• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Modware Heat Guzzler Socket A Heatpipe HSF $4.99 + S&H

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Originally posted by: KF
Originally posted by: mikeford
... or maybe just shake it to see if it sloshes.
Click to expand...
Interesting idea.

I still have an extra heat pipe heatsink loose from a retail Opteron I recently got. It has four copper tubes. I shook it, and no audible sloshing, and it doesn't feel like it has anything sloshing in the slightest. As expected. Like I said before, you would not expect any noticeable liquid in these things. The water is a rarefied gas at below atmospheric pressure, plus a trace of wetness absorbed in the wicking surface.

I'm not about to cut a tube open to see if anything is visible. The fool that did this originally didn't find any liquid, and I see no reason it wouldn't be the same. If he had troubled to read the article he linked to, he might have had a clue.

Let's face it, the guy was a screwed up. Getting the heasink on wrong (against the step or the clip backwards) was one of the common screwups with socket A. It only tilts the heatsink slightly, and you usually can't see it (without a dental mirror.)
Click to expand...

Mine actually mounted "backwards". The heatsink is stepped, but for me it only went on with the step facing the other way. The flat part gets nowhere near the raised part of the socket. I also have a shim installed, which probably saved my core.

Even if mine's backwards, it fits fine that way, and as an added bonus it puts the fan on the other side. This way the flow goes CPU FAN -> Heatsink -> PSU FAN -> OUT. Nice and efficient 🙂
 
Originally posted by: sjwaste

Mine actually mounted "backwards". The heatsink is stepped, but for me it only went on with the step facing the other way. The flat part gets nowhere near the raised part of the socket. I also have a shim installed, which probably saved my core.

Even if mine's backwards, it fits fine that way, and as an added bonus it puts the fan on the other side. This way the flow goes CPU FAN -> Heatsink -> PSU FAN -> OUT. Nice and efficient 🙂
Click to expand...

Well, if it works it works. But what would happen is the V bend of the reversed clip that presses on the heatsink would be off to the side, past the core, and the heatsink would tilt slightly off the core, and the temperature would soar. It is only making good contact on one edge. Usually the clip can removed and reversed. Usually the fan is the only thing keeping it from siliding right out. Sometimes they fall out, or if you take off the fan, and the clip comes off, you don't know which way the clip should go. It is not symmetrical.

AMD's bumpers are firm enough to keep the heat sink from tilting so much that the edge of the core gets crushed. The way mine got crunched (just the edge) is the heatsink had a round base that missed the bumpers. After that I moved the bumpers (slicing them off with the adhesive using a razor blade.) The CPU worked OK anyway, and it remains in daily use by the person I gave it to. That was back when 800Mhz was just below the bleeding edge 1Gs. I had a couple more Athlons after that, and the HS clips only got tougher. You could only get them on with the aid of a screwdriver, and some people had the clip cut the nub of the socket right off. Then they went to triple slotted clips that grabbed all three nubs. My other cores never got crushed, and it wasn't because of my skill. I am terribly clumsy. The bumpers protected them. Otherwise, I no doubt would have have crunched them all. 🙂
 
Originally posted by: KF
Originally posted by: sjwaste

Mine actually mounted "backwards". The heatsink is stepped, but for me it only went on with the step facing the other way. The flat part gets nowhere near the raised part of the socket. I also have a shim installed, which probably saved my core.

Even if mine's backwards, it fits fine that way, and as an added bonus it puts the fan on the other side. This way the flow goes CPU FAN -> Heatsink -> PSU FAN -> OUT. Nice and efficient 🙂
Click to expand...

Well, if it works it works. But what would happen is the V bend of the reversed clip that presses on the heatsink would be off to the side, past the core, and the heatsink would tilt slightly off the core, and the temperature would soar. It is only making good contact on one edge. Usually the clip can removed and reversed. Usually the fan is the only thing keeping it from siliding right out. Sometimes they fall out, or if you take off the fan, and the clip comes off, you don't know which way the clip should go. It is not symmetrical.

