Zds Project Log

As I needed a break from stretching acrylic, I laid my eyes on the rest of the case - what parts have been planned with enough to work on. And yes, I could implement DVD drive covers, had I exact plan about the top cover decoration. And as artwork and lighting tie together, I needed to prototype the lighting to see how it looks.

I had planned to control HDD and DVD activity lighting with just basic components like capacitors, resistors and diodes, but I felt them a bit hard to prototype. as I do not have too much different kind of electrical components lying around, so I should have had to buy dozen of components just to see what I could do with them.

Thus I went and purchased PIC microcontroller, construction kit for PIC programmer and miscellanous supplemental tools:

Here is close-up of the PIC chip itself:

Now to soldering the programmer kit.. :-).

Here is one of my ideas about how to cover the DVD drives:

The rectangles are outlines of the drives. The idea is that the areas marked with darker red are windows showing innards of the drive and brighter red ones are just lit acrylic. My plan is to light the sections based on drive activity; the more activity, the more sections are lit.

I have placed the windows so that only thing visible is the rotating disk

As the previous reservoir image was only a technical drawing, I drew a quick scetch of how the reservoir might actually look like:

The idea is that the main body is of clear plastic, and then hose connections and upper and lower parts are covered with black plastic that looks partly organic and is covered with shiny lacquer.

Now that I’m taking a break from being nasty to innocent acrylic, here are some blueprints about my future plans.

First chart of overall placement of the components, from the front:

And then reservoir in particular; the version that contains peltiers:

The latter one is a cross-section from up looking down. Bright red marks the hot side of the peltiers, deep blue the cold side.

I started by cutting the other end. This time I was wiser and drilled holes in the ends of the cuts to make it less prone to ripping:

As you can see, from proper angle yesterdays achievements do not look so bad:

I tried new trick and used iron wire to force the acrylic to the inmost point:

And found out that it was a bad idea; wire left just too heavy marks:

To lessen the need of muscle force in stretching the acrylic down I clamped 12.5 kilos worth clay with spring to the border of the acrylic:

After many hours and lots of sweating, cursing, iterating and trying out new tools to punish the acrylic with, here I am:

And without the mold:

It still has lots of irregularities and the concave parts are not yet as deep as I had hoped for, but I am not desperate yet. It gets better every hour.

However, I am already considering doing a second one with different methodology and taking the one that looks best.. This one screams “home-made” right now with big neon sign, and that’s not what I’m aiming at. Sigh.

Enough planning and experimenting. Going for the real thing:

I used clamps to add weight to the borders so I didn’t have to bend so much by hand. The back clamp is to keep the thing in balance:

And then pliers to do the stretching:

Added two more screws and washers to keep middle part of the acrylic flat:

First round of the front part done. Still lot to do, it’s bumpy as anything, but correct shape is rougly there:

Then I had to cut the acrylic to be able to do the ends from the same piece:

And ended up finding that acrylic rips very easily when in plastic form:

Oh well. I knew that is was propably too ambitious to try to do it from one piece. Back to plan B, using three pieces; one for the middle and two for the ends:

And now back to real modding. No more just archives and slideware :-).

I bought two frames to see if they could help in stretching:

And then went on and used PVC film to find out few different ways not to do the stretching:

Finally had time to edit these images for web. These show the upper shroud mold in the finished condition:

As you can see, the surface is a bit rough, but I do not think it matters much. The acrylic is quite stiff stuff, even when heated up, it will not replicate little ripples here and there. What needs to be correct is the shape, and especially the protruding parts, since they will stand most of the stress.

Here is a close-up showing the three layers of materials in the mold:

As one of the disks in our home server broke before x-mas, I had to shut down and open up the home server to swap the drive. Now that I had it open anyway, I thought it would make sense to upgrade the hardware at the same time. Just swap in new MB, CPU, memory and the replacement HDD and start it up again… Bzzzt. Boy, I was wrong. But well, more on that story later on, now I have got it up’ n’running again.

Here is the server in the old setup:

The old machine was AMD K6 450MHz class veteran, and even while it had almost 400M of memory, it’s IO bandwidth just couldn’t keep up with the four 7200RPM IDE drives in two physical (four logical) controllers. So I gathered stuff I had lying around; obsolete Duron and Athlon XP 2000+ processors given by friends, DIMM module that didn’t work reliably at DDR400 anymore and Chaintech Socket-A MB from online auctioning servive.

To swap the MB I had to rip out almost all the guts, and as I didn’t want to do it multiple times, I first built the new system in a classic ghetto way:

After lot of hassle, tuning with Ubuntu Live CD (very nice tool also for configuring hard drives..) and trial with another Socket-A MB from another friend I ended up packing all the data to the two Seagate drives, 120G and 160G in size.

