]

try to replace mosfet backplate to something harder to avoid bending of MOBO.

[Diverge]

Do you have a DVM (meter)? If so, try to find what the block is shorting too. Set it to continuity mode, and put one probe on the block, and check out ground and +12V. those would be the first ones I'd check as thats what the 8-pin cpu power plug is (yellow = +12V, black = ground)... and that is used by the cpu regulator circuit (which your block is cooling).

you don't need to be powered on to find the short. so a meter should be a safe way to find the short without damaging your board again.

[tiborrr]

As Diverge suggested measure where the block is possibly shorting out . I will try to get a board on monday, install the EK-R3Gene block and test it out. No promises though, I have to see if the board is in stock somewhere in the vicinity.

Best Regards,
N.

[MrToad]

Do you have a DVM (meter)? If so, try to find what the block is shorting too. Set it to continuity mode, and put one probe on the block, and check out ground and +12V. those would be the first ones I'd check as thats what the 8-pin cpu power plug is (yellow = +12V, black = ground)... and that is used by the cpu regulator circuit (which your block is cooling).

you don't need to be powered on to find the short. so a meter should be a safe way to find the short without damaging your board again.

Thanks for the suggestion after reading it I went "duh, how didn't I think on that"

As Diverge suggested measure where the block is possibly shorting out . I will try to get a board on monday, install the EK-R3Gene block and test it out. No promises though, I have to see if the board is in stock somewhere in the vicinity.

Best Regards,
N.

That's really kind of you

I'll take the fluke out to play tomorrow, now I have a 7 year old walking around the house singing "bored, bored, bored, bored..." if we don't go out in the next 15 mins I'm gonna turn murderous

[amajed]

I know its a stupid question but I was stupid enough to forget about using them, did u use some sort of standoffs ? u put them between the block and the motherboard , a plastic standoffs like this:

only its lower , like 2mm or 3 , they make sure the block isn't losing and also not making any unnecessary contacts

[KingLex]

I just ordered the board + block. I'm getting a little worried now
I think I will wait before installing the block when it arrives.

[Vetalar]

@Diverge
since block bolted to board with metal bolts through metalized (and grounded i think) holes in MoBo there MUST be "ground" on the block but no 12v.

[MrToad]

I know its a stupid question but I was stupid enough to forget about using them, did u use some sort of standoffs ? u put them between the block and the motherboard , a plastic standoffs like this:

only its lower , like 2mm or 3 , they make sure the block isn't losing and also not making any unnecessary contacts

Not a stupid question, but yes, I did remember to use the standoffs

In fact I love standoffs and don't miss the times when blocks didn't have them at all.

I think I remember Naekuh championing for them when only one manufacturer seemed to give them any relevance

Either way it's too late now to start tinkering about with the multimeter... Tomorrow after work I'll give it another shot...

Can't say buying a Rampage III Extreme and another block didn't cross my mind tho I'm really starting to lose it with this setup, but things are so tight that changing the board would mean re-thinking the whole thing (I'm using a Lian-Li PC-B10 and a Mo-Ra3 Pro 4x140 so I could fit two pumps, two 5870s and other stuff inside the case...)

Either way, tomorrow more and better

[Diverge]

@Diverge
since block bolted to board with metal bolts through metalized (and grounded i think) holes in MoBo there MUST be "ground" on the block but no 12v.

Nope, it doesn't have grounding at those holes. In fact, the stock heatsink uses plastic washers to keep the screw from touching/scratching any nearby traces. The only holes that have grounding are the motherboard to case holes.

I have a board here, and just looked at it. I been debating weather to get this block or not... Or if I even want to watercool the northbridge/voltage regs.

[MrToad]

And... board #3 down the pan... I was going to check with the multimeter, but before that I did one last "dry run"... Big mistake...

It will no longer POST. It stops on the CPU led, and doesn't follow through. I've checked the CPU and is fine on a different board (phew!). If I lose this 980X I'd be pretty miffed...

This time there's no chipping on the NB or any other discernible faults, but still dead as a dodo...

I have one last R3Gene (the first one that died I actually managed to process an RMA due to the fact that ASUS is one of our suppliers and the rep was feeling kind that day), gonna think very very carefully what my next step is... If the air flow wouldn't be so poor in the area I would just leave it stock and don't look back...

EDIT: I've taken some pics of the dead board just in case they reveal anything I've missed or there are any differences... Apologies for the bad quality of the pics, photography is not really one of my strengths

[MagisD]

Save a board hang a damn fan there.

