qubit

Board swap in MaXLine II (300GB)

Recommended Posts

Long story*, but I fried the PCBs on two Maxtor HDs. I was able to successfully recover the data on one of them (the older one: 160GB D540X) by buying the same model on eBay and swapping the boards. But the 300GB Maxline II is considerably rarer. In fact it's unselectable in StorageReview's reliability survey, to my chagrin. (I really don't understand why StorageReview won't put drives in the reliability survey that they haven't been formally reviewed.)

Although it showed up once on eBay, I passed up buying it because I wanted to wait and see if the board swap on the first one worked (I hadn't yet bought the needed hex screwdriver). But it hasn't shown up since then. The closest match since then has been the 250GB MaXLine II.

So I'm wondering, what are the chances that a board swap would work from a 250GB drive to my 300GB one, as long as both are MaXLine II? My drive is 5A300J00816R4 (Mfg. Date 29MAY2004), and below the power and IDE connectors it says "CADBBSMT", "D6FAA" and "1705404CMY ER"; presumably those relate to the PCB. I've heard the term "PCB uplevel" but I've never seen it defined... what is it, exactly?

Any help would be appreciated. (But please don't tell me I need to back up my data. I'm well aware of that.)

*I had sector 0 on my 80GB boot drive go bad. Windows hung in an infinite loop because it refused to time out trying to read the partition table of that drive. Linux spent a considerable time on bootup trying to read the table; it did time out, but it dealt with bad sectors very poorly. It insisted on reading data in 8-sector chunks even if I turned off readahead with hdparm, so if I had ghosted the partition using Linux, I'd have lost 8 sectors for every single isolated bad sector (there turned out to be three of them; two were in the NTFS $Logfile, and an old one was already in $Bad).

So it was back to Plan A, somehow getting Windows to boot so I could use WinHex to copy an image of the partition while logging bad sectors. Foolishly, I booted with the drive's power unplugged then hotplugged it after the boot finished. The system instantly powered down the moment the connectors even began to barely touch, and when I booted up after that, the drive that had been slave on the same interface made a repeated head-park sound. Amazingly, the "hotplugged" drive itself was unaffected.

On a later boot, another drive (the maxline) ceased to be recognized or spin up at all; the strange thing is, it was on an entirely different interface (the first one was connected to motherboard IDE, and the second was connected to my Promise card).

I realized that I should've disabled the Promise IDE card in Device Manager, rebooted, then re-enabled it after booting had finished. Eventually I did get the data off the 80GB drive with help from that method. I wish it had occurred to me before I tried the crazy idea of hotplugging.

Share this post


Link to post
Share on other sites

There is no difference in PCBA between different capacity sizes of the same model. I think the Maxline II was essentially a "Calypso" model, aka DiamondMax Plus 9. So you might also try swapping PCBA from one of those. There should be a revision # printed somewhere on the PCBA, ideally you'll want those to match.

Even with matching PCBAs, there's a chance the boot code flashed on the PCBA doesn't match the drive, in which case the drive might not spin up, or might click on the latch when you apply power. In that case, try another one.

Good luck, don't waste too much money on this :)

Share this post


Link to post
Share on other sites

Thanks for the information and advice, Tarjan. And the good luck. :)

A very good match just popped up on eBay, so I grabbed one. The "5A300J00816R4" is exactly the same and the manufacturing date is just 1 month earlier. If the PCB swap works, I can use it (swapped back) as a receptacle for the recovered partition! (Which will be convenient, because there's no loopback mounting in Windows. And making an image will be faster than copying the files — which is important if the drive catches fire after a certain amount of time.)

But if that hadn't popped up today, I probably would've gone for a 250GB model on your recommendation. :)

Edited by qubit

Share this post


Link to post
Share on other sites

Tarjan: "I think the Maxline II was essentially a "Calypso" model, aka DiamondMax Plus 9."

MaXLine +II and DiamondMax +9 should be identical with different in-factory testing. MaXLine II and DMax +9 not.

