[PLOG] Mac G5 --> PC Conversion

[angelkiller]

Are there some pinions or gearwheels or whatever you call them missing? If you find out the problem you can salvage parts from some other drive to fix this one.

EDIT: Oh, this mod looks really good. Keep it up!

I found the problem with the optical drive. There's a small motor in the front that's responsible for ejecting the tray. Well the band that connected it to another gear(?) was missing. So the motor turned, fruitlessly. At first I tried to use Stretch Magic, but the knot got in the way. So after some thinking, I went to my small Lego stash. And I got a rubber band. Yes a Lego rubber band. (an official Lego part btw) Well, it fits and even better, IT WORKS! Legos FTMFW!! Luckily this was an easy fix. Could've been much worse.




The rad brackets need a bit of tweaking before they'll fit. No work tomorrow, so I should be able to get some stuff done!

Radio Shack closed half hour ago, so I can't get the resistors I need for the IR thing. I'll be there first thing tomorrow though.

 

[BrooksyX]

Awesome! I had to replace a couple rubber bands in some Xboxs recently. Just ended up using some plumbing rubber bands. Cost like $0.75 for a two pack.

 

[Cuzza]

Lol... when I read that first post about the drive issue I thought, "I bet there's a broken belt in there".. you can fix a lot of things with rubber bands.

When you say the resistance needs to be substantially over 48 ohms, how much is substantially? 75 ohms? 100? 1000?

I think 100 would be ok, but I'd go with 1000 to start with. You don't want any residual current going on so start high, if it doesn't work come down. (kind of guessing here, I want to try this myself)


I think what Ahhzz is trying to say is that the LED may not work if the voltage is too low, not that you would know because it's IRand you won't see it! You have to be careful to get the voltage in the right range to the LED. Too low, don't get over the bias voltage. Too high, fry it.

Then within that range tune the brightness so that your thing is sensitive enough to work but doesn't open the drive every time you move your head.

 

[angelkiller]

GREAT NEWS!! THE IR SWITCH WORKS!! Pics and vid when I get the wiring finalized!

Now for some tweaking. Right now I'm using the 1k ohm resistor and it's wired up just like the picture Cuzza provided near the bottom of page 3. Now the thing is that I have to hold the detector directly in front of the emittor in order to get the drive to open. If I lower the resistance, will that allow for less light to activate the switch?

Edit: I don't think the resistor makes any difference on how much light is needed to trip the switch. I changed to the 100 ohm resistor and got the same results. As long as it's above 64 ohms, I think it's ok. (maybe)

What is happening is that the IR LED is only bright at its tip. And the detector is oblivious to any IR light not at its tip. So the only way that this works is that the IR LED is facing the detector. There's about a 30 degree window where the detector can pick up the LED. Just a quick pic to illustrate:

The detector must be right in front of LED in order for it to work. It doesn't pick up anything off center. Maybe I need more IR light? Or maybe I need to modify the detector so it'll pick up stuff that's not directly in front of it? Any ideas? Thanks.

 

[Cuzza]

yeah try lowering the resistance on the detector. then try upping the voltage on the emitter, it says 1.7V max so go to that.

 

[angelkiller]

Ok, I put a 22 ohm resistor on the emitter, giving the emitter 1.7v. The resistor in the detector circuit is 100 ohms. Everything seems more or less the same.

This is frustrating. I was really banking on the detector picking up the IR light reflected off of my hand. I was actually more concerned about it being too sensitive.

 

[Cuzza]

hmmmmm.... you might have to build a more complicated circuit for the detector, with some transistors and such to amplify the effect

Since the detector is working in the switch modeit needs a certain amount of light to activate and it's not getting enough right now. but you need the switch effect but with a lower threshold. i'll think about it.

EDIT..

This is the one you want. It uses an NPN transistor. When the detector senses the light the voltage at the base of the transistor drops off, so it switches off and you get voltage to Vout.

