Need help with 17" iMac G4 1.25GHz motherboard repair

[CooperBox]

I was recently preparing the subject model iMac to install an SSD (which I've performed several times successfully on another 17" and also 20" iMac G4).
Unfortuntaely on this occasion whilst trying to prise-off a rather tight connector I damaged 2 adjacent components on the motherboard.
See following photos:

i) Is anyone capable of identifying these components please.
ii) If so from where can they be purchased?
iii) Can anyone suggest from where I may be able to get these installed.
I'd have no problem in removing the motherboard and shipping within Europe to facilitate the repair.
Any help would be greatly appreciated.

 

[philgxxd]

These are coils hence the four connecting pads. If anyone else is popping up I have the tools to do the repair. I once bought a G4 Cube from you that is still in my possession.

 

[CooperBox]

Hi Philipp, yes I well recall the sale of the Cube. How is it performing?
Thanks for info on the coils. Are replacements fairly easy to obtain?
If so could we work out a plan for me to get the MoBoard to you for repair? I was gutted when I realised I'd done that damage during disassembly, and would love to get this iMac up and running again with a new SSD inside.

 

[philgxxd]

Hi again. I just tried to search for a schematic but didn’t find anything yet.
The connector above the coils is for the LCD panel?
Could you give me the logicboard number so that searching for it maybe bring up the schematic?
Otherwise I would assume that the 4 coils are just of the same value and I could measure them when there are here.

 

[Needleroozer]

Hi again. I just tried to search for a schematic but didn’t find anything yet.
The connector above the coils is for the LCD panel?
Could you give me the logicboard number so that searching for it maybe bring up the schematic?
Otherwise I would assume that the 4 coils are just of the same value and I could measure them when there are here.

According to the 820-1501-A schematics, this connector (J12) is indeed for the TMDS link to the internal display. That common-mode choke (L76) is labeled as having 90Ω impedance, so the trick is choosing one with the proper frequency response to match its friends on the logic board.
Interestingly the schematic shows an alternate configuration wherein 0Ω resistors are used rather than the common-mode chokes, possibly for development purposes.

 

[CooperBox]

Many thanks to both for your feedback. I'm sure that Needleroozer"s attachments will prove valuable here and for others with similar iMac G4 queries.
@philgxxd: Again thks for your comments. I'll try and confirm the logicboard number tomorrow, plus any other macro photos that may prove to be of assistance.

 

[CooperBox]

I confirm that the iMac G4 1.25GHz logicboard p/no. is 820 1599-A
@Needlerooser: Your 820-1501-A schematics would appear to refer to the G4 1.0GHz model but is nonetheless interesting especially the info on the common-mode choke (L76). These four components appear to be common accross both logicboards.
A search has turned up the following schematic, sheet one of which gives mention of my 820 1599-A logicboard.
https://www.electronica-pt.com/esqu...k,953b20147a6a6ecaa57ef587d1cfe148/no_html,1/





Had a rummage about in my 'spares' room and located another G4 1.25GHz iMac purchased long since and forgotten about. It booted but to a screen with garbled artifacts. Unsure of the problem, I delved inside and took a few photos but couldn't identify any obvious defects. I'll show this board in a follow-up post.

 

[CooperBox]

Here are a few photos of the other iMac G4 1.25GHz logicboard 820 1599-A in my posession. I cannot see any obvious signs of capacitor bulging or leakage - that's just about how far my experience goes in terms of logicboard inspection for on-screen graphic artifacts.








@philgxxd: Please let me know if you think you may be able to procure these L76 and L74 components to perform a repair. As we're both in Europe I could also send the 2nd logic board for trouble-shooting and (hopefully) eventual repair. I would of course pay for any parts and reasonable cost of repairs and return postage.

 

[philgxxd]

I'm having problems to find a replacement without knowing more specs of the component.
I would like to continue to discuss the matter a little more with you guys before anything else. Maybe @Needleroozer can give more insight and guide us to a suitable replacement.
As a last resort you could of course take the donors from the "garbled screen" logicboard.