Sent from my X10a using Tapatalk

[Vetalar]

some pics of the dead board

maybe some damage on the backside? there is 1/3 CPU MOSFETs overhere...
as i'm already told - i'll give up after 2nd kill.

interesting: how many such blocks produced by EK? and is there more negative feedback on them???

[MrToad]

Save a board hang a damn fan there.

Don't think is not tempting, but I'm very obstinate... Sometimes too obstinate for my own good.

maybe some damage on the backside? there is 1/3 CPU MOSFETs overhere...

Nope, I've already checked that... Actually is the NB again... There is a small area on the surface, towards the middle, that doesn't look right... It's like it's suffered smallpox, if you know what I mean. My feeling it's that's shorted again and blown the NB.

as i'm already told - i'll give up after 2nd kill.

And common sense says that you're right... But there's a little insidious voice in my head that keeps pushing me to give it another go...

interesting: how many such blocks produced by EK? and is there more negative feedback on them???

As far as I know I'm the only one having the problems, haven't read any other posts with issues, what has me wondering over and over again "what am I doing wrong?"

EDIT: Anyway I'm waiting to see if Tiborr has anything to add before I go ahead with board #4...

[MagisD]

Definition of insanity is repeating the same actions over and over expecting different results.

Quick question are you useing same block to kill all these boards?

Sent from my X10a using Tapatalk

[Creekin]

but I'm very obstinate... Sometimes too obstinate for my own good.

no kidding? me too, it's a good thing, just costly

But there's a little insidious voice in my head that keeps pushing me to give it another go...

I hear that voice too,....annoying little sod isn't he!

what has me wondering over and over again "what am I doing wrong?"

When i hear that ^ little voice, I put the current plan aside and start from the beginning again. Throw out all pre-conceived idea and theories and start from scratch.

Try to come at it from a different angle, If you take a different path and arrive at the same destination, then it is not you that is wrong, but the vehicle.
edit:

Definition of insanity is repeating the same actions over and over expecting different results

yeah what he said ^

Sorry to be so cryptic, but it works with most problems not just pc's
Hang in there! this is a better drama than Lost!
Can't wait to see who dies next

[MrToad]

Definition of insanity is repeating the same actions over and over expecting different results.

Quick question are you useing same block to kill all these boards?

That's one definition... Another one is:

"What if I told you insane was working fifty hours a week in some office for fifty years at the end of which they tell you to p1ss off; ending up in some retirement village hoping to die before suffering the indignity of trying to make it to the toilet on time? Wouldn't you consider that to be insane?" (Steve Buscemi as Garland Greene in Con Air )

Either way, the answer is no. This is the 2nd block and I haven't repeated my steps over and over again, is just that the difference between "the block is too loose" and "damn, now is too tight and I've shorted/killed something" seems to be in the region of decimals of mm...

[...] When i hear that ^ little voice, I put the current plan aside and start from the beginning again. Throw out all pre-conceived idea and theories and start from scratch.

Try to come at it from a different angle, If you take a different path and arrive at the same destination, then it is not you that is wrong, but the vehicle.[...]

I hear you, that's why I've asked my gf too to look into the problem too. She knows very little about computers, but her logic is sound.

Hang in there! this is a better drama than Lost! Can't wait to see who dies next

Hmmm... Probably my bank balance, but we'll see in the next episode

[psygnosi]

Temporary Solution

I bought a 2nd Rampage III Gene and the EK Nickel/Acetal block to build. Since the 1st board and block are working fine, I did not want to take any chances so I basically taped the whole 2nd EK block(including the channel area and bottom part). I also taped the whole mosfet backplate as well. Now I put some thermal paste on the northbridge and put the taped up block on it. I cut a perfect rectangle on the part where the thermal paste made contact with a xacto knife. Then I got the thermal pads put them on the bottom of the block. I cut around the pads so that the tape underneath can be removed. I did the same thing to the Mosfet backplate. Last but not least I put red paper washers(screw head) on the bottom of the mosfet backplate.
Doing this eliminated any chance of the block causing the short. Only exposed area is the northbridge. The system booted and I was happy. I further went ahead and used the xacto knife to carefully cut around the line where the acetal and nickel met and removed the excess tape on the acetal area. Until someone can actually find where the exact location of the short is, this is possibly the only solution for now. Thanks and hope this helps you guys.

Psygnosi