Near disaster with an Apogee GTX

[hellcamino]

I've been using an a Swiftech Apogee GTX for about 9 months now as well as Primochill Ice for coolant.

In the last couple of days I have noticed my cpu overheating, when I tested the overclocks I wasn't breaking 55c and suddenly I am getting forced cpu shutdowns due to temps!

I checked and sure enough I have no coolant flow...*sigh* many people told me I was crazy for running so much from a single pump and maybe they were right.

So I removed the pump after pinching off it's connecting hoses and set it up next to my kitchen sink with a jumped PSU and a 2 gallon bucket full of water. Once primed the pump would move 2 gallons of water in about 25-30 seconds, that sounds about right according to specs for an MCP655-B pump so I put it back in the case and made sure there was no air left in the system . No flow...

Now I'm getting serious here, I ordered another pump as well as full cover gpu blocks for my 4870X2's and more MX-2 ( I will review the gpu blocks later) .

Now I dissasembled the entire system and drained everything and started cleaning and inspecting all my parts. Everything seemed fine until I got to my Swiftech Apogee GTX block, I had been blowing through all blocks to ensure proper flow and when I tried to blow through my cpu block I almost popped an eyeball out due to unexpected resistance...Great I found the problem I thought and set about looking down the barbs with a flashlight and saw it was packed full of blue stuff. Amazing concidence here the blue stuff matches the color of my coolant perfectly! I tried letting straight hot water run through the block and worked my way up to 100psi compressed air to no discernible effect. So I disassembled the block to see if it was even salvageable and it in fact is NOT salvageable. The gunk I could clean out relatively easily now but the corrosion was terrible having eaten a hole almost all the way through the top shell of the block as the plating was disintegrating.

Here are some pics :

First the top after being cleaned of blue gunk. Look at the lower right and you will see the hole and nearer the center you can see where the plating is flaking off.



Now the bottom or cpu contact side.



In conclusion I'll never be using the Primochill Ice coolant again, as this coolant was changed only 4 months ago I'm thinking the recommended 6 month change interval is a joke. The Apogee GTX I'll be attempting to rma shortly.

Maybe I should start selling Engine Ice in 2oz bottles and call it something spiffy like...Swiftech does with hdrx aka antifreeze lmao, I spoke to Gabe on the phone today while getting the rma setup for my Apogee GTX and he let slip that it's valvoline racing coolant "officialy endorsed by NASCAR". I laughed myself half to death over that one!

I already knew it was just antifreeze in a 2oz bottle ( what a money maker ) but he started telling me how much better it was than any other coolant additive. Gabe in fact stated that it was made in the millions of gallons and was in use for many different cooling applications including race cars....

Well I immediately called BS on that one as I grew up building/working on race cars as a member of a pit crew and for over 20 years now that I can attest to the only additive you are allowed to run in NASCAR is in fact water wetter! Yep you guessed it! The same exact stuff people use in their pc's coolant! I have personally seen this stuff work to fantastic effect in a race car with a 30 degree drop in temps within 5 minutes of pouring it in.

Gabe says BS on that one of course but I nailed him on the Hydrx = antifreeze question and for those of you who have been duped into buying antifreeze in tiny bottles...LOL hit me up I will so send you some antifreeze in a little bottle for HALF the price of Hydrx! I will offer a special DEAL if purchased by the gallon even lmao!! I will even paste a label on it that says "better than snake oil" or "not antifreeze" ROFLMFAO!!!!!

On a serious note I spoke to both Gabe and Michelle at Swiftech today and you definitely want to speak only with Michelle as Gabe will bend your ear with his facts and figures and "we are engineers" crap talk forever it seemed like. Gabe also insisted that I had been doing something wrong with my Apogee GTX since lapping it dropped my temps....Interesting point there as lapping it dropped my cpu temps at load by 15c! The "bowing" concept he is working with would be great if you didn't bow the hell out of your board by the time you got the block tight enough to achieve proper contact.

A positive to this is that Michelle said she would ship a new top for my Apogee GTX today. The downside is that they want me to ship my old top to them. WTF for ? I could send them a pic that would be quite obvious but Gabe apparently wouldn't believe it and insisted that I should go look at what "real systems" are like over at XS....Which is where I would supposedly find that my case is incredible isolated. I just Google'd Apogee GTX corrosion instead and immediately came up with a ton of threads on XS where people had the exact same problem I did!