AMD's bumpers are firm enough to keep the heat sink from tilting so much that the edge of the core gets crushed. The way mine got crunched (just the edge) is the heatsink had a round base that missed the bumpers. After that I moved the bumpers (slicing them off with the adhesive using a razor blade.) The CPU worked OK anyway, and it remains in daily use by the person I gave it to. That was back when 800Mhz was just below the bleeding edge 1Gs. I had a couple more Athlons after that, and the HS clips only got tougher. You could only get them on with the aid of a screwdriver, and some people had the clip cut the nub of the socket right off. Then they went to triple slotted clips that grabbed all three nubs. My other cores never got crushed, and it wasn't because of my skill. I am terribly clumsy. The bumpers protected them. Otherwise, I no doubt would have have crunched them all. 🙂
Click to expand...

I hear ya on that, thats why I have a shim installed. My core does have a chipped corner, likely from a long time ago when I installed the original HSF. I was thinking it looked backwards, but I couldn't get it to center properly the other way, so maybe its the way its designed.

 
Pictures I saw had the fan pointed toward the side of the motherboard. What sort of before and after temps are you getting?
 
Mine does point the fan towards the mobo (towards the AGP slot).

My before temps were awful. It would idle at like 40 and hit load at about 65 using a cooler I got from SVC, which was slightly better than the stock cooler. Now I'm oc'ed to 2.1 GHz, idles at 30-34 depending on ambient and doesnt break 45 under load. In passive mode w/ a 120mm case fan underneath, it got to 55c or so running prime95.

I really need to voltmod this board and see if I can hit 200FSB.
 
My northbridge I think its called chip was pretty hot, 114f on the heatsink, so I was thinking of sticking some old 486 fan on it first.
 
Just to give an update, so far so good on running this HSF passively. I didn't prime it for an extended period, but i did keep college football on (720p ATSC feed) for about 4 hrs yesterday and the temps never broke 50C.

Today I'll keep it on a 1080i feed during football and see how hot it gets. This is an HTPC so really it has to be stable in those situations, so I probably won't bother priming it for 24 hrs.

My setup is like this: 120MM Yate Loon mounted in front of the HSF --> No fan on heatsink --> out to PSU. I also have one more 120mm yate loon to blow air from the front of the case over the two tuner cards and passively cooled Geforce 6200.

I'm very happy with this heatsink. Definitely a steal at $12 to my door.
 
I dont think "fake" or not fake is an applicable distinction, they're just poor quality heatsinks designed to look better than they actually perform.

For a decent 'sink you will need much thicker aluminum if not copper, and especially in the former case, good bonding from the heatpipes to the fins, not just slipped on. If the fins are aluminum they also need be thicker than these are.

I doubt it's copper plated aluminum too, more likely just a polished anodized finish.

It's a $5 'sink though, what'd you expect? It's not like they're going to take a > 75% loss to make, distribute, sell. It's a bit amazing that they can even sell a poor 'sink this fancy for $5.
 
Originally posted by: mindless1
I dont think "fake" or not fake is an applicable distinction, they're just poor quality heatsinks designed to look better than they actually perform.

For a decent 'sink you will need much thicker aluminum if not copper, and especially in the former case, good bonding from the heatpipes to the fins, not just slipped on. If the fins are aluminum they also need be thicker than these are.

I doubt it's copper plated aluminum too, more likely just a polished anodized finish.

It's a $5 'sink though, what'd you expect? It's not like they're going to take a > 75% loss to make, distribute, sell. It's a bit amazing that they can even sell a poor 'sink this fancy for $5.
Click to expand...

Read my impressions. It's not a poor or even mediocre heatsink. It's excellent.

I'm running it with the included thermal paste and an aluminum shim on a Barton 2500 OC'ed to 190FSB (2090 MHz), passively cooled with a 120mm yate loon nearby in the case.

How much better than that do you want?
 
Originally posted by: sjwasteRead my impressions. It's not a poor or even mediocre heatsink. It's excellent.

