Welding on a soldered fixture would cause problems since it gets hot enough during welding to melt any solder and the whole thing would fall apart during final fabrication. The tubing joint is also included in the weld that attaches the riser to the tube and I don’t want any contamination either. 100% correct, luvbug
i could solder that together in 1 hit , but i am sure he knows what he is doing . i only asked lol .... nice bit of modding there m8 , i think by the look of it this will be a s**t hot project . keep up the good work
The whole thing could have been soldered but there were a few important reasons why it wasn't: 1. The joint between the risers and the pipe is an important surface as far as thermal conductivty is concerned. By welding it the whole assembly is rendered into a single piece of copper which eliminates any unnecessary interfaces. Consider also that the top of the riser is tangent to the cross section of the pipe and that soldered joints tend to be thin. By welding the joint I can get a larger area at the point at which the riser and pipe contact each other. Not that performance is all that important in something like this, but if I'm going to go to all this trouble I might as well go all the way. 2. In order to get the whole assembly into the PSU I have to include the pipe joint in the riser joint which would mean some possible nastiness. 3. Since this is a showpiece the consideration of aesthetics is also important. Keeping everything consistent looks better. Besides that, welded copper isn't something that you see all that often so it's cool to include.
after reading what you just said m8 i strongly aggree and now understand why u did it that way . and yeh i gotta admit i have never seen copper welded
A tiny update this time. I've been very busy between my two main projects, real work and sitting around, but I made a little progress with the case. Here's the finished waterblock being mounted up in the PSU. I got the female fittings soldered on and spent about an hour brushing the oxide coating off the thing. It fits very well but it seems that the hardware holding the MOSFETs and such is going to be impossible to reinstall without desoldering the whole lot and screwing them to the heatsink outside of the PSU. I'm not really up for surgery like that at this point so my plan of action is to use AS adhesive instead. I did a couple test fits and it seems like it will work very well especially since they just have to stuck in place and don't support any weight. I've also got the little insulating sheets sandwiched in there just to be on the safe side. This is the mold for the aluminum handle that I want to cast. I'm going to do a split mold sand cast so the mold needs a little work so that it releases properly but it shouldn't be that big a deal. I just have to try to get the mold made before the school's metal shop stops pouring for the year. So that's all for now. I'm still trying to figure out exactly how I want the frame rails designed and not making much progress. As soon as I decide on something that looks pretty good I either decide it won't work out or come up with some great new idea that necessitates a redesign. Hopefully I'll figure something out soon and start sticking some metal together.
Sorry that is what I meant about the mosfet carrying a charge. The guy that did this mod fried his main board as he forgot that the mosfets are live. Sorry again i missed the thread but I can't see which ones ive posted in due to the stupid forum icons.
New Update Time I revieved a big box of stainless steel stock from OnlineMetals a few weeks ago but hadn't time to do anything between final projects, real work, and graduation activities (I'm an official High School graudate now!). Now that I have time I have to get some work done so I can get this thing together. As an incentive I've decided that I'm not allowed to buy any computer parts until this is done, so BF2 will have to suffer through 512MB of RAM for another few weeks. I've changed the design around a bit so that the IO guard has room to allow for big plugs and the radiators now mount on a slant on the outside so that they can get some good airflow without reducing the security of the box with air screens. It also simplifies construction and makes the whole thing look a bit nicer. Here's what the enderlying frame looks like now. The body panels will follow this shape closely. Here's the main bottom piece. It has three main purposes, the first is to hold the entire case together, the second is to armor the lock mechanism, and the third is to be water reservoir. It's 1.5" square tube, about 19.5x14.5 inches so it holds quite a bit of water. All the frame pieces are TIG welded and then ground down smooth. And here's the frame upright. This was such a pain to make. Since my bandsaw is handheld all the angles have to be scribed with a protractor, cut wrong, and ground to fit right which they never really do. The real hard part is trying to keep the entire thing square in both uprightness and tube rotation while getting all the angles exactly right so the assembly comes out to the right final dimension. I even had a lot of trouble just getting the rectangular base square. I made it in two right angle sections, neither of which were square by the time I finished welding them no matter how much I tried to keep them in shape while welding. They both had to be heated up and tweaked square. When the two pieces got put together the assemply wouldn't sit flat so I had to clamp it down to the table and heat up all the joints with my rosebud torch. I'm pretty sure all this trouble is caused by the contraction of the welds as they cool. At least it's all of all acceptable squareness and size now. Since water distribution and the reservior is integrated into the frame I also had to make a final decision on the flow. The bottom piece and uprights will be connected (you can see the connecting holes in the picture) as a return/reservior while the top piece that runs the length of the case will be segregated as a distribution header.
