Baffled by Toshiba Mk5065GSXF readability issues

LarrySabo

Member
I don’t know if the drive was dropped or not. The partition table is not found on MacBook Pro from whence it came, or by DFL. Lots of bad sectors at beginning LBAs and it really struggles to read everywhere. Heads 2 & 3 fail head test and heads 0 and 1 struggle to read throughout drive -- in spite of pristine-clean and undamaged heads viewed under the microscope.

Swapped heads with two other donors and no change. Platters look clean but I was surprised at the amount of wobble. However, donors that test healthy have similar amounts of wobble. (You can see the reflection of the clean chamber filter screen oscillate as the platters are rotated.) Donors work normally with their heads reinstalled in their HDAs. An exactly matched donor just 5 weeks older than the patient won’t even calibrate. I tried the donor PCB with the patient ROM and no change.

The drive has been imaging with heads 2 and 3 disabled and it’s just inching along at 63% of maximum LBA. With ~30% or LBAs read, a raw scan of the clone provides many usable JPG files but not the couple of critical files. At the current rate, it will take another two weeks to complete the scan using just heads 0 and 1.

Anyone have any suggestions or know what the hell is going on? I thought maybe a platter swap might be informative but want to get as much data as possible before trying that, given my dismal record.
 

Sam

Member
Does DFL support building a virtual translator? That might help reading speed.

The donor heads producing the same results suggests highly that the heads are not at fault but if you try patient heads on one of the 2 compatible donors you tried it would verify that for good.

I haven't seen many 2.5" drives with any noticeable platter wobble. Are you spinning the platters with a torx driver inserted in the center screw?

Another way I've checked for spindle problems is by looking at lateral wobble. Rock the platter back & forth and watch how much the edge travels in between the parking ramp slots. Do it with a donor at the same time for comparison. I've had a couple cases--dropped Passports--where there was just a little more rock than there should be and a platter swap solved the case.

I've also seen faulty magnets--also on dropped Passports--where one of them had just barely started to "peel" off. A magnet swap solved those cases (but man it took a while to figure those out).
 

LarrySabo

Member
Thanks for the suggestions, Sam. Yes, I used virtual translator and set DE to automatically apply it whenever a reset occurs. I rotated the platters using tweezers in one of the cover cap holes, IIRC. Not sure it would be different if I just used the centre screw but will try that next time I have the cover off. Will check lateral wobble, too. I never thought about the magnets -- good suggestion. I'll try swapping them and see if that helps.

With heads 2 and 3 not reading almost at all, even with head swaps, it makes me think there may be ROM corruption at play (adaptives/DAC). Either that, or the wobble at the top platter has more effect than at bottom platter. Anyway, thanks for the suggestions. I'll update with any new findings.
 

pclab

Moderator
If I remember right, I had a cse once with a Toshiba that was slow reading and using terminal it was way faster. I will try to find it.

Enviado do meu MI 5s através do Tapatalk
 

LarrySabo

Member
Thanks, Nuno. I don't know how to read from the UA by the COM port on DFL-SRP and I don't think I want to invest the time to figure it out for this case. Maybe another time. The cloning is 65% through the drive reading from just 2 heads and was progressing at about 1% per day. that means it would take another month to finish that pass but I just cancelled it because of so many unreadable sectors. I'll try some of Sam's suggestions, or just take what I have so far and give up the case.

The link in your follow-up post just opens the following: "The requested topic does not exist."
 

LarrySabo

Member
Here is a video of the platter wobble I mentioned. I couldn't detect any wobble judging from the separation between the ramp and the platter, but there is clearly wobble there in both drives. Scanning the user area of the donor showed no issues at all -- it's a healthy drive.

View attachment Platter wobble, patient on left, good donor on right.zip
There is no wobble detected when the spindle is pushed/pulled so I think "wobble" is not the correct word to use. It seems to me the spindle axle is simply bent. The effect is the same, as far as the heads are concerned, because the platter surface raises and lowers as it rotates, which must make servo tracking a challenge.

I just swapped the magnets and no change at all.
 

Sam

Member
The downward pressure from the tweezers is creating the "wobble" effect you're seeing--I can see horizontal travel on the platter edge at the very beginning of the video before you even start to spin the platter.
If you were to spin the platter with a torx driver inserted into the screw in the middle I suspect you wouldn't see the wobble (unless the spindle IS actually bent).
You could check for lateral travel by using a second set of tweezers in the opposite hole you've got the first set in and alternate downward pressure. You'd be able to see the travel between the parking ramp slots.
 

LarrySabo

Member
You are right, Sam. I can see the platters deflect as I press down and release pressure with the awl, and pressing down alternately on opposite sides of the spindle with tweezers does make the platters visibly raise and lower when viewed at the ramp. So much for my platter wobble theory. :lol:

BTW, I swapped the patient heads to the donor HDA and a UA scan is horrible. A heads test comes up happy though, where they didn't in the patient. I think I need a bigger hammer. :roll:
 
Top