Depends.

IDE drives, usually no.

MFM drives? In theory, yes.

wierd_w wrote on 2025-02-24, 18:24:

Depends.

IDE drives, usually no.

MFM drives? In theory, yes.

I was afraid that might be the case. Mine are IDE/SCSI drives. The drives got noisy not because of wear, but because they were in storage for ages. Is there a chance they will clean up by themselves, or should I just move them back to storage for emergency use? I'm afraid most drives of this age have this problem.

I didn't get very far with MFM drives either. Or anything early SCSI or IDE that's based on MFM/RLL tech. Even those drives usually have the bearing either inaccessible (unless you open it and remove the platters - NOT recommended), or in the few cases where it's sort-of possible (ST-225 IIRC) the bearing is protected by dust cover and only very thin oil can, eventually, seep into it. And that's not a good lubricant for these bearings, but better than having it dried out I guess.

Most loud HDDs are like that because the main spindle bearing(s) is shot already. It would require replacement to fix it, not really possible without clean room, experience, probably a small press and obviously spare parts. Packing the old bearing with fresh grease (after cleaning it out) is more likely but would require all of the mentioned things except spares, and quite frankly would be a short-tem fix only. And there's other problems these old clunkers can develop like sticky heads (actually, platters).

TL;DR: Keep using it until it's dead. Unless it's so loud you can't stand it - I have a specimen like that, might be still usable inside a case so that's why I'm keeping it for now.

Deunan wrote on 2025-02-24, 18:38:

I didn't get very far with MFM drives either. Or anything early SCSI or IDE that's based on MFM/RLL tech. Even those drives usually have the bearing either inaccessible (unless you open it and remove the platters - NOT recommended), or in the few cases where it's sort-of possible (ST-225 IIRC) the bearing is protected by dust cover and only very thin oil can, eventually, seep into it. And that's not a good lubricant for these bearings, but better than having it dried out I guess.

Most loud HDDs are like that because the main spindle bearing(s) is shot already. It would require replacement to fix it, not really possible without clean room, experience, probably a small press and obviously spare parts. Packing the old bearing with fresh grease (after cleaning it out) is more likely but would require all of the mentioned things except spares, and quite frankly would be a short-tem fix only. And there's other problems these old clunkers can develop like sticky heads (actually, platters).

TL;DR: Keep using it until it's dead. Unless it's so loud you can't stand it - I have a specimen like that, might be still usable inside a case so that's why I'm keeping it for now.

I have three specimens like that, and when I say loud, I mean about 50-60 decibels.

I think this last 40Gb Maxtor drive either got more silent, or my ears got used to it, but the computer has a particularly noisy CPU fan (old Intel boxed socket 370 fan), and with the HDD on it I couldn't hear the fan. Now I kind of can. But the sound is extremely unpleasant high-pitched grinding.

Deunan wrote on 2025-02-24, 18:38:

I didn't get very far with MFM drives either. Or anything early SCSI or IDE that's based on MFM/RLL tech. Even those drives usually have the bearing either inaccessible (unless you open it and remove the platters - NOT recommended)...

I've had a bit of luck using a fine Teflon based lubricant - you used to be able to get a tiny tube of it
a "Radio Shack" and I still have a couple...

A lot depends on how accessible the bearing is ... if it is, sometimes you have to
remove a circuit board and/or a sticker/dust cover .. but if you can find the end of
the spindle, a small amount of lubricant, let sit for a day, a small amount more..
then run the drive for a few days upside down (assuming the spindle end is on the
bottom - the idea being to use gravity to help a little to draw the lubricant into the
bearing).

No idea how "real" this technique is, but it's what I did on a couple occasions where
the drive did actually improve...

On the subject of "opening" drives -- many years ago I had a much older (but small capacity
drive - so internal parts would have been bigger) that developed "stiction" ... it just wouldn't
spin up sometimes ... over time it got a bit worse, and I kept it loose but connected just outside
the system.. I could often make it work by rotating it back and forth about the vertical axis a
few times...

Eventually it got so bad that it was unusable ... I decided "hey this thing is trash anyway -
why not open it and see what it's inside" - which I did - the head was suck in place... it
took surprising little force to free it, and you could see a "spot" on the platter where it has
stuck .. still in the "it trash anyway" mindset, I polished the platter, put it back together
and wanted to see "how bad my damage would affect it".

Very surprisingly - it spun up, I didn't hear anything crunching on the platter/head, so
I tried formatting it - which worked! - and I used that drive in a very non-critical role
for a couple more years ... but just to be clear, I DON'T RECOMMEND DISMANTLING
A DRIVE - I'm sure newer drives with far smaller construction inside would have no chance
of surviving!

Dave ::: https://dunfield.themindfactory.com ::: "Daves Old Computers"->Personal

Wouldn't recommend to lubricate the spindle motor bearings either. Maybe if you have some sort of clean room, a steady hand and a lot of time/patience.

The only thing I ever lubricated on a hard drive, was the stepper motor.

