Contribute
Register

Ersterhernds iMac G5 20 Inch A1076 Project

Status
Not open for further replies.
I'm trying to put in a laptop board myself. Its powered by 19.5V. Could I bypass the 19.5V PSU and instead use one of these step up or step down voltage converter regulator sold on eBay to power the board from the existing iMac PSU?
 
I'm trying to put in a laptop board myself. Its powered by 19.5V. Could I bypass the 19.5V PSU and instead use one of these step up or step down voltage converter regulator sold on eBay to power the board from the existing iMac PSU?

The power step-down shouldn't be too much trouble, but the challenge will lay in how to actually turn the G5 PSU on. You probably don't want to leave it fully powered all the time but rather to sit in a standby state until the power button is pressed to turn on the computer.

I accomplished this by using two power supplies (see page one of this thread) and a relay wired to the PS_ON pin of the G5 PSU for it to properly power up. On my system, its the NUC 19V PSU and the NUC Front Panel Header pin 9 that provide the power to get it all started. Kiwi and MacTester used a more elegant solution of a microcontroller to accomplish this. Either way, turning the G5 PSU on and off is critical to the exercise and you'll need to decide how you're going to do it.

Once you figure out how to do this, the step down part shouldn't be too much trouble.

Cheers!
 
I'm not as smart as you guys to figure that stuff out. Obviously you do not want the PSU on all the time. So how is this handled on the original iMac then? I'm hoping to mimic same with a laptop board?

It would be nice then just to use the PS from laptop but then not enough juice for backlight. My idea is to put the MB which really narrow, and the usb/lan/audio mini card, and somehow connect these to the existing I/O. The display soldering I think I can handle. I just got to figure this power supply distribution before I embark.
Could you tell what type make relay u used?
 
Last edited:
I'm not as smart as you guys to figure that stuff out. Obviously you do not want the PSU on all the time. So how is this handled on the original iMac then? I'm hoping to mimic same with a laptop board?

It would be nice then just to use the PS from laptop but then not enough juice for backlight. My idea is to put the MB which really narrow, and the usb/lan/audio mini card, and somehow connect these to the existing I/O. The display soldering I think I can handle. I just got to figure this power supply distribution before I embark.
Could you tell what type make relay u used?

I'm not sure how Apple did it but probably much the same as its been accomplished on this forum, except integrated into the logic board circuit.

The relay I used is found at this link. When 5V is applied to one end of the relay, you have the option to either open or close a separate circuit at the other end. Its an isolation relay, so theres no electronic connection between the two sides, as is optically controlled. In my application of it, the 5V from the NUC pin 9 activates the relay when the power button is pressed, and grounds PS_ON within the iMac PSU, thereby turning it on.

This relay proved to be the perfect solution to the problem. It essentially does the same function as physically inserting a paper clip between PS_ON and GND in the iMac PSU. When the NUC powers down, the 5V supply from pin 9 ceases, and the relay opens the circuit between PS_ON and GND, the same as removing a paper clip between the two pins.

Hope this helps. Sorry about the delay in response.


Ersterhernd
 
Last edited:
Thanks Ersterhernd for the replies. Im starting to order some parts now. Boy you guys are right. Space is at a minimum. What makes it hard is the centralized 2 main fans. I've been trying all sort of schemes for days. I think I've settled the best layout possible. On the HDMI conversion did you use the 1K resistor?

As far as PSU. Since I'm not using a NUC board I don't have on/off pins. My idea is to utilize the laptop power on circuit board to turn on the MB. Even if I after relocated the power button. Then figure out how I can get 5V to this relay from the board to get the other PSU going. All I have is a cpu fan connector, micro SATA on board connector, and a battery female push pin connector underneath MB. Perhaps steal the 5V from USB? Are all USB powered by 5V? Does it have to be 5V, can it be 3V relay? Testing battery terminal, I got 3+V on 3 pins.
When you spliced the AC lines from the NUC adapter in the iMac PSU, does that mean the NUC power adapter inside the iMac is always on when the iMac is plugged in?

If I were to use lets say only the PSU from laptop 90W or upgrade to 120W, can I split the barrel, and use one to MB and the second out to a boost 19V to 24V to inverter? Does splitting cause 19V 3.5amp to drop?
 

Attachments

  • IMG_20160618_202926 copy.jpg
    IMG_20160618_202926 copy.jpg
    150.4 KB · Views: 181
Last edited:
Thanks Ersterhernd for the replies. Im starting to order some parts now. Boy you guys are right. Space is at a minimum. What makes it hard is the centralized 2 main fans. I've been trying all sort of schemes for days. I think I've settled the best layout possible. On the HDMI conversion did you use the 1K resistor?

As far as PSU. Since I'm not using a NUC board I don't have on/off pins. My idea is to utilize the laptop power on circuit board to turn on the MB. Even if I after relocated the power button. Then figure out how I can get 5V to this relay from the board to get the other PSU going. All I have is a cpu fan connector, micro SATA on board connector, and a battery female push pin connector underneath MB. Perhaps steal the 5V from USB? Are all USB powered by 5V? Does it have to be 5V, can it be 3V relay? Testing battery terminal, I got 3+V on 3 pins.
When you spliced the AC lines from the NUC adapter in the iMac PSU, does that mean the NUC power adapter inside the iMac is always on when the iMac is plugged in?

