• Hello, Guest! Welcome back, and be sure to check out this post for more info about the recent service interruption and migration.

New Hybrids for the Macintosh Portable M5120?

techknight

Well-known member
I am starting this thread because we are at the point where this needs done. 

I also want a record for links to other peoples projects on that so they don't get lost. I know there is one floating out there, but alas... I forgot to save the link to it. 

Anyways... I have a prototype M5120 and sure enough the hybrid is shot. And I am fresh out with no real way to fix it. first it was not wanting to turn on, etc... I was able to fix that but its stuck in charging mode which means the PMGR IC switches to a different ADC scale. so if the battery is about 6.1V or lower, it shuts off saying its too low, no reserve power. So basically the window comparator isnt latching. Could be the oscillator inside the IC, or it could be the reference comparison voltage has changed/drifted out of spec due to resistors changing in value. This is also dangerous for the battery, because the charging circuit will NOT shut off when it reaches full charge. 

The trouble with the hybrid ICs is the cap juice gets into the ceramic substrate, and eats away/changes the laser etched carbon film resistors that they print onto the ceramic substrate. If those values change, it throws everything off, especially because there are analog voltage references, and resistor divider/feedback networks for the window comparator. Those resistors change in value, the whole thing is just screwed. 

That one project I saw awhile back, the resistor values logged looked suspicious so I am going to say that his Hybrid is also bad and he has some bad resistor readings. 

the Hybrid that was on this board, a 100K ohm pull-up resistor was reading about 86K after I pulled the hybrid out of circuit. and before I heated everything up, it was around 4K. So I know this is an issue. 

So, curious. has anyone worked on this project yet? I dont want to do a whole bunch of work and re-invent the wheel if someone has already done this. 

And, would anyone want to pick up the ball once the design is tested/complete, and make assembled PCBs? Also, how big is the demand? Just curious. 

 
Last edited by a moderator:

maceffects

Well-known member
I'd be willing to buy some after produced.  I have a dead M5120 and my friend has a stack of dead ones I suspect are related to the hybrid.  I personally lack the design experience to do much, but I am happy to have things produced/distributed.  How big the market is questionable.  It could be quite large given how many issues these have, however, are people willing/able to spend the money is where I might have some doubt.  Though, to be fair, the price of the Mac Portables have been shooting up. 

 

androda

Well-known member
The repo I know of with information on the hybrid is here.  Might be the one you're thinking of. https://github.com/ppieczul/macintosh-portable

Makes me wonder whether the backlit portable's system could be reverse engineered and used instead.  Obviously we lack the schematic on the backlit portable so we can't immediately see what differences there are in the rest of the regulator circuit.  Seems like Apple wouldn't have much incentive to rework the whole power distribution system: only that which was necessary to get rid of the hybrid.

I'm pretty sure that my friend whose portable is afflicted with (at least) a bad hybrid would be interested.  That portable was the first computer he bought with his own money and he wants to preserve it.

So I'm willing to help where I can, with testing prototype hybrid boards and such.  We could go as far as making a hybrid test board which has the same downstream power components and a nice breakout of places to check voltages and signals, to connect the test versions to (and not risk blowing up real portables in the process).  Don't have a ton of PCB design experience, but that grows with time.

 

androda

Well-known member
As a next step to my suggestion about reusing the backlit portable's non-hybrid setup, I spent an hour or two tracing it out and recreating in kicad.  This comes from a backlit portable which I just recently replaced capacitors on.  Quite happily, it works fine now.

Here's a picture from my kicad work.  Note that U1M and U5L have totally the wrong symbols and look like resistor packs.  I'm pretty new to kicad and couldn't find better symbols for them.  Also, I'm totally unsure what the values of these resistor packs are.  Looking up their numbers didn't help.  Measured one in circuit and it came out 100k, so take that as you will.  Also note that D200 is an unknown dual diode, not sure of which way it points or whether it's a zener.

Fortunately, I did manage to identify several signal and power lines.  Signals found: A_D, HICHG, CHRG_ON, +12_GEN, -5_EN, HD_PWR

Need to spend some time comparing which signals the hybrid emits / uses against what I discovered on the backlit portable's board.

BacklitPortableHybridTrace.png

 
Last edited by a moderator:

LaPorta

Well-known member
This is all way out of my league, but I certainly would be willing to purchase one or more to future-proof my Portable.

From what you describe as the issue, I am now amazed and thankful mine survived, and I replaced the caps with polymer ones.

 
Last edited by a moderator:
Top