Canon AS-100

[NeXT]

This is a system I've been chasing for quite a while (it looks cool but my god I'm now aware how useless it is) and during VCF PNW I struck a deal to pay and pick it up after the show from the storage unit it's lived in for a few years. Needed a bit of a cleanup but it's nearly complete.

So the idea is that the machine launched a short while after the 5150 launched and everyone was still trying Their Own Thing to make an office/home computer that was running MS-DOS (or CP/M-86...or the P-system....), so it is not IBM PC compatible and for me at least that immediately throws a lot of cool ideas out the window. More on that later. I'm aware of a half-dozen machines kicking around on the internet. I don't think they sold well.....

This particular machine is the AS-100C, which means it's the color display model which uses the NEC D7220 graphics chip with 96K(?) of dedicated video memory and the main CPU is an NEC licensed 8088. Base memory has been upgraded form 128kb to 512kb, it has the RTC option and it appears that it does not come with a BASIC in ROM and instead drops you to an x86 debugger monitor if it can't find anything to boot.

Ports seem to be parallel and the floppy interface which are standard, plus two additional serial port cards and the hard disk interface card, which is so simple, I'm pretty sure is just a buffer and bus extender and the drive controller/formatter lives in the disk box, which I don't have.

The external box sitting immediately next to the system is the dual DS/DD 5.25" disk box which is pretty much necessary if you don't have the hard disk box with the one floppy drive. There's four keyholes and matching posts on the system/disk box so they can lock together but the cabling for data/power are on the back. The 8" disk box however is purely optional and does not attach to the other disk box or the system.

The missing bits of hardware are:

-The combination DS/DD 5.25" floppy drive and 10mb (formatted 8mb) hard drive which again I have the interface board for, so this used to have that.
-Your choice of black ribbon dot matrix or color inkjet printer, which they expected you to purchase to the point there's four indents in the top plastic of the main system SPECIFICALLY for the printer feet to sit in.
-A weird little knob-like device which has confused me and The Internet for so long we all thought it was a jogwheel but after reading pretty much all of Canon's documentation I can now (explicitly) call it The Chode Joystick because it's just an 8-direction joystick with three buttons and the flattest, stubbiest grip I've ever seen. Depending on if you are in a graphics mode or a text mode it functions slightly differently. I have never seen one outside of Canon's marketing literature.

Snuci has a system as well and he has imaged his floppies and documentation. More material is available over at oldcomputers.dyndns and combined we have MS-DOS 1.25 (so that means there's some dark magic at play to support a hard drive and you basically have no other Quality of Life things you'd expect from DOS), CP/M-86, Canon's Canowriter word processor and Canobrain which seems to be a sort of combination programming language, graphing application and spreadsheet application. There's no other software currently known to exist, specific to the AS-100, unless I've missed something which leaves you with the long list of programs written that are MS-DOS compatible. (and by that there's nearly no killer-apps that make that list)
For documentation thankfully we have books for all of the above, plus the service manual. I have also attached a dump of the BIOS.

Current status: It doesn't boot.
While cleaning I did see that both 5.25" drives have a number of very leaky capacitors on their boards. The system can spin the motors but you hear no head action or the drive light turn on so they likely need to be serviced.
I also do not yet have a DOS disk correctly written out. There was a discussion that started here that sorta died out - https://forum.vcfed.org/index.php?threads/canon-as-100-floppy-disks.54857/
Snuci has imaged his disks (again, thank you) with the Applesauce but both his .IMG and .IMD files seem to not be something I can write out (same with TD0 images from the other site) and I can't verify if they are correctly written because of above drives needing work.
PJ-1080S's seem to show up from time to time but it looks like it's a game of throwing lowball offers and hoping someone takes it. I made one attempt and rather than accept/reject the offer they pulled the listing. >:T

 

[ldkraemer]

Here are the CP/M definitions I conjured up for the Canon AS-100.

/* Not in 22DISK */
BEGIN CAN1  Canon AS100 Testformat DSDD 5.25"  CP/M86 - 512 x 8
DENSITY MFM, LOW
CYLINDERS 80 SIDES 2 SECTORS 8,512
SIDE1 0 1,4,3,6,5,8,7,2
SIDE2 1 1,4,3,6,5,8,7,2
ORDER SIDES
BSH 4 BLM 15 EXM 1 DSM 139 DRM 127 AL0 080H AL1 0 OFS 4
END

# CAN1  Canon AS100 Testformat DSDD 5.25  CP/M86 - 512 x 8
diskdef CAN1
  seclen 512
  tracks 160
  sectrk 8
  blocksize 2048
  maxdir 128
  skew 3
  offset 18432
#  boottrk 4
  boottrk 0
  os 2.2
end

# libdsk
[CAN1]
description = CAN1  Canon AS100 Testformat DSDD 5.25  CP/M86 - 512 x 8
sides = alt
cylinders = 80
heads = 2
secsize = 512
sectors = 8
secbase = 1
datarate = DD

# flashfloppy/GOTEK
[CAN1]
cyls = 160
heads = 2
secs = 8
interleave = 3
bps = 512
id = 1
#rpm = 360
rpm = 300
rate = 250
mode = mfm
iam = no

I also have a ZIP file of software, that has directory listings.