MaXLine II is a 5400rpm drive with ball-bearings, MaXLine +II is a 7200rpm drive with fluid-dynamic bearings. These drives have totally different spindle motors, so I doubt their PCBs (and thus spindle motor controller chips) are interchangable. I don't suggest overclocking the harddrives. :P

Tarjan, you should give a little more though on what suggestions you are about to give other people. That one was quite a bad one, IMO.

qubit: "But if that hadn't popped up today, I probably would've gone for a 250GB model on your recommendation."

It a good thing you didn't. You PCB and/or spindle motor and/or R/W head and/or signal amplifier might be up the smoke if you did try DMax +9 250GB's PCB on a MaXLine II 300GB. They are mechanically very different so I see a high probability of some extra damage to occur if a PCB swap was tried. Normally a PCB swap wouldn't harm the PCB or the drive (*), just remain unreadable, but in this case I wouldn't be that certain about it.

(*) Assuming of course that it's just a firmware mismatch and the heads don't receive a static shock during swapping operation.

Edited by whiic

Share this post


Link to post
Share on other sites
It a good thing you didn't. You PCB and/or spindle motor and/or R/W head and/or signal amplifier might be up the smoke if you did try DMax +9 250GB's PCB on a MaXLine II 300GB. They are mechanically very different so I see a high probability of some extra damage to occur if a PCB swap was tried. Normally a PCB swap wouldn't harm the PCB or the drive (*), just remain unreadable, but in this case I wouldn't be that certain about it.
I disagreed with Tarjan about the DMax +9 and MaXLine II being the same or even similar. What I meant is that I might've tried a 250GB MaXLine II.

whiic, do you agree with Tarjan that "There is no difference in PCBA between different capacity sizes of the same model," in the case of the MaXLine II in particular?

Edited by qubit

Share this post


Link to post
Share on other sites

"do you agree with Tarjan that "There is no difference in PCBA between different capacity sizes of the same model," in the case of the MaXLine II in particular?"

I wouldn't bet on it either way. If 250GB variant is created by disabling some of the heads of a 300GB variant, couldn't using a 250GB variant's PCB on a 300GB variant give you access to a limited amount of capacity? The lost sectors would also distribute quite evenly over the original accessable space, so not much data to be recoverable.

Also, platter densities may vary between variants. 250GB MaXLine II might be just 3 platter, 6 heads @83GB/platter where as 300GB MaXLine II would be 75GB/platter. Maxtor annouced a 320GB MaXLine II (80GB/platter) but they had trouble getting it to production.

(Or, MaXLine II 250GB might be a 4 platter, 7 heads version, with 75GB platters short-stroked from 262GB to 250GB. I really don't know.)

Share this post


Link to post
Share on other sites
If 250GB variant is created by disabling some of the heads of a 300GB variant, couldn't using a 250GB variant's PCB on a 300GB variant give you access to a limited amount of capacity? The lost sectors would also distribute quite evenly over the original accessable space, so not much data to be recoverable.
Hmm. I guess I sorta had the idea that the PCB might read data on the drive's geometry (along with most of the firmware) from its own head/surface 0. But your supposition is perhaps more likely, because a drive manufacturer or refurbisher might want to mark head 0 to be skipped.

Has anyone ever tried swapping PCBs from different-capacity models in the same line?

Share this post


Link to post
Share on other sites

Sadly, the board swap didn't work on the MaXLine II. The eBay drive worked fine, and after the board swap it didn't spin up and wasn't recognized at all (which is exactly what should've happened). My drive tried to spin up, but instead made weird noises (I couldn't hear them too well because of all the fans in my case) and not only prevented its paired master from being detected, but both drives on the other onboard IDE channel as well!

My hypothesis at this point is that the board has to be either exactly the same, or slightly newer so that it's backwards compatible. The Maxtor 4G160J8 swap worked. The drive from eBay I used for that swap was newer and had a newer PCB code. However, the 5A300J0 from eBay has a mfg date 44 days older than mine. Of course, that's only two data points so far...

Share this post


Link to post
Share on other sites

Just an idea: the death of your MaXLine II might not be related to the incident that killed the drive on another IDE channel.