I'm running it with the included thermal paste and an aluminum shim on a Barton 2500 OC'ed to 190FSB (2090 MHz), passively cooled with a 120mm yate loon nearby in the case.

How much better than that do you want?
Click to expand...


I read your impression and no offense intended, but you don't really have passive cooling if you put 2 x 120mm fans in a HTPC. You shouldn've have had the prior temps unless the prior heatsink were really bad or you don't have a hardware encoding tuner card, instead a high load like capturing straight to MPEG4.

Is it possible another heatsink is even worse? Yep.
 
Originally posted by: mindless1
Originally posted by: sjwasteRead my impressions. It's not a poor or even mediocre heatsink. It's excellent.

I'm running it with the included thermal paste and an aluminum shim on a Barton 2500 OC'ed to 190FSB (2090 MHz), passively cooled with a 120mm yate loon nearby in the case.

How much better than that do you want?
Click to expand...


I read your impression and no offense intended, but you don't really have passive cooling if you put 2 x 120mm fans in a HTPC. You shouldn've have had the prior temps unless the prior heatsink were really bad or you don't have a hardware encoding tuner card, instead a high load like capturing straight to MPEG4.

Is it possible another heatsink is even worse? Yep.
Click to expand...

My intent was never to set up a test bench here. I took a gamble on a $12 heatsink for my HTPC (and yes, my prior temps were so awful it was shutting down via the BIOS kill at 70C) and got lucky, I suppose.

As previously mentioned, I'm running HDTV through this. A Barton 2500 at stock speeds is at 100% usage decoding 1080i feeds, slightly lower usage OC'ed to 2.09 GHz. Believe me, its loaded up. Maybe not on the encoding end, but the nice thing about HD is that it comes encoded over the air for you. Decoding is another story and was the reason I even bothered to mildly (read: no extra voltage, just a straight FSB bump) OC this CPU.

The case is nearly inaudible (2x120mm yate loon fans) and there's nothing strapped to the CPU. I guess I might as well pull the PSU fan too, then? To me, this is passive cooling on the CPU. I'm pretty sure anyone running something like a Scythe Ninja is going to have a couple of case fans and still call it passive. The key is there's nothing with zero clearance on the heatsink pushing air through narrow fins to make a lot of noise.

Regardless, call it what you will, it is extremely quiet (virtually inaudible from a few feet away) and still keeps the temps below 55C under load. I'm not posting on SPCR, so I feel no need to get into semantics. My setup is well documented in this thread.

Please don't mislead others by suggesting it's poor quality unless you own it or have used it. This is a hot deals forum and I'm no overclocking zealot, and cant pretend to be. There are forums for those if you want to express an opinion that is pure conjecture. I'm only reporting back that I found this to be a hot deal and have given plenty of justification for the purpose of providing others with evidence that this is indeed a great value and not of "poor quality".

I come to hot deals for that sort of info, so I hope that I've given a bit of that back. I'm by no means offended, but I do want to emphasize my actual experience with the product.
 
Originally posted by: mindless1
I dont think "fake" or not fake is an applicable distinction, they're just poor quality heatsinks designed to look better than they actually perform.

For a decent 'sink you will need much thicker aluminum if not copper, and especially in the former case, good bonding from the heatpipes to the fins, not just slipped on. If the fins are aluminum they also need be thicker than these are.

I doubt it's copper plated aluminum too, more likely just a polished anodized finish.

It's a $5 'sink though, what'd you expect? It's not like they're going to take a > 75% loss to make, distribute, sell. It's a bit amazing that they can even sell a poor 'sink this fancy for $5.
Click to expand...

What is the basis for repeating what was unsubstantiated in the negative article linked to? The negative article was by someone misinformed, if not deluded, so what he had to say has nothing to recommend it.

Let's take the claims one by one.

