Regarding the installation of a Hydro Series on LGA775, I did this today, with an H105 (Asetek big brother to the H75) in fact.
I actually bought the H105 some time ago for this but never got around to installing it. I was going to buy the Asetek LGA775 install kit from eBay as well but realized it was going to cost me nearly $30 after shipping to Canada. I have several Asetek-based coolers though so I remembered that the older Asetek backplates already have provisions for LGA775. It's just that the top retainer/pressure plate don't have inserts for this socket (the inserts work two ways, for just about every socket but LGA775!). I'm not exactly sure why they did this as compatibility for 775 seems all too easy--it's just the mount points are slightly further inwards.
Anyway I pulled the backplate from an LGA1156 system I have using the old version and repositioned the little screw collars to the S775 position. On that system I just used the limited-travel backplate from the H105, without issue, keeping the rest of the hardware on that the same (top-plate, inserts, mounting screws).
Since the newer top plate has smaller holes, it's easier to drill those out than to try to figure out how to possibly mount the older top-plate/screws (with the inserts) to a 775. I'm not sure why the person above had troubles drilling out the plate--it's pretty thin/flimsy and drilling was fairly easy with titanium-coated bits. I just drilled out each side closer inwards. On the newer CoolIt-derived units (like the H60 2013, H80i, H100i) I'd imagine drilling the top-plate is just as easy but you'd need the generic backplate as mentioned. On the older CoolIt ones (original H100, etc.) I think they actually have enough provision already to mount on 775--these were actually marketed with 775 support IIRC.
Anyway I realized I had no idea what happened to the proper mount standoffs (the ones with M3 threads on either side) for my H105! So I had to devise something else. The standoffs for LGA2011 (Edit: These are the AMD standoffs, the black ones!) have one side threaded M3 or whatever (that can go into the backplate) and the other side is 6-32 imperial thread, which presumably threads into the AMD board when you're installing on that. So what I did was I just got some Nylock-insert 6-32 nuts from Home Depot (Nylock is not required but it was only 3 cents more per nut, and hey they can't back out :cool: ). These were used in place of the thumbscrews and easily hand tightened with an appropriate socket. In other words I basically used the AMD standoffs in reverse/upside down. Note: This would be totally unnecessary if I had the proper standoffs, I just couldn't find them--all that is normally required is a proper backplate and drilling out the top-plate.
I couldn't use the top-plate retaining ring (which is funny because I always forget to put that thing on Asetek coolers anyway, lol) as it would have hit capacitors around the socket. That plastic ring is by no means necessary though, it just makes install slightly easier. Got it on, tightened down the nuts I got from HD and done--installed perfectly fine. I have attached pictures of my handywork! :D:
The results were actually pretty great! I also have a Q9550 (in the form of a Xeon X3360), but it's OC'd to 3.6Ghz. This machine was previously struggling with an old low-profile heatpipe cooler from Thermalright (was pretty good on a C2D, not so good on an OC'd Quad). Previously I couldn't get the temp under 70C under LinX loading (in fact it went upwards of 80 under sustained load). I kept using the machine as it was but was never comfortable with the temps on it, under heavy load.
Now with the H105 under LinX load for 15-20 mins, it won't even get to 60C! I actually have a slight mixup of fans, I have one 2700 RPM SP120 (included) on the top but on the bottom I have an SP120 Quiet (1500 RPM). Basically SpeedFan is regulating them such that they basically have the same speed range despite one being voltage-controlled and the other PWM. If the temp gets above 60C on Core 0 (which is ~5 degrees higher than the rest--always has been on this CPU), the top will ramp to 100%, but it rarely ever does that as even under LinX it won't get there easily and if it does it's only a few seconds. But at ~1500 RPM on both fans, it's enough to keep the temp at around 57-58C on the high-reading core. In regular use the fans run much slower and the problem now? My PSU fan is louder than everything else, LOL!
Idle is a little different. Since C2Qs don't have the idle benefits of new processors and there's no dynamic voltage (not on my board anyway) I have to have it set to 1.25625V in BIOS to run stable at that clock (after vDroop it's actually just around 1.18V under load). So the idle temp is 34-39C depending on what core you look at since Core 0 always reads higher than the rest on my CPU. A bit higher than the idle temps for most other Hydro units on Nehalem and later CPUs but ok. The load temps are fantastic, I have to say. Definitely happy putting the H105 on that machine :D:
Anyway,I just don't understand why these coolers all dropped LGA775 support since all it needs, is a backplate and top-plate that has provisions "further in" than they have existing. It's not like they have to include all-new hardware here! But no matter, it's super easy to mod the newer top plates, and if you have an old Asetek-unit backplate or get a generic one, you'll be totally in business for LGA775 CPUs, provided you don't have any capacitor-clearance issues, which you probably won't on the Asetek ones if you don't use the installation ring. The newer CoolIt ones, I'm not sure as the square blocks may have more issues there but one would have to try with their particular unit and board to find out.