Your MaXLine doesn't spin up using either PCB. The new PCB might work with the drive... I wouldn't be certain that there's anything bad with the old PCB (though it is perfectly possible). Does it make the "weird noises" with both PCBs? If yes, the spindle might be stuck. Weird noises may come from spindle motor trying to spin up and/or actuator trying to vibrate the R/W head to break the stiction. They might be ticking, beeping, whining, etc. noises.

You could try to twist the drive in your hand while you power up the system. That way you add the your twisting force and inertia of spindle & platter assembly to HDD own attempt to break the resisting force. This is of course highly non-recommended for working drives but unless you're ready to spend much money for data recovery specialists, you've got nothing to lose.

Share this post


Link to post
Share on other sites
Your MaXLine doesn't spin up using either PCB. The new PCB might work with the drive... I wouldn't be certain that there's anything bad with the old PCB (though it is perfectly possible). Does it make the "weird noises" with both PCBs?
With the original PCB it is completely unresponsive. It's as if the IDE cable is plugged into nothing. No sound at all, and the BIOS acts as if it's not there. I even tried plugging it into a power cable from a different computer, and put my ear close to it, but heard nothing (and felt no vibration). The eBay drive acts the same with that PCB in it.

With the PCB swap, the behavior is entirely different. The BIOS pauses for detection while the drive tries to spin up, but instead of spinning up it makes weird noises. Eventually, the BIOS detects nothing at all in any of the four IDE "slots". But with the drive unplugged, the BIOS detected the drives on the other three slots just fine.

I'm afraid that after those "weird noises", the heads were not parked... I'm afraid to move that drive any more than necessary. Is my fear justified?

How hard would it be to figure out what part of my drive's original PCB was fried, and to swap/resolder only those ICs that need it? And what tools would I need for that?

Edited by qubit

Share this post


Link to post
Share on other sites

OK. So maybe the old PCB received a shot bad enough to damage it and it died during next reboot. (Was it a hard or soft reboot? Did you turn the power off, i.e spin down the disks?)

The new PCB hasn't been damaged by the incident, so it tries to spin it up. Either the PCB is incompatible or the spindle is stuck for some reason (stiction, jammed bearings, whatever). If it was a bad head (possibly damaged by the electrical shock), could that prevent the spindle from starting? Because wouldn't that usually cause a "click-of-death" (i.e spindle spinning, actuator moving back and forth recalibrating itself to no avail).

Why would both the PCB and the spindle die at the same time? Spindle motor shouldn't be as sensitive to electrical shocks as head are, so being connected to a PCB while PCB receives a shock would seem an unlikely cause for spindle not spinning up. So, my guess would be that the PCB was damaged and due to the damage, it was unable to park the heads when the power was removed. There's two separate ways of parking the head:

1) by host issuing STAND-BY or SLEEP command the the HDD and HDD controller responding to the command by parking the head.

2) power is removed from the HDD: most PCB logic falls asleep but some kinetic energy is still stored to the spinning spindle assembly. That rotational energy is transfered back to electricity when no electricity is received from PSU. This energy is used to move the actuator to parking position. (I don't know whether it includes logic or whether it's just constant voltage applied to actuator wirings. Using logic would seem more friendly but would require part of the logic chips to be powered by the spindle "dynamo".)

If during the last power down the computer didn't issue a STAND-BY or SLEEP command just prior to power down(*), AND the automatic head parking with spindle motor residual energy failed, then your head are not parked. It would also explain the spindle not spinning up. If heads land outside landing zone (where the surface is rough**), the spindle might get stuck.

(*) Issuing either of the two commands to stop the spindle stop is sometimes recommended by HDD manufacturers (like Hitachi) as this ensures write cache is empty when powered down. Also it might reduces wear and tear.

I don't know if computer manufacturers follow their wishes but at least some systems spin down the HDDs prior to powering down. Spindle stop command could be issued by the OS or by the hardware (motherboard).

