Laing pump failures

[Top Nurse]

Thought it would be nice to have a record of Laing pump failures since there sure seems to be quite a bit of buzz about this lately. If you have had a pump failure please tell us:

• What pump model you were using (model number).
• Whether it happened at initial use or some time after pump was installed.
• Any noticeable side effects before failure (sound, vibration, magic smoke, etc.)
• Any mods made to the pump.
• Approximate date of failure.
• What size tubing used.

Before anyone starts ragging on me about this I would like to state that the Laing pumps have a reliability issue. Reliability in a computer is valued, by most, in just about any computer system. If the pump fails to start you may get a seriously damaged computer like when a fan fails on a CPU heatsink. For those who doubt this I suggest that you do a simple Google search with the terms "Laing Pump Failures."

References: Pump of Choice?; MCP355 Failures; DDC/DDC 2 Failures

 

[silverJON]

last night

Laing DDC pump with Petra's top. It doesn't always start up and sometimes it randomly stops, but this time it stopped and the CPU temperature shot up, the PC died before I got the chance to kill it, and it melted my beloved AN8-32X's MOSFETs.

THANK GOODNESS my Opteron 170 LC9BE (2.97GHz Orthos stable) survived, although I don't know whether she'll ever overclock again.

The problem isn't my power supply - the pump doesn't always start up first time on an otherwise totally unloaded Corsair HX520, or any other of three or four power supplies I've used.

Can I go to anyone to get a motherboard replacement, or does this count as my fault? I've mauled the motherboard's heatsink system far too much for it to still be in warranty.

 

[Top Nurse]

Sorry to hear of your misfortune. I'd be pissed if my computer fried over a water pump that is hyped the way it is around these forums. However, I am planning on using a DDC soon as a test bed for trying out the AC Poweradjust unit. Supposedly it can be set so as to do a power boost at startup that alleviates the pump doesn't turn on problem. So it just stopped after it was already running well?

 

[mwin]

I had a Laing D4 that I ran 24/7 at 7V for about two years without a problem. I replaced it with a D5 that I run at the 2 setting. I've had it for about a year and a half, but it doesn't run anywhere near 24/7. I've got a DDC+ with Petra's top on it that I've run since October. It doesn't have to go 24/7, either. The only pump I've ever had die on my was a CSP-MAG from Danger Den (I don't know if that's a Laing pump, or not...). It only lasted a few months before it started rattling and then a few weeks later crapped out completely.

 

[Cathar]

Part of the problem with the popularity of a brand of pump, is that even a 1% failure rate will get more people putting their hands up saying that their pumps have failed than a less popular pump brand with a 10% failure rate.

According to various distributors I've spoken to and gotten the inside word on, the 6 month failure rate really is down around the 1% mark. The problem had been investigated, and there was a very high suspicion that failures were linked to the power supply, with poorly regulated "bumpy" power being the primary suspect.

The use of modded tops also increases the load on the pump's electronics in ways that few people outside of the pump manufacturers can really appreciate. If you use a pump with a modded top fitted, it really is a case of "Caveat Emptor".

Almost all motherboards made in the last 3 years support a thermal shutdown mechanism, and such has been around for 5 years from various motherboard manufacturers. As CPU's got hotter and put out more heat, heatsink failure became a real concern. Heatsink fans also do fail for various reasons. If your motherboard supports a thermal shutdown mechanism, you are extremely foolish if you allow your system to run without that feature enabled.

Back to Laing pumps though. (The CSP-MAG is NOT a Laing pump btw). Laing pumps have almost cornered the market in the USA, and in other countries, with their low noise and good pumping power characteristics. They are THE PC watercooling pump of choice for almost all systems sold.

The CSP-MAG's had shocking failure rates. Something like 50% within 12 months I've heard. Where were the witch hunts against the CSP-MAG's? Well, there weren't many witch-hunts, because the pumps were cheap, and everyone knows that you get what you pay for.

So, what are we trying to achieve here? It's a fact of human nature that people will complain loudly when things go wrong, but largely be silent when they're happy. If they don't notice the pump (because it's working), they have nothing to talk about. A pump fails and somone doesn't enable their thermal shutdown mechanism on their motherboard, and the angered individual feel all too justified to splash their experience across all boards, sometimes under different pseudonyms.

The truth lies with the distributors. They are the ones that really know what the story is.

If we're puzzling over whether it's wise to buy a pump with a 1% failure rate, it's also worthwhile considering that the pump can be replaced. A pump failure is not the worst thing that can happen. A radiator fan failure can also be catastrophic. Even worse in fact 'cos the water temperatures can rise causing pressure to build, and split a hose and spray boiling hot water all over the inside of the system. That may also cause pump failure too.