I think that's how it works, never really understood transistors, mainly because I haven't tried since I was 10 years old messing about with my electronics kit (back then I had no idea)

Have to work on getting those resistances right though....


EDIT #2: Pretty sure that's wrong, don't use that circuit!


EDIT #3

Try this

 

[angelkiller]

Alright!! I got it working how I want it to. The last circuit Cuzza made worked.



Video Linky!!!

Great thanks goes out to Cuzza.

 

[Cuzza]

Nice work, good soldering. I need to get one of those soldering stations. And learn to solder properly. I suck.

 

[angelkiller]

Nice work, good soldering. I need to get one of those soldering stations. And learn to solder properly. I suck.

Oh I don't think I solder that well. Isn't there some technique that you're supposed to use? Like heat the component not the solder or something? Ah well.

The soldering station is definitely recommended for those who solder more than twice a year. Before I had to hold the iron in one hand, the component in another hand, and the solder in my other hand. Yeah.. that didn't work out too well.

 

[Ahhzz]

Very Sharp. Sorry Angel, for lack of reply, was extended AFK Very impressed mate, may have to try playing with one of these myself now!!! I've got an extensive background in electronics, but haven't used it in years, and some of the possibilities rolling thru my head are wild heheh

 

[Cuzza]

Just got around to watching the video.

if it is too sensitive just bump up that 500-ohm resistor

 

[angelkiller]

Very Sharp. Sorry Angel, for lack of reply, was extended AFK Very impressed mate, may have to try playing with one of these myself now!!! I've got an extensive background in electronics, but haven't used it in years, and some of the possibilities rolling thru my head are wild heheh

Thanks!

Just got around to watching the video.

if it is too sensitive just bump up that 500-ohm resistor

Sorry for asking again, but by how much? I don't know how the resistance affects things. Will 50 ohms make a difference? 200 ohms?


And I'm having a problem getting the radiator brackets mounted. The res is mounted on the top chamber separator and the rad has to line up with the res. (The barbs of the rad and res will be about 1cm apart. Not much room for error) But right now the brackets are too large by less than 1cm total and they push up on the top separator plate, causing it to flex. Ideally, the top of the brackets (the short part) should be about a few mm lower than where it is now.

The problem is that making that change. I don't have any tools to make an exact bend in the aluminum. I just used pliers and the carpet for fulcrum. Bending the bracket back straight doesn't work. There's still a tiny bend that I can't get out. So now I'm stuck. I need to make a very fine adjustment in the bend of the bracket, but I don't know how to/can't get the bend how I want it. Any suggestions?

You can refer this post for pics of the bracket.

 

[Cuzza]

i don't know what to suggest, but if you can't do it yourself, pay someone to do it.

as for the resistor, probably serious changes, i'd go to 1000-ohm, if that's too much come back down. trial and error. or get a pot and use that to vary the resistance, then swap it for what you need.

 

[angelkiller]

I'M GONNA REVIVE THIS THREAD! CLEAR!!

Whew, that was close. I almost let the thread die. Can't let that happen again!



First, I played with stock Apple fans again. I could make a thread dedicated to how those fans work... It's really weird. Ah well, I wired the fans up to 5v via Molex.
Details learned about Apple stock fans:
All the fans have 2 positive leads and won't work with just one. For example the 80mm fan in the top chamber has 4 leads, 12v, GND, RPM, 12v, in that order. Both 12v pins must be connected or the fan won't start. The only stock fans that I'm going to reuse are the rear 80mm ones. It too is wired weirdly. Both fans are already wired together, 4 wires from each fan, and 1 6 pin header. (some wires join) The setup is basically the same as the other fan, just two of em. Each fan has it's own 12v and RPM wire. The fans share their 2nd 12v wire and a Ground. After playing around, I've found out some things. First by unplugging the shared 12v line, you stop both fans. If you unplug one of the individual 12v lines only 1 fan stops. (depending on which wire you pull) Maybe this is used for some kind of fan control? Also, if you take the sticker off of the hub, there are only three exposed points on the fan's PCB. You can take voltage measurements from all three. I forgot what they are exactly, but they're all somewhere between 3V and 4v. Also, I noticed that the fans do not respond to voltage changes. I tried changing all the 12v lines to 5v. No (perceptable) change in rpm or noise. (I had to eye it) But I measured those 3 points again, and they were exactly the same! BUT they fluxuated a bit more slowly. At 12v the voltage would stay pretty constant, it would change by .01v every so often, but would jump right back pretty quickly. At 5v, The voltages were slower to change. This leads me to think that the fans have their own voltage regulation circuit built in and spin at a constant speed regardless of input voltage. But what use would that be? It would make sense for them to be controllable. I know all that info was a bit unnecessary, but I thought I'd document my findings, as this info is hard to come by on the net.