(**) Unlike with airplanes, heads have to land on a rough surface. The polished part of the platter, where data is resided, is so smooth that if the heads landed there, the heads would become stuck as there's no air between it and the sliders. Capillary action drawing protective lubricant into any voids below the slider would probably make things even worse. Well, I don't know the physics related to the stiction.

I just know that in order to spin the platters up you'd need extra force. Maybe even more force than a perfectly functional PCB could offer. That's why I suggested twisting the HDD around spindle motor axis. If you twist it around any other axis, that's OK too because the heads because even sideways forces would help to break the stiction. I wouldn't intentionally twist it around actuator axis as actuator is a light assembly and twisting around it wouldn't likely be of any help. Spindle motor assembly is much heavier. (Also, if you manage to break the stiction, it's better to have the platters move below the heads than the heads move above the platters. Sideways motion over a immobile platter could cause scratching.)

But the most important thing is: power has to be applied to the HDD when twisting it. Otherwise you might unstick it, but when the platters come to a stop a second later, they'd be stuck again. You'd want it to continue spinning and back-up the contents. You need power. And you need the data cable also, because while the heads not parking might be caused by a bad PCB, that's not certain. You wouldn't like the HDD to spin up, then power down, and when powering up the next time with data cable attached, notice it's stuck again.

IMO, you've got nothing to lose if you aren't going to contact a professional data recovery service. You might as well try. Twisting the HDD wouldn't harm the functioning PCB. It could damage the media but what good is a non-scratched media if the platters aren't spinning? Without some expensive data recovery specialists: nothing.

So my suggestion is, twist the dead drive unless the data on the platters are very important.

____________________

But before you start twisting the HDD, describe the noise the drive makes. Does it sound like a phazor noise? Something from Star Trek?

A sound sample: phazor noise

One of the several previous discussions.

One PDF on stiction.

Share this post


Link to post
Share on other sites

The PDF suggests not to connect the data cable. This is undestandable suggestion because it's easier to twist the drive with less cable attached. There is a possibility to use power wires much longer that the data cable and power wires are thinner as well, making twisting much easier.

But for this being of any use, it's required that the next time it spins up, it also parks the head when it spins down. Otherwise it wasn't of any use.

But if the new PCB is identical enough and working OK, parking the heads are unlikely a problem anymore.

Share this post


Link to post
Share on other sites

Wow... thanks, whiic. I appreciate your providing me with all that information. I must admit it's not what I wanted to hear. I had my hopes set on the data recovery being a simple matter of finding a close enough match for the PCB, and other posts I've read indicated to me that finding a perfect match is a difficult issue for some drives.

On the other hand, even before your July 27th post, I was afraid that my using a mismatched PCB had done something weird to the drive and left it unparked. But I thought that parking a drive was merely a safety precaution to protect from bumps and during transport; I merely feared that I would have to handle the drive very gingerly when swapping its PCB a second time and remounting it. Reading your post suggests to me that (with modern drives at least) parking is absolutely required to prevent data loss and head stiction. Am I correct in this interpretation?

If the heads are currently unparked (for whatever reason) does that absolutely mean that some kind of mechanical intervention will be needed to force-start it, and platters will be inevitably scratched?

The weird sound my PCB-swapped Maxline II makes doesn't sound at all like the phazor sound effect you linked. It's much quieter, and as far as I remember it was an intermittant, fairly constant-pitched high-pitched tone, maybe with a little variety thrown in. My case is very noisy so I will have to isolate that drive and plug it into a different case to get a good recording... and I'm very afraid of moving the drive around more than necessary. I'm also afraid that a second attempted spin-up with the possibly mismatched PCB will damage it even more. Right now, I have it unplugged but still mounted in my case, and removal is a two step process (cage, then drive).

How sure are you that a mismatched PCB isn't the issue, and why? It still seems likely to me that that's the issue rather than stiction or an unparked head. Even with the previous Maxtor PCB-swap that worked for me, I think there was a slight incompatibility... when I wiped the drive with randomness (after recovering all the data on it), it developed hundreds of somewhat-evenly-distributed bad sectors and then proceeded to do so again after I repeated the process. (I expected it to find a few bad sectors and remap them, and then not develop any more for a while. That's how it behaved before the PCB swap.)

