[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: The (Mis-)Adventures of Dave or A Long Update and a Few Questions

Dave Rawson wrote:

As promised, here's a bit of what I've been doing and how it's turned out.
I try to work on the basis of all's well that ends well, so perseverance is a necessity for me. I'll often struggle with something's that meaningful to me, far past rational return for effort, save for satisfaction.
So I can't tell you how carefully I worked with my failing Franklin Ace 1000 as I tried to bring it to usability as key after critical key stopped working. It had been 15 years since I last experimented with using it, and I remembered little at first. Rather than waste those precious keypresses, I started researching the treasure trove of information in the CSA2.Faqs and my own stock of old notes from back in the day.
Looking at the ADT instructions for cold-metal bootstrapping, I realized if I had the right cable and/or combination of adapters lying around, I might cobble together or already have a null modem. That would allow me to control the Franklin from the PC keyboard and the failing keys would be moot!
I had about a half dozen cables, including a laplink, a bi-directional, some db25 to db9, some db25 to db25, a db 25 to 25 piece that said null modem on it showing the pinout relations, and a pinout checker thingee with lights that glow red or green to indicate activity or conflict. That was encouraging enough that I determined to test the pinouts of all the cables to see what was what. Easier said than done.
Bear in mind, this is excruciating delayed gratification for me. I haven't had the luxury of extra surface area to even possibly set this up in over five years. Nevertheless, I found my trusty little meter I use for continuity testing and gathered all the cables and db25 to db9 reducers and set up to tediously chart them for the first time, vowing to store the results with each piece no matter its usefulness to me at that time. I didn't even want to calculate the pin combinations that I'd need to do since that wouldn't speed things up at all. It would take what it would take.
Except that it didn't. It took more and worse. The probes on my little meter were too large to make meaningful contact. Even an old VOM I had didn't fit. So it was the bent paper clip device that I'd often used before, except now there were two of them, one on each end of the cable and a probe friction leveraged into each clip whilst holding the cables with whatever free fingers were left over, then testing each of the 9 pins 25 times (for the 9 to 25 cables) and testing each 25 pin 25 times (for the 25-to-25 cables). Ungainly, frustrating, tedious hour upon hour, but dead-on accurate and I-never-have-to-do-that-again-thank-God!
After a bit of puzzling over the accumulated maps, it looked like I could cobble what I needed, but found I had no terminal program since I hadn't used HyperTerm in ages on the XP setup I intended to use, and I couldn't locate where it resided or from where I could install it. No matter, I found a copy online, installed it, discovered it was a proprietary version (HyperTerminal, not HyperTerm) with a 30-day trial. I uninstalled the program <hock-ptoie!>, but it took extra pieces with it, it turns out, so I had to reinstall it with one less day to try it.
So I hooked the PC to the Franklin via my AppliCard terminal mode. The lights of the activity checker glowed an encouraging green. But now, of course, the "#" key choose this moment to fail. I mean all I needed was a friggin' "IN#3" and wonders would unfold. But no, after 20 years of waiting for this moment, one keystroke stood between me and success. In frustration, I stopped babying the key and really pressed down on it, the thumb from one hand pressing the index finger of the other for extra force. And it worked! But for how long? Would it work again? 

What a great account, Dave!

Though it never seemed to be relevant, I should tell you that most
keyboard failures are just contact oxidation as a result of disuse.
Therefore, cleaning the contacts will often restore them to operation.
And the simplest way of cleaning is pressing the key repeatedly with
moderate pressure until it works--and by repeatedly, that could be
a few hundred presses!