The need for clean room for hard drives is overrated unless you're dealing with critical data on modern large capacity drives. Any dust, unless it's in excessive amounts, that falls on the platter will be pushed by the centrifugal force and caught by the dust filters on the edges of the drive, at least in relatively modern (IDE) drives. Hard drives are not airtight and they are not dustproof. I have opened a hard drive to fix a stuck head (on older drives the rubber around the stopper tends to turn sticky) more than once, with positive results. I was surprised to find that data recovery people with very good track record mostly do not use a clean room either.

songoffall wrote on 2025-02-24, 20:50:

The need for clean room for hard drives is overrated unless you're dealing with critical data on modern large capacity drives. Any dust, unless it's in excessive amounts, that falls on the platter will be pushed by the centrifugal force and caught by the dust filters on the edges of the drive, at least in relatively modern drives. Hard drives are not airtight and they are not dustproof. I have opened a hard drive to fix a stuck head (on older drives the rubber around the stopper tends to turn sticky) more than once, with positive results. I was surprised to find that data recovery people with very good track record mostly do not use a clean room either.

Second that. The hard drive is already dieing so nothing is lost trying.

I've revived some striction this way, the drives affected still run fine.

Be careful with the lubricant, if it touches the surface the data area is gone.

waterbeesje wrote on 2025-02-25, 06:58:

Second that. The hard drive is already dieing so nothing is lost trying. […]
Show full quote

songoffall wrote on 2025-02-24, 20:50:

The need for clean room for hard drives is overrated unless you're dealing with critical data on modern large capacity drives. Any dust, unless it's in excessive amounts, that falls on the platter will be pushed by the centrifugal force and caught by the dust filters on the edges of the drive, at least in relatively modern drives. Hard drives are not airtight and they are not dustproof. I have opened a hard drive to fix a stuck head (on older drives the rubber around the stopper tends to turn sticky) more than once, with positive results. I was surprised to find that data recovery people with very good track record mostly do not use a clean room either.

Second that. The hard drive is already dieing so nothing is lost trying.

I've revived some striction this way, the drives affected still run fine.

Be careful with the lubricant, if it touches the surface the data area is gone.

That's what I'm worried about - getting liquid on the platter. But I was thinking of a different solution. I have some hard drives where the spindle motor is fine, but the platters or the heads have degraded. Maybe I could just transplant the spindle motor from them?

songoffall wrote on 2025-02-25, 07:28:

waterbeesje wrote on 2025-02-25, 06:58:

Second that. The hard drive is already dieing so nothing is lost trying. […]
Show full quote

songoffall wrote on 2025-02-24, 20:50:

The need for clean room for hard drives is overrated unless you're dealing with critical data on modern large capacity drives. Any dust, unless it's in excessive amounts, that falls on the platter will be pushed by the centrifugal force and caught by the dust filters on the edges of the drive, at least in relatively modern drives. Hard drives are not airtight and they are not dustproof. I have opened a hard drive to fix a stuck head (on older drives the rubber around the stopper tends to turn sticky) more than once, with positive results. I was surprised to find that data recovery people with very good track record mostly do not use a clean room either.

Second that. The hard drive is already dieing so nothing is lost trying.

I've revived some striction this way, the drives affected still run fine.

Be careful with the lubricant, if it touches the surface the data area is gone.

That's what I'm worried about - getting liquid on the platter. But I was thinking of a different solution. I have some hard drives where the spindle motor is fine, but the platters or the heads have degraded. Maybe I could just transplant the spindle motor from them?

Never tried that, it involves handling the spindle and disc plates.
This could be possible i guess... If you can find the exact same motor. The hard disk needs to set the rotation speed as intended, within specs, voltages, current, measurement from the motor. And even then i can imagine there is a need for calibration.

It is possible to relubricate sealed (ZZ) ball bearings, but the easy method is a bit of a hit-and-miss operation and will not get rid of old dried up grease. There are some things that needs to be taken into account too:

- It's likely already too late. Relubrication will of course not fix wear caused by earlier inadequate lubrication.
- Even old hard drives are extremely sensitive to particle ingress. You'll of course have to open the HDD and expose platters and mechanism to access the bearings. It might be hard to achieve a proper clean room environment at home.
- Tolerances are very small, it might be problematic to get everything properly aligned when reassembling without specialised equipment.
- You generally need very steady hands and great care. The inside of an HDD is mechanically sensitive, especially the I/O arm is easy to break.
- Lubricant compatibility if you go for the easy method. The original lubricant is unknown. Fully synthetic oils are however usually compatible with just about anything.

If the alternative is replacing the HDD, and you've got nothing to lose, it might be worth a shot though.

Once you've removed the bearings from the HDD, the proper way of relubricating them is to remove at least one of the dust shields (they're normally press fitted in the outer ring), clean the bearing thoroughly, and fill it to ~30% with a suitable grease before re-fitting the dust shield(s).

The easy method is to put the bearing in a synthetic oil bath, heat it all up to ~150-200 deg C (or whatever the specified max temp for the oil is), and let it all cool down back to room temperature. The initial temperature rise will make trapped air in the bearing expand and partly escape. When it all cools down, the remaining air contracts, causing an under pressure, that will suck oil into the bearing.