The moral of the story is use motherboard thermal protection shutdown ALWAYS, if you have it. If the pump fails, you can get it replaced because the support IS there.

If you're super paranoid about it, then just stick with what you know, and stop hunting witches.

 

[Ranker]

Thank you Cathar for supplying a voice of reason. Gabe, CEO of Swiftech, confirmed the same failure rate.

 

[Erasmus354]

This thread is prototypical TN flamebait. Cathar's response is all that is needed for this, please lock it admins.

 

[Top Nurse]

Almost all motherboards made in the last 3 years support a thermal shutdown mechanism, and such has been around for 5 years from various motherboard manufacturers. As CPU's got hotter and put out more heat, heatsink failure became a real concern. Heatsink fans also do fail for various reasons. If your motherboard supports a thermal shutdown mechanism, you are extremely foolish if you allow your system to run without that feature enabled.

If you're super paranoid about it, then just stick with what you know, and stop hunting witches.

Unfortunately my A8N-SLI Deluxe doesn't have that capability to shut down the computer and I'm not wanting to change it anytime soon. However, I do have a flow meter installed and integrated to my Aquaero and have it configured so that if the flow drops to less than 50 liters per hour it will shut down my computer. Others aren't as fortunate.

As far as the 1% figure goes I didn't know that Laing released these figures.

I am curious as to why you are here berating this thread as a witch hunt when another thread of similar content has been going on here for some time now without so much as a whisper from you. What I am interested in finding out is under which circumstances these pumps fail. The one item I added is the tubing size to see if using smaller tubing had some effect on this, which wasn't covered in that other thread.

This thread is prototypical TN flamebait. Cathar's response is all that is needed for this, please lock it admins.

Yes, once Cathar has spoken everything is right, huh? Nobody, even me, is right 100% of the time and neither is Cathar. How long did we go about saying that tubing size did make a huge difference in cooling capability? How long did everyone point to Cathar's research and tell us that we were stupid to use small tubing? Now Cathar says something else and everyone automatically jumps on the bandwagon. So who do we believe? I for one take anyone's "research" here in this forum and elsewhere with a grain of salt.

I believe what I see on my own computer and when it doesn't match up with common thought I point it out. I read all over the place that there shouldn't be a difference of more than 1 C in a proper cooling system. I get a solid 1.5 C difference between my radiator input and out put at idle. That increases to a little over 2 C under load. So I mention this and am told I'm lying even when others point out similar results. The problem here is that too much is based on theory and not enough on practical observation. Since most people don't have the capability to measure water temps they just assume that what has been written is correct.

 

[Erasmus354]

No TN, it doesn't matter if Cathar said what he posted, or someone else. It was an intelligent reply which addressed the irrational / flamebaiting concerns of your original post in a manner which should satisfy the point of the thread.

As for your second paragraph, links? I only ever base temperature differential numbers off of pure thermal capacity and flow rate calculations. It is a simple scientific, not theoretical, matter that at certain flowrates, for a given amount of heat input there will be a given temperature difference. Temperature difference all depends upon flowrate and heat input. The more heat you put in the higher the temperature difference. The slower you flow the higher the temperature difference. I suspect you had pretty low flow rate and pretty high heat input (a long loop) which would be the cause of your large temperature difference. There is also the slight problem of the likely +/- 1-2C error range on the temperature probes you used.

 

[Top Nurse]

No TN, it doesn't matter if Cathar said what he posted, or someone else. It was an intelligent reply which addressed the irrational / flamebaiting concerns of your original post in a manner which should satisfy the point of the thread.

As for your second paragraph, links? I only ever base temperature differential numbers off of pure thermal capacity and flow rate calculations. It is a simple scientific, not theoretical, matter that at certain flowrates, for a given amount of heat input there will be a given temperature difference. Temperature difference all depends upon flowrate and heat input. The more heat you put in the higher the temperature difference. The slower you flow the higher the temperature difference. I suspect you had pretty low flow rate and pretty high heat input (a long loop) which would be the cause of your large temperature difference. There is also the slight problem of the likely +/- 1-2C error range on the temperature probes you used.

Sorry, but I don't see any flame baiting here. I just asked a simple question and explained my rational for it. I also included some reference material just so you wouldn't think I was dreaming up something to flame Laing pumps.

I already thought about the variance range on my thermal probes last week. I switched them and they were within .1 C of my previous figures. Actually I was running about 160 LPH with my Aquastream pump and I only have four (4) water blocks running. A CPU, NB, GPU, and a VR block. Nothing out of the ordinary.

Are these the pics you want or did you have something else in mind?

 

[Digger]

Dear Mods , Please don't lock this thread as I for one am curious about how many [H]ers have experienced DDC failures .

 

[Erasmus354]

160 LPH = 44.444 grams of water per second.

It takes about 4.186 joules to heat 1 gram of water by 1 C (or take away that much to cool it).