Best regards,

David

Share this post


Link to post
Share on other sites

P.S. Note that buying another Maxtor 5A300J0 on eBay (this time trying to get a more perfect match) is still an option for me...

Share this post


Link to post
Share on other sites

I bought another 300GB MaXLine II on ebay. I tried to first ask the seller what the PCB version was (on the sticker underneath the IDE connector) but got no response. I bought it anyway. (What I did know was that it had a manufacture date 2 months later than my original drive, which was promising — the PCB swap that didn't work had a date 1 month earlier.)

I was dismayed to find that this second ebay drive, just like the first, had "D4FAA" as the PCB version (at least that's what I'm guessing it is). It's weird that my original drive would be "D6FAA" although chronologically between the two in terms of manufacture date.

Anyway, the result was better with this second PCB swap; the drive spins up fine, and when I power down, it seems to park its heads. However when the BIOS tries to detect it, and later when the operating system also tries, I hear a click that sounds sort of like a head park.

Here are audio files, with background, fan and motor noise filtered out:

access attempt by BIOS

access attempt by BIOS - triple click

access attempt by Windows

At this point I have several mutually exclusive theories...

  • The access attempt during the first PCB swap damaged the drive, and it would have worked fine if I'd done the PCB swap with the 2nd ebay drive to begin with. It's not stiction, because the drive spins up. That leaves platter scratches as a possibility.
  • The PCB version (or what I think is such) is irrelevent; each PCB is programmed with specific parameters tailored to the drive to which it is coupled, and it so happens that the 2nd ebay drive matches mine more closely, but it's still not close enough.
  • A matching PCB version of D6FAA would work. (But then why did the newer D4FAA board work better than the older D4FAA board?)

At this point I'm again thinking about probing the fried D6FAA board to see what part of it is damaged, and pirating parts from one of the other boards to fix it (by resoldering). Does anyone have any thoughts on this?

(Note that both ebay drives work perfectly with their own PCBs. I haven't tried swapping the boards between them... admittedly that would provide more clues, but it could also ruin those perfectly good drives.)

Edited by qubit

Share this post


Link to post
Share on other sites

"I haven't tried swapping the boards between them... admittedly that would provide more clues, but it could also ruin those perfectly good drives."

Desoldering & resoldering PCB components from one board to another al also quite risky.

Yep... those audio files sound pretty much the same as my first MaXLine II produced when trying to access bad sectors. If it clicks when Windows, then BIOS has recognized it, eventually. But is it recognized as MaXLine or by the factory nickname? (i.e is there firmware corruption)

Do you see anything in Disk Management? Unallocated space? Any errors logged in Event Viewer?

Maybe you have bad sectors over the boot sector, partition table, FAT or MBR, that causes read attempts to damaged sectors every time you boot your system.

If Windows doesn't recognize the contents of the disk, maybe you could clone it sector-by-sector to a working one. Then possibly try to recreate boot sectors, etc. with software file recovery tools. (I.e repairing the copy on a good drive. Do not attempt to write anything to the damaged drive.)

Share this post


Link to post
Share on other sites
Yep... those audio files sound pretty much the same as my first MaXLine II produced when trying to access bad sectors. If it clicks when Windows, then BIOS has recognized it, eventually. But is it recognized as MaXLine or by the factory nickname? (i.e is there firmware corruption)
Oof, I forgot to say that it's not detected. At all. The only sign that it's plugged into the IDE cable is that when the BIOS or OS try to detect it, it makes that sound.

From what I know of hard drives, that could mean that there's a "bad sector" on a critical area of the drive, like in the firmware or sector mapping tables stored magnetically in the drive, right?

But it could also mean that the swapped PCB still doesn't match the drive's setup in some way... maybe it's trying to load that information from the wrong physical sectors (and by physical I mean truly physical, not LBA).

