IBM PC AT Service and Repair Video From 1984

[Homer J Simpson]

[AF6LJ]

I forgot my safety glasses

[tom66]

I didn't see them mention the design flaw with the power connectors on the main board. Get them in the wrong order, since they're both 6 pins and interchangeable, and the board fries...

[AF6LJ]

I didn't see them mention the design flaw with the power connectors on the main board. Get them in the wrong order, since they're both 6 pins and interchangeable, and the board fries...

I am sure the tranees were schooled on that bit of information.
I suppose yhou could consider it a flaw, lots of gear have interchangable connectors, that can cause problems. This was from a time before every eight year old was building their own PC to play Quake 3 on.

[xrunner]

FRUs and CRUs - got it! 

[helius]

At 16:00, the TOY clock battery is a Li-MnO₂ type, what Panasonic calls "CR". These were considered new at the time: they were first marketed in Japan during the mid-'70s.
At 18:00, you can see that the DRAM chips are stacked piggyback.

[bitseeker]

That was fun. I thought Superman was going to fly in during the title sequence.

Come to think of it, I may still have an IBM PC service manual somewhere. Came in a binder that fit into a hard sleeve. Fancy.

[German_EE]

At the time of the IBM AT I was living and working in the UK and a friend of mine gave me a faulty CMOS battery and just said "have fun". The warning not to dispose of the battery in fire was very specific so I decided to give it a Viking funeral.

In the UK they celebrate something called Bonfire Night every 5th November, it's similar to the 4th July celebration except that they also light a large fire and have lots of treats like treacle toffee. Eager to follow the local tradition I joined in a Bonfire Night event with some friends and had lots of fun, but by 10pm the party had died down and the fire was just a pile of embers, this was when I tossed the battery into the fire. The initial few minutes were disappointing. There was a brief flame as the wires burned away and then a longer burn as the plastic case met its end. Then it happened.

Do you remember the end of 'Close Encounters of the Third Kind' when the mothership opened up? That bright light with the aliens hidden in the glare? Well, it was like that only brighter, and there was the loudest bang I have ever heard. For the next ten seconds I could see nothing as my eyes recovered but eventually I managed to see again and I realized that the bonfire was gone, there were only a series of small fires around the garden due to the scattered embers.

A few years later I spoke to an industrial chemist and he reckoned that I was very lucky. The contents of the battery would have melted in the heat of the fire and then exposed to super-heated air when the case eventually ruptured. The chemical reaction would then be quite violent.

[helius]

Better to watch videos than to risk life and limb with batteries, imho.

This cell ruptures very suddenly when recharged with 50V (not even reverse charged, apparently):
[this type is very strongly reinforced, which is why the release is so sudden. it's good because they hardly ever leak spontaneously]


This is what it looks like underwater at 400fps:

[continuo]

Meanwhile, at the Lithium Cell disposal facility in Mexico...   

[pickle9000]

20 MEG HARD DRIVE!!!

More space than you'd ever need in a lifetime.

[bitseeker]

20 MEG HARD DRIVE!!!

More space than you'd ever need in a lifetime.

And lighting-fast access times via stepper motor head positioning. Ah, those were the days.

[AF6LJ]

20 MEG HARD DRIVE!!!

More space than you'd ever need in a lifetime.

And lighting-fast access times via stepper motor head positioning. Ah, those were the days.

And you would never need a drive larger than 23MB.
I fondly remember the days of running Debug C:800 to set the drive up.

[bitseeker]

I fondly remember the days of running Debug C:800 to set the drive up.

debug g=c800:5 

or c800:ccc on RLL controllers

Wow, you're really rattling old brain cells now, Sue.

[AF6LJ]

I fondly remember the days of running Debug C:800 to set the drive up.

debug g=c800:5 

or c800:ccc on RLL controllers

Wow, you're really rattling old brain cells now, Sue.

That's It!
My first hard drive was a 65MB drive on an XT clone I had to do a low leve partition since the OS couldn't recognize a drive that large. (DOS 5.0)
 I had the computer for three years and in 1990 bought the drive and controller.

[bitseeker]

My first hard drive was a 65MB drive on an XT clone I had to do a low leve partition since the OS couldn't recognize a drive that large. (DOS 5.0)
 I had the computer for three years and in 1990 bought the drive and controller.

Mine first "IBM PC" was an AT clone, but I could only afford a 20MB (Seagate ST225) hard drive. Holy smokes you can still buy them on eBay!

After that, I upgraded it to a 72MB Miniscribe. Full height, 5-1/4", boat anchor of a drive.

[Halcyon]

After that, I upgraded it to a 72MB Miniscribe. Full height, 5-1/4", boat anchor of a drive.

Similar to this drive I have?....
(Shown here next to a 2.5" disk with over 2300x the capacity. It still works fine to this day!)

[bitseeker]

Similar to this drive I have?....
(Shown here next to a 2.5" disk with over 2300x the capacity. It still works fine to this day!)

Yes, similar size. Here's a photo of the 40MB version (pretty short defect list on that one). The 72MB model has the same housing.

[German_EE]

Given the current state of hard drive technology with 6TB 3.5 inch hard drives I wonder how much we could now store in a 5,25 inch full height device? Every PC I have ever seen has had multiple 5,25 inch drive slots just sitting there and doing nothing apart from (maybe) giving a DVD or Blu Ray drive a home.

I am however glad that MFM and RLL drives are no longer around and I can't remember the last time I had to set heads, cylinders and sectors when doing a drive install.

[AF6LJ]

Imagine one 5.25" drive four platters each being 6TB...

[helius]

Larger platter diameter limits the rotational speed that can be safely achieved. And rotational speed is the defining factor for access latency, so a larger diameter platter will have poorer latency than a smaller one, given the same motor and head technology.

[zapta]

The video mentions a clock backup battery, what was used in previous models to keep the clock running?

[German_EE]

Previous machines (the IBM PC and the IBM PC-XT) didn't have CMOS memory or a clock chip, if you wanted the time then you looked at the clock on the wall. There were some add-on cards that offered a clock but this meant using up one of the expansion slots.

[bitseeker]

The video mentions a clock backup battery, what was used in previous models to keep the clock running?

Nothing. If you powered it off, you'd have to set the time again after booting.

[AF6LJ]

Larger platter diameter limits the rotational speed that can be safely achieved. And rotational speed is the defining factor for access latency, so a larger diameter platter will have poorer latency than a smaller one, given the same motor and head technology.

Which is why those big drives with many platters use to interleave between sides and platters.
The stepper motors took half a day to get from the inside to the outside of the platters.
Thank GOD for IDE and the soft'er tracks and the meter movement like head actuator.

I am spoiled now, I could never go back to using a spinning drive for my boot up drive, cold storage of date... Yah, that wouldn't be a problem.

Having an SSD is like going from a floppy drive to a hard drive in the old days.