To have a true temperature drop of 2C at that flow rate you need to be taking out 372W (watts = joules per second) of energy from the water.

Personally, that seems like a bit much for an A64, Geforce 6 series, nForce4 NB and some Vregs.

No theory here TN, simply incontrovertible physics....

 

[Ranker]

But according to TN, ones own "results" are above any sort of bias that would manifest itself in stuff like Math and Physics, such as the techniques that you and Cathar utilize.

 

[Top Nurse]

Well here is another pic of the power conversion part of the Aquaero. Please keep in mind that this is at idle and just surfing the web. I have the voltage to the CPU maxed out on the MB bios.

Personally I would say that this power measurement (per AC manual) may be a bit off as they have no control over what kind of additives people put in nor the percentages. I assume that they set it at the recommended AC Fluid to distilled ratio.

 

[R1ckCa1n]

But according to TN, ones own "results" are above any sort of bias that would manifest itself in stuff like Math and Physics, such as the techniques that you and Cathar utilize.

Why don't you give us your results, Ranker?

As for me, six months and running strong with a DDC1 w/ alphacool top, dead silent and 6mm tubing.

 

[Top Nurse]

Why don't you give us your results, Ranker?

As for me, six months and running strong with a DDC1 w/ alphacool top, dead silent and 6mm tubing.

I don't think he will ever make public anything about his rig because it is a figment of his imagination. In fact recently he really got his panties in a bunch when I asked him about it. How long ago was it that he was going to post pics so we could STFU? Seems like it has been almost two (2) months now and the boy has been doing back flips and flip flops ever since.

So do you think the DDC on 6mm tubing has anything to do with it not crapping out or would tubing size just be a lark?

 

[Arcygenical]

~30 D5 pumps in service. 1 returned due to failure after 8 months.
~12 DDC pumps in service. 0 returned.
4 DDC+ pumps in service + tops. 0 returned.

1 failure out of 46 Laing products used. For this failure...

-D5, swiftech variety, MCP655.
-8 months.
-Not to my knowledge, according to the customer it just never started working one day... We tried to revive it by pulsing the voltage, shaking it, hitting it, to no avail.
-The original seal that I included under the pump was in place, so I know it was not removed or modified in any way.
-February of this year.
-7/16 ID

 

[R1ckCa1n]

Ahhhhh..... Data! Thanks Arcygenical

I really think the only pump to worry about is the CSP-MAG. Boy did they get a block eye.......

 

[Top Nurse]

~30 D5 pumps in service. 1 returned due to failure after 8 months.
~12 DDC pumps in service. 0 returned.
4 DDC+ pumps in service + tops. 0 returned.

1 failure out of 46 Laing products used.

Now we are getting somewhere! My feeling is that if one person posts a failure then there re probably a lot more out there that never report them. Now we have a much bigger statistical model to work with. I'm thinking that a DDC might be more in the line for me in FF for my GPU and CPU/NB circuits.

 

[Cathar]

Unfortunately my A8N-SLI Deluxe doesn't have that capability to shut down the computer and I'm not wanting to change it anytime soon. However, I do have a flow meter installed and integrated to my Aquaero and have it configured so that if the flow drops to less than 50 liters per hour it will shut down my computer. Others aren't as fortunate.

Then that's an adequate step to take. That sounds like a decent solution in the face of a motherboard without a thermal shutdown.

I'm sure that there are software thermal shutdown programs about too. Even if the pump stops or the radiator fan fails, you've got a few minutes before things get nasty, and a software thermal shutdown package should have sufficient time to do its thing.

As far as the 1% figure goes I didn't know that Laing released these figures.

Laing themselves don't, and I never said that they did. Laing don't direct sell, but rather operate through distributors and reseller. I said distributors and resellers reported the 1% rate.. Since they're the ones that have to handle warranty returns, when they tell me what their return rate is, it'd be a pretty big logical leap to assume that such was not indicativee of the actual return rate when we're talking about percentage failures of many thousands of units shipped.

I am curious as to why you are here berating this thread as a witch hunt when another thread of similar content has been going on here for some time now without so much as a whisper from you. What I am interested in finding out is under which circumstances these pumps fail. The one item I added is the tubing size to see if using smaller tubing had some effect on this, which wasn't covered in that other thread.

The difference there is that the author of that thread didn't start with a stated presumption based upon anything other than his direct experience and data gathered (although he didn't disclose his sources).

Such polls can be worthwhile, but I question the legitimacy of the results of asking a forum, where it has been demonstrated in the past that the majority of people who respond to such questionnaires are those who feel compelled to do so, typically with a prior failure as motivation.

If the poll was somehow compulsory, rather than voluntary, then that'd be a better way to gather the data, and that's where I was coming from.