The fins need to be thicker:
Do they? They are very thin and flimsy on the excellent Thermalright I have and the AMD retail HS I have. Why didn't Thermalright and AMD use thicker fins if thicker would be better? I can't believe they couldn't have made them at leat twice as thick if it would have helped. They'd still be very thin. They look as thin as they could make them and still be able to handle the HS with the supporting system they have (and need.) Maybe thin is the way it needs to be if the known good ones have thin fins. How thin is too thin?

The fins should be copper:
The fins seem to be aluminum on the Thermalright and AMD, so aluminum must work well enough, although I imagine copper would be better.

The fins aren't bonded well:
Aren't they? The fins on the Thermalright and AMD look like they are pressed on, not soldered or welded. But it could be they just did it so neatly you can't see any sign of it. Pressing does make good contact. House wiring uses nothing but pressure contact, and good electrical contact is more difficult than good thermal contact. Back during the space race, they invented wire-wrapping to get around unreliable solder joints. Industrial computer equipment used to be wire-wrapped. I still have a collection of wire-wrap sockets in my "junk" bin.

I'm not sure why the base was thought to be copper. Sjwaste said it looks like anodized aluminum.

Anodizing vs plating:
They are analogous except for the direction of the the ions in the process. The added material deposits at the anode (positive side) with anodizing, and it goes to the cathode (negative side) with plating. (Aluminum, so I understand, is not very amenable to plating, so it is anodized.) In either case, there is not enough material to affect thermal conduction. It is mainly for appearance and corrosion resistance. Plating something does not make it conduct heat better. It may help the pressure contact between two materials.

The price:
Maybe it's too loud?

Everything I have ever found out about the cost of production vs the selling price tells me that the differential is enormous for "high end" items. The fewer the people they market it to, the worse it is. Let's say an amplifier sells for $10,000, then the cost of manufacturing per item might be $1000. If you can't sell 'em at a premium price, you just dump 'em for what people will pay, and move on.

 
Originally posted by: KF What is the basis for repeating what was unsubstantiated in the negative article linked to?
Click to expand...

What is the basis for pretending otherwise?

Thermalright 'sinks are not that thin. "Kinda" similar thickness is not the same thickness.

Who said they should be copper? Not I, but they should in fact EITHER be copper, OR thicker.

No the fins aren't bonded well, how can it not be obvious? You are making arbitrary nonsense comparisons with things like household electrical wiring.

Anodizing vs plating is NOT the same, completely different process, different reason, and different result. It's mainly there to make the 'sink look like it's not bare aluminum even though bare aluminum conducts better than that does.

Plating does in fact help conduct to whatever degree the plating itself is a better conductor. Granted that may be a small % difference, too trivial to even consider in many cases but nevertheless, does have the potential to change conduction or radiation.

I am apathetic whether you agree, but the sad part is you pretend to know all towards a delusion about a 'sink that is mediocre at best.
 
Originally posted by: mindless1
Originally posted by: KF What is the basis for repeating what was unsubstantiated in the negative article linked to?
Click to expand...

What is the basis for pretending otherwise?

Thermalright 'sinks are not that thin. "Kinda" similar thickness is not the same thickness.

Who said they should be copper? Not I, but they should in fact EITHER be copper, OR thicker.

No the fins aren't bonded well, how can it not be obvious? You are making arbitrary nonsense comparisons with things like household electrical wiring.

Anodizing vs plating is NOT the same, completely different process, different reason, and different result. It's mainly there to make the 'sink look like it's not bare aluminum even though bare aluminum conducts better than that does.

Plating does in fact help conduct to whatever degree the plating itself is a better conductor. Granted that may be a small % difference, too trivial to even consider in many cases but nevertheless, does have the potential to change conduction or radiation.

I am apathetic whether you agree, but the sad part is you pretend to know all towards a delusion about a 'sink that is mediocre at best.
Click to expand...

How would thicker fins help? Thicker fins means you can stack less of them in a given space, resulting in less surface area. And that's beside the point, because the fins aren't that thin. I'd argue they need to be thinner on this particular heatsink.

I'll concede that thermal conductivity would be better if it were copper, but it conducts well enough for my passive configuration, so it must be pretty good.

