Page 1 of 1
WD 4 tb with bad heads eats new heads, hot swap?
Posted: 16.07.16, 01:18
by osity
I got a clicking WD 4tb that had bad heads. Once I got the donor heads in the drive worked and had ID for about 5 minutes but then started clicking again after trying DE. New heads are dead now.
Second donor has arrived but I'm concerned that I'll experience the same results after install. Rather than just killing more donors heads, Is this an opportunity for a hot swap or smart hot swap?
Re: WD 4 tb with bad heads eats new heads, hot swap?
Posted: 18.07.16, 19:25
by DataMedics
If you backed up all tracks & modules, you can certainly give it a try. In some cases you can still read some heads that way. Just be sure to edit the headmap in RAM after you spin it up, so you're reading just one head at a time. These newer ones tend to be problematic when not all heads are functional.
Re: WD 4 tb with bad heads eats new heads, hot swap?
Posted: 18.07.16, 22:26
by osity
try a hot swap or a smart hot swap?
I tried a hot swap and when i get to the microjogs sections where you average module 47 the donor doesnt show any microjogs... all the heads show 0x0000.
Do i still continue with make averaged? Something doesn't seem right...
Re: WD 4 tb with bad heads eats new heads, hot swap?
Posted: 19.07.16, 14:08
by DataMedics
I really wouldn't be worrying at all about microjogs, that's the least of your concerns. You should try smart hotswap. Write all tracks to the donor drive, put patient PCB onto it. After it ID's edit headmap in RAM, put it to sleep, and transfer it over. Then see if you can read sectors from whatever head you leave enabled.
However I wouldn't be surprised if it doesn't work. I haven't had a great deal of success smart hot swapping WD 4Tb's.
Re: WD 4 tb with bad heads eats new heads, hot swap?
Posted: 19.07.16, 16:22
by AJ2008
Most new WD drives have adaptive values inside ROM module 4F.
Take a known working 4TB, and write 4F from donor of same version. If it continues to ID and remain accessible after repowering, then there are no servo tables in 4F. But if access and identification is lost, then hotswap wont be possible.
Re: WD 4 tb with bad heads eats new heads, hot swap?
Posted: 19.07.16, 16:58
by osity
Data Medics: I can't write patient tracks onto donor. The patient is clicking.
AJ2008: The donor drive's ID continues to be accessible after writing 4F to it from another drive.
Re: WD 4 tb with bad heads eats new heads, hot swap?
Posted: 19.07.16, 17:41
by DataMedics
Didn't you back up the modules & tracks during that 5 min after it ID'd? Should always be the first thing you do after you get a drive working.
Re: WD 4 tb with bad heads eats new heads, hot swap?
Posted: 19.07.16, 18:40
by osity
I backed up the SA modules and ROM.
I always skip backing up the the tracks on all my jobs/drives because of the excessive time it takes to do (and the additional stress on drive). Do you back up the tracks on all your jobs?
Re: WD 4 tb with bad heads eats new heads, hot swap?
Posted: 20.07.16, 09:21
by AJ2008
What is SA Directory address at patient? Set this same on donor. Rewrite donor SA to new location and boot donor from here. Now, make straight hotswap and try to read SA modules from patient SA. Maybe this will not work, but without patient adaptives it will be impossible to get data access following hotswap process.
Did you adjust head map on RAM of patient to check if one of the service heads will read?
Re: WD 4 tb with bad heads eats new heads, hot swap?
Posted: 20.07.16, 09:26
by AJ2008
osity wrote:Source of the post I backed up the SA modules and ROM.
I always skip backing up the the tracks on all my jobs/drives because of the excessive time it takes to do (and the additional stress on drive). Do you back up the tracks on all your jobs?
I did not read this post fully...
If you have patient resources then use you can attempt hotswap, but you should change donor DIR address to match patient. When donor is ready you can make hotswap with `03' '0C' '31' '36' '4x' modules from patient
Re: WD 4 tb with bad heads eats new heads, hot swap?
Posted: 20.07.16, 18:57
by osity
Sadly after not getting anywhere with the how swap procedure on Ace's site I tried to write back everything and now my donor spins up... but then just spins down.
it seems like Ace's article is actually a "smart hot swap" and not a "HOT SWAP" as the article is titled.
Re: WD 4 tb with bad heads eats new heads, hot swap?
Posted: 20.07.16, 19:13
by DataMedics
osity wrote:Source of the post I backed up the SA modules and ROM.
I always skip backing up the the tracks on all my jobs/drives because of the excessive time it takes to do (and the additional stress on drive). Do you back up the tracks on all your jobs?
I do usually backup tracks, but usually by composite instead of individual heads. I do it just for the convenience of smart hot swapping should the need arise later. Coincidentally I've got a 3Tb which is in the same boat as yours which I'm going to be attempting that later today. Fortunately I did back everything up before it started clicking and spinning down again.
You can do it by modules instead of tracks, just more of a hassle as the DIR module rarely matches. So if you write all modules at once, some will be stored in the wrong places. As AJ2008 referenced, you need to first make sure that the ROM references the same DIR location in both drives (or change it if it doesn't). Write the new DIR module to the drive, restart it, restart the utility, possibly do LDR & DIR upload (using the patient DIR), then write all modules. It can get messy, which is why tracks are easier, you only have to adjust the DIR location in ROM sometimes.