iMac G4 15" [Nickjf20]

[nickjf20]

Update:

I've soldered up the 2.5mm Pro Speaker connector to the TA2024 (same chip as original iMac) amplifier (eBay special). All works, but soldering isn't fantastic. I nabbed a fine tip temperature controlled iron for the rest of the mod.

Sound quality isn't as rich as via the original iMac - there's a TAS3004 on the iMac before the TA2024 that is presumably setup to act as an EQ/DSP. This can be emulated with software EQ, but I won't be bothering as the sound is 'good enough' as is.

The two remaining unsoldered pins on the 2.5mm connect to a Dallas DS2430A 1-Wire (same system as the DS18B20 temp sensors) EEPROM device in the Pro Speaker 'splitter'. It just reports a unique 64bit registration number and other read/write data in a 256bit EEPROM. It only has 2 connectable pins. Picture found elsewhere

Whether it acts as a simple serial number for the speakers, or it encodes some form of revision number to allow different parametric EQ curves depending on the batch of drivers used, I don't know. Nothing to do with blocking audio from non-apple products though, that's for sure.

The 3 USB ports are soldered as well - the 5V lines have a diode incase I need to add an external 5V line to the shared USB power in the future. Apparently hubs that switch between external adapters and bus power use this setup.

Ran the Arduino (with servo going), whilst charging the MotoX & eReader and there were no overcurrent errors on the PowerMac, hopefully the BRIX won't need an external 5V supply either. Time will tell.

Still to solder: Power Switch and Ethernet

 

[MacTester57]

SMC is now soldered up, just a few more headers to add. This is identical to MacTester's SMC, albeit with 1 fan header and no tach feedback. So of course, thanks to MacTester & Kiwi

Hi nickjf20

Thanks for the mention. You have already made some good progress.

Good to see, that all the work of Kiwi, ersterhernd and me is useful for you.

I'm writing this answer on my iMac G5, which is still working great on 10.10.4. The internal slot-in DVD drive is now also working, which means, that this project is now definitely finished.

Good luck!

MacTester

 

[nickjf20]

Hi MacTester,

All your mods are fantastic, it's really good of you (all) to share and document everything so well. I'd be clueless without it all.

I do have a question or two for you though!

On the PSU connector, am I right in thinking that the PNP simply inverts what an NPN would do in the same position?
i.e. pulls the PSU pin low when D4 is low
I'm not sure what the 100nF cap is for either?

I'd like to know a little about the touch sensor setup you used as well - does the ground plane sit insulated behind the two sensors?

Thanks in advance

---------------

All the ports apart from the line in (mic) are wired up now

I've cut an acrylic panel to hold everything, but the hole saw I used has made a real mess of things ... might write it off and sort something with TheLaserHive instead

 

[nickjf20]

Everything was soldered up and complete, connected it all up, and without setting the downconverter correctly, I sent 12V to the LCD panel accidentally for a minute or so (expects 3.3)

Can't get an image to display at the moment, but the display still reports as 'iMac' in System Preferences and shows the correct resolution. Perhaps there's hope, perhaps not

---

I'll leave it until the morning, but I'm not hopeful. Lesson learned, I suppose.

Here's the adjustable culprit

Anyone have a spare 15" LCD panel

 

[kiwisincebirth]

I do have a question or two for you though!

On the PSU connector, am I right in thinking that the PNP simply inverts what an NPN would do in the same position?
i.e. pulls the PSU pin low when D4 is low
I'm not sure what the 100nF cap is for either?

Essentially yes you are correct. These types of transistors (BJT) switches on when a voltage differential is applied causing a current to flow. In the case of PNP the current must flow from emitter to base. For NPN the opposite is true, the current must flow from base to emitter.

In testing without the capacitor, The computer would experience strange hangs/shutdowns. This was assumed to be due to small voltage spikes on the base turning the transistor off, causing short loss of power to PC. The capacitor absorbs these spikes preventing the issue.

Bummer about LCD

Kiwi

 

[nickjf20]

Thanks Kiwi, your help, as always, is much appreciated!

I've managed to source a 17" from a friend that I'll be collecting sometime this week, so the project is back on, and better!

---

The extra time has allowed me to finish the Tarkus speakers

Just need fully routing and painting now

 

[nickjf20]

Took the iMac LCD apart as a last ditch attempt to see what was wrong, more for interest than any hope of revival.

It turns out there's a fast acting SMD fuse just after the 3.3v line in on the LCD PCB. Checked it, and there was no continuity.
Jumped it with a thin piece of wire, and the screen is working again. I'll add a 1.5A fuse lower down the line.

Success!

There is some corruption in the way of flashing white dots in places of high activity i.e. youtube videos
Presumably a shielding issue.

 

[golfleep]

Great build, glad you got the screen working again!

 

[nickjf20]

Great build, glad you got the screen working again!

Cheers Golfleep

----

The original 3.5mm headphone is now working with jack detect, switching the audio from the internal amp via the relay pictured. I picked up the 4.5v versions as someone was selling them substantially cheaper than the 5v versions, I don't see why running it at 5V should be a problem

There's a lot of noise when connected to the BRIX, but nothing when connected to a phone or similar. Presumably the BRIX needs grounding, or I need to lift the ground of the BRIX audio out. I think Esterhernd had the same problem in one of his builds.

The BRIX drives the amp far louder than the MotoX - at full BRIX volume the little pro speakers sound fantastic, and play clean.

 

[MacTester57]

Good progress you made! It's good to hear, that you've fixed the LCD

There's a lot of noise when connected to the BRIX, but nothing when connected to a phone or similar. Presumably the BRIX needs grounding, or I need to lift the ground of the BRIX audio out. I think Esterhernd had the same problem in one of his builds.

I had a similar problem. It's essential, that all analog audio devices have the same, stable GND potential and that you don't have a GND loop. The solution is documented in this post:
http://www.tonymacx86.com/imac-mods/130197-mactester57s-imac-g5-17-a1058-model-11.html#post1043277

MacTester