How long did everyone point to Cathar's research and tell us that we were stupid to use small tubing? Now Cathar says something else and everyone automatically jumps on the bandwagon. So who do we believe? I for one take anyone's "research" here in this forum and elsewhere with a grain of salt.

Things change. The research was valid at the time it was conducted, but as things change, it became out of date and invalid, and so I updated it. Is there a problem with this? The only problem I can see is that some people like to live in the past, rather than accept that everything is dynamic.

Since most people don't have the capability to measure water temps they just assume that what has been written is correct.

Most people, even when they do use thermometers, don't take the time to calibrate them against each other, or against a common reference. Most just use the CPU case temperatures and ignore ambient conditions.

The engineering theory behind what is presented has been researched heavily over the last 100 years, and things like nuclear reactors get built on the back of such research trusting that the calculations & extrapolations hold true, and they do. The problem often comes when people armed with cheap uncalibrated thermistors with 2C accuracy, but a 0.1C resolution, make observations without first accounting for the inaccuracy in the equipment they are using. I personally consider most people's temperatures that they report on-line to be only correct within + or - 3C. In order to derive a more accurate picture of what is occurring, we often need to take it to the engineering theory level, and extrapolate from results that were obtained using correct procedure and equipment.

I make no apologies for trying to clarify the picture of what is occurring when everyone's running around with dicey readings clouding the story. Get 20 people online with their results, and view it with a personal bias, and you can always find some results that will agree with your bias, allowing you to dismiss the others.

I'm not interested in bias. I'm interested in deriving the facts, through a combination of research and applied theory. If people want to continue their bias based fights, that's their concern. I speak to those willing to listen, and to those willing to understand that while what makes up a "perfect system" is not written in stone, the underlying engineering principles are.

 

[Ranker]

Such polls can be worthwhile, but I question the legitimacy of the results of asking a forum, where it has been demonstrated in the past that the majority of people who respond to such questionnaires are those who feel compelled to do so, typically with a prior failure as motivation.

If the poll was somehow compulsory, rather than voluntary, then that'd be a better way to gather the data, and that's where I was coming from.

I agree. I mentioned this in the other thread over on XS. The only people that are going to post on that sort of thread are either those who experienced failures and want the world to know about it and the forum regulars. It'll skew the results and even then the sample size wouldn't be large enough to make any sort of statistical analysis.

The engineering theory behind what is presented has been researched heavily over the last 100 years, and things like nuclear reactors get built on the back of such research trusting that the calculations & extrapolations hold true, and they do. The problem often comes when people armed with cheap uncalibrated thermistors with 2C accuracy, but a 0.1C resolution, make observations without first accounting for the inaccuracy in the equipment they are using. I personally consider most people's temperatures that they report on-line to be only correct within + or - 3C. In order to derive a more accurate picture of what is occurring, we often need to take it to the engineering theory level, and extrapolate from results that were obtained using correct procedure and equipment.

Yet there will always be a subset of people who claim that their findings supercede any sort of mathematical or physical law without even recognizing that their equipment is incapable or unsuitable for making such an analysis. Even worse is their stubbornness when confronted about such a possibility.

I'm not interested in bias. I'm interested in deriving the facts, through a combination of research and applied theory. If people want to continue their bias based fights, that's their concern. I speak to those willing to listen, and to those willing to understand that while what makes up a "perfect system" is not written in stone, the underlying engineering principles are.

And it's upon this principle alone that the majority of the watercooling community regards you highly. Those that attack you or your findings usually do so because your findings/observations reflect poorly on whatever issue it is that they cling to. However, what I find worse is when people pick and choose what it is they support from all of your findings: only choosing what suits their agenda or beliefs. It's amusing since your methodology remains the same through all of your work.

 

[Top Nurse]

And it's upon this principle alone that the majority of the watercooling community regards you highly. Those that attack you or your findings usually do so because your findings/observations reflect poorly on whatever issue it is that they cling to. However, what I find worse is when people pick and choose what it is they support from all of your findings: only choosing what suits their agenda or beliefs. It's amusing since your methodology remains the same through all of your work.

What I find amusing is that brown spot on your nose.

 

[Ranker]

What I find amusing is that brown spot on your nose.

When you have nothing to add, you resort to personal flames. Should I say that same about you when you agree with others? Your tactics are getting old TN.

 

[Top Nurse]

Should I say that same about you when you agree with others?

I rarely agree with anyone.

 

[Arcygenical]

I rarely agree with anyone.

Agreed

Updating my post to answer some of the bullet points in the op

 

[joecop120]

If you want to know, I applied for a job at the company that did the reliability testing for Laing DDC pumps. I don't know the results of their tests, but the company is Cooligy, based in CA. (I got a job at a different pump company)