The fins aren't bonded but fit tightly to the heatpipe with a fair amount of contact area, just like a lot of high priced heatsinks on the market.

Why don't you spend $12 and test this thing yourself? That way you can substantiate your claims of "mediocre" and poor quality" with some evidence. This is hot deals, not the zealot den. Leave the conjecture elsewhere.

The stupidity is that you're laying a heavy opinion without any obvious experience with the product. That or design and sell your own since you've asserted quite a bit about what works.
 
Be fair on price too, this is a closeout of Socket A stuff, the last chance not to sell it for scrap metal in a couple more months. It doesn't matter what it cost to make it, that ended years ago.

Personally if I could leave myself a little note a month ago it would say, don't spend a penny on the old socket A systems, buy a 939 PCIe. I didn't and I am in about $135 of improving that isn't feeling like great value. So $12 more into the blue spiraling water isn't part of the plan, unless it looks like it really does a decent job, and so far it sounds marginal.
 

>What is the basis for pretending otherwise?
That's it? Thanks for the clarification. You have no knowledge of this heatsink whatever? Is that right?

>Thermalright 'sinks are not that thin. "Kinda" similar thickness is not the same thickness.
So then what is the gauge of the Thermalright XP90 vs this one? Any idea, roughly? Is it about half? 80% 90% 99% Any idea what the optimum gauge might be? How is it determined what the ideal gauge shold be?

>Who said they should be copper? Not I, but they should in fact EITHER be copper, OR thicker.
How about neither, which was the point?

>No the fins aren't bonded well, how can it not be obvious? You are making arbitrary nonsense comparisons with things like household electrical wiring.
Since you have no knowledge of this heatsink whatever, how could it it be obvious to you? I don't get it.

It is not reasoning by analogy. My point was that pressure does make excellent contact and in fact experts prefer that technology over alternatives. I don't see how you missed it.

>Anodizing vs plating is NOT the same, completely different process, different reason, and different result. It's mainly there to make the 'sink look like it's not bare aluminum even though bare aluminum conducts better than that does.

>Plating does in fact help conduct to whatever degree the plating itself is a better conductor. Granted that may be a small % difference, too trivial to even consider in many cases but nevertheless, does have the potential to change conduction or radiation.

I'm not sure what the argument is on this one. As I said, the difference is which electrode the ions go to. Look it up. Functionally, coating metal is almost exclusively to reduce corrosion, which in turn improves appearance. So the function of anodizing and plating is the same. Of course it doesn't make the least bit of difference for heat because the anodized coating is so microscopic as to be translucent (which is how they get those peculiar metallic colors.) All metal oxidizes immediately after exposure to the air and so has a coating on it. Metals used for plating just stay looking nice because the oxide which forms is thin, even, translucent, and very impervious to passing oxygen. Nowadays any finish aluminum, even silver colored, is anodized to protect it, because exposed aluminum gets looking nasty fast. You can't have bare aluminum that is exposed to the air. It is either coated destructively with aluminum oxide by being in the air, or protectively by being anodized. In other words, an anodized aluminum surface has less coating on it than a "natural" one. And plated sufaces also have a coating, due to oxidation, which is what protects it, and which may well be thicker than an anodized one. The reason they don't plate aluminum, I believe, is that it does not easily accept plating, and it is just the opposite for many other metals, so they plate those. That's it.


>I am apathetic whether you agree, but the sad part is you pretend to know all towards a delusion about a 'sink that is mediocre at best.
Is it supposed to matter that you are apathetic? You did not substantiate anything. I just find technology interesting and like examining it. All I did was point out what was unsubstantiated. That tends to make people mad, I know, but I tried to tone it down.

You didn't mention what my supposed delusion was. I think I'm being realistic and accurate about the technology. I'm not making any claim whatsosever about this particular heat sink. This could be rotten a heatsink, but if it is, there is nothing at all in what you said that demonstrates it.

