by J_Th4ng » Wed Nov 11, 2009 9:39 pm
Part III: Assembly – Putting it all togetherOkay, so once all of the components arrived, I was ready to start with the build. Being a typical PC geek, I thoroughly enjoy this part of the process, it’s one of the few activities where I am happy to take my time, and make sure that everything is put together exactly how I want it. I also tend to get very absorbed when doing this, as such I attempted to take photos to track the process, but it appears that I forgot for long periods of time during the build. Anyway, I’ll post what I have.
First, my rig as it was:
The first thing that I had to do was remove my old system from the case. Actually, this isn’t true. The first thing that I really had to do was make sure that the kids were all securely in bed, as the last thing I wanted were those extra pairs of ‘helping hands’ that invariably end up destroying fragile computer components.
Anyway, once they were safely locked up bedded down, I set to stripping down the old PC. An easy enough job to do, and one that always amazes me with the amount of dust and other crap that has accumulated inside the case over time.
Once I had removed all of the old components, I was left with an almost empty Chassis. I say almost empty because there was still amount of dirt and grime in there, as well as the watercooling unit, which I didn’t intend removing if at all possible.
This led me to my first unpleasant task. Due to the change in waterblock, I needed to drain the watercooling system. I also would not be able to close up the loop immediately afterwards, as it’s never advisable to have the CPU block piped in before it’s mounted on the CPU. What this meant was that I had to remove the old waterblock, drain the water into a bowl, and try and ensure that every lost drop came out so that I didn’t have any stray water droplets landing on my new motherboard as I mounted it into the case. This turned out to be fairly messy, but not overly problematic.
Once that was done, I set about seating the CPU into the motherboard, and getting the waterblock securely mounted onto the CPU. First step in the process was to attach the tubing to the CPU block (once again you don’t want to do this once the block is mounted on top of the CPU as it takes a bit of force to get the tubes securely attached to the mountings). Once completed, I attaché the backplate to the motherboard, so that the waterblock could be mounted. Once this was done, I dug out an old tube of ‘Arctic Silver 5’ thermal compound, and smeared a thin layer over the top of the CPU. Seating the CPU was a breeze, and the process was completed by placing the waterblock on top of the CPU, and screwing the retaining bracket into the backplate that I had attached earlier. The mounting bracket screws have spring tensioners, which is good as it ensures that you can’t over tighten the mounting bracket and thereby damage the CPU.
With the CPU and waterblock securely attached onto the motherboard, I was ready to mount the motherboard into the chassis. Fortunately, the alignment of the mounting risers was identical to my old motherboard, so none of these had to be repositioned. This time I also managed to remember to attach the motherboard backplate (that silvery thing with holes cut into it that covers all of the connections on the back of the case) before mounting the motherboard. My last build had me removing the damn motherboard afterwards to that I could put this piece in. Anyway, the board went in nice and easily, and it was then a simple job to attach the three RAM modules in their triple channel configuration. So far so good!
It was at this point in proceedings where the value of the HAF-932 chassis came to the fore. The case being a full tower design, there is plenty of space to work within it. Also, having a bottom mounted CPU means that there is space at the bottom of the motherboard to work in when attaching all of the peripherals which always seem to have their headers at the bottom of the board (USB, Firewire, SATA, IDE, Audio headers and all of the case switches).
Before I put all of the other bits in, I reconnected all of the piping for the watercooling unit. No errant drips had come out so far, but from this stage onwards I was going to be flipping the case around a bit to get everything else in, and didn’t want to take any chances.
Mounting hard drives into the HAF-932 is a doddle. Unlike most conventional cases, the drives are mounted sideways into the chassis, by means of clip and slide out trays that need no screws (thumb or otherwise). This sideways mounting has the added advantage of leaving the cable mounts for power and motherboard connection on the side of the case, behind the motherboard tray. The only slight complication that I had with this build was the fact that the Intel SSD is a 2.5” drive, not the usual 3.5” drive. Fortunately I had remembered to buy a special mounting bracket for the SSD, so it was easy to attach this to the drive, and then the drive was easily placed into one of the sliding HDD trays.
Obviously optical drives can’t be mounted sideways (unless you want the drive tray to come out on the side of the case). Even still these are easily mounted into the HAF with retaining clips – again no screws are required.
Once all of my drive bays were populated, I installed the graphics card and sound card. Originally this was the X-Fi platinum card and break out box, but this was subsequently replaced with the Xonar Essence ST.
