Need advice on how to best clean/de-rust an IBM 5051 w/o damaging it any further

[DDS]

I've been following this thread hoping to see some solid resolution of something that a common collector can do with rusty cases, but so far it has been all over the place.

So, yes, the ideal solution is to take it to a professional, and have them blast/repowder it. Perhaps a few people here have that kind of equipment, know other that do it, or have piles of money, but not most.

So what exactly should a common person do within reason? (things one can get at Home Repo, etc)

I would hope that some commonly available rust remover, plus some kind of spray paint would be a serviceable solution? At least one that would improve the appearance, and halt rust? Would any of this prevent a professional form re-doing it later?

Firstly, forget removing the rust without damaging the existing paint. Metallic oxides are pretty tough. That's why we use them to make sandpaper. Second, plan on any rust you merely cover up coming back at some point.

Sooooo.

Do whatever you like. Its your computer. But if you want it to look right after you do it, and still look right six months after that, you're going to need to (one more time):

1. Soda blast. The only step that most folks probably can't do at home. Or sand it. Time vs. Money. vs. Elbow Grease. You get to choose. Use a grit size that won't leave huge scratches in the steel. Your primer and top coat may or may not flow out enough to fill them.

2. Neutralize. The shop that does step 1 can do this for you. If you sanded the rust off you won't have anything to neutralize. Just wipe it down with a tack rag.

3. Prime with a good primer compatible with step 4.

4. Top coat with a good paint compatible with step 3.

Steps 2 - 4 can be done at home with any weak acid, lemon juice, household vinegar, whatever, and rattle cans from your favorite hardware store. Be advised that as soon as the steel surface is exposed to atmospheric oxygen and humidity it will begin to oxidize. This is one of the best reasons to have it all done at a shop that has the materials and equipment to do it right. Again, your computer, your wallet, your shortcuts, your results.

But if it looks like "heck" in 6 months you might be questioning why you didn't do it right in the first place. And remember that 90% of a good looking paint job, on your house, your car, or your computer, is a direct result of proper surface prep.

One more tip. If you really really care how it comes out, practice with your selected technique on some scrap metal before you risk boogering up your "soon to be beautiful" computer case. Maybe practice several approaches and different materials and see what comes out best for you.

Good luck!

 

[paul]

So, with the faulty cap clipped out does the board work now? May as well find out before you go to the trouble of replacing them as they are not critical items at this point in the game.

The relay looks is a DPDT 5V coil with 14-pin DIP base, should be easy to find - but why bother for now unless you plan to use the cassette port? It can't hurt to leave it in - or unsolder it. Seems like a waste of time trying to repair such a delicate item. Also, while it's out might be a good time to wash the board.

The motherboard does look a bit rough, unfortunately, not a bad idea to look for an alternative.

 

[deathshadow]

The primary ingredient in WD-40 is (about half) is CAS 64742-47-8, which could be called "white kerosene", basically a fuel. A little less of the other half is a light petroleum oil, what's left usually includes a surfactant (detergent), which probably helps to break the surface tension of any water.

Which is why WD-40 IS NOT A LUBRICANT!!!

It might displace water (what the WD part stands for) and break down rust, but it IS NOT A LUBRICANT!!! It's a CLEANER.

Sorry, in my spare time do bicycle repairs out of my garage and the #1 thing people screw up wheel hubs and chains with is constantly hosing them down with WD-40. A situation made WORSE by the fact it also breaks down grease. (confirming that detergent part of it's makeup) so while it might spin free for five minutes, go five miles and you're doing nothing but turning the bearings into fine dust.

That said, there's NOTHING wrong with using WD-40 as a rust removing CLEANER, the trick is like any other paint remover is to then WASH it with a detergent (normal dish soap is fine) and LOTS of water when done. Honestly regardless of what you use to remove the rust, going at it with a detergent, a scotch-brite pad, and rinsing the crap out of it to make sure all the chemicals and particulate are removed is pretty much a "must do" before you even THINK about applying primer, much less paint.

It might seem counterintuitive to wash steel with water since most people associate rust with water, but if you towel dry, air-dry in the sun for at LEAST 4 hours (or 10 minutes in a 240 degree hotbox), PRIME IT WITH A PROPER PRIMER -- you'll be fine. (It's like explaining to people how stainless steel rusts faster if left in salt water than normal steel or even just plain iron.)

