How much pump is enough? How much is too much?

[pdf27]

Originally Posted by MaxxxRacer

Cathar, thanks for this amazingly great info. I am using it in my pump reccomendations article over at XS. Here is a link. Please check it out when you get a chance and let me know what you think. I sited and gave you some thanks in the thread for your info and linked to this article that you made.

From a quick look through it doesn't look like many of the people over there have got the concept that pump heat is bad for their systems - there were several people who were going straight for the biggest pump they could find
Thanks for trying to spread a little truth and light about though!

[Razor6]

Originally Posted by unregistered

Panworld has no MTBF #s, part of why Swiftech sells Laing pumps now

How does Laing know the MTBF of their pumps, they're just estimates right? Wouldn't they need to wait 6-7 years after development to specify the true MTBF?

[aaronspink]

Originally Posted by Razor6

How does Laing know the MTBF of their pumps, they're just estimates right? Wouldn't they need to wait 6-7 years after development to specify the true MTBF?

Anyone quoting MTBF is either using statistical modelling (get X number of pumps for y amount of time, crunch numbers) or accelerated testing. Don't know what you would do for accelerated testing for a pump, but for something like a cpu, you do heat cycling and radiation acceleration along with voltage.

Aaron Spink
speaking for myself inc.

[BillA]

the second story of the office building at the plant in Hungary is nothing but hundreds of pumps on test for years
generally extrapolation of one log cycle is the norm is regression testing (per my experience)

[Bugsmasher]

Just for info on the 10PI/15PI. According to Panworld they use a loose figure of 20,000+ for thier MTBF number.

[Bugsmasher]

Just for info on the 10PI/15PI. According to Panworld they use a loose figure of 20,000+ for thier MTBF number.

[BillA]

this is bullshit
a MTBF is the product of a statistical calc
loose figure my ass

I have FAR more experience with Panworld, and their pumps, than anyone here
and my experience with Laing is slowly building
there is no (reliability) comparison, they are worlds apart

[Razor6]

Is Panworld that bad or is Laing that good? With the data in this thread showing the 50Z as the ideal pump I would suspect that there are going to be alot more people looking to purchase it. Would you say that's a bad idea?

[Bugsmasher]

That was sort of the odd thing when I heard Panworlds numbers. According to those who have used them they seem to be exceptionally dependable pumps. As stated many times by people in this thread the MTBF numbers are pretty much conjectures and extrapolations. As such one company may claim X MTBF while another decides to play it safe and claim X-20% MTBF because when all is said and done....its not a proven number.

By reputation I believe they are both good durable brands depending of course on which pump you select. Panworld did indeed run into some problems with their original 600 series due to the pump not being designed for the type loads and stress that watercooling setups put on it. They have since the original run changed the bearings setup in their 600 series and its supposed to be rock solid now.

I think the 'larger' Panworld pumps are supposed to be exactly or near exactly as dependable as Iwaki pumps. In short they have quality written all over them.

The Laing lines seem to have had quite a bit of success as well. Once again a quality pump company.

Bottom line being I think when it comes to durability they are very close to being a wash. It really boils down to which you would prefer according to their numbers. Both the distributors and manufacturer reps I have talked to for Laing and Panworld have stated they are extremely interested in the watercooling market and as such plan on backing their products to the hilt.

[aaronspink]

Originally Posted by Bugsmasher

That was sort of the odd thing when I heard Panworlds numbers. According to those who have used them they seem to be exceptionally dependable pumps. As stated many times by people in this thread the MTBF numbers are pretty much conjectures and extrapolations. As such one company may claim X MTBF while another decides to play it safe and claim X-20% MTBF because when all is said and done....its not a proven number.

MTBF is not conjecture and extrapolation. There is a lot of hard science and testing involved in coming up with a true MTBF number. Accelerated testing, large lot testing, stress testing, spot testing, etc. It is generally not cheap (tying up upwards of hundreds or thousands of parts and related equipment for long periods of time). MTBF numbers also tend to be "downgraded" by a manufacturer to account to variations they couldn't tests for.

If someone is quoting an MTBF number they better damn well be able to demonstrate how they arrived at the number in a reasonable manner in a court of law OR they have really really bad legal advice.

