D-Den 8800 quality and corrosion (pics)

M.D.K.

Limp Gawd
Joined
Aug 31, 2004
Messages
282
So I look over at the front of my computer to see one of my Aquatubes looking like this:


(There's actually a tubemeter half way back that isn't even visible :eek: )

Crap, gotta take apart one of the loops and fix the problem. I'm using LOTS of AquaComputer's Glycol solution in my system, but apparently that isn't enough to keep all that corrosion from happening. My loop has Aluminum, Copper, and a large amount of Stainless Steel.

Anyway, when I got to the 8800 block, I popped it open to find this:



Aside from the large amount of corrosion and sludge buildup, the next thing I notice is the horrible machining job done in making these blocks.

Look at the material left over at the ends of the "channels":


The other end:


If you look at the top of the block (the delrin piece) it's just wide open over the core area, meaning most of the fluid is just going to flow right past the utterly pathetic channels cut into the core plate.




So, apparently $150 at DangerDen these days gets you a piece of junk hardly suitable as a paperweight. A shame too, since I really liked their blocks in the past:


So, anyone wanna trade their EK block for a nice, broken in DangerDen block? I'll clean it off I promise!


We'll see how the inside of my radiator fared when I get to that.
 

Erasmus354

[H]F Junkie
Joined
Mar 12, 2004
Messages
9,450
I agree that the DD block is pathetic, and very dissappointing. However....according to all the AC fans on this forum corrosion doesn't exist! surely you must be confusing corrosion for something else!
 

Arcygenical

Fully [H]
Joined
Jun 10, 2005
Messages
25,065
However....according to all the AC fans on this forum corrosion doesn't exist!

Haha. I was thinking the same thing buddy, I was thinking the same thing.


All that goop looks disgusting btw.
 

M.D.K.

Limp Gawd
Joined
Aug 31, 2004
Messages
282
That goup would be buildup of AC fluid and particles from corrosion.

Edit: Pulled the block from the card, I gotta laugh at the core contact on this thing:


Also I think it was "hand lapped" with a piece of steel wool.
 

Merc

Limp Gawd
Joined
Mar 11, 2005
Messages
409
What temps were you seeing with that DD block? I'm using an MCW60 on my 8800GTS and its working great. At 675/1.90 max GPU Temp (diode) was 45c after a few rounds of 3DMark06. I have a buddy who is trying too get his hands on a DD for his GTX and I suggested he save himself a few bucks as the Swiftech block works like a champ.
 

Arcygenical

Fully [H]
Joined
Jun 10, 2005
Messages
25,065
What temps were you seeing with that DD block? I'm using an MCW60 on my 8800GTS and its working great. At 675/1.90 max GPU Temp (diode) was 45c after a few rounds of 3DMark06. I have a buddy who is trying too get his hands on a DD for his GTX and I suggested he save himself a few bucks as the Swiftech block works like a champ.


Merc, what do you have on the RAMDAC chip by the DVI connectors?
 

Merc

Limp Gawd
Joined
Mar 11, 2005
Messages
409
One of the Swiftech heavy copper heatsinks. I used a little AS Ceramique in the middle of each chip and then a dot of Superglue at the corner of each HS to mount them. RAM temps as reported by Everest Ultimate are in the low to mid 30s but I have a lot of air blowing in there.
 

Arcygenical

Fully [H]
Joined
Jun 10, 2005
Messages
25,065
By RAMDAC, I meant that other open die chip to the right of the main chip... The higher one.

Yeah, my RAMDAC's HSF got up to 80c with one of those HSFs...

with an IR thermometer too. I had to make a HSF outta aluminum for that little bastard.
 

Merc

Limp Gawd
Joined
Mar 11, 2005
Messages
409
It seems to be working but I don't have a temp for that particular chip. Everest Ultimate reports the GPU Diode and case temp, ambient temp and GDDR temps.
 

M.D.K.

Limp Gawd
Joined
Aug 31, 2004
Messages
282
I run for silence, so I wanted a fullcover block since there are no fans in the case. At idle it does 55C-57C, not much over 65 at load. (660/1100)

Anyway, its time to clean this mess up.

Grab your tools: 1 old toothbrush, 1 bottle of ketchup (complete with Lycopene propoganda on reverse side):


Don't forget to take your pimpin' watch off, this could get messy.


Apply liberal amounts of ketchup (the more the better!) and start scrubbing! Here's a shot after the first pass at just the right side:


Just look how much more shiny the right side is!