I will reassemble my Apogee GTX block but will never use it in my main benching rig after a narrow avoidance of disaster, I'll throw it on my Abit IP35 Pro and see if I can run some 5ghz benches with an e8600. I'll also be tearing the block down each and every time I change the coolant.

As a disclaimer I have conducted no scientific test on Switech's Hydrx and I'm sure they probably put a few drops of special sauce in each little bottle so they can claim it as their own.Thermaltake Mozart Tx case
PC Power & Cooling 860 Turbo
Asus P5Q
X3360 @ 4.054ghz
4x1gb Mushkin @954mhz
Sapphire 4870X2cf w/Ek blocks
x-fi fatal1ty Pro
3x74gb raptors RAID0
lcs :
MCP655 pump x2
Dtek Fuzionv2 w/quad nozzle
EK nb block
enzotech slf-1
swiftech micro res
Thermochill 120.2 radiator
Swiftech 120x2 radiator

[fatman]

wow good thing you caught it when you did. lmao

[Xilikon]

Military plating specifications my ass... All it need is just a pinhole and the plating become totally useless. Good thing Swiftech made a new design without aluminium this time with the GTZ.

[migueld]

Why don't you get the copper top? I'd be skeptic about using alu again :-/

BTW, if you like blue, then I recommend that you get Pentosin G11 which I've been using for about a year, and it's crystal clear, no stains or gunk.

[hellcamino]

I am considering either Pentosin or antifreeze currently but as for buying the copper top...Forget it! I already bought a junk part that didn't last and I learned my lesson too! Never reward a company that sells you a bad product by buying another part from them to fix it.

The saying goes :screw me once... shame on you... screw me twice ...shame on me. With the attitude I got from Gabe on the phone I'll never buy anything from Swiftech ever again.

[hellcamino]

What exactly are you testing and will you be posting results ?

[kevinbo03]

I want to test the effectiveness of additives, And see how well they reduce corrosion in Alu/Copper loops. Basically, I want to see if it's feasible to run a super ghetto LC loop on crunchers, and just use tons of additives to offset the corrosion.

A half dead GTX is perfect since it's already primed for corrosion, And would be a real world test. I'd post results, though i sure wouldn't be no scientist. This would mostly be a proof of concept for myself before making an investment.

[kevinbo03]

And i just want to say that while i personally dont think the GTX was a smart move on swiftech's part, I've had many other of their products that work great. MCW60, MCR's, And MCRES just to name a few. GREAT products. I think they've (obviously) strayed away from copper/alu mixes, And thats good. Dont make a blanket decision based on one fault- The recovery says everything.

[Bobly]

Man that kinda sucks... however they're a couple of things that really bugged me in your whole post.

The Apogee GTX I'll be attempting to rma shortly.

Was it actually something wrong with the GTX, because it sounds like you're putting all the blame on the coolant, or maybe the simple fact your coolant didn't have any anti-corrosion to avoid this... Either way you shouldn't really be RMA'ing but heh, ethics differ (as we saw in the RMA ethics thread).

On a serious note I spoke to both Gabe and Michelle at Swiftech today and you definitely want to speak only with Michelle as Gabe will bend your ear with his facts and figures and "we are engineers" crap talk forever it seemed like. Gabe also insisted that I had been doing something wrong with my Apogee GTX since lapping it dropped my temps....Interesting point there as lapping it dropped my cpu temps at load by 15c! The "bowing" concept he is working with would be great if you didn't bow the hell out of your board by the time you got the block tight enough to achieve proper contact.

A positive to this is that Michelle said she would ship a new top for my Apogee GTX today. The downside is that they want me to ship my old top to them. WTF for ? I could send them a pic that would be quite obvious but Gabe apparently wouldn't believe it and insisted that I should go look at what "real systems" are like over at XS....Which is where I would supposedly find that my case is incredible isolated. I just Google'd Apogee GTX corrosion instead and immediately came up with a ton of threads on XS where people had the exact same problem I did!

More importantly, this bit just made me cringe... I mean, where do I start?