I have some experiments in mind with lubricating the platters. In fact I already did some on a dead HDD just to see how well the head can cope with lubricant applied in a few spots rather than on the whole surface. This was not for any loud operation, rather for tackling the problem of some older HDDs that get "sticky" due to age and bad original lubricant choice.

With what little experience I have working on HDD mechanical parts I really don't see any possibility of replacing the motor. It's often interference fit into the case, though sometimes it's held by screws - but even in this case these are only accessible with platters removed. Even MFM drives can have servo data on platters (even the cheap and popular ST-225, or NEC D51 series) so removing the platters might just kill the HDD. In fact platters, plural, is pretty much a no-go for more modern drives like IDE or SCSI, unless you have some custom device to hold them together and prevent misalignment. I've tried sticky tape, it's not impossible but quite difficult to pull off. And speaking of, removing and reinstalling the heads is another issue without custom made tools. Any attempt to disassemble the head block or removing platters from stepper-motor based HDD will introduce head offset or wobble or both. Only the dumbest MFM drives can survive that (ones that do not use any form of servo data) and data will be lost due to required low-level format afterwards. No matter how well machined, a replacement motor will come with a sligthly different spindle column and the platters will now have some, even minor, track positioning issues.

The YT videos showing HDD being fixed are quite misleading. First the modern HDDs with servo data embedded into the tracks can deal much better with platter or head swaps, and even that will require some software tools to perform initial calibration. And that HDD will not "work" like this afterwards, all that's good for is getting the data out, platter by platter. Yes, there are some people who took the platters out, replaced the bearings or the notorious rubber head swing limiters and were able to put it back together and make it work. But these are not the norm, these are exceptions. So while yes, if the drive is already considered unusable you can try and lubricate it, or whatever - but in 95%+ of cases it won't work better afterwards. It will be properly dead.

Deunan wrote on 2025-02-25, 11:44:

I have some experiments in mind with lubricating the platters.

I personally think that will cause more issues than it solves, even if it might initially seem OK. Some oils on the wrong type of surface can have a corrosive effect, and the gap between the head and the cylinder/track currently being read is so narrow that oil would create friction, as opposed to the intent of reducing it, as it gradually gets warm and dries out in the air inside the drive (unlike the parts that do need lubricating, which are not constantly exposed to air). I've not studied this stuff for decades though, so I may be talking out of my arse, but I wouldn't risk it on a working drive or one I intend to try and salvage data from.

Oh I'm aware of the risks. But the one HDD I was about to try it on for effect had finally got stuck so bad that one of the heads got ripped off. I didn't tap on it hard enough I guess. So one way or another these die, might as well try. But I agree that this should be done only as last resort, otherwise it might make the HDD unusable. Right now I have a few "problematic" drives but none so bad that I would want to continune my experiments.

Current idea is:
- use something PTFE based, very small amounts and only on the inner tracks / landing zone
- the lubricant only needs to prevent sticking when the HDD stops spinning, one it does spin up the head will slide on air gap as it should
- the problem is multi-platter HDDs with multiple heads, very hard to get proper access, preferably try it on single platter HDD first?
- heads will transfer some of the lubricant all over the platters, long-term effects are unknown for now

As for the OP's problem, some old HDDs can get noisier if the lubricants dried out and/or got packed into lumps. Warming the drive a bit might help but within reason, 60C would be max and the drive should not be operated above 40C. But putting some heat into it and letting it run for hours afterwards should get the lubricants back into operating range and consistency - if at all possible. On HDDs with stepper motor for head positioning it's also good to give that motor a very good exercise. Lubricating those is usually easier if need be, most of the time both rear and front bearings can be accessed (though it will also require opening the driver protective cover). Same issue as before - only thin oil will work and it's best not to overdo it.

debs3759 wrote on 2025-02-25, 13:59:

Deunan wrote on 2025-02-25, 11:44:

I have some experiments in mind with lubricating the platters.

I personally think that will cause more issues than it solves...

If you're willing to take chances (ie: open the drive - but why not as it's only going to become worthless)

I do recommend "polishing" the platter, perhaps with a very small amount of penetrating
lubricant on a cloth - the idea is to remove the tar that is building up without damaging the
platter - this did work for me... Afterward you can clean it with something completely benign
if you worry about that sort of thing.

I've not tried, but I doubt just lubricating the platter would help, nor do I think it would
cause much damage - It's going to get "flung off" as the drives spins up and absorbed
into the filters ... (if you use enough to block the filter it could cause other problems)

Dave ::: https://dunfield.themindfactory.com ::: "Daves Old Computers"->Personal

I'd be afraid of head crash/fatal stickage at velocity.

The head does not need lube. It floats on an 'air bearing' created by the platter moving underneath, pulling the air in the drive enclosure with it, 'balling up' in a small vortex as it collides with the head. The distance is very small, and oil or other sticky 'gunk' can be enough to slam into the head, instead of this air, and cause the fatal stickage.

Cleaning the platter will be difficult for the same reasons-- touching the head in any way is bad juju.