All done!


If you're wondering why the outer edges of the channel aren't shiny, its because D-Den didn't mask of the channels when they media blasted the outside of the block. A little bit of preparation on their side would've saved me a bit of scrubbing, and probably saved the block from the bit of corrosion that I couldn't get out. I didn't want to risk tearing the rubber seals while scrubbing, so I had to leave a bit of crud on the edges. I'll probably be replacing this block anyway.

Now, onto that radiator! All i have to do is remove a few things....




...this could take a while.
 

Merc

Limp Gawd
Joined
Mar 11, 2005
Messages
409
I don't envy you MDK. Watercooling is great but working on it can be a hassle to say the least.
berserk.gif
 

madmat

2[H]4U
Joined
Jun 3, 2005
Messages
2,372
MDK, why didn't you just pull the O-ring out of the base before cleaning it? Then you could've scrubbed to your heart's content and not had to worry about tearing anything.
 

Merc

Limp Gawd
Joined
Mar 11, 2005
Messages
409
DD is making a killing on that block. It doesn't look like the machining is at all intricate. Very rough in fact. I'd think they'd at least polish the channel. How is the surface on the GPU contact area?
 

Erasmus354

[H]F Junkie
Joined
Mar 12, 2004
Messages
9,450
DD is making a killing on that block. It doesn't look like the machining is at all intricate. Very rough in fact. I'd think they'd at least polish the channel. How is the surface on the GPU contact area?

As long as there isn't scraps hanging off it is better not to polish. A rough finish has more surface area, and better cooling, than a smooth polished finish.
 

madmat

2[H]4U
Joined
Jun 3, 2005
Messages
2,372
As long as there isn't scraps hanging off it is better not to polish. A rough finish has more surface area, and better cooling, than a smooth polished finish.

Yep, just like porting and polishing cylinder heads on cars. Intake ports are left rough to cause the fuel/air mixture to break up more thus staying atomized. The rough texture in the channel breaks up the boundary layer in the water thus eliminating laminar flow which aids in picking up heat from the block by allowing more water molecules to come into contact with the channel.
 

M.D.K.

Limp Gawd
Joined
Aug 31, 2004
Messages
282
DD is making a killing on that block. It doesn't look like the machining is at all intricate. Very rough in fact. I'd think they'd at least polish the channel. How is the surface on the GPU contact area?

I assume you mean the dry side? Its horrible.


^^No, not at all. Besides creating more friction on the water (you want the water to flow evenly through the block), the mediablasted areas corrode easier.


n00btard: I have 2 loops.


Arcygenical: It's a work in progress. My case is 70% handmade and only 80% finished. When its done, the cable management will be much better.
 
Joined
Dec 13, 2004
Messages
820
Get yourself one of these badboys, in stock at Petras.

75105282.jpg



I do have a question about the EK though, I notice the DD is gasketed all the way around on the inside of the water channels as well as the outer edge, The EK only has a gasket around the outer edge, whats to stop the water from just flowing over the middle piece of the EK?

BTW, image socket has to be the slowest image host on the planet.
 

M.D.K.

Limp Gawd
Joined
Aug 31, 2004
Messages
282
the material in the center should be flush with the clear top, so even without a gasket hardly any water is going to get in there. I don't know why D-Den went with a center gasket, maybe to keep water from sitting in the center and corroding?


And yeah, imagesocket is horribly slow, I'm working on getting a new host.
 

madmat

2[H]4U
Joined
Jun 3, 2005
Messages
2,372
^^No, not at all. Besides creating more friction on the water (you want the water to flow evenly through the block), the mediablasted areas corrode easier.

Ummm...I don't think so. if the idea was to get smooth flow through the block there wouldn't be fins or channels or other turbulence inducing items in water blocks. The entire idea with a water block is to expose as many water molecules as possible to as much of the surface of the block as possible. If the block was all smooth and pretty inside like the EK you'll end up with a boundary layer over the smooth areas which act as a blanket insulating the surface of the block from the rest of the water which in turn hinders the transference of heat from the block to the water. If you don't want corrosion in your loop then stop mixing metals in it and it won't be an issue. The EK block will do OK over the die since it has the wave shaped fins but over the ram? Nah, might as well stick to air.

BTW, don't use vinegar on stainless steel, it will eat it. the higher the quality of the stainless (you can tell the quality by the magnetic properties) the faster it will eat through. I worked in a machine shop that produces filters for various media and we used stainless tubing for everything except vinegar bearing fluids, for those we used high carbon steel.
 