Before you power the PSU, go over the thing with an ohmeter, and check the resistance between the tab(little metal bit on the mosfet) and the cooling block. A couple of the mosfets will have live 336V wall current on it (directly rectified 240 VAC). You mess up, you get a BIG spark, and blow the breaker, and probably your PS. Hey! this guy is WCing his PS too http://forums.bit-tech.net/showthread.php?t=82322&highlight=processor .
For some reason I got motivated to work again so I have a good update this time. Pictured above is the mounted lock pack with a mockup rod inserted to square everything up. The hole in the frame that the tube goes through has a liner tube welded in so that it's water tight. I have some 1/2" 304 stainless rod along with some nylon plain bearings coming from McMasters to build the actual locking mechanism. The case will unlock when the lock pack retracts the bolt and the rod is pulled outward. This will disengage the four (or six, I haven't yet decided how much work I want to do) door locking bolts along the sides of the case. That part of the job will get very interesting with all the little levers involved with making everything move. Next up on work accomplished is the main frame of the case getting finished. It's all done, I just have to pressure test it in the pool now. The thing weights in just over a feather at about 15 pounds. And here's my work from today. I made templates from aluminum sheet with the beverly shear and had the steel panels cut at the sheet metal shop. All six panels (front, back, two to shield the top and bottom of the locking mechanism on the bottom, and the two side doors) are 1/8" steel plate and weigh in at about twenty pounds for the whole lot. Somehow I'm going to have to figure out how to bend and cut the two side panels. Hopefully I'll be okay between my rosebud torch and plasma cutter. One more bit of good news, I'm getting my new (to me) Bridgeport milling machine on Saturday. It's a Series 1 9x42 table unit from the 70s with a digital readout. It's the final essential piece of equipment in my shop. Now I'll just have to try not dropping it (2400lb) when picking it up with the tow truck winch and setting it on the trailer to take home.
hmm, a 35 lb case.... guess you won't have to worry about that at lan parties! oh, and incidentally, congrats on your milling machine
Looking nice tho, maybe you'll want some detachable weels/casters if you ever plan on going to a lan party...
I've got some big news! First is that I got my Bridgeport and it's great. It's accurate all around and cuts anything like nothing. Second is that my friend Matt is restoring his old VW bug Baja in my garage so I'm going to be doing a lot more work on this case since I have to be out there to help him along. Here's picture number one of the update: Expect to see many more custom waterblocks coming out of my shop. Second picture: I got the front of the case all ground down and added two more holes for the front lock handle. The holes don't go all the way through, they are tubes that about 3/4 of the way through the frame tube and are plugged at the end. This will allow not only a cool looking handle but also to stop a potential burgular from twisting the handle which could damage the lock because of the large lever arm. Here the actual handle is getting set up for the inital tack welding: Then welded: Then sanded down: Add the the central locking bar that extends into the case in the middle, and use the emilling machine to make a little slot for the lock bolt: Top view of the handle installed and locked: And a bottom view: Next on the to do list is finishing the rest of the lock mechanism. I think it's going to turn out really great now that I've the mill to notch and slot the various bars and levers, but only time will tell. I've decided to postpone pressure testing until I some more parts mounted so I can install the bungs for the water barbs since each poses another leak risk. Hopefully I'll have another update tomorrow.
As we all know, the most important part of a mod is meticulous planning, right? Okay then, at least it's pretty important when designing a lock mechanism with a ton of parts that have to be machined and positioned precisely, so that's what I did tonight. Three and a half hours of figuring and tweaking in Solidworks and I have this: A little explanation: The rod poking out the bottom is the one you pull to unlock it. It directly pulls the two back bolts that fork off. The two front bolts are connected by a lever which acts to reverse the pulling direction to disengage them. Since the lever normally works on a circular path the bolt and main rod must allow for a changing horizontal distance. This is done by adding an intermediate bar and slotting the lever for the fulcrum which modifies the circular path to be flat on main rod side. The best part is that it all checks out in the simulation software, which means that if I happen to build it somewhat correctly it might work beautifully. All that's left is just taking all these technical drawings and, you know, turning them real (machinists take note, mad pun yo!). Seriously, what could be better than spending several days of my free time machining tiny stainless steel parts that will be encased by armor plating never to have anyone to lay eyes upon them again. Oh well. Who knows, maybe you'll get to see something physically done next update.
The point of the mechanism is for all four locking bolts to move at the same time. The bolts shoot through a plate with a hole in it that's attached to the inside of each of the main doors (two on each) but I haven't quite figured out how I want to make them since they have to be a particular shape to clear the frame tubes so I haven't given a picture of how it all comes together yet. Hopefully it'll all make sense when some more parts get finished.
