Groups | Search | Server Info | Login | Register

Re: I could use a hand locating hardware bits and pieces and/or technical documentation for repair.

Newsgroups	comp.sys.3b1
Date	2015-12-21 10:45 -0800
References	<11f11849-2c2c-433a-ad2f-555403129d14@googlegroups.com> <slrnn79ivu.nhp.BPdnicholsBP@Katana.d-and-d.com>
Message-ID	<c86d2f5f-f461-4d3f-b0ad-1e395abb58d1@googlegroups.com> (permalink)
Subject	Re: I could use a hand locating hardware bits and pieces and/or technical documentation for repair.
From	Chris Smith <protheus@byteorder.net>

Show all headers | View raw

On Friday, December 18, 2015 at 9:17:10 PM UTC-6, DoN. Nichols wrote:
> On 2015-12-19, Chris Smith <protheus@byteorder.net> wrote:
> 	Have you washed it?  Clean tap water can't do much damage to it
> compared to what the "unspecified liquid" has already done.  I did this
> with a different keyboard which suffered from "unidentified liquid"
> damage from a squirrel who was resident in the house.  Luckily, I
> discovered it rather quickly, so the wash and let dry worked well that
> time.  
> 	Strip the case open.
> 
> 	Turn on warm to hot water in the shower.  (Hot will be better,
> but warm easier to control things by hand.)
> 
> 	Rinse it for quite a while, ending with water as hot as you can
> manage, then shake it, and stand it on end for a day or two to let it
> dry thoroughly.

Well, the answer is yes and no.  I've cleaned the exposed and nearly exposed surfaces as well as I could with isopropyl, but I hadn't decided yet to try immersion or just spraying the thing down as you suggest.  It has been on my list of things to attempt, because I thought the exact thing you did:

> 	You probably have some of that liquid under the chips, which is
> why you want to rinse it quite a wile and then let it dry.  (Rapid
> drying could be done with a vacuum chamber, but most houses don't have
> that. :-)

Or if not liquid, perhaps some conductive left-overs.  So... maybe I'll give that a shot to start with.

> 	A couple of days ago there was a request for schematics of the
> power supply, and I mentioned that the power supply and the video board
> in the monitor assembly were purchased in rather than built as part of
> the computer, and the vendors did not supply the schematics.  I *think*
> that they keyboard may be the same.

I can tell you that the logo on the PCB is not an AT&T one, and it has somebody else's model number to go with it. :)

> 	Can you back up about a week in this newsgroup and check for the
> URL?  I never noted it down, because I have the printed manual, but it
> would be worth your while to download and print out.

Thanks.  This is something that will be great to have, even if it doesn't have the information I'm currently looking for.

> >        I might not mind building a converter.  I'd be very tempted to
> > buy a spare keyboard at this point, even if the original can be
> > recovered.
> 
> 	Note that there were at least two versions of the keyboard.  One
> has the tactile modifications of the index finger home keys (F and J) as
> vertical grooves to the left of the letter, and the other has similar
> grooves under the letters.  IIRC, one had a better feel than the other,
> but you first want something which *works*.

I've only ever seen the one with the vertical troughs cut into the J and F keys, but I've seen perhaps less than a handful of keyboards for this thing.

> 	The emulation would be a trick with modern drives.  (Hmm ...
> perhaps easier using something like an Arduino and thumb drives. :-)

> 	The old MFM drives sent a raw signal from the read head, and
> depended on logic in the computer (the WD101 or for modified systems the
> WD-2010 chip) to convert that into bytes for transfer to memory.  You
> first have to tell the drive which cylinder you want to step to and
> which head within the cylinder.

> 	Todays drives (IDE, SCSI and the like) accept a request for a
> specific sector number (counting from zero at the beginning) and figures
> out internally what cylinder and head to look for) and then spews out
> the sector's contents as a sequence of bytes.  I think that translating
> between the two would require another controller chip like the WD-1010,
> and a smart board (Arduino, Raspberry PI, or other) to ask the modern
> drive for the data and make it look like the old MFM drive.  Not a
> trivial task.

Right.  I was hoping some of the cheap embedded ARM machines with GPIO included might be fast enough to fake the old protocol and read/write the result to a flash device.  If you can spend $20 on a Raspberry Pi and some miscellaneous hardware instead of $90 on an antique drive that's maybe already going bad from ebay, it may be a good option.  Though -- all other things being equal -- I do prefer actual hardware. :)

The closest I've seen is the emulator device AJ mentioned, and I hadn't seen it before now.  Looks like it depends on having a small computer system hooked in and also some dedicated hardware.

> 	Once you get it working (or find another), then comes the trick
> of observing it on a 'scope to try to figure out the encoding.  If you
> do find it -- please post it for future reference.

Well, I'm fortunately married to a woman who shares some of my hobbies.  She has one (an entire second 3b1) that we actually managed to buy old stock from somebody many years back.  Even still has the shipping container for it I think.  She hasn't started hers up in a while either, but her keyboard is probably fine, and I could probably borrow it for a bit in a worst case.  It wouldn't get my keyboard working, but it would allow me to interact with the computer and maybe poke at it with what tools I have available.  Unfortunately I don't have an oscilloscope or a logic probe at the moment.

Thanks for the help.  I'll try a more thorough cleaning and see how it goes.  Not too much to lose.

Chris

Back to comp.sys.3b1 | Previous | Next — Previous in thread | Next in thread | Find similar

Thread

csiph-web