It was time for these old workhorses to retire:

They all all Deskstars, 20G, 30G and 40G in size. The bottommost (40G) is the only one that’s broken for sure; the two others are just suspicious and slow enough so that I decided the additional 50G of disk space just isn’t worth all the troubleshooting. 24/7 server use is tough job for such old beasts, after all.

With them I could also swap out this nice piece of hardware:

It’s four device Ultra ATA 133 controller I had used to have dedicated lane for each of the four drives. Now that I have only two disks, two IDE connectors on the MB are enough.

The server is also *a lot* quieter now; I almost feel like keeping it in our study for easier tuning ;-). I need still to swap the CPU fan to some thermally controlled model and maybe do some tricks to the PSU fan, but the HDDs are now well below the ambient noise.

Based on feedback here is the revised plan:

I would especially like to thank Fozzy and Wolfe for their insight.

After getting lots of recommendations against using peltiers I went and studied few tutorials about DIY phase change cooling and browsed through available pre-assembled solutions. The conclusion is that they just do not fit to this project, neither physically nor to the target noise level and the budget.

So, it’s either TEC+water or just plain water for this project.

The plan above has few advantages: it requires minimal amount of modifications should I go or not go for TECs and it does not have any valves. Thus TECs could be completely controlled by just some PWM controller with relays based on set target water temp.

As TECs work as insulators when not powered, and temperature gradient between radiator inlet and outlet is pretty low, I think this solution would not severely impair the efficiency of the water loop even when running with all TECs off.

So, I think this is good enough plan to go with for now. I’ll revisit it when it’s time to actually connect the blocks, if needed. The peltiers are not a big investment, as I can get all the 4 I plan to use for some 80 euros, so I might anyway give them a try.

Electricity is not a big issue either, since the plain water setup should be more than enough to keep the system cool in desktop use and I am gaming only some 5-10% of the time my system is running (24/7).

Thanks again, guys!

Even thou I was reminded by several people that peltiers are very inefficient in cooling per watt ratio, I couldn’t resist doing another design using them. This one eliminates need for two waterblock but sports two valves:

Here are two designs in my mind for single loop:

The crucial difference between them is that in the first one both pumps are working in parallel and pushing through CPU block (which might be pretty restrictive, like Swiftech Storm) while in the second one one pump is pushing and another pulling from the same block.

I think the latter one might work better with hi-restriction block like Storm, but as I haven’t seen such anywhere else I am wondering if it has some fundamental problem?

Inspired by the CoolIT integrated TEC CPU cooling solution I started to wonder about using double water loops to take advantage of TECs when needed.

The main reason why I had decided not to use TECs was that if you install them on your CPU and GPU you need to give them power all the time to prevent them from working as a insulation.

Then I came up with this idea:

As you can see, the idea is to have TECs between two water loops and then have a valve to control whether water gets to “direct” contact with another loop or not.

To balance the restriction for the two pumps I chose to put the components that can run hotter or can do without so much cooling to the second loop.

So, when not needing the absolutely highest clocks, I could run without TECs and have quite low power consumption (like 200W for the whole system under load) and then when wanting to get all the speed I can get I could turn the valve, power the TECs and let them drain another 200-300W.

This of course puts requirements for the radiator quite high, but I have plenty of headroom there.

Now I just have to find out does this make any sense..

Just a small update. I’m still busy as hell at work, and right now I need to do some more planning to continue concrete modding work. I tested stretching 1mm PVC film over my upper shroud mold and found it to be *very* hard job. So I am considering doing the final acrylic one from several pieces, and in order to know where I want the seams to be, I need to decide how the surrounding parts will be fitted.

Anyway, my order from Delvie’s Plastics finally arrived. Since the goods weight several kilos, I asked for them to ship it with the cheapest airmail method and yes, it really was slow. I guess the x-mas time has slowed postal service down even from the regular level.

But here they are:

Acrylic hinge parts, rods and sheets. Nice.

The smaller sheets from closer distance. Pretty dull, huh:

..until you give ‘em some UV light:

Delvie’s offers nice assortments of all their four fluorescent acrylic sheets, and one of them I got.

I also got the Seagate Barracuda 160G 7200.9 I ordered a while ago. Here are the Seagate siblings: the old 160G 7200.7 PATA and the new 160G 7200.9 SATA:

As you might remember, I will swap the PATA drive in to our home server to replace a failed RAID part there. The SATA one will become my new system drive as soon as I get the operating systems moved from PATA drive there.

And for the record, here is the HDD cooling/silencing system from my current rig:

Drive sits inside the cage and then both stand on the rubber feet at the bottom of the case.

Unfortunately, as SPCR found out, the 7200.9 has sharper seek noise than the 7200.7 one, even thou is has fewer platters. Oh well.. It’s still very quiet, so I should have no problems damping it.