I also got some screws for the second HDD. It slides in and out of the HDD cage with ease.


I also got the PSU open and out of it's casing. I plan to mount the PSU PCB on the bottom of the case. (not directly on the bottom) However, ran into one snag. Part of the filtering stage is soldered right onto the power plug. Since the pics were taken, I've cut the main lines right above the PCB, so I could access the components on the plug without the PSU PCB attached. I'm not exactly sure how I'm going to handle the AC plug yet.



Most notably, I finally got the radiator mounted. It's not 100% perfect, but it's pretty damn close. It's in, which is progress to me. I had to get a friend to help me make the rad mounts because I don't have the right tools. With a drill press and a vice, making the mounts was easy. These are the 3rd pair I've made. These are made of steel (rather than Al) and I've polished them just like the top separator plate.


So everything is starting to come together! Now it's time to get the motherboard mounted. I'll be working on that next. Should be done in the next 2 days or so. Scratch that, it WILL be done in the next 2 days. I need to get this done before school starts back. 10 days left.

 

[Cuzza]

excellent. that is pretty strange with the fans though. maybe you already covered this but why are you taking the psu out of its case? does it not fit with the case on?

 

[kyle2020]

/subscribed.

 

[steelkane]

nice res-rad setup,, looks awesome

 

[phanbuey]

yes... I was wondering what happened to this build... MOARH PICTURES!

 

[angelkiller]

excellent. that is pretty strange with the fans though. maybe you already covered this but why are you taking the psu out of its case? does it not fit with the case on?

Yes it does fit in there with the case on. However, I was going to cut unnecessary cables because there's no space for hiding excess cables. So since I was going to hack it up, I might as well take it out of the case... (bad reason, I know) But a better reason is that the mounting holes are on the 'back' (exhaust grill) of the PSU. I need the PSU to mount horizontally in the case in order to fit. (ie exhaust grill facing rear of case) There's literally nowhere to attach the PSU unless I made my own bracket. But seeing my previous attemps at brackets... yeah. I could just drill some holes in the bottom of the PSU cover and screw it to the bottom of the case, but if I go through all that effort, I might as well take it out and use the PSU PCB holes. Hopefully that made sense, kinda hard to explain without seeing it. If you saw it you'd understand.

But I still unsure about the filtering components on the AC plug.

yes... I was wondering what happened to this build... MOARH PICTURES!

As you wish laugh


So right now I'm letting the JB Weld dry under my mobo standoffs! First, I got my dremel and scuffed up the areas where the standoffs would go. Distemper said in post #20 that he(/she) didn't scuff the aluminum, but he(/she) was using 9 standoffs and I'm only using 6, so I need the strongest bond possible. Then I mounted 6 of the shorter standoffs onto my motherboard. I got the JB Weld and put a decent sized blob on each spot in the case and lined the graphics card up and placed the mobo attached to standoffs on the JB Weld. And I added to weights to give it better contact. Unfortunately, I can't tell how good of contact the standoffs are making with the JB Weld... So we'll find out tomorrow I guess. Tomorrow, I'm going to add some more JB Weld around and on top of the standoffs for added strength/security. Pic to illustrate:

 

[p_o_s_pc]

what are the specs of the rig?

 

[angelkiller]

what are the specs of the rig?