Once a key is working, it will usually continue to work if it is being
used with moderate frequency.  This is what "wiping contacts" were
designed to do.  ;-)
Then an epiphany! If I could figure out a bit of what the ADT code that would be ported over was designed to do, and scavenge what I needed from that, I could write a little program on the PC to (once it was saved to disk on the Apple side) boot into the key-combinations I would need to reconnect the machines one the Franklin's "#" finally gave up the ghost, hoping that it hadn't already. If I could manage that, then at least I'd have a lifeline no matter what (I thought).
Well, I managed another squeeze out of the failing "#" key, managed to send the program over and save it to disk and it booted right up and reestablished the connection automagically. Hoo-ray!
From there, I managed to bootstrap the early ADT program over and get it to boot!
But alas, alas, alas... I was brought to ground when I realized that it was the Franklin/Apple side that initiated transfers and my keys didn't work!
So, I have avoided eBay like the plague. Totally uninterested in it. Didn't want to go through the learning curve. I hate learning curves involving money and experimenting, inevitable mistakes and loss.
But, as you all saw, I managed to buy an old Apple set-up and, going with advice, found an SSC and bought that, too. About 10 days later of interminable wait, all arrived and I unpacked and pondered what I had and how to put it all together. I sussed out the useful cards, identified and bagged the less immediately useful cards and software and turned on the cardless Apple. It powered on. Check.
I hooked up my old monitor that I now knew (from 10 days before) still worked. 40-column Video worked. Check.
I strapped in my Franklin drives I knew worked and booted up the bootstrapped ADT. Check-o-roonie!
Excited, I found the settings for the SSC and set them to 9600 baud as a start. But the, the SSC was in Slot 2 and recommended "Modem" mode, where the Applicard was in Slot 3 and used "Terminal." This was just enough of a change that the null modem cable I had cobbled seemed to no longer work or at least wasn't making connection with the ADTPro on the PC side.
So I reached over, twisted the cable to better see the transmission lights on the pin checker, which pulled on the SSC which wasn't properly seated and it rose up at the case end at about a 45' angle and the monitor filled with gibberish. I had, in 30 years of fiddling with computers their innards, their cards, their in-and-outs, finally fried something.
The Apple would no longer beep. It would power on, but it wouldn't beep, nor boot.
So, you can imagine this moment. Was I fated to fail?! This was more than a bit much. This was self-inflicted carelessness born of caution. Not wanting to force the SSC (as I might now) or accidentally crack the motherboard, I apparently had, as I said, not properly seated it, because it had slipped like butter on ice and just that quick.
The morning I started my hunt and over time bought two Apple ][es and two more SSCs. I now have a working system with an installed SSC and yesterday ported over the back sides of my disks from which I had earlier copied with the FC5025 (purchased in the interim). Running at ADT's top speed, the sides port in 28 seconds, 2 seconds faster than Mike's measurement of the ProDos DC5025! WOW! 

Yes, the FC5025's host program is terribly unoptimized.  It should
easily be able to read a whole track in just two rotations, but because
of a performance bug it is taking 5+ for ProDOS and 12 for DOS reads!


This is glorious!

So all's well that ends well (so far).
I tell you all this because if I can't laugh at my own despair, then I'm down for the count and because I have a couple of questions related to my earlier disaster.
1) I likely fried the original SSC when I fried the Apple. Can I test it without risk by putting it in the working ][e? That is, I can't imagine any harm it could do if it's dead or if it's working, but you can see I don't trust my instincts since I have a working system and why risk that on unvalidated instinct alone? 

It may very well be OK.  The Apple was fried by the slanted fingers on
the corner of the card shorting across a couple of connector contacts.

With any luck, the card survives...

After an incident of electronic violence, there's always some doubt
about the extent of the damage and whether it might prove lethal to
other gear.  At some point, when your curiosity exceeds your caution,
you'll insert it in a working Apple and test it.  ;-)

This is not guaranteed to be safe, but is probably fine as long as
the contact fingers are in good shape and nothing on the card is
obviously fried.
2) What did I likely kill in the Apple I fried? Is there something I could pull from my own ancient Apple ][+ to replace whatever was blown? 

It's hard to be sure, but I'd guess that the +12 got momentarily
connected to the data lines, frying the tranceiver IC.  Perhaps
someone here with access to the schematics can tell you which IC
that is and where it is on the main board.  With any luck, only
one thing was zapped.
3) The ADPro transfers are so rapid, that after a number of them, it actually started my Franklin drive chirping when copying blocks from around 40 to 200. It's short of a screaming or squealing sound, but definitely a high-speed rubbing of some kind that can't be good. Could it be overheating somehow? Or dust or grit from disuse being ground up? Ideas? 

It may just be the old drive belt squealing, or it may be the head
or pressure pad.  Does it squeal when operating without a disk in
it?


Thanks as always for your kind help, advice and encouragement.
And my golly, thanks so much to the folks who have made all this possible: the Faqs, ADT, CiderPress and so-on. Bless you all for making this available! 

Whoo-hah!


I'm glad to hear that you are succeeding in recovering all your data!

And, again, what a great storyteller you are--with all the agony and
the ecstasy!  ;-)

-michael

NadaNet and AppleCrate II: parallel computing for Apple II computers!
Home page: http://home.comcast.net/~mjmahon

"The wastebasket is our most important design
tool--and it's seriously underused."