H50 Hydro

[pluke the 2]

Hey guys,

I was wondering if anyone has experienced the same problems with the Corsair Hydro H50 unit. I have a I7 920 with P6T Deluxe V2 with Vcore @ 1.192 OC'd to 4.035 and while running Prime95 Temps are getting up to 71C. I have very adequate cooling in my Xaser Case (fully modded) and I have taken out my heatsink and re seated it twice. I have applied the perfect and right amount of Artic Silver 5 both times having the same results. Please any help would be much appreciated. Anyone put a higher CFM fan on it? PS I have seen great reviews on this heat sink and it has worked really well for many people, I would like for people who are considering to buy this look into this forum and do some homework. If anyone can give any more input for me and to help others.

thank you,

pluke

 

[pluke the 2]

Anyone?

 

[Shadow703793]

^+1. The H50 doesn't have enough rad to cool the i7. You need AT LEAST a 220 rad to cool an i7 well.

 

[Conumdrum]

Peeps pushing the overclocks like you are won't find much comfort in the H50. It's a tiny radiator. You can improve cooling by getting some MUCH higher CFM fans in push/pull. You'll still get about the same performance as a high end air cooler with HS push/pull fans.

People who have real watercooling have at LEAST a 120x2 sized rad, a MUCH more effective heatsink, a more powerful pump, and MUCH fatter hose.

Of course thats:
$50 rad (easily $100 for a GREAT rad)
$65+ pump
$75 CPU block (my new XT was $83 shipped)
And 20-30 more in hose etc.
Ohh fans too.

You get what you pay for......................................... My I7 965 stock at 81F ambient room temps loads at 55C. I'd be in the low 60's and VERY quiet if I had your clocks.

 

[pluke the 2]

Mmmm, I was kind of hinting toward the folkes that do have H50 w/ I7 920 getting nice temps, I do appreciate all of your takes.

 

[rubix_1011]

It's still a piece of crap, regardless.

 

[Shadow703793]

^+1. A TRUE 120 or a Meg beats it hands down.

 

[pluke the 2]

Finally fixed it,

INTEL I7 920 OC'D TO 4.1 @ 1.92 LOAD @ 52C.

HYDRO H50 CORSAIR

Pluke -

 

[Viking445]

What was the issue and what fixed the it? I'm considering an H50 for my Phenom 965 BE rig.

 

[pluke the 2]

The issue was not enough thermal paste from the copper heatsink to my CPU. See the below thread:

FIX = after re-seating my Hydro H50 Copper Heatsink and applying Artic Silver 5 Thermal Paste for the 3rd time. Also look into the AMD bracket you have to buy (Additional Purchase) that doesn't come with the Hydro unit... you can see this thread here, it's basically where I screwed up...

Here is the thread:

http://www.tomshardware.com/forum/255294-29-asking-huge-favor

Pluke

 

[Conumdrum]

Wait, you have your CPU at 1.9 volts?

Prime 95 small ffts for 30 min? And whats your room temps?

Look at page 22 and 23.
http://download.intel.com/design/processor/datashts/320834.pdf

I smell fish...............

 

[Shadow703793]

^+1. There is NO WAY you can run at 1.9v with out LN2/DIce.

 

[pluke the 2]

uhh, 1.192* no fish for us sir

 

[Shadow703793]

^ lol, thought so.

 

[paul mohr]

I thought about getting the HD50 but none of the reviews I read on it were very promising. It seems the small sealed water cooling units really don't compete well with a better air cooling system. Before I dropped 80 bucks on of these water coolers I think I would jury rig a cage fan blower and just pump a butt load of air into an open case.

Probably not the most quiet option though lol.

This whole cooler thing seems like it would be fairly straight forward, not like its rocket science or anything. Yet I have read lots of reviews where a simple inexpensive air cooler out performs a huge and expensive radical looking air cooler. You would not think that would be the case but when you read two or three comparisons from different testers you have to start to believe what you see.

I think the HD 50 would be an ok option if you were running a moderate system and just wanted to lower temps a bit. And you didn't have room for a huge heatsink or wanted something really quiet. IF you could get it on sale. But if your trying to cool something that gets really hot I just don't think it will hang with a better rated air cooler.

Another thing to consider when you are looking at reviews or posts on a forum is that most of us don't use monitoring equipment like a test lab would. We are just using some software that tell us what the sensors in our computer say. And these are not always real accurate. Mine shows me temps from 20C at idle and 35C under a load with my AMD. For all I know the temps could be wrong and I might really be running 50 or 60C. So take some of these temps you see with a grain of salt.

What you can look at is the temperature differences. "Not accurate" is not really the best way to put it. Not calibrated would be better. From what I understand the temp difference will be accurate. Just the actual temp may not be correct. Say at idle it says 35C, well it might actually be 40C. But then when the temp goes up to say 50C it will really be 55C. The temp reading will be off, but the temperature spread is accurate. It will be off by the same amount regardless of what temp it reads.

This is how it was explained to me anyway. So sometimes it might be better to look at the temp differences rather than the actual temperature readings.