PSU|400W OCZ Fata1ty
Mobo|Gigabyte Ga-P35-DS3L
CPU|Xeon 3210 (B3 Q6600 with 8x multi)
RAM|2GB (2x1GB) Crucial Ballistix DDR2-800
Graphics|Asus 9600GS0
Storage|74GB Raptor
|1TB Hitachi
Optical|Stock Apple Superdrive
Watercooling |D-Tek Fuzion
|Danger Den CPX-Pro
|Black Ice GT Stealth 240
|Koolance 80mm Inline Res

Oh, and thanks to all the people that complimented me. I'm sorry if I forget to respond to them, but I do read and appreciate them.

 

[Cuzza]

i think i understand. you just be careful with those exposed circuits mate

 

[angelkiller]

Does anybody actually read post titles?


So the motherboard standoffs are JB Welded to the case! Then I added some more JB Weld for added strength/security.


Next, I started working on the wires for the front panel connectors. So I cut all the wires I needed, stripped the ends off, and added a bit of solder to each end. I also took the black plastic connector holder thing off the front panel.


I ended up with all of this ^^


I was gonna go ahead and solder them to the front panel, but I think they should be sleeved. I think all the wires would look a little out of place without being sleeved. The problem is that if I order some, the earliest I'll have it is by Wednesday. And it's not worth overnighting it imo. Plus, I can't seem to find any anyway. Anyone know where I could get some? (local or online)


I was going through some spare parts, and I found a usb bluetooth adapter. Since I found it, I thought it could be useful. I had planned to put it right on the USB header. You can see the 4 extra wires. Problem is that solder won't stick to the gold contacts. Anyone know how I can get that BT adapter connected? The casing was destroyed in order to get the PCB out.

 

[angelkiller]

Obviously this isn't gonna get finished before school. So expect slower updates from now on. That said, I'm in the finishing stages of the project, so there's not much serious stuff to do. Anyway, more pics.


I drilled holes for the PSU and got that mounted in the case. I also cut the stock I/O ports out of the case. It hurt to do because the case was flawless, but I had no choice.


Next, I focused on the PSU. I cut off the wires that I wouldn't be needing and wrapped thier ends in heatshrink so they don't short anything out. Then I shortened the 4pin ATX connector since it only needed to be like 2 inches. However, I really botched the soldering job up imo. I had to solder and unsolder all 4 connections about twice each. The first time the heatshirnk I used wasn't large enough to go over the soldered connection and the second time(s) I was rushing and forgot to put the heatshrink on before I soldered. Idiot me. So hopefully all the connections turn out to be ok... they're gonna be carrying ~4-6A of current. Hopefully the resistance won't be to high.


Last, I had to cut the frame of the rear 80mm fans because it interfered with the I/O ports on the mobo. This was a miserable experience. Anyone who's cut plastic with a dremel knows my pain. Plastic does not cut... it melts.... Long story short, cutting plastic with a dremel is hell and will always come out very messy at best. Of course the type of plastic makes a difference... the thicker and denser, the worse. Of course, Apple uses the thickest and densest plastics ever. [/rant]


But, all in all, I'm making progress. Here are some shots with the mobo and PSU mounted. If I wanted to, I could just as easily throw in the HDDs, the optical, the rad, the res, and the pump. So just kinda imagine all that in the system.


Concern #1: PSU cooling. There will be 2 120mm fans in the front of the case and 2 80mm fans right next to the PSU. Do you think I need additional cooling? I was thinking about getting 2 more 80mm fans and putting them right in front of the PSU. But do you think I could get away without them? Need to now ASAP cus I'm ordering tomorrow morning. (thanks!)

Concern #2: USB bluetooth adapter. In my last post, I showed a USB bluetooth adapter that I had found. I'd like to use it because I've just found a wireless Mighty Mouse for $30. However, solder won't stick to the USB contacts on the adapter. The casing for the adapter has been destroyed. Is there any way to get that adapter wired up?

Still more to come. I promise.