The Feser Company X-Changer 120.3 Pressure Drop Results

[Martinm210]

Now, someone send a PA120.3 to Martin

Danger Den might be helping me with loaner sample...I'll keep you posted..

[Xilikon]

That's great. I'm in favor of using the same testbed so we can have the same point of reference

[NaeKuh]

Now, someone send a PA120.3 to Martin

*sigh*

dont look at me. Im already trying to work out something with martin on him borrowing my RD-30.

Just worried he's gonna get in trouble with the wifey.

[Martinm210]

*sigh*

dont look at me. Im already trying to work out something with martin on him borrowing my RD-30.

Just worried he's gonna get in trouble with the wifey.

Yeah I need to setup my pump test rig in the garage and maybe run a scrap board through the table saw every now and then.

[dexaroni]

Man, its really great to see a radiator with some kind of branding on it and not looking OEM, it gives the WC market a more mainstream appeal.

Anyways, for that price, it better perform well compared to the PA 120.3.

[Sparky]

Assuming thermal dissipation is about equal with a thermochill (just assume for a minute):

Do you want to spend $130-ish for a thermochill or $180 for the feser rad that looks like it is in a bit nicer finish? Is the finish worth $50 to people?

I was surprised at one thing in the chart, I thought the HWlabs rads were more restrictive than the swiftech rads, not the other way around According to the chart I have about 1 PSI drop in my loop from the MCR320.

[nikhsub1]

Martin PM Marci, he may know what barbs were used in the PA... I would BET they are the plastic BSPT barbs that used to ship with them...

[Waterlogged]

Assuming thermal dissipation is about equal with a thermochill (just assume for a minute):

Do you want to spend $130-ish for a thermochill or $180 for the feser rad that looks like it is in a bit nicer finish? Is the finish worth $50 to people?

I was surprised at one thing in the chart, I thought the HWlabs rads were more restrictive than the swiftech rads, not the other way around According to the chart I have about 1 PSI drop in my loop from the MCR320.

Sparky Sparky Sparky.

That's a BIX (orange) in the chart to your MCR (apple). I'm sure if you looked at a graph with the MCR and GTS, the MCR would slaughter the GTS. BIP and MCR would probably be close though

[Sparky]

I thought it was all the hwlabs rads, not just the GTS that were really restrictive (aside from the GTX).

My bad.

[NaeKuh]

I thought it was all the hwlabs rads, not just the GTS that were really restrictive (aside from the GTX).

My bad.

nah, the BIP is actually pretty good.

The GTS is thin, i have one for tight spaces.

The PA is thick and long. <dont get sick thoughts people>

The GTX is thick and regular.


This radiator looks like a beefed up version of the GTX Gen2. I wonder why DD isnt making one themself. It should be cheaper cuz its stateside.

[Martinm210]

Sparky Sparky Sparky.

That's a BIX (orange) in the chart to your MCR (apple). I'm sure if you looked at a graph with the MCR and GTS, the MCR would slaughter the GTS. BIP and MCR would probably be close though

Yeah, I've asked in several emails to get some pressure drop data from HWlabs without luck. The ONLY data I've ever come across is from this french review:

http://translate.google.com/translat...language_tools

That dinos22 led me onto back in July last year. Exemplary testing for sure.

Unfortunately I've never found any sort of pressure drop test results for any of the other HWlabs radiators. CrazyJoe is working on testing his 480GTX though.

So YES the BIX 320 is the only radiator from HWlabs that I have ever found a PD curve for. If anyone knows of any other's I'd be very interested.

HWlabs mentioned they were updating their site December/January with more info, but I havn't seen anything yet.

This review seems to indicate that the GTX does in fact have a lower pressure drop, but flow rate tests can't be used for anything more than a "yeah it's less restrictive" type of comparison.

http://translate.google.com/translat...hl=en&ie=UTF-8

Anyhow, I'd like to compare with more but information is not available.

And radiators are all generally fairly low restriction compared to blocks, the graph is just sized to fit the data.

[Xerran]

SUBSCRIBED. I can't wait for the results, I'm especially eager to see how copper effects performance.

[Xilikon]

Martin PM Marci, he may know what barbs were used in the PA... I would BET they are the plastic BSPT barbs that used to ship with them...