Erasmus354

[H]F Junkie
Joined
Mar 12, 2004
Messages
9,450
MDK, did the slime by any chance look like this?

slime.jpg


That is the pic of someone who was also using AC Fluid. It seems like that slime isn't an uncommon problem with AC Fluid.
 

chanchan

Limp Gawd
Joined
Sep 14, 2005
Messages
482
Sorry what is AC fluid? And that is naaaaasty!

I hope I don't get that crap with my setup -____-
 

bob

2[H]4U
Joined
Feb 13, 2002
Messages
2,971
MDK, you might want to rehost your pictures to a server thats not on dial-up... Its been over two minutes now and the first picture I clicked on isnt even half loaded.
 

Bbq

King of Charts
Joined
Feb 21, 2006
Messages
12,272
AC fluid is the little eyedropper-container of fluid you mix with distilled, at 2%, and it should prevent corrosion.

I'm guessing your corrosion problems are because someone at AC didn't do their job properly and didn't anodize the Tube' properly.
 

BrandonB

2[H]4U
Joined
Jan 25, 2007
Messages
2,487
well thanks for that... ill be getting the EK block :)

Let me know what temps you get with the EK block, I'm surprised that my koolance block performs better than the DD block haha, I get idle temps of 40c, load is around 50c
 

Teknokid

[H]ard|Gawd
Joined
Feb 17, 2007
Messages
1,292
yeah no probs! although i just blew all my money on new psu's for my rig. i also got a triple rad. by april i should have ek block, 250ml res, new tubing and remote control (speed adjustable) fans on my rad :)
 

BillParrish

Supreme [H]ardness
Joined
Aug 25, 2006
Messages
7,520
Those metal chips on the lip where the mill cutter either started or finished it cut is totally unacceptable. Easily clog a rad channel if one came loose and if one was running around loose in my pump impeller !! :eek: Rest of it I dont care to debate, but those chips are horrible.

DD shame shame, go beat the crap out of your machinist, any kid in high school metal shop knows better than to leave that stuff hanging off a piece of work.

Edit: I have sent DD a msg concerning this thread
 

Stevennoland

Limp Gawd
Joined
Jan 5, 2006
Messages
418
Look at the material left over at the ends of the "channels":

I'm actually am a big fan of DD products, but this picture does them no justice. Those burrs are completely unacceptable. If I were you, I would contact DD and let them know about this. I would want to know about quality issues like this. I'm ashamed for DD.
 

Merc

Limp Gawd
Joined
Mar 11, 2005
Messages
409
I understand what everyone is saying about laminar flow and boundary layers and you are all absolutely correct. I just think that if you are designing a block with one continuous channel it would be better to break the flow up in an engineered way rather than leave it to random roughness from a less than accurate cut.

It is obvious that the channel was blasted in the same way as the rest of the block. With that technique there is no way of knowing how strong any of those little peaks of material are. Put a flow of heated liquid on it and let it run continuously and erosion will eventually find the weak ones, break them off and put them in your loop where they'll end up in some constricted space, like an intricately cut CPU block, where they can do the most damage.

A much better method is to machine that channel down to a decent smoothness and strategically place turbulence generators in the stream to break up the boundary layers. Also, there is already quite a bit of turbulence just from the liquid entering the block through the nozzle. A smooth surface would serve to calm that water down a bit for entry into the channels over the GPU area. So, though I agree on boundary layers needing to be broken up and the theoretical benefits of a randomly blasted surface making more contact room I ain't buying it. There was only one thought in this design, saving manufacturing costs.

Worse yet, liquid coolant will eventually wear that channel smooth and all that abraded material will be scattered throughout your entire watercooling loop polluting those blocks that were well machined, wearing away seals and breaking down your pump. After all, you have injected a metal abrasive into your system. Better manufacturers smoothly machine the metal knowing this.
 

Shoggy

Sales and Support at Aqua Computer
Joined
Feb 14, 2005
Messages
621
I'm guessing your corrosion problems are because someone at AC didn't do their job properly and didn't anodize the Tube' properly.

AC does not anodize their products itself. It's done by Rohde AG. That company is known for their high quality and is mainly anodizing for visual and medical products as well as measuring systems.

I don't know the price for a aquatube but only the anodizing of a aquagraFX block costs ~5 Euro. Other companys in Germany would do that for 0.50 Euro... - but not at that quality level.