More CP/M Definitions on github.

https://github.com/ldkraemer/CPM-Floppy-Definitions


Larry

 

[stepleton]

You find this same badge (minus the AS-100 bit) on the PJ-1080A inkjet, which I think you mention above. (I see PJ-1080S but I'm guessing that's meant to be A?)

 

[NeXT]

Yes, sorry. I wrote that up with three things going at once and got my models mixed up.

 

[NeXT]

So, the pointing device.

Like I said, it's not a rotary encoder or any sort of analog control device at all. Likewise you don't rotate it at all and use a button for example to shift what axis it works on, because that would be stupid. You can get an idea of how it works from the marks going off in different directions but when you look into the documentation it becomes very clear it's just tapping into various keys in the keyboard that become directional keys when you enable Cursor Lock. This also means that unlike the mouse port on the Olivetti M24/AT&T PC 6300 and similar machines you cannot substitute it for a quadrature mouse.

Coupled with the weird connector it uses to attach to the keyboard I'd almost suspect any that were sold have over the years been confused as control hardware for Canon's camera or other media control equipment but searching for the A-1100 finds nothing. It just bounces back to the same AS-100 literature.

 

[whartung]

That joystick is very cool.

Something that repeats faster than the keyboard, and also lets you intermix, say, "up" and "left" so you can go diagonally.

If the software can keep up, it gives you almost GUI-ish positioning and scrolling speed "for free", while keeping the keyboard experience the same.

 

[ldkraemer]

I found a typo in the libdsk Definition. Cylinders should be 160.

# libdsk
[CAN1]
description = CAN1  Canon AS100 Testformat DSDD 5.25  CP/M86 - 512 x 8
sides = alt
cylinders = 160
heads = 2
secsize = 512
sectors = 8
secbase = 1
datarate = DD

Larry

 

[snuci]

I also do not yet have a DOS disk correctly written out.

I am certain you know this but the drives are double sided, quad density so you'll have to write a DSDD floppy with a 1.2MB drive and make sure double stepping is off and there was a data rate adjustment needed? Check IMDA on that image and it will tell you what to change.



I pulled out my AS-100 and used IMGDISK to write out the MS-DOS diskette that you have images of from my site just now and it works fine so I can confirm at least that image is good. I suspect they are all fine.

I also have the 37 pin hard drive adapter in mine but it certainly didn't have the hard drive option. On quick glance, that looks suspiciously similar to the hard drive adapter for the Mindset computer. It connects to a Xebec controller via a 37 pin cable that, in turn, connects to an MFM hard drive. If the service manual has the pinout for that 37 pin connector, I can check if they are the same or similar.

The PW-1080 printer normally sites on top of the computer. The Centronics cable interfaces are right close to each other so a short ribbon cable (6 inch?) is needed. I didn't have that cable but seeing your floppy drive cable, I don't have that so I must have been using the Centronics printer ribbon cable for the floppy drives. They seem to perfectly fit so either mine is wrong or yours is wrong. As long as it works, it doesn't matter. I will need to make an additional Centronics ribbon cable when I find cheap connectors.

EDIT: Never mind my ramblings. The Centronics ribbon cables have a different number of pins. The floppy cable has more pins so disregard.

Here's what it looks like:




Hope this helps.
Santo

 

[NeXT]

Your ramblings are more than appreciated.
I didn't get the ribbon cable to connect the drives together but I did have a couple of very short SCSI cables. Since presumably the cables would be straight-through with no twists or repinning I considered it safe to use those rather than crimp a new cable.

I am certain you know this but the drives are double sided, quad density so you'll have to write a DSDD floppy with a 1.2MB drive and make sure double stepping is off and there was a data rate adjustment needed? Check IMDA on that image and it will tell you what to change.

This I did not know! o: That explains some of the weirdness we talked about earlier but I got lost on the last thread where people were discussing the disk images.

I pulled out my AS-100 and used IMGDISK to write out the MS-DOS diskette that you have images of from my site just now and it works fine so I can confirm at least that image is good.

I pulled out the Portable 386 and tried that initially but it did not like the images, though perhaps that's the controller not liking the workaround with a 1.2mb drive or the different data rate you mentioned.

 

[snuci]

I pulled out the Portable 386 and tried that initially but it did not like the images, though perhaps that's the controller not liking the workaround with a 1.2mb drive or the different data rate you mentioned.