When it comes to painting it, I would NOT be using some crappy spray enamel. While the end result might not be 'authentic' per-se, I'd probably be looking for something a bit more rugged and longer lasting like a BBQ paint.

Though really something like a steel case SHOULD be powder coated and baked, not painted. You paint it, well... there's a reason you're more likely to find rust on the black painted parts of a 5150 than you are the outer beige if it hasn't been scratched up.

... and in the case of a bicycle, it's also ok to use WD-40 as a CLEANER or to free up stuck, parts as well -- just be sure to add lubricant AFTER!!! If you spray it on the hubs, crack them open and repack the bearings with grease (green marine bearing grease is cheap and works best) when done. With the chain, skip all the fancy lubes and go for 3 in 1 for clear weather riding, 10:30 engine oil for wet. WAY more reliable than the fancy expensive bike shop lubes.

Well, unless it's an internal gear hub, then you just have a long day ahead of you...

-- edit -- Oh and when it comes to painting, for **** sake do NOT try and do a retouch when it's that badly damaged, either strip the whole thing to bare metal or please, or don't bother. You have rust UNDER the paint that's spread from the parts you can see, you trap it inside the paint things are just going to degrade faster than if you left it alone!

 

[ropersonline]

So, with the faulty cap clipped out does the board work now? May as well find out before you go to the trouble of replacing them as they are not critical items at this point in the game.

Hang on, are you saying I could just remove the faulty cap and try another power-on test without it, before the replacements arrive? (Or maybe even with the faulty tantalum cap experimentally replaced with that elco [which way/with which pin going where?] that I mentioned?) Could you elaborate on the "not critical" point? What's the role of those C7 caps there now and what would happen without one (or with one temporarily replaced w/ an elco)?

The relay looks is a DPDT 5V coil with 14-pin DIP base, should be easy to find - but why bother for now unless you plan to use the cassette port? It can't hurt to leave it in - or unsolder it. Seems like a waste of time trying to repair such a delicate item. Also, while it's out might be a good time to wash the board.

Here's the PCB pattern underneath the DIP relay:

    Ø O Ø O Ø O O Ø

    Ø O Ø O Ø O O Ø

Only the Ø's are actually wired through though. I'm not sure if this counts as a 16-pin base or an 8-pin base. Probably 16-pin. I'm also not sure what the non-connected soldering pads even do – maybe they're really manufacturing artefacts? The identification of the thing as DPDT is helpful; thank you very much. I have been able to figure out that the thing connects to a +5VDC trace from the PSU, but I got confused and thought this had something to do with the keyboard; thanks for pointing out that it's actually just cassette-port related. Could you elaborate on what this thing actually does?

As for obtaining a replacement (if necessary), are you saying that any physically compatible +5V 16-pin DPDT DIP relay should do the trick? Does it have to be a coil relay – I'm asking because I've seen completely enclosed ones as well, and I'm not sure if they are coil relays or something else.

The motherboard does look a bit rough, unfortunately, not a bad idea to look for an alternative.

Granted – but I don't like the waste, cost, and heartless/giving up aspects of that solution. If I can reasonably get this to work without buying a new board, or worse, new board and case, I will.

Thanks also to deathshadow, whose advice I've read with interest.

 

[Chuck(G)]

Sigh, WD-40 does contain a light oil--MSDS says a little less than half by weight. I suspect it's probably less than 5W--great for sewing machines.

I'm old enough to remember the homebrew formulas for lubricating bicycle chains--make a slurry of melted paraffin and 30W motor oil--or even boiling chains in paraffin It lasted longer than straight oil. There were also the folks who made half-and-half mixtures of kerosene and motor oil and just doused the hell out of the chain periodically. Then there were the specialty lubricants--TriFlow, Krytech, baking with moly disulfide, etc. Chains (and everything else) just get dirty and the best thing is to clean and re-lube. I will confess to using Boeshield T-9 from time to time, however.

But I wouldn't use WD40 to abate rust. If you're going to use oil for the job, use a good penetrating oil and then degrease everything in sight. Paint doesn't stick well to ruse or oil.

If you can find a good epoxy rattle-can paint in the color that you'd like, it works better than the run of the mill stuff. You usually use it for repainting chipped areas on appliances such as clothes washers. But you won't get the texture you're after.