One site tested this heat sink and found it among the best (in 2004.) Another tester found that it almost didn't work, and "exposed" it (as a fraud), but discredited himself as a doofus, at least in my eyes, with the heatpipe foolishness. It is easier to believe the doofus botched something than to believe two tests could be legitimately so contrary.
 
I bought three of these. Finally have a heat sink that cost less than my processor. I get 35C on a Duron 1800 without much thermal grease just idling on the bios screen (managed to accidentally wipe the stuff that it comes with off on my shirt, leaving whatever stuck to the processor), whereas I was getting I guess around 70C with a thermaltake volcano 6 and loud as hell. Oh yes it is a fubar to get onto a ECS k7s5a mb. I will try again with arctic silver but I'm a little gun shy because on the first go it wouldn't post, plus I need to find a hard drive to use.

clearly the base is not metallic copper, you'd have to be a real rube to think so. Whether or not it has any fluid in the heat pipes is unknown, but they sure took pains to seal up the ends of the pipes. I could do without the blue leds on the fan, which is okay noise-wise, not any louder than a stock Barton 2500+ fan. obviously with the fan off it's dead quiet.
 
Wanted to give an update.

Running in my HTPC, Barton 2500+ OC'ed to 190 FSB on ECS N2U400-A. Removed 70mm fan. Running 2x120mm Yate Loon case fans.

It's been encoding XVID continuously in this "passive" (someone suggested case fans make it not passive, you decide). Hovering steady between 49 and 50C.

Incredible purchase at $12 and still available. If you have a Socket A system on a stock cooler and want to OC a bit, go for it.
 
yenno it says right on the manufacture's page that the base is not copper:
Heat Sink : Al Fin + Advanced AL Base + Triple heat pipes
 
I was looking for a decent heatpipe cooler for my Barton, since the Silentboost with the Coolermaster Aero7 blower is a bit too loud for my taste. I have a cheap, Logic generic case, with a front, back, side and top 80mm case fan setup. Light duty surfing sees the CPU @ around 46C, which is too high. Since this is an older system, I couldn't really justify the $60-$100 CAN cost of a high end heatpipe, so I looked around and found the $5 special at Directron. Since the shipping cost was more than the price of the cooler, and since the cooler itself was so cheap, I have ordered 5, the plan being to resell the extras on the local free classified website for a few bucks to recoup the cost. Hey, if it's a pile of junk, I'm out $5. It may be touch and go whether or not it can cool better than the Silentboost. As far as the reviews go, I take them all with a grain of salt, since the results vary so widely on ANY product, you have to assume some reviewer incompetence if a test goes really horribly wrong, when others find the product adequate to stellar in its performance. Let's give thanks for the crappy reviews. It probably resulted in sales going into the toilet: Presto, $4.99 heatpipe heatsink...

I think a nice new Coolermaster Centurion 5 case might be coming my way soon. Looks like a decent box for the money. I'm eager to see how everyone does with their heatsinks. Good luck. Even if it's a bust, it's just a bit of fun. If it works well, there's the satisfaction of getting a great deal. It's a no lose situation really. Anybody can shell out big bucks for the top of the line of anything and brag about the performance of their system.

Oh yeah, I'm running a Barton 2500 Mobile at 2450 MHz, at 1.8V. I have had it over 2600 MHz, but stability is not great. It will run reliably over 2500 MHz, but cooling is the big issue.


Don't forget, there's also the AMD 64 version for only $2 more if anyone is interested. I believe the Intel one is sold out.
 
Ok. I'm a believer. My temperatures have dropped between 8-10 degrees. over the Silentboost. I had to draw file and lap the base a LOT though, it was very uneven. Overall, I am impressed, a killer deal for $4.99!
 
On a duron 1.8 GHz not over clocked or anything, I am getting steady 19C 100% cpu. This is with the factory thermal grease and fan. Haven't tried with arctic silver yet (don't know where I put it). Without the fan one of the temperature sensors goes up to 70C after a day or so while the other one stays at 19-20C. This is on an ECS K7S5A MB case fan in front and one on p/s with the case open.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top