 

[philgxxd]

@Needleroozer
Would it be sufficient to measure the inductance to find a suitable replacement?

 

[CooperBox]

I'm having problems to find a replacement without knowing more specs of the component.
I would like to continue to discuss the matter a little more with you guys before anything else. Maybe @Needleroozer can give more insight and guide us to a suitable replacement.
As a last resort you could of course take the donors from the "garbled screen" logicboard.

Ref your last comment I would of course ultimately agree to this if no alternative solution could be found.

 

[Needleroozer]

I'm having problems to find a replacement without knowing more specs of the component.
I would like to continue to discuss the matter a little more with you guys before anything else. Maybe @Needleroozer can give more insight and guide us to a suitable replacement.
As a last resort you could of course take the donors from the "garbled screen" logicboard.

It took a bit of finagling, but I managed to find a boardview for the 820-1599-A PCB which you can open in OpenBoardView (compressed and attached as a .zip because the forum wouldn't let me upload a .bvr). Unfortunately, neither the schematic nor the boardview give us much to go off of in terms of the actual component name, but the footprint is that of an 0805 (2012 metric) component.

@Needleroozer
Would it be sufficient to measure the inductance to find a suitable replacement?

Common-mode chokes are typically specced for a singular impedance at a given frequency. 100MHz is typical, but to know for sure you'd want to do a frequency sweep from 0 to at least 100MHz so that you could compare the impedance-versus-frequency plot against commercial options...
After running parametric searches on Digikey and Mouser (90 Ohms at 100MHz, 0805/2012 package), there aren't too many results for whatever reason, so you may wind up being able to differentiate based on the DC resistance of the chokes on the PCB. These range from 0.2 Ohms to 4 Ohms, however, so you would either need very good multimeter leads or a four-wire resistance measurement setup to characterize that accurately.

If I had to choose one, I'd probably go with these due to their low DC resistance and relatively flat impedance with respect to frequency:

ACM2012-900-2P-T001 TDK | Mouser

ACM2012-900-2P-T001 TDK Common Mode Chokes / Filters 50V 400mA 90ohms AEC-Q200 datasheet, inventory, & pricing.

www.mouser.com

Another interesting finding is that those unpopulated zero-Ohm resistors (R603, R594, etc) have footprints on the bottom of the board directly underneath these common-mode chokes.

So if you have some zero-Ohm 0402s, you could pull the chokes off and install eight 0402s down on the bottom of the board instead. The chokes are there for signal conditioning rather than ESD protection, and this is all inside the iMac anyway, so I think it's probably fine to swap them out for jumpers.


Alternatively, if you have any dead iMac G5 motherboards, they seem to have the same 90-Ohm chokes on their internal TMDS connector, so you could harvest spares from there as well (I checked this on an 820-1540 schematic, but I'm pretty confident they'll all be the same).

 

[philgxxd]

Thanks so much @Needleroozer. You went to great length to explain that in every detail possible.
I think we can do this. Thanks again for the effort!

 

[philgxxd]

I would love to have a set of 4-wire kelvin probes for my DMM but they are so damn expensive. 260$ (11059A Keysight)
Looking into third party. Any recommendations?

 

[philgxxd]

Another interesting finding is that those unpopulated zero-Ohm resistors (R603, R594, etc) have footprints on the bottom of the board directly underneath these common-mode chokes.

So if you have some zero-Ohm 0402s, you could pull the chokes off and install eight 0402s down on the bottom of the board instead. The chokes are there for signal conditioning rather than ESD protection, and this is all inside the iMac anyway, so I think it's probably fine to swap them out for jumpers.

Is bridging with solder blobs or fine wires not an option as this would be very easy to do for @CooperBox ?

 

[CooperBox]

@Needleroozer:
I really must thank you for the time you've taken to assist here in providing such valuable information. It's truly much appreciated!
Similarly for @philgxxd for progressing same.

