What's new

Preparing a Hikaru BGA Chip for Re-installation

I've been trying my hand at BGA rework, not Hikaru hardware but Naomi which is pretty similar.

jovy.jpg

Modified Jovy 7500.

microscope.jpg

Microscope camera to watch the chip drop into place. :D

reballed.jpg
 
Last edited:
How is there no trace of solder on the chip at all, that's crazy. It was obviously never soldered properly from the factory =O, shame you're in the states that would be a fun one.
 
Ace` said:
How is there no trace of solder on the chip at all, that's crazy. It was obviously never soldered properly from the factory
Click to expand...
that is 100% the problem with every Hikaru produced as far as I know.
from what I've heard In production they failed to properly prep the chip surfaces so the chip separates from the solder quite easily.

I think when mounted inside of a cab and shipped they're probably find because any shock from the cab being rolled over bumps isn't going to cause shear stress across the BGA solder, but when the PCBs are removed from the cab and shipped, all bets are off, if it's even set down roughly on it's side that giant heat-sync is going to rip the chip right off of the board from the sheering force.

most of the survivors I've seen have had GOBS of hot-glue poured around all of the chips.
 
twistedsymphony said:
from what I've heard In production they failed to properly prep the chip surfaces so the chip separates from the solder quite easily.
Click to expand...

Yup ^ oxidation of pads, lack of heatsink structural support, concussive damage... this is stuff Ken has mentioned in various posts over at KLOV.

1746219522275.png

1746219611841.png

1746219655979.png
 
I'd love to have a hikaru someday.
But someone has to be willing to dedicate a setup to fixing them, and then a multi is viable.
I am certain there's money in it, there's more parts boards than working ones laying around.
My concern is how many got thrown away.
 
Just to follow up on this, the conclusion is, if you reballed all the ics, which tbh is quite a bit of work, then Hikaru could be slightly mishandled like naomi and still be rock solid?
 
They didnt wet the pads and used the wrong type of solder or soemthing like this. Meaning the bond for the solder is barely holding on. This is why bumps have the potential to break these solder connections on most if not all of these that have not been reballed. Leaded solder is more ideal but even leadfree would yeild better results than the factories mess up.

Ken told me that sega put out a memo when they just came out saying they are aware all the boards were defective and to RMA them until extra spare boards ran out of stock and sega wasnt planning to rerun a new production of boards. AKA they immediately gave up on the platform after the factory mistake or they only built to order with the games they sold and expected to sell and then gave up after finding out.
 
GeekMan1222 said:
They didnt wet the pads and used the wrong type of solder or soemthing like this. Meaning the bond for the solder is barely holding on. This is why bumps have the potential to break these solder connections on most if not all of these that have not been reballed. Leaded solder is more ideal but even leadfree would yeild better results than the factories mess up.

Ken told me that sega put out a memo when they just came out saying they are aware all the boards were defective and to RMA them until extra spare boards ran out of stock and sega wasnt planning to rerun a new production of boards. AKA they immediately gave up on the platform after the factory mistake or they only built to order with the games they sold and expected to sell and then gave up after finding out.
Click to expand...
I didn't know the whole story. Crazy mistakes....
 
Here's an idea.
If they all go pop, can we just create a better pcb?
Maybe just ship the chips and get a new hikaru back.
 
The chips are manufactured with copper pads. There are two ways to go about manufacturing the pcbs:

1. Ideally the bga chips are pre-balled. This means the copper pads are tinned by default when the balls are added.

2. You can ball the pcb itself and install the chip on top of the pre-balled footprint before going in the oven. This would require the chip AT LEAST be tinned. That didnt happen.

Method two is probably NEVER done but seems to be what happened with Hikaru but for whatever reason the chips were never tinned. Even if you have an undamaged chip it is very hard to tin it because of the years of oxidation on the copper.
 
You must log in or register to reply here.
Back
Top