Once all of these components were installed, but before mounting the PSU, I connected all of the cables. SATA cables to drives and motherboard. Case fans to the motherboard fan headers, the audio and USB front panel connectors, as well as USB backplates mounted to motherboard headers. Watercooling radiator fans connected to the watercooling main unit, and of course the power switch, power LED, reset switch and HDD activity LED.
Once all of the functional components were installed, I installed the PSU. As with most modular PSUs, there are a few cables that are hard wired to the unit, and a bunch of removable cables which are only mounted in the event that they are needed. Happily, all of the hard wired cables were needed for my installation, meaning no wasted cables left in the case. These were a 24pin motherboard power connector, an 8pin motherboard 12V connector, and two six/eight pin PIC-E connectors. That had my motherboard and graphics card sorted.
I needed two optional SATA cables, as I have six drives (2xSSD, 2xHDD, 2xOptical), and each of the cables have a choice of 2 or four connectors.
I also needed one optional Molex cable, to power my watercooling pump and for extra power required on the sound card. Once again, thanks to the design of the HAF, all of these cables were able to run behind the motherboard tray, well out of the way.
Once everything was properly connected and double checked, I closed up the back panel of the case. Or at least I tried to... There were so many cables running at the back, a lot of them on top of each other that the side panel couldn’t fit.
As with a lot of the world’s problems, this was quickly resolved with the help of a roll of duct tape, and some cable ties. Who cares what it looks like back there, that side of the case has no need for prettiness or airflow.
With all of the bits connected, and the case partially closed up, all that remained was to turn on the system and see how it worked. Well, not quite. There was one more rather crucial task remaining. At this point, my watercooling unit was missing the vital ingredient. Water. I unlatched the reservoir/pump unit, slid the unit forward out the front of the case, and opened the water entry plug. With the aid of a funnel, I poured the water back in to the unite, noting with satisfaction that it flowed fairly well into the pipes to fill the unit. Not completely, of course, but enough to be sure that there would be water in the system when I fired it up. What I now needed to do was to turn on the juice, and then top up the reservoir as the pump moved water through the system.
So I took a deep breath, and pressed the power switch. Nothing. I instantly panicked. What had I done wrong, why wasn’t it working? Had I cocked it up completely? Nope, it turns out that I hadn’t actually plugged it in. This super new PSU of mine doesn’t actually generate its own electricity, it still needs to be connected to the wall
After I corrected my error, I pressed the power button again. Light came on and fans started to spin. The inside of my case was bathed in a sickly green glow from the watercooling radiator fans. I started celebrating. Then I started panicking again. No water was circulating through the watercooling loop!! Effectively, I was running my new i7 CPU without any cooling.
Switched off the power, and started checking all of my connections. Everything was properly wired up. I knew that there was power going to the pump, because the radiator fans are powered from there, and they were spinning (and glowing green). Had I broken the pump somehow?
A short while later, I found the problem. Somehow, I’d managed to connect a molex connector to the watercooling pump the wrong way round. As anyone who’s actually read this far will know, molex are supposed to be shaped so that they can only fit one way round. I’d managed to force the connector in the wrong way around. Interestingly, this stopped the pump from working, but the radiator fans still worked just fine. Very odd.
Anyway, I turned the connection around, and fired it all up again. Success. The pump started pumping, and the air in the loop disappeared rapidly, replaced by green water. I breathed out again.
Bleeding the air out of a watercooling loop is actually quite a tricky business. The first thing that one needs to do is ensure that the reservoir is at the highest point in the loop. A basic knowledge of physics (or a tendency to fart in the bath) is enough to tell anyone that air likes to move upwards through water. If for example your radiator is at the top of the loop (as when it’s mounted in the top of your case), the air will travel there, and not come out. This will obviously mean that your radiator won’t work too well, as heat won’t be removed from the water. What one actually has to do is detach the radiator from the case, and put it at the bottom of the loop, shaking and spinning it around a bit to dislodge all of the air out of it. As you do this, the air will travel through to the reservoir, and be released (much like a fart bubble exiting the bath). Whilst doing this, the water level will drop, so you need to top up the reservoir while doing this, until no bubbles remain.
Anyway, so I did all of the above, and finished the hardware build by closing up the side of my case.
The entire process had taken me about 5 hours. It was now about 1:30am on a school night, so I packed up all of my stuff, moved the PC back into the study, and called it quits for the night.
[b]Part III: Assembly – Putting it all together[/b]
Okay, so once all of the components arrived, I was ready to start with the build. Being a typical PC geek, I thoroughly enjoy this part of the process, it’s one of the few activities where I am happy to take my time, and make sure that everything is put together exactly how I want it. I also tend to get very absorbed when doing this, as such I attempted to take photos to track the process, but it appears that I forgot for long periods of time during the build. Anyway, I’ll post what I have.