If I were to use lets say only the PSU from laptop 90W or upgrade to 120W, can I split the barrel, and use one to MB and the second out to a boost 19V to 24V to inverter? Does splitting cause 19V 3.5amp to drop?


Yes I used a 1K resistor on the HDMI. I've done several of these machines and settled on a method that although wasn't pretty to look at, worked very well and was considerably easier dealing with the required soldering of such tiny wires.

The link is here.

There must be a way to wire the iMac power button to the laptop board power button. There are traces on the corner piece of the iMac board (you've cut it already) that trigger on/off. You'll need to test for zero resistance while pressing the button to find the correct traces. Its a feature I'd go to great lengths to preserve.

The relay I used is definitely 5V as per the specs. USB will give you that, but stays powered during sleep mode, although it may work to accomplish on/off to the relay.

The NUC power brick remains 'on' the same way it would if plugged into the wall in a normal NUC application. Its designed to do this, not to be unplugged after every use. The NUC PSU was just small enough to fit and actually touches the inside of back cover of the iMac when installed.

Splitting your 19V and stepping up to 24V at the barrel connector will feed the full 24V to the inverter all the time. I think the 24V line in the iMac PSU only puts out 20V or so when 'off'. You may need to figure out some way to circumvent that for proper inverter life. I'd try to re-use the iMac PSU if at all possible. It not only will feed you all the required voltages, but it contains the AC power socket that you'll require for use of the iMac power cord.

The pic you've attached of your layout shows that your laptop board uses about 2x the real estate of the NUC. I'd continually test the installation of the back cover throughout your project. It MUST fit on properly when complete, the heavy hinge assembly takes up a lot of room, and there cannot be any shorts to the metal case.

Good Luck and keep going!


Cheers
Ersterhernd
 
Last edited:
The link is here
Wow, if thats the case maybe I should get these jacks then. See pic.

There must be a way to wire the iMac power button to the laptop board power button. There are traces on the corner piece of the iMac board (you've cut it already) that trigger on/off. You'll need to test for zero resistance while pressing the button to find the correct traces. Its a feature I'd go to great lengths to preserve.
OK there is hope. I've looked I think at Kiwi solder hookups to the cut board to power button, I'm trying to locate the area but it seems that area on mine is a little different. Will keep looking.

Splitting your 19V and stepping up to 24V at the barrel connector will feed the full 24V to the inverter all the time.
I forgot about that. That can't be good. Although I did order a 2 lamp inverter 1x4" board like for $4. It only needs 12V. As anyone tried converting to LED backlight on this?

The pic you've attached of your layout shows that your laptop board uses about 2x the real estate of the NUC. I'd continually test the installation of the back cover throughout your project.
The board is very very slim. Everything is flat. I have put the back, and it doesn't look its hitting anything. That board is great, even has a m-sata slot. I'm even thinking of bringing the bezel camera from laptop, and put in on the iMac .
 

Attachments

  • Screen Shot 2016-06-19 at 2.14.09 PM.png
    Screen Shot 2016-06-19 at 2.14.09 PM.png
    133.4 KB · Views: 148
Hi Ersterherd. Maybe you can shed some light. i'm confused. So I did the tracing on corner board where the iMac power switch is, getting 0 resistance. I'm assuming I do the same and look for 2 leads on ribbon cable on the power board to get same thing 0 resistance. I only get 0 resistance with first lead on ribbon and ground to standoff hole, see pic. Every other lead on the ribbon has some reading. So when I apply the wires from iMac board to test it turns on the laptop without pressing the iMac power switch. Why is that happening?

pw board.jpg pw wire.jpg
 
Hi Ersterherd. Maybe you can shed some light. i'm confused. So I did the tracing on corner board where the iMac power switch is, getting 0 resistance. I'm assuming I do the same and look for 2 leads on ribbon cable on the power board to get same thing 0 resistance. I only get 0 resistance with first lead on ribbon and ground to standoff hole, see pic. Every other lead on the ribbon has some reading. So when I apply the wires from iMac board to test it turns on the laptop without pressing the iMac power switch. Why is that happening?

View attachment 201061 View attachment 201060
You only want zero resistance when the imac button is pressed. I'm not sure how the circuit on your laptop board works. On the nuc I wired the button directly to the power button pins on the front panel header.

It's unlikely that a FPH exists on a laptop though.
 
You only want zero resistance when the imac button is pressed. I'm not sure how the circuit on your laptop board works. On the nuc I wired the button directly to the power button pins on the front panel header.

It's unlikely that a FPH exists on a laptop though.

The power switch measures 0 resistance both when pressed and not pressed. I assume I did it wrong then. So to get this right, I need some resistance on 2 holes and when button pressed goes to 0 then correct?
Yeah no such luck with FPH on the laptop. I'm determined though...:problem:
 
Status
Not open for further replies.
Back
Top