Hmm.. When you change the value from 250kbps to 300Kbps here (highlighted):



your screen, when writing, will show the changed value of 300kbps in the upper left corner and on the "0/0" line like this. Yours still shows 250kbps. Maybe double check that? If it's not that, you might need a different computer. This one is writing okay with those values.



Santo

 

[NeXT]

Yes it was a config issue on my end. Matching your settings write a disk out fine and I can read it through DOS.

The problem is however it's still not booting, so unless something weird is going on that IMD doesn't catch while verifying I'm strongly suspecting the capacitor issues I saw on both drives may be causing problems. I'll have to look more into this.

 

[NeXT]

Okay I finally stopped procrastinating and specced/ordered out the capacitors for the Canon MDD220 (K-61221-05) floppy drives.

Main board
-------------
100uf 16v x 1 - P15796CT-ND
47uf 10v x 5 - P962-ND
33uf 25v x 1 - P977-ND

Motor board
--------------
0.47uf 50v x 2 - 732-9579-1-ND
4.7uf 35v x 1 - P979-ND
33uf 16v x 1 - P809-ND
10uf 16v x 6 - P807-ND

The leakage was especially bad on the main board as opposed to the motor board but I'm opting to service both anyways.

 

[NeXT]

And they are in.

The main board uses capacitors from Panasonic and they had all very clearly leaked. Most read poor or ESR but none actually measured bad. The motor control board uses capacitors from Nichicon and I found none of those had either leaked or tested bad. The affected traces were checked and so was the nearby write protect sensor on both drives.

Unfortunately this did not solve out no-boot issue either.
So we know Snuci's images are good and when written out even on my Compaq Portable 386 the contents are readable. The drives on the Canon are recapped after being obviously in need attention and while dismantled the heads were cleaned and the slide rails serviced, so neither should have an obvious problem at this point. I could clearly see there's a head load solenoid on the drives but we never hear it click, but we do hear the motor spin, so why is it taking the signal to start the motor but it's neither engaging the solenoid or turning the IN USE light on the front of the drive on? Is there more to the 50 pin Centronics cable?

Edited: I can confirm the floppy drives are *not* the issue at this point.
The manual for the MDD-220 doesn't seem to exist online but after some experimentation I found turning SW3 on...SW3...to ON the head load worked correctly on the Disk-O-Matic to let me read a floppy written on another machine, then hammer it with IMD to make sure it could do everything else. They're fine.

Though, note the green light on the drive. So we SHOULD be hearing the head load solenoid and the light turn on and it isn't, so this means we absolutely have a problem upstream of the drives.

 

[snuci]

Did you check the drive jumpers against the Field Service manual here? https://oldcomputers.dyndns.org/pub...n_as-100m-c_field_service_manual_(bw_ocr).pdf

 

[NeXT]

I can see there is both dip switches for the floppy drives (they don't tell WHAT they do but just their factory positions), there's is another block inside the 5" drive housing and they are set to..

00000101

...which should be correct for now as I'm working without the 8" drive box attached.
The switches on the system board do not seem to affect floppy operation that I can tell.

It's gotta be the cable. A SCSI cable should be passive but that's the only variable now unless it's a bigger hardware failure. Let me go digging into storage and see if I can find something else.

 

[NeXT]

It WAS the SCSI cable!! >:U

I was lied to! This cable worked fine on the mac but I guess something about how the ends are terminated either omits or bridges pins that are shared. I have no idea. I swapped it out for an AMPhenol terminated cable where the plastic ends are terminated per-wire and that one worked. System booted and both drives responded.
the keyboard however is unhappy but it seems to be coming back with exercise.

 

[NeXT]

Digging through my pile of SCSI cables, I found two really nice AVID branded 10" cables that let me attach both the 5.25" abd the 8" drive box together with the system, then I got to play with what formats the 8" drive particularly likes.

It wants Double-Sided Double-Density media. For me at least that's a really weird type because I only have single density disks hanging around and I can't seem to get FORMAT to properly identify them. I have one DS/DD disk on hand and it's bad in the middle, so I'll have to go bother Glitch or someone to see if they got any more tested disks in stock.

 

[vwestlife]

It wants Double-Sided Double-Density media. For me at least that's a really weird type because I only have single density disks hanging around and I can't seem to get FORMAT to properly identify them. I have one DS/DD disk on hand and it's bad in the middle, so I'll have to go bother Glitch or someone to see if they got any more tested disks in stock.

Try bulk erasing them. That's a trick I've used a few times to successfully "rejuvenate" old floppy disks, especially those which were previously formatted using a different drive.

 

[seat safety switch]

We might be able to bodge a Gotek in here temporarily for diagnosis. I'll often do that if I suspect the drive (or my disk-writing abilities) at all.