 

[Agent Orange]

I'm old enough to remember folks putting kerosene in an automotive crank case to "smoke 'er out". Probably not a good idea for hemi or a LS1, but it worked on those old in-line 4 and 6 cylinder clunkers.

 

[Chuck(G)]

Not to mention plugging a leaking radiator with potato soup...

The old-time bush pilots used to buy WD-40 by the gallon or drum and mix it in with the fuel. The claim was that it prevented the engine from fouling or some such thing.

 

[ropersonline]

Oh shit, I only just realised I transposed the 51 and 50 in the headline. It's an IBM 5150 of course. Sorry about that.

I should have noticed, especially since right after flubbing that headline, I got it right again:

Need advice on how to best clean/de-rust an IBM 5051 w/o damaging it any further

Hi, I have an IBM 5150 that's been out of commission…

PS: Looks like there is an IBM 5051... of sorts.

 

[ropersonline]

I'm still hoping for some re-confirmation from paul or somebody else as to whether it's really okay to clip or elco-replace the blown tantalum cap for testing, but here's what I've figured out in the meantime:

I've also found that (+/-)12 V seem to get used very rarely on the board, mostly just on the ISA slots. But in trying to look at the board's traces starting with the P1/P2 connectors on the board (for the P8/P9 plugs from the PSU), I've noticed something that seemed odd to me:

There are no traces for either the +5V or the ground. Here's what I think might be the explanation:

It looks to me as if the PCB of the 5150, rather than being just double-sided, could actually be multi-layered and contain both a ground plane and a +5V plane sandwiched inside. These seem to be fairly uniform layers, not traces, with only the vias being left out (except for those that are meant to connect to that layer). Does anyone know for sure? Is my above picture accurate? I've never looked at this stuff before in this detail. Is this fairly typical for a mainboard? For most PCBs, even? Or are early IBM PC mainboards exceptions to the way most boards/PCBs are generally designed?

 

[Chuck(G)]

Yes, the 5150 PCB is multlayer--as are pretty much all of the PCs after it. It just makes sense--you "bury" the supply and ground planes and reap the benefit of much lower resistance between components on those rails.

As far your +12 and -5, those matter a lot if what you've got is one of the 16-64K board (i.e. uses 16K DRAMs). Those are three-supply chips. Otherwise, probably not so much for the motherboard--mileage varies on what ISA boards are being used.

As a rule of thumb, most serial I/O (RS232C) boards will use +5, +12 and -12; very few boards will use -5.

 

[ropersonline]

Yes, the 5150 PCB is multlayer--as are pretty much all of the PCs after it. It just makes sense--you "bury" the supply and ground planes and reap the benefit of much lower resistance between components on those rails.

TIL.

As far your +12 and -5, those matter a lot if what you've got is one of the 16-64K board (i.e. uses 16K DRAMs). Those are three-supply chips. Otherwise, probably not so much for the motherboard--mileage varies on what ISA boards are being used.

As a rule of thumb, most serial I/O (RS232C) boards will use +5, +12 and -12; very few boards will use -5.

My board is a 64/256K one. It's a Ⓑ 5150.

Because the blown tantalum cap sits on the +12 trace…

…I've figured out that that +12V trace leads pretty much just (probably only) to the ISA slots. It's available on the back of peripheral cards, on the ninth pin from the slot end of the card:

I have a CGA-compatible card (the Hercules Color Card, which is not a Hercules Graphics Card though it is a graphics card made by Hercules, the company), and because I also have a real IBM CGA card and the Hercules Color Card isn't looking too hot anyway, I thought I might even be willing to risk the latter. So I looked closely, and figured out that on it, the only thing that connects to that +12V pin is the light pen connector, which not being present means nothing's connected to that at all.

So that should mean I should definitely be safely able to retest the board with the blown cap clipped out or replaced, right? As long as I don't connect anything other than the PSU/mainboard, HCC/monitor and keyboard, right?

Could someone elaborate on those "parasitic effects" that supposedly occur when a tantalum cap is replaced with an elco? I've only recently heard about it, but I don't really know what I'm talking about – or reading.

 

[Chuck(G)]

Yes, if you have the 64-356 model with no ISA cards using +12, you should be safe.

Tantalum caps are still electrolytic, but are better at bypassing high-frequency RF. You could probably replace most of them with plain old aluminum electrolytics shunted by a 0.1 uF ceramic and get pretty much the same effect.

