Rant mode on, Lytron fittings
 

The fittings on the Lytron heat exchangers are 37deg. JIC flare fittings. The fittings are designed for high pressure hydraulics, and they are leakproof up to over 100PSI in most settings. Unfortunately, the fittings are f*cking expensive.

2x 37degJIC male insert-> elbow 1" pipe thread cost me $37.66. They are machined steel, treated to be near inert to corrosion, and the smaller ones were more expensive. You CAN NOT get these in copper or aluminum since the fitting is designed for high pressure use only. Why they would put this on a radiator not rated for that pressure is beyond me, but I just wanted to warn you all who are thinking about those radiators. It would be better to cut the fittings off and solder on copper pipe, if your budget can't handle $18+ fittings.

Ohmigod am I p!ssed.

yeess, those heat exchangers are for oil

Eh, so what? I guess I don't have to worry about it busting under the strain from my pond pump, right? Who cares if it's for oil ... I bought it cos its 1337 and looks sexy in a thong. I don't think I could say the same thing about a heater core, so NOOGE!

ahh, I misunderstood
I thought you said pissed, but it was pleased

carry on

I'm pissed about the cost of the fittings, not my luvley radiator ... though in the end this might end up being a good thing. Since they fit 1" thread, I can hook them up to some bigazz PVC and they bring me closer to finishing. It just means I'll have to really bargain hunt for some of the other stuff.

The 4000 series are designed for deionized H2O, the 6000s for water, and the oil coolers for... airspirit has a 6320(?) so it was made for water. However, the oil coolers have much better performance/size because of the flat tubes with microchannels. The 4XXX and the 6XXX coolers have near-identical performance. Use the Lytron heat exchanger selector to check it out.

I know I'm biased as I'm selling one of these things,
but koslov's comments are quite correct
(obviously Lytorn says the same)

less eye candy,
better cooling,
MUCH better flow/smaller pump

I'm right now running that radiator in my system with a 350 GPM pump. I used to idle at around 43C. I've yet to see this over 37C. It looks ghetto because my Chieftec full tower is sitting inside the box I made with the lid leaning on the top of the case, and the "football" (1.5 ft x 4" PVC tube, T fitting at bottom in 1" PVC for use as a res/airtrap) is sitting on top of the whole mess.

This is not giving me near the restriction that I expected from the charts. I'm getting very good flow through the system, better than with my BIX, and I have two more 90deg turns in this than before. Something tells me that the flow chart on Lytron wasn't calculated for water ... but I could be wrong.

Oh, and did I mention my temps are down even though I still have large amounts of air bubbles in my system? Since there is more water mass in this, I'm going to give it a couple of hours before I do a little dance, but I think this is going to work nicely.

About the fittings, why didn't you go for cutting and welding/gooping on different fittings? That's what I plan on doing with mine, as it has weird 90's at the inlet and outlet that would make routing tubing difficult. So, I'll just cut 'em off, and goop on some larger barbs (ID of channels is 5/16).

OT: When calculating pressure drop using a specific pump, I saw something on the Lytron pages where you can plot the estimated pressure drop of a cooling loop on a pump's PQ graph and draw a straight line to the origin. Wherever the PQ curve and the line intersect, that gives the actual pressure drop. Is this accurate? Can someone explain how this works? Here's the link.

In the end I want this to be a professional job. I don't want to get into soldering fittings and that ... I'd rather use the fittings there to idiot-proof it, since I won't be doing any customization, and also because if I DO want to change this in the future, I want to have all of my options, not just the options left to me by my hacksaw.

I've peaked out at 42C, as opposed to 48-50C like before (this is after max CPU use for 15 minutes). I am happy with this. Now its time to assemble the thing right ... once I finish my parts acquisition ... and maybe I'll have all the bubbles out by then just in time to put more in!

If you're prepared to spend $38 on 2 fittings then you'll have to!! :p ...

Old temps:
Idle->43C, Peak->49C
New temps:
Idle->36C, Peak->43C

Six degree drop, and this is using less airflow than what I had through my BIX (I had 2x120mm@12V on the BIX, I have a weak desk fan on Low blowing across the Lytron, not nearly enough air for this radiator). Due to the construction of the radiator, I still have large quantities of air in my system (probably in the top of the radiator), and I'm still trying to drain it all out. I expect my temps to drop another degC or two once that happens (that has to be very inefficient, lowering my heat transfer surface area in the radiator).

Now that I have my proof of concept done (the radiator will work fine), I'm going to pick up the first desktop case next week and put my first machine (the current one that is on) in the loop correctly with the water tower air trap, using just the 350 GPH pump I have now. Hopefully I'll have the second on the loop with a larger pump by mid December, and the third on by early January. Once I have the cooler box internals put together, adding another machine to the loop will be a one hour chore at worst. I'm really itching to get started.

BTW, it's been a while since I've seen pics of a system as ghetto as the one I'm running now. Even though it works very well, it sure is uglier than sin.