but that is what marketing is all about...
if you can take a number.. like watts and then multiply by 3600 to give watt/hours then they have a bigger number to make their claims with...

unless they make a statment that you can actualy prove that is wrong...
then they can be sued... and if they can't be sued.. then they will make that statment..
hell... a marketer will tell you that the sky would be more blue if you bought their product if they could get away with it. :evilaugh:

Thing is, W/h is a rate of change of heat dissipation, while Wh is a total energy dissipated in an (unspecified) time. Hence, these are meaningless...

To plumb it in parallel or series is quite easy to decide. Just buy some cheap vinyl tubing and try it both ways, see which way its flowing faster through the CPU block. Give it the old bucket test.

Too bad about the Black Ice radiators. Are you sure there isnt a core or two to fit? BIX seem like kind of a rip off to me.

Hmmm...
They're changed their site. They used to state measurement conditions on every page that had ratings for their radiators.
The one for the BlackIcePro (120mm but thinner than the BIX)
still does *100 CFM, .28inH2O fans, 6 l/min flow rate. The missing ones are (IMHO) a simple oversight - stuff gets lost nearly every time a web site gets re-designed.
I've been using the BIPs because they seem to do a decent job if you're watercooling for low-noise.
Without getting into the inaccuracies of stating CPU diode temps) I just swapped a BIP out for a chevette core + criticool shroud in my "main" PC.
Reported temps (adjusting for ambient, which have gone from 20C to 24C -not that my wall thermometer's particularly accurate either) are 1C cooler - which means they are essentially the same.
Bear in mind that I'm "off the bottom of the chart" in terms of airflow and air pressure (single panaflo on rad undervolted to 5V).

That is NOT what marketing is about. Bill gives out real numbers, and gives us the test conditions. However, most people don't give a **** about the test conditions, they just see the 100W rating he gives to the Micro, and think, "Hey, that's good enough!" "Marketing," like that of HELabs, is the worst part about watercooling.

SnowRider
you say you have read up on the subject; your questions conceal the expected results
go here and read the article on the Thermochill rads

as Groth stated, undefined conditions = bs marketing

bobkoure
and the temp rise across the wet side ?
dry side ?
suggest you read that rad article also

the BIX is expensive...
but essential in where size was the major factor... only factor.
the other option that should be considering when you plumb everything in...
how many bends and length of tubing you will need...
I have seen certain recommended flow paths... but if the recomended flow paths cause to many bends... make your own flow path...
like AntiBling said.. the bucket test is truely the only way to test your flow rate.

Ok, I've already read through the "Online Database for Heatercores: Ben" thread which poses my next question. I can't seem to figure out how to actually buy or look at the heatercore I'm interested in. For example, I like the dimensions of this heatercore but I don't know how to do anything with that information. Help?? :shrug:

*edit* I really should've waited to post my questions about radiators until this question got answered which will definitely limit my choices on rads/heatercores

2-756 is the Fedco number.
9.5 and 5.375 (aka 5 3/8") are the dimensions.
1.625 is the thickness (i.e. 1 5/8").
0.625 (twice) is the inlet and outlet diameter (5/8").

I knew the dimension part of it (sorry, I guess I didn't clarify that) but didn't know what to do with these #'s: 306 | a304 | 2-756
Ok, the 2-756 is the Fedco #....... what's that:shrug:?
What are the other #'s for and how would I go about buying this heatercore if it was the one I wanted? Thanks

Umm... We know for SURE that more water will flow through series, unless we're talking outrageous restriction like an X-flow. Parallel offers lower inlet temperatures.

Erm... You sure about that? If you've got two similar blocks, the flow resistance from them in parallell will be less than that for them in series. While this cuts the flow rate through the paralell blocks, it increases the system flow rate as a whole. If you split the flow at/after the CPU block then surely the CPU benefits from the increased flow while the GPU/NB/Other gubbins (which is less temperature sensitive) has reduced cooling.
This is all theoretical (I have yet to build a water cooled system) but it appears to make sense to me.

i note that still omits the key to their overblown ratings... 40C differential over the air... :rolleyes:

Fedco are one of the firms that make heatercores, each of whom have their own numbering system....
this extremely useful site lets you find the others:
http://www.autopart.com/radiatorinfo/heatercoretech.htm

enter the fedco number, change manufacturer to fedco and click continue... it also gives you a drawing from the GDI number...