I thought the plastic barbs was added by DD, not Thermochill ?? BillA is from the UK ?

[Martinm210]

FYI,
Here is swiftech's published curve, it's for the new G1/4 style which matches what I posted. The older 3/8" NPT barbed version did have less pressure drop, but the newer G1/4 also has a little better thermal performance.


Here is the 220 showing the thermal changes between the barbs, I still don't quite understand how smaller barbs do that, but it's good!

I really admire swiftech and thermochill for posting this sort of good useable data to help us ofigure things out.

[Marci]

Martin PM Marci, he may know what barbs were used in the PA... I would BET they are the plastic BSPT barbs that used to ship with them...

Not a clue I'm afraid - it was a long time ago. IIRC, I sent him our "older" G3/8" metal barbs (the ones that require PTFE) when I shipped the rads to him to test, but whether he used them or not I couldn't say (memory crap these days)... either way, he'd have accounted for the barbs' own pressure drop when he produced the figures... he always does...

thought the plastic barbs was added by DD, not Thermochill ?? BillA is from the UK ?

Correct - plastic barbs were never added by us... DD supplied them with the rads. In the UK, we sold them without barbs and let customer make the choice. Some bought TEFEN's blue 3/8" BSP fittings, some bought our metal 3/8" BSP fittings, some bought the O-ring fittings that are now available from us...

BillA = USA (Texas area to be precise when last I heard from him, but he was off to do something with irrigation systems or summat I think so maybe somewhere else now).

I really admire swiftech and thermochill for posting this sort of good useable data to help us ofigure things out.

That's cos BillA tested independently years ago (when he first reviewed HE Series radiators posted at overclockers.com), then went to work at Swiftech and produced their data, then went to work for Coolingworks (and produced their data), then, once he'd left both companies (thus assuring neutrality of reviews) was commissioned to produce our data once again for the PA series. All credit for those companies releasing that data goes to BillA for actually collecting that data for us to publish.

Now that Bill has retired from the scene, you won't find similar data that can be directly compared to any of his... it HAS to be tested on the same test rig under the same conditions for any of it to be directly comparable - remember that important detail. Swiftech data can't be compared to PA series data or HE series data or Coolingworks data... all 4x sets of data were produced on different testbeds and thus can't be directly compared to each other accurately. If thermal testing is to be done on Feser's new rad, whoever does it should, at the same time, test all others to produce comparable data. I suggest "acquiring" all triple rads for a bulk test session on a single testbed in a controlled environment.

The data used in the Cooling-masters article for the HWLabs series was also produced by BillA. David wrote the review etc, but the thermal and pressure drop data was all provided by BillA.

ALL c/w radiator data that is used as the foundation of most watercooling calcs tends to have been produced by BillA. I don't think anyone else has produced reproduceable accurate c/w data since (Bill's testrig was worth thousands and consisted of VERY good equipment, including a small wind tunnel and enviro chamber, and could produce accuracy to 4 decimal places).

And, yes, 5 deg C improvement is a hell of a lot to pull over the PA Series unless using what we'd consider to be "noisy" fans... in which case yes, it's easily doable. I'm still a firm believer in Cathar's work, and strongly believe any improvement over the PA Series at it's "optimal" airflow range will be in the very small fractions of a degree (ie: a difference of miniscule proportions difficult to measure). As Cathar stated, the brick wall has been hit when it comes to radiator performance, and I stand by his statement fully, until proven otherwise by testing as thoroughly (and appropriately - no 500cfm fans and 30lpm liq flow escapades) as that done by BillA.

[Kilyin]

It appears FrozenCPU has decided to stop stocking the Thermochill PA120.3 and will now only carry the Feser Xchangers.

I noticed they had removed the 120.3 from their web page so I sent them an email about it and they confirmed, no more PA120.3. Seems kind of self defeating on their part, and one less U.S. source available.

[ward0]

It appears FrozenCPU has decided to stop stocking the Thermochill PA120.3 and will now only carry the Feser Xchangers.

I noticed they had removed the 120.3 from their web page so I sent them an email about it and they confirmed, no more PA120.3. Seems kind of self defeating on their part, and one less U.S. source available.

The same thing is happening in Europe

[Baleful]