First, my rig as it was:
[img]http://i171.photobucket.com/albums/u314/Dr4g0nn3/IMG_8709.jpg[/img]
The first thing that I had to do was remove my old system from the case. Actually, this isn’t true. The first thing that I really had to do was make sure that the kids were all securely in bed, as the last thing I wanted were those extra pairs of ‘helping hands’ that invariably end up destroying fragile computer components.
Anyway, once they were safely locked up bedded down, I set to stripping down the old PC. An easy enough job to do, and one that always amazes me with the amount of dust and other crap that has accumulated inside the case over time.
Once I had removed all of the old components, I was left with an almost empty Chassis. I say almost empty because there was still amount of dirt and grime in there, as well as the watercooling unit, which I didn’t intend removing if at all possible.
This led me to my first unpleasant task. Due to the change in waterblock, I needed to drain the watercooling system. I also would not be able to close up the loop immediately afterwards, as it’s never advisable to have the CPU block piped in before it’s mounted on the CPU. What this meant was that I had to remove the old waterblock, drain the water into a bowl, and try and ensure that every lost drop came out so that I didn’t have any stray water droplets landing on my new motherboard as I mounted it into the case. This turned out to be fairly messy, but not overly problematic.
[img]http://i171.photobucket.com/albums/u314/Dr4g0nn3/IMG_8711.jpg[/img]
Once that was done, I set about seating the CPU into the motherboard, and getting the waterblock securely mounted onto the CPU. First step in the process was to attach the tubing to the CPU block (once again you don’t want to do this once the block is mounted on top of the CPU as it takes a bit of force to get the tubes securely attached to the mountings). Once completed, I attaché the backplate to the motherboard, so that the waterblock could be mounted. Once this was done, I dug out an old tube of ‘Arctic Silver 5’ thermal compound, and smeared a thin layer over the top of the CPU. Seating the CPU was a breeze, and the process was completed by placing the waterblock on top of the CPU, and screwing the retaining bracket into the backplate that I had attached earlier. The mounting bracket screws have spring tensioners, which is good as it ensures that you can’t over tighten the mounting bracket and thereby damage the CPU.
[img]http://i171.photobucket.com/albums/u314/Dr4g0nn3/IMG_8712.jpg[/img]
With the CPU and waterblock securely attached onto the motherboard, I was ready to mount the motherboard into the chassis. Fortunately, the alignment of the mounting risers was identical to my old motherboard, so none of these had to be repositioned. This time I also managed to remember to attach the motherboard backplate (that silvery thing with holes cut into it that covers all of the connections on the back of the case) before mounting the motherboard. My last build had me removing the damn motherboard afterwards to that I could put this piece in. Anyway, the board went in nice and easily, and it was then a simple job to attach the three RAM modules in their triple channel configuration. So far so good!
It was at this point in proceedings where the value of the HAF-932 chassis came to the fore. The case being a full tower design, there is plenty of space to work within it. Also, having a bottom mounted CPU means that there is space at the bottom of the motherboard to work in when attaching all of the peripherals which always seem to have their headers at the bottom of the board (USB, Firewire, SATA, IDE, Audio headers and all of the case switches).
Before I put all of the other bits in, I reconnected all of the piping for the watercooling unit. No errant drips had come out so far, but from this stage onwards I was going to be flipping the case around a bit to get everything else in, and didn’t want to take any chances.
[img]http://i171.photobucket.com/albums/u314/Dr4g0nn3/IMG_8715.jpg[/img]
Mounting hard drives into the HAF-932 is a doddle. Unlike most conventional cases, the drives are mounted sideways into the chassis, by means of clip and slide out trays that need no screws (thumb or otherwise). This sideways mounting has the added advantage of leaving the cable mounts for power and motherboard connection on the side of the case, behind the motherboard tray. The only slight complication that I had with this build was the fact that the Intel SSD is a 2.5” drive, not the usual 3.5” drive. Fortunately I had remembered to buy a special mounting bracket for the SSD, so it was easy to attach this to the drive, and then the drive was easily placed into one of the sliding HDD trays.
Obviously optical drives can’t be mounted sideways (unless you want the drive tray to come out on the side of the case). Even still these are easily mounted into the HAF with retaining clips – again no screws are required.
Once all of my drive bays were populated, I installed the graphics card and sound card. Originally this was the X-Fi platinum card and break out box, but this was subsequently replaced with the Xonar Essence ST.