Aaron Spink
speaking for myself inc.

[BillA]

yup, a technical description needs to be based on facts

at our request, Laing is putting together a consumer oriented summation of test programs documenting the MTBF value
PanWorld does not have these numbers because the testing was (apparently) never done

Swiftech has experience with thousands of these pumps and I can state unequivocally that there is no comparison between the 2 products' reliability
but hey, some use Danner pumps also

[JoeKamel]

Originally Posted by unregistered

Swiftech has experience with thousands of these pumps and I can state unequivocally that there is no comparison between the 2 products' reliability
but hey, some use Danner pumps also

When you were selling the MCP600s, were they purchased direct from panworld or were they sourced from a distributor that rebranded them?

[BillA]

direct, I am speaking from experience
they are a 'good' pump, but lack the reliability of the Laing where such is a consideration
but given my experience I would really double ck the app and pump suitability
- a closed loop WCing system is not an aquarium or pumping from a sump

[DDogg]

Originally Posted by unregistered

direct, I am speaking from experience
they are a 'good' pump, but lack the reliability of the Laing where such is a consideration
but given my experience I would really double ck the app and pump suitability
- a closed loop WCing system is not an aquarium or pumping from a sump

As I understand it, the AquaXtreme 50Z-DC12 is a 'version 2' of the MCP600 that was created after there were some problems with MCP600 V1 (please correct if I am mistaken).

That leads to the question - How long has the version 2 of the AquaXtreme 50Z been available, and is there any anecdotal experiences from users of the newer version pump failing for whatever reason?

[BillA]

jesus christ, I handle all warranty claims for pumps
I do suspect that Swiftech has sold many more thousands of these Rev.2 pumps than Cooltechnia (probably ever will)

re-read my posts

[DDogg]

Originally Posted by unregistered

jesus christ, I handle all warranty claims for pumps
I do suspect that Swiftech has sold many more thousands of these Rev.2 pumps than Cooltechnia (probably ever will)

er-read my posts

Sorry to aggravate you, Bill. I did not know that Swiftech had sold the version 2, pardon my ignorance and thanks for the reply.

Very Sincerely, Jesus christ

[BillA]

my apology
I too often presume that 'all' are familiar with a pump so often discussed
I sourced it, and with the agreement of PanWorld put the pump into an application for which it was not suited
lots of failures ensued, the pump head got redesigned; Ver.2 eliminated the pump head problems
remaining are some undefined (to me) electronic problems which seem to result in a few failures
as I said, a 'good' pump, just not on a par with Laing from a reliability aspect

[DDogg]

Your reply is much appreciated.

I'll bet those few lines represent a lot of hair pulling and a few choice words on your part. Must have been hell for a few months. I now understand why the MTBF is so critical for your company. Good luck with the new Laing product, I'm sure it will do well.

[BillA]

and many thousands of $s 'fixing' a problem not apparent initially,
lessons in prudence re pump selection

[dacooltech]

Originally Posted by DDogg

As I understand it, the AquaXtreme 50Z-DC12 is a 'version 2' of the MCP600 that was created after there were some problems with MCP600 V1 (please correct if I am mistaken).

That leads to the question - How long has the version 2 of the AquaXtreme 50Z been available, and is there any anecdotal experiences from users of the newer version pump failing for whatever reason?

DDogg the answer to your Q is no. We don't have any experiences with the newer version of the pump (AQX-50Z) failing for whatever reason. If customers do have bad experiences, with out a doubt they don't hesitate to share it with others over the forums.

[BillA]

somewhat different experience than ours with the same Rev.2,
granted we have a much larger number of pumps in service (several thousands)
while we have not been selling the MCP600 for some months, I still get 1 or 2 every month of the Rev.2
(8 in the box now since the last shipment to Chris)

the failures are perplexing; most are dead while others (~50%) develop an intermittent run/erratic start problem
this situation has existed from the start and despite MANY discussions remained unresolved
- perhaps they have improved since we stopped buying ??

[dacooltech]

Originally Posted by unregistered

jesus christ, I handle all warranty claims for pumps
I do suspect that Swiftech has sold many more thousands of these Rev.2 pumps than Cooltechnia (probably ever will)

re-read my posts

WTF Bill. what the hell do you think you're doing?