15C temp drop from lapping? Yeah right, did you forget to mention that between unlapped and lapped you also changed from using chewing-gum to MX2 as thermal paste?

When you RMA or ask for a replacement it's standard procedure to ship the old one back, some companies will let you ship back the old one after the receive the new one but most go as far as wanting the old one first, it's the price you have to pay to have stuff fixed for free.



Bottom line, I feel for you, really, what happened to you sucks, but don't act like a 13 year old about it

[hellcamino]

As a matter of fact I have been using MX-2 for TIM for quite some time now and it's all I ever used on that block too...I also stated that this block wouldn't ever go back in this rig so I could truly care less about the rma. My coolant did in fact have corrosion inhibitors and the only affected part was in fact the Apogee GTX block itself!

Acting like a 13 year old ? W/E dude, it sounds like you need to read up a bit before you tell me I'm full of it as this is a very common problem. Your comment about lapping makes me LOL as I can only assume that you have never done it yourself!

[Repliquant67]

I had a apoge gtx with inovatek anti-corosion liquid during 1 year and have a begining of corosion at the level of the fittings.

[BulldogPO]

I´m so glad that I bougth Copper Top to my GTX
Now I´m hoping that Swiftech will make Nehalem Socket 1366 mounting kit for this.

[Repliquant67]

With the price of the copper top, you buy a better waterblock ...

[BulldogPO]

I did get my Copper top really cheap.

[hellcamino]

With the price of the copper top, you buy a better waterblock ...

Indeed I did, a D-tek fuzion v2 with a quad insert.

[Solarfall]

wow that looks nasty..i've never had such issues, im using distilled water and thats it..

[Sailindawg]

I just put together my water cooling loop. Two things I stayed far away from, mixed metals and any proprietary coolant. I'm just running distilled water with PT Nuke. I don't care about bling, only good cooling results. I guess if I was to useany additive it would probably be Pentosin. I tried Hydrix , but was not that impressed with it, plus it's green.

It seems you're sorting it all out. I was pretty amazed at the accelerated rate of corrosion in that water block.

[CyberDruid]

Yeah but we keep hearing straight from the horse's mouth that GTX blocks are fine they never corrode! How could this be

Maybe one day Swiftech will pull that piece of junk from their otherwise excellent line. If I were the "man" I would have dropped the GTX from the line the FIRST time this was posted. And you know the Stealth is the same POS.

C'mon Gabe....wake up and smell the galvanic corrosion.

[Xilikon]

Man that kinda sucks... however they're a couple of things that really bugged me in your whole post.

Was it actually something wrong with the GTX, because it sounds like you're putting all the blame on the coolant, or maybe the simple fact your coolant didn't have any anti-corrosion to avoid this... Either way you shouldn't really be RMA'ing but heh, ethics differ (as we saw in the RMA ethics thread).

I beg to differ on this. Technically, Gabe is telling everyone it will be fine till the face is red. It's obvious the plating failed so it's grounds for a RMA but be honest with Michelle of Swiftech about that. If they truly care for customer care, they will replace it with another block (hopefully not a GTX, maybe a GT or a GTZ).

It's very hard to deliberately corrode a block since you will need to have lots and lots of patience. Nobody sane will let a corroding block for that long in a loop so this show he's unaware of this until he disassembled.

[Trigger5521]

My GTX..same thing different color. ALU = NFG

[DeathWalking]

Holy , dude, that's pretty bad.

[Navanod]

Nothing wrong with PC ICE my friend...the fact that your other copper components didn't get ruin by that alu top on the GTX means your coolant did its job.
But when its DIRECT CONTACT between the alu and the copper on that block, no amount of coolant corrosion inhibition's gonna stop that GC rating through

[Broken]

Wow, Swiftech pulled a Koolance. I wouldn't even use that block anymore, unless they man up and send you a copper top, which should have been on it in the first place, or at least delrin.

[[XC] riptide]

I mustr confess I also had a GTX that went to absolute crap after a long time 24/7 running. It got completely blocked. The GT's are still running fine however,

[T_Flight]

Aluminum.

As God as my witness I will not get Al near my system.

It will be brass and copper for me only. No feerous metals, and no Al...at least I better never find any in there.