Share this post


Link to post
Share on other sites

Or maybe it's looking for the firmware / mapping tables in the right place, but trying to parse them in the wrong format... who knows what kinds of things Maxtor may have changed between PCB revisions. :-\

About the detection thing... I also have a seven year old IBM Ultrastar DMVS-36D that no longer spins up (it tries but fails). Interestingly, it is still detected, but as "DMVS" instead of "DMVS-36D" (and of course the sectors are inaccessable). I wonder why the Maxtor doesn't do something similar.

Share this post


Link to post
Share on other sites

You could try the PCB #1 and #2 now that you managed to spin up the drive. At least it not stuck (anymore). It's unlikely though that it will help but there's a slight chance.

"Oof, I forgot to say that it's not detected. At all. The only sign that it's plugged into the IDE cable is that when the BIOS or OS try to detect it, it makes that sound."

If it's not detected by BIOS, there's not much possibility to recover anything by software measures.

How did your Deathstar data recovery project succeed eventually?

Share this post


Link to post
Share on other sites
You could try the PCB #1 and #2 now that you managed to spin up the drive. At least it not stuck (anymore). It's unlikely though that it will help but there's a slight chance.
PCB #2 is the one that managed to spin up the drive... I strongly suspect that PCB #1 would still just make a soft high-pitched whine like it did before. Of course the scientific thing to do would be to try #1 again just to see if it does do as before... and I could record + post the sound.
If it's not detected by BIOS, there's not much possibility to recover anything by software measures.

How did your Deathstar data recovery project succeed eventually?

Exactly... I was only able to recover data from my Deathstar because it was still detected and accessable.

Thanks for wishing me luck before :D My recovery effort panned out beautifully. I was able to reverse engineer the drive's RLL and ECC schemes, and combining this with filling in data from context was enough to recover a good amount of data. I still haven't recovered all the bad sectors that are recoverable and important to me... so I've kept the ghosted partition online, un-resized and un-defragmented to allow me continue the project when I want. It's the only FAT partition I still actively use (not counting flash cards)... it's not a carbon copy of the bad drive anymore, but unchanged files still occupy the same clusters.

I find it very unfortunate that when the ATA spec was extended to use 48 addressing bits, READ LONG and WRITE LONG were not included. There are no 48-bit versions of those commands. So I wouldn't be able do the same thing with a >128GiB drive, sadly.

Share this post


Link to post
Share on other sites

"PCB #2 is the one that managed to spin up the drive"

I meant that #1 was the original, #2 was the first swap and #3 was the one that managed to spin it up.

Why I think trying them again could have a small probably of success is because IF it somehow didn't manage to park the head and they are not parked, the original or replacement #1 could work. Why I'd estimate the likelyhood of working to be very small, is because I don't believe there's much difference in how much torque the spindle motor can create. If you didn't twist the HDA while applying power to the drive, the likelyhood that there was stiction present is very low.

Switching back to PCBs #1 and #2 (assuming the current one is #3) is not the first thing to do. I'd try to find as many computers as possible and try to get it recognized plugged into different motherboards and PCI controllers. The head are easily damaged due to electric shock, so I would try to avoid swapping PCBs any more than necessary.

Share this post


Link to post
Share on other sites
I meant that #1 was the original, #2 was the first swap and #3 was the one that managed to spin it up.
Oh! I guess I thought of it is "#0" without really realizing it ;)

I did try the original again, and it behaved the same as before. It's like it's not even receiving power. (I'd do some simple tests with my multimeter, but it has a fuse I need to replace and I keep forgetting to do so.)

I'd try to find as many computers as possible and try to get it recognized plugged into different motherboards and PCI controllers.
You've got a point. The fact that the detection attempt has some effect could mean that a different ATA controller may report what's going on instead of simply coming up empty. But I'm very pessimistic about the chances of actually recovering data that way... I can't think of any reason it could work.

Do you think it'd be safe to scan both sides of a PCB in a flatbed scanner? I'd like to scan two and compare them. I'm pretty sure it'd be safe if I put the PCB inside a hard drive bag, but that would tint and degrade the scan.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now