How the F do you know how many pumps Cooltechnica sells? What are you, our CPA?
I can tell you exactly how many pumps Swiftech sold, and how many blue pumps Pan World is left with. Is it what you want?

somewhat different experience than ours with the same Rev.2,
granted we have a much larger number of pumps in service (several thousands)
while we have not been selling the MCP600 for some months, I still get 1 or 2 every month of the Rev.2
(8 in the box now since the last shipment to Chris)

the failures are perplexing; most are dead while others (~50%) develop an intermittent run/erratic start problem
this situation has existed from the start and despite MANY discussions remained unresolved
- perhaps they have improved since we stopped buying ??

Cut the crap Bill will ya? Swiftech didn't stop buying the pump due to reliability concerns, but you lost the deal due to other reasons which -unlike you- my business ethics won't let me spell it out here.

If people have problems with a product as always they don't keep it to themselves and will post it on the forums.

and many thousands of $s 'fixing' a problem not apparent initially,
lessons in prudence re pump selection

Yea right, a company is going to spend thousands of $s 'fixing' a problem on a product and then they're going to stop buying it. who do you think you're fooling Bill? You should run for president.

Geez, what's this?
F this man. I won't let Cooltechnica offering a freakin' single Swiftech product again.

Can anyone remind Bill that ProForums is not SwiftechForums, and pimping or vendors bashing other vendors' products is not allowed.

[BillA]

hmm, was describing Swiftech's experience
no more, no less
and I believe I allowed as to the possibility of additional improvements after Swiftech stopped buying

and I did NOT bash YOUR product Bruce
again, I commented on Swiftech's ongoing experience with the MCP600

sorry if I jerked your chain Bruce, but we have an experience here also
the initial question was if the Rev.2 pumps had any failures
you said no, your experience
I said yes, my experience

[bobkoure]

Originally Posted by dacooltech

Yea right, a company is going to spend thousands of $s 'fixing' a problem on a product and then they're going to stop buying it.

This is not necessarily absurd behavior.
I'm pretty sure we've all gone through the stage of having had an old / high maintenance vehicle - and that at some point we stopped throwing money at it, even though we'd already spent lots.
Business-wise, I've been in this situation in the software world (particularly around purchased source code). Sometimes, even though you have the source and your development team has spent hundreds of hours fixing it, it's still a rational decision to decide that this code is simply a specification and re-coding from scratch is the long-term best choice (the stuff has to work and not just in "typical" situations).

I am not saying that Swiftech dropped these pumps for that reason (I have no idea what really happened), but that it's not inherently stupid business behavior.

[dacooltech]

Bob if we were talking about an old / high maintenance vehicle, or something else you may be right. Maybe I wasn't specific enough.
I was specifically talking about the MCP600 / AquaXtreme 50Z, and I have no idea whatsoever what exact problem Bill is talking about and where / how Swiftech spent thousands of $s ??? Guess, that's something only he can explain.

AQX-50Z is a perfectly good product, outperforming the muffled D4, DDC and other pumps in its category. Oh yea guess why Bill tries to paint a picture as if this pump is not well suited for our application. As shown in this particular thread, it's the best freakin' pump for WCing.

I mean come'n, if this pump is a failure as Bill tries to put it, are we crazy to put 2yrs warranty on it? I am not talking about the manufacturer's warranty either, i mean we as in Cooltechnica warrants the AQX-50Z-DC12 pump for 2yrs.

The only difference that I can think of is that Swiftech supplied the MCP600 with 1/4" ID reducers -since they offer 3/8"ID 1/2"OD tube only- , and we did not use el cheapo reducers whatsoever. I don't know if this has anything to do with whatever problem Bill is talking about. but hey if they had thousands of $$s to spare, maybe they should get the NPT threaded version of the pump, and spent some money on female 1/2"NPT by 1/2"OD quick-connects, instead of trying to save a few bucks by going with el cheapo 1/4"ID nylon reducers.

This's the version of the pump that we offer for 1/2"OD tube with quick-connects:


Swiftech never offered a version of this pump with quick-connects.