If none of that made sense maybe someone can explain it better lol.

Paul

 

[pluke the 2]

Great explanation.

 

[xbonez]

a lot of people don't seem too happy with the H50. personally i'm using it with my Phenom II X4 965 on an MSI 790FX-GD70 and am very happy with it. I have OC'ed my processor to 3.8ghx at 1.475V, and during a Prime95 stress test, temps never exceed 48-50 degrees. The AMD 965 has a threshold of 62 degrees, so I'm way under the limit.

I use the radiator with two Yate Loon 120mm fans in a pull-push config

 

[Conumdrum]

xbonez, so you had to mod the fans to get great temps. Yea, thats what I hear a lot. It isn't bad, it really isn't watercooling, it is as effective as a good air cooler. Thats about it.

Physics comes into play here. Radiators need x surface area and good airflow. To get really good temps if your running a HOT rig you need more cooling. The H50 is a great start into the world of watercooling. Too bad you can't do anything with it once you get the bug.

Again, a WC quality pump and res is over $70, rads start at $50, good CPU blocks start at $70.

 

[Conumdrum]

uhh, 1.192* no fish for us sir

Ambient temps and for how long did you run prime 95 small ffts? If thats how you ran it and your ambients are 75F+ you got a GREAT chip! Ohh what is your proggy to measure temps?

 

[paul mohr]

Let me explain this a better way.

I have another post where I reviewed a cheap cpu cooler from a large chain store. In this post I listed temperatures with different configurations and what the temperatures where with the stock cooler. The temps were like 30C idle and maybe 58C under a load for the stock cpu cooler. After installing the the new cpu cooler they dropped to 20C at idle and 35C under a load.

You might see that and think "WOW 35C under a load, I need one of those!" The thing to remember is not to look so hard at the actual temperature readings. My sensor could be wrong giving me low base readings or my particular CPU or set up might run really cool to begin with. Don't look at someones temps and think getting the same part they have will duplicate those exact temps. Every system and chip is different. And the ambient temp and voltage is going to play a big impact as well. 99 percent of the time you are not going to cool lower than ambient temperature with an air cooler. And if your computer requires slightly more voltage to run stable at a certain speed it is going to run warmer too.

The more important thing to look at is the temperature differences between stock and after the new cooler was installed. And the temperature spreads between idle and a full load. The impressive thing about the above numbers for my system isn't that I run at 35C under a load because that might not be the correct temperature. The thing to look at is that I had a 10 degree drop at idle and over 20 degrees under a full load. And the spread between idle and full load went down to about 15 degrees. And that was with a cheap mediocre cooler, there are better coolers for not much more money.

So the question to xbonez is what temperatures were you running before you spent $80 on the HD50 and added two high performance fans? And how is it installed in the case? And how much of a drop in temps did you really get when you switched?

Some will look at the temps of 50C under a load and not be too impressed. Unless it was running at 70 something before. Which I hope it wasn't considering you said your limit is 62C. If you were running at the limit of say 61-62 degrees before and the HD50 only netted you a 10C or so drop that isn't all that impressive for the money invested. You could get a 30-50 dollar air cooler that would do better.

No one is saying the small sealed water coolers don't work, because they do. Just the price is a little high for the performance they offer.

Paul

 

[pluke the 2]

Ran Prime around an hour - hour thirty Ambient Temp : Mid-Low 70’s. Real Temp, Core Temp, Speed Fan (which HAS DIFFERENT READINGS, so I turned it off), Also A Thermocouple to measure the different temp passing through radiator @ load & @ idle.

 

[Conumdrum]

Great, so now we know 'the rest of the story' looks like ya did real well.

 

[xbonez]

xbonez, so you had to mod the fans to get great temps. Yea, thats what I hear a lot. It isn't bad, it really isn't watercooling, it is as effective as a good air cooler. Thats about it.

yeah, but then each fan cost me only 3 bucks. But its true that the H50 is as effective as the high end air coolers (TRUE, V10 etc.), but not comparable to a full watercooling system

 

[paul mohr]

I wouldn't even call it as effective as a "high end" air cooler. From the reports I have seen it is about as effective as a mid grade air cooler for almost half the price.

Paul

 

[xbonez]

Let me explain this a better way.

So the question to xbonez is what temperatures were you running before you spent $80 on the HD50 and added two high performance fans? And how is it installed in the case? And how much of a drop in temps did you really get when you switched?

Paul

Sorry, I missed your question initially.
I built my system with the H50, so I never really used the stock cooler at all to be able to compare temperature drops. My ambient room temperature is generally 75 degrees farenheit (24 deg celcius) and my procesor idles at 35 deg.
Yate Loons, while being fans with really high CFMs, are in fact very cheap which is why I got them. I did not try using the radiator with the single default fan so again, I can't compare.
I currently have my radiator attached to the back of my case (Antec 1200) with two fans attached to it in pull-push. Its overall effect is intake in to case. I have another fan installed in the rear that intakes too. 3 fans in the front are all intake too. Exhaust is courtesy a 200mm fan mounted at the top of the case and a 120mm fan at the side of the case.