If you really think the covering is faulty than send a mail to support@aqua-computer.de. I'm pretty sure that they are very interested in your aquatube IF it is faulty. Only a few days ago Stephan wrote that they analyze faulty parts together with Rode AG.

And always remember what a child has told us: there is no corrosion :D
 

M.D.K.

Limp Gawd
Joined
Aug 31, 2004
Messages
282
Shoggy: The goop in the block looked very similar to that. Mine wasn't that bad, but I bet it would have been if I had let it go any longer. This loop was only running for 2 MONTHS!

Are you saying I should avoid the AC fluid? I have a bottle of similar stuff from Innovatek that would probably work. The strange thing is that the other loop in my system has almost the same materials in it (they both have an Aquatube, one of my radiators, a blackice radiator, and D-Den blocks) and I haven't had a single problem with it.


To everyone's comments about the burrs on the block: I completely agree, and thats the reason I posted the thread. Unfortunately I don't have anything else to use at the moment, so the D-Den block has to go back on.

Edit: Merc I don't think there was any engineering influence on the finish quality of this block. The whole thing just screams of "how many of these can we make and how cheap can we make them", which is really unfortunate because I really, really used to like DangerDen.
 

FlimFlam

Limp Gawd
Joined
Jan 5, 2005
Messages
490
You might want to try Fluid XP. Yeah, it is expensive, but I like it over everything else I have tried. I used to go with usually a zerex/distilled water mixture, and then I have tried other things like Primochill, and Danger Den's mcp fluid (i forgot if it is the 5 or 40 stuff), but DD's fluid had a sediment on the bottom and suspended in the coolant.

At any rate, Fluid XP is more viscous and you may have a negative delta T, but it is really maintenance free. I have had it in a system for like 2 years without issues, and spilled like 18 oz on two cards, a mobo, and ram which were fine. Also, I try not to mix metals in blocks and rads without lots of additives.
 

M.D.K.

Limp Gawd
Joined
Aug 31, 2004
Messages
282
Yeah, I was going to get some FluidXP, but no one had it in stock with some other parts I needed.

Anyone have any tips for cleaning out the inside of a BlackIce 240GT?
 

n00btard

Gawd
Joined
Dec 10, 2005
Messages
922
some 13-year old kid on XS ("serialk11r") said he flushed his radiator with Listerine, and load of crap came out.

but he's 13-years old and his English is almost another language, so I'm going to leave the choice of trying Listerine to you.
 

R1ckCa1n

Supreme [H]ardness
Joined
Jul 22, 2004
Messages
4,355
Crap, gotta take apart one of the loops and fix the problem. I'm using LOTS of AquaComputer's Glycol solution in my system, but apparently that isn't enough to keep all that corrosion from happening. My loop has Aluminum, Copper, and a large amount of Stainless Steel.

The key word is LOTS and is the root of your problem. Too much AC Fluid is worse than not enough as it leads to build up. For example I can get four to five fills with one bottle of AC Fluid. I am guessing that loop had 1/2 to a full bottle used.

I agree that the DD block is pathetic, and very dissappointing. However....according to all the AC fans on this forum corrosion doesn't exist! surely you must be confusing corrosion for something else!

When used correctly, corrosion is the last concern. But yes, DD's quality is in the toilet as has been for years. Another reason I would rather air cool than use a DD block.

MDK, did the slime by any chance look like this?

slime.jpg


That is the pic of someone who was also using AC Fluid. It seems like that slime isn't an uncommon problem with AC Fluid.

Another example of either too much AC Fluid or too much Glow Motion.

AC does not anodize their products itself. It's done by Rohde AG. That company is known for their high quality and is mainly anodizing for visual and medical products as well as measuring systems.

I don't know the price for a aquatube but only the anodizing of a aquagraFX block costs ~5 Euro. Other companys in Germany would do that for 0.50 Euro... - but not at that quality level.

If you really think the covering is faulty than send a mail to support@aqua-computer.de. I'm pretty sure that they are very interested in your aquatube IF it is faulty. Only a few days ago Stephan wrote that they analyze faulty parts together with Rode AG.

And always remember what a child has told us: there is no corrosion :D

Again, when used correctly there should be no problems.
 

theelectic

Limp Gawd
Joined
Oct 30, 2006
Messages
362
some 13-year old kid on XS ("serialk11r") said he flushed his radiator with Listerine, and load of crap came out.

but he's 13-years old and his English is almost another language, so I'm going to leave the choice of trying Listerine to you.

He is the reason ignore lists were created.
 
Top