Ok thx. That'd help except for the fact it didn't seem to be working. I entered the Fedco # switched the manufacturer to Fedco and clicked continue but just got a page where it looked like information should've been but wasn't there. I'll keep trying but I think they might be having problems at that site maybe? Thx again.

anybody have a copy of the images like they use to have in leakycar.com?
I'd host them with the table hosted here

I didnt say that quite right. What I meant to say was try this:


CPU in series with GPU and NB

vs. this

CPU in series with GPU and NB in parallel


and see if the CPU flow is significantly slower when all are in series. If so, choose configuration 2. If not, leave them all in series. Hope I explained that ok. I tried putting some arrows and all in the post but when I posted it they wouldnt line up right.

Significantly slower? No... Try to work some numbers. The pressure drop is half of that of the NB or GPU block @ system flow. Assuming we're using Swiftech's, at 7.2" drop at 1GPM. You're using a Mag 3. Your CPU block is a Swiftech as well, at 8.6" drop. Work out the difference in CPU cooling. I doubt it'll be measurable. Your NB and GPU will both suffer from the reduced flow more than they'll gain from the lower inlet temperatures. You're correct, although stupid. :p J/K, obviously. The CPU gains are small, if measurable.

Really? 40C? Yow!
It's surprising to me that they bother to specify 6 l/min. I had the impression (from Bill Adams' Radiator Heat Dissipation Testing) that, above a certain fairly-small minimum, coolant flow didn't matter that much - particularly for the flat-tube radiators.
40C... so in a 68F room, the coolant would be 60C/140F? Of course, to the extent that you weren't trying to dissipate at their rated number, that temp should fall.
OK guys - you're totally right. There might be numbers, but they're not all the numbers and so qualify as "marketing-speak". Maybe they planned to remove these partial specs from all the web pages and the one I ran across was an oversight...
IMHO, folks choose these radiators (and the thermochill 120x1) because theyfit. I don't think a "real" set of performance curves would hurt sales at all - and would help with things like fan selection.
I've a couple of PCs "out there" with BIPs in 'em (coupled with Eheim 1046, Panaflo L1A at 5V, Swiftech "old style" center inlet blocks). For silencing a system, these work great (one of the above systems is in a recording studio - not on the same side of the glass as the mikes, but noise is still an issue).
Bob
PS: I'm a software guy who got into this because noise really bugs me...

i believe it's a standard differential for domestic radiators (i.e. the ones in your central heating system) where 40C differential isn't quite so off the scale....

Hey guys, what do you all think about the new DD Maze 4 chipset waterblock? I've been looking at it and I think it looks pretty nice. Plus it has a mounting option for the stupid chipset hooks on my mobo :mad: :D

*edit* BTW, I still can't get that heatercore cross reference site to work http://www.autopart.com/radiatorinfo/heatercoretech.htm. Is everyone having this problem or is it just me because I'm doing something wrong.

All I can say about the chipset block is: "It's about time." It has an incredibly low pressure drop, and cooling that is good enough for the chipset.

works ok for me (on a 2-342) - could it be that the core you were looking at isn't in their database?

Yeah it does work for you. I just tried that heater core (2-342) and it worked fine. Why is it that the only heater core I pick, out of that list of hundreds is the one they don't have :cry: :cry:. I thought I tried a different core before just too see if it worked and it didn't though hmmmmm...... that must've been before I knew what #'s to plug in.

*edit* Ok..... I just tried it with other core #'s and so far your's is the only one that actually brought up any results. What a helpful tool :mad: lol.

*another edit* I found some that work. It seems that most of the 2-700's don't work, grrrrrrr.

While we're on the topic of heater cores I might as well ask..... does anyone know of a heater core that is equal to or a little less than these measurements - 11"L x 5 3/4"W x 2"D? The 5 3/4" measurement is really cutting it close :eek: because the core would have to fit in my upper drive bays which I'm sure all of you know are also 5 3/4". I'm basically looking for one that can fit two 120mm fans, and has both barbs on one end. So let me know if you know of one that is close to these sizes. Thx