 

[paul]

Such detail! I would have accidentally set the board on fire by now and moved on to the next project

The relay looks is a DPDT 5V coil with 14-pin DIP base, should be easy to find...

I think this is the closest current datasheet for a replacement relay. Just match the contact form and pinout with a 5V coil and purchase from Digikey or such. But as you can see in the schematic, it simply switches the cassette tape interface between record and playback and you could just leave it out for now.

http://www.omron.com/ecb/products/pdf/en-g5v_2.pdf

 

[modem7]

But as you can see in the schematic, it simply switches the cassette tape interface between record and playback and you could just leave it out for now.

To be noted is that a missing relay has a minor (and ignorable) consequence: a "131" error produced at power-on of the 5150 motherboard. This results when the cassette data loopback test done by the Power On Self Test (POST) fails. Diagram at [here].

 

[ropersonline]

I've much more to report (though the news isn't really good), but I'll get around to that when I get around to it. For now, I'd just like to ask about an inconsistency I found. As best as I could determine, the only +12V tantalum capacitor on the 5150's board is the one next to the mounting hole in the corner between the power connectors and keyboard/cassette connectors – the one that blew up on mine (for images see above).
That's ONE 10µF 16V tantalum capacitor on the +12V trace. (There are other 10µF 16V tantalum capacitors on the board, but they're not on the +12V line.)
However, the logic diagram in IBM's Technical Reference has this (highlighting mine):

So according to the logic diagram, there should be five 8.2µF tantalum caps, but there really is just one 10µF one.
Does one 10µF cap somehow correspond to five 8.2µF ones? (Is there a logarithmic scale involved?) Does anyone have any explanation for, or even just thoughts on this inconsistency?

 

[ropersonline]

Tantalum caps are still electrolytic, but are better at bypassing high-frequency RF. You could probably replace most of them with plain old aluminum electrolytics shunted by a 0.1 uF ceramic and get pretty much the same effect.

At the peril of getting on your nerves with amateur questions, I haven't really understood what you said there:

1. bypassing hi-freq RF: Is this bypassing in the ignoring or in the shunting sense? Does this mean tantalum caps are resilient against any influence of hi-freq RF, or does this mean they will let hi-freq RF flow past unencumbered (as a bypass around a town would)? Does it mean something else entirely? I haven't really understood.

2. elcos shunted by a small ceramic cap: What does shunted mean in this context? Again, I haven't really understood.

Thanks for your time and help.

 

[ropersonline]

http://www.omron.com/ecb/products/pdf/en-g5v_2.pdf

Thank you! Would it make a difference for our purposes whether one buys the H1 high-sensitivity version or the standard one? I'm just asking because the H1 version also is on ebay.

 

[Chuck(G)]

Shunted = in parallel with.

Bypass = shunt to ground. There are two reasons to put a capacitor on a supply rail. The first is decoupling--that is, to decouple the (slow) effects of switching on the power supply, so that they don't interfere with switching., The second is bypassing - getting rid of high frequency signals that are riding the supply rail.

 

[paul]

... the only +12V tantalum capacitor on the 5150's board is the one next to the mounting hole in the corner between the power connectors and keyboard/cassette connectors ...

I would expect that the schematic is intending to imply that those caps are placed one each adjacent to each bus connector, rather than next to each other.

My "B" board has two (three pin) unused cap locations (marked "CX") one each between every two bus connectors. Only one cap is installed.

I'm guessing that, once tested, they found the additional caps were not required for noise suppression and the staff in the design department were so excited about releasing the world's first IBM PC they simply forgot to notify the staff in publications about this trivial detail.

But, as noted before, you can test the basic functionality with no cap in place. An since we are all on the edge of our seats wondering if your board is going to work, I would hope you do this soon!

 

[paul]

...Would it make a difference for our purposes whether one buys the H1 high-sensitivity version or the standard one?..

Hmm, according to the spec sheet the 75477 driver can sink 500mA so you could use either version, which draw 30mA and 100 mA respectively. The only downside to the H1 is that it's fractionally slower, which doesn't matter in this application. Personally, I'd get the H1 version if it's no more trouble or significant cost, but both will work.
You could measure the coil resistance of the old part with an ohm meter and determine which part is a better match to those resistance values on the spec sheet.