​

​

 

[Needleroozer]

Is bridging with solder blobs or fine wires not an option as this would be very easy to do for @CooperBox ?

Totally an option as long as he removes the common-mode chokes first, yep. Little pieces of wire would be easier in my opinion since a 0402 footprint is a little big to bridge with surface tension.

 

[Needleroozer]

I would love to have a set of 4-wire kelvin probes for my DMM but they are so damn expensive. 260$ (11059A Keysight)
Looking into third party. Any recommendations?

Unfortunately my home gear isn’t fancy enough to have 4-wire measuring capabilities, so I don’t have any specific experience in this realm. The last time I used four-wire measurements was in a circuits lab in college…
With that being said, these guys look alright and review well:

AST Labs Test Lead Kelvin Clip to 4-channel Banana Plug (3 feet)

Automotive Aerospace Telecommunication

www.amazon.com

I’m not sure if the tweezer-style probes would be able to grab the contacts on the bottom of the chokes without flying off or shorting to the other side, though. If you wind up having to solder diode legs on to have something to grab, then traditional alligator clip leads might be just fine.

 

[CooperBox]

Totally an option as long as he removes the common-mode chokes first, yep. Little pieces of wire would be easier in my opinion since a 0402 footprint is a little big to bridge with surface tension.

@philgxxd
As my soldering skills are virtually non-existant, is the above something I could task you with? As already mentioned I'd reimburse for any parts purchased plus reasonable cost of repairs and return postage.
Cheers!

 

[philgxxd]

Sure, I’m happy to help out. I’m out of town currently for the rest of the week so I will come back to you shortly.

 

[CooperBox]

Sure, I’m happy to help out. I’m out of town currently for the rest of the week so I will come back to you shortly.

Ok, thanks!👍

 

[CooperBox]

As mentioned above, both 17in 1.25GHz G4 iMac's in my possession have logicboards p/n. 820 1599-A.
Now just to confuse the issue, I forgot that I had a 3rd iMac G4 1.25GHz stored away (fully functional) which today I opened up and was surprised to find the logicboard p/n. 820-1501-A. This being the same p/n that needleroozer showed the schematics for in post #5 above.
Thinking that particular board was special to the 1.0 GHz 17in model, I re-assembled the base and booted the iMac to confirm exactly if this was a 1.0 GHz 17in or 1.25 GHz 17in model. The system report confirmed it as the latter. Photos of the board follow:

The highlighted circle shows a different configuration to the 820 1599-A board

You’ll note that the 4 common-mode chokes appear to be absent on this 820-1501-A configured board. Is this related to the use of 0Ω resistors alternative schematic configuration referred to by needleroozer?

 

[Needleroozer]

As mentioned above, both 17in 1.25GHz G4 iMac's in my possession have logicboards p/n. 820 1599-A.
Now just to confuse the issue, I forgot that I had a 3rd iMac G4 1.25GHz stored away (fully functional) which today I opened up and was surprised to find the logicboard p/n. 820-1501-A. This being the same p/n that needleroozer showed the schematics for in post #5 above.
Thinking that particular board was special to the 1.0 GHz 17in model, I re-assembled the base and booted the iMac to confirm exactly if this was a 1.0 GHz 17in or 1.25 GHz 17in model. The system report confirmed it as the latter. Photos of the board follow:

View attachment 2308885

The highlighted circle shows a different configuration to the 820 1599-A board

View attachment 2308886

View attachment 2308887

View attachment 2308888

You’ll note that the 4 common-mode chokes appear to be absent on this 820-1501-A configured board. Is this related to the use of 0Ω resistors alternative schematic configuration referred to by needleroozer?

There’s a good chance that they’re simply on the other side of the logic board.

 

[CooperBox]

There’s a good chance that they’re simply on the other side of the logic board.

Ok, thanks.

 

[philgxxd]

@CooperBox I left you a PM some days ago.