Once all of these components were installed, but before mounting the PSU, I connected all of the cables. SATA cables to drives and motherboard. Case fans to the motherboard fan headers, the audio and USB front panel connectors, as well as USB backplates mounted to motherboard headers. Watercooling radiator fans connected to the watercooling main unit, and of course the power switch, power LED, reset switch and HDD activity LED.
Once all of the functional components were installed, I installed the PSU. As with most modular PSUs, there are a few cables that are hard wired to the unit, and a bunch of removable cables which are only mounted in the event that they are needed. Happily, all of the hard wired cables were needed for my installation, meaning no wasted cables left in the case. These were a 24pin motherboard power connector, an 8pin motherboard 12V connector, and two six/eight pin PIC-E connectors. That had my motherboard and graphics card sorted.
I needed two optional SATA cables, as I have six drives (2xSSD, 2xHDD, 2xOptical), and each of the cables have a choice of 2 or four connectors.
I also needed one optional Molex cable, to power my watercooling pump and for extra power required on the sound card. Once again, thanks to the design of the HAF, all of these cables were able to run behind the motherboard tray, well out of the way.
[img]http://i171.photobucket.com/albums/u314/Dr4g0nn3/IMG_8710.jpg[/img]
Once everything was properly connected and double checked, I closed up the back panel of the case. Or at least I tried to... There were so many cables running at the back, a lot of them on top of each other that the side panel couldn’t fit.
As with a lot of the world’s problems, this was quickly resolved with the help of a roll of duct tape, and some cable ties. Who cares what it looks like back there, that side of the case has no need for prettiness or airflow.
With all of the bits connected, and the case partially closed up, all that remained was to turn on the system and see how it worked. Well, not quite. There was one more rather crucial task remaining. At this point, my watercooling unit was missing the vital ingredient. Water. I unlatched the reservoir/pump unit, slid the unit forward out the front of the case, and opened the water entry plug. With the aid of a funnel, I poured the water back in to the unite, noting with satisfaction that it flowed fairly well into the pipes to fill the unit. Not completely, of course, but enough to be sure that there would be water in the system when I fired it up. What I now needed to do was to turn on the juice, and then top up the reservoir as the pump moved water through the system.
So I took a deep breath, and pressed the power switch. Nothing. I instantly panicked. What had I done wrong, why wasn’t it working? Had I cocked it up completely? Nope, it turns out that I hadn’t actually plugged it in. This super new PSU of mine doesn’t actually generate its own electricity, it still needs to be connected to the wall :sneaky:
After I corrected my error, I pressed the power button again. Light came on and fans started to spin. The inside of my case was bathed in a sickly green glow from the watercooling radiator fans. I started celebrating. Then I started panicking again. No water was circulating through the watercooling loop!! Effectively, I was running my new i7 CPU without any cooling.
Switched off the power, and started checking all of my connections. Everything was properly wired up. I knew that there was power going to the pump, because the radiator fans are powered from there, and they were spinning (and glowing green). Had I broken the pump somehow?
A short while later, I found the problem. Somehow, I’d managed to connect a molex connector to the watercooling pump the wrong way round. As anyone who’s actually read this far will know, molex are supposed to be shaped so that they can only fit one way round. I’d managed to force the connector in the wrong way around. Interestingly, this stopped the pump from working, but the radiator fans still worked just fine. Very odd.
Anyway, I turned the connection around, and fired it all up again. Success. The pump started pumping, and the air in the loop disappeared rapidly, replaced by green water. I breathed out again.
Bleeding the air out of a watercooling loop is actually quite a tricky business. The first thing that one needs to do is ensure that the reservoir is at the highest point in the loop. A basic knowledge of physics (or a tendency to fart in the bath) is enough to tell anyone that air likes to move upwards through water. If for example your radiator is at the top of the loop (as when it’s mounted in the top of your case), the air will travel there, and not come out. This will obviously mean that your radiator won’t work too well, as heat won’t be removed from the water. What one actually has to do is detach the radiator from the case, and put it at the bottom of the loop, shaking and spinning it around a bit to dislodge all of the air out of it. As you do this, the air will travel through to the reservoir, and be released (much like a fart bubble exiting the bath). Whilst doing this, the water level will drop, so you need to top up the reservoir while doing this, until no bubbles remain.
Anyway, so I did all of the above, and finished the hardware build by closing up the side of my case.
The entire process had taken me about 5 hours. It was now about 1:30am on a school night, so I packed up all of my stuff, moved the PC back into the study, and called it quits for the night.
[img]http://i171.photobucket.com/albums/u314/Dr4g0nn3/IMG_8718.jpg[/img]