Universal Serial Interface Card Model J13-A for Apple IIe

13 posts / 0 new
Last post
Offline
Last seen: 1 week 4 days ago
Joined: Aug 4 2015 - 14:30
Posts: 150
Universal Serial Interface Card Model J13-A for Apple IIe

Some years ago I bought an Apple IIe which I'm now beginning to restore.

Along with it came some cards, but without any documentation.

One of these cards is this serial interface card.

Does anyone have any manual or documentation or information about it?

 

Googling it I only found one place where another guy also asks about it, but no answer except about the Super Serial Card which this is not.

https://www.retrocomputacion.com/forum/foro-general-apple/81431/mi-apple-ii/?p=2#post-81634

 

What I want the most is a description of the DIP switches.

I have not tested it yet, but testing will be difficult without knowing how to set the switches.

 

Here is a photo of it.

Offline
Last seen: 6 hours 54 min ago
Joined: Jul 5 2018 - 09:44
Posts: 2536
It is probably more or less

It is probably more or less compatible with the older Apple Communications Card.  It uses the 6850 which is suitable mainly only for devices 1200 baud or slower because it has no buffering at all (yes I know people will tell me that it can be used faster, but there are better choices).  The dip switches are obviously used to set the baud rate and probably also stop bits and parity.  Unfortunatelly I don't know what those settings are, although some guesses could probably be made.  Or it could be figured out by trial and error by attaching something like a terminal, or a computer running terminal software and then trying different baud rates there until data was moving cleanly.

 

 

Offline
Last seen: 6 hours 54 min ago
Joined: Jul 5 2018 - 09:44
Posts: 2536
Also, someone on here sold

Also, someone on here sold one...  but like 16 years ago...

 

https://www.applefritter.com/content/sold-cards-apple-brand-hispeed-scsi-820-0153-ca-serial-interface-card

 

 

 

Offline
Last seen: 6 hours 54 min ago
Joined: Jul 5 2018 - 09:44
Posts: 2536
The seller is still active,

The seller is still active, but the buyer hasn't been seen here in 7 years.  Just on an outside chance I sent him a private message to see if he is still around and if he has the manual.

 

Offline
Last seen: 6 days 21 hours ago
Joined: Aug 18 2017 - 16:53
Posts: 164
You can find it out. The 6850

You can find it out. The 6850 doesn't have an internal baud generator. There must be a clock signal at pin 3 (receive) and 4 (transmit).

I've never seen this card before nor the backside. So that's what I think:

The LS163 (pin 2) is connected to a clock signal from the slot interface. The output (pin 11, 12, 13, 14, or 15) is connected to the CD4024 (pin 1). The CD4024 is a binary divider with 7 stages.The outputs (3, 4, 5, 6, 9, 11, or 12) are connected to the dip switches. One must be switched on to connect this clock signal to the clock inputs of the ACIA 6850.

 

Regards

Ralf

Offline
Last seen: 6 hours 54 min ago
Joined: Jul 5 2018 - 09:44
Posts: 2536
I agree with you Ralf, those

I agree with you Ralf, those two chips are most likely what is generating the baud rate signals to the 6850.  My guess given the age of the card is the bit that is flipped on is probably for either 1200 or 2400 baud, those are the more common speeds of cards used in the //e era.  Possibly it was last used for a printer given that the plug is in the connector not marked modem.  6850 cards are fine for use with printers since high throughput isn't a critical thing and handshaking is fine to prevent data loss.

 

 

Offline
Last seen: 1 week 4 days ago
Joined: Aug 4 2015 - 14:30
Posts: 150
RalfK wrote:I've never seen
RalfK wrote:

I've never seen this card before nor the backside. So that's what I think:

 

Here is the backside.

Offline
Last seen: 1 week 4 days ago
Joined: Aug 4 2015 - 14:30
Posts: 150
RalfK wrote:There must be a
RalfK wrote:

There must be a clock signal at pin 3 (receive) and 4 (transmit).

I put my oscilloscope on pin 3 and 4 of the 6850.

This is the result.

I tested with with only one of the switches in ON position each time.

Except for switch 8 which seems to act as some kind of modifier.

Least significant digits fluctuated a little.

SwitchKhz - Switch 8 OFFKhz - Switch 8 ON
7 on2.43181.7477
6 on4.86343.4955
5 on9.7266.9911
4 on19.45313.982
3 on38.90627.964
2 on155.62111.86
1 on311.20223.72

 

 

 

 

 

 

 

 

 

Edit: Sigh - this table looked really good in the editor. Now it looks just bad.

There seems to be a pattern here, although I am not quite sure how to interpret these numbers.

Switch 7 which gives 2.4318 KHz could perhaps be 2400 baud?

To use this with ADTPro I would need 19200 or 115200 baud I think. But how do I get that?

Offline
Last seen: 6 hours 54 min ago
Joined: Jul 5 2018 - 09:44
Posts: 2536
You'll probably need to make

You'll probably need to make sure you use a cable with flow control because a 6850 won't begin to handle 19200 let alone 115200 without losing characters unless some kind of handshaking is happening.  Really even a 6551 can't with only a single byte buffer.

 

Here's the 6850 datasheet.  If you look at the section on clock signals, they can have a 1, 16 or 64 multplier applied.  That might make the table make more sense?

 

http://dunfield.classiccmp.org/r/6850.pdf

 

 

Offline
Last seen: 6 days 21 hours ago
Joined: Aug 18 2017 - 16:53
Posts: 164
tolderlund wrote:Switch 7
tolderlund wrote:
Switch 7 which gives 2.4318 KHz could perhaps be 2400 baud?
 
 

 That's ok :-) AFAIK +/-3% tolerance is allowed. Every value is within that tolerance.

 

 

Edit: 19200 Baud is 4 on and 8 off,

115kBaud is 2 on and 8 on,

the other switches are off. The firmware code will use a 6850 internal divider of 1, and nor 16 neither 64.

 

Offline
Last seen: 6 hours 54 min ago
Joined: Jul 5 2018 - 09:44
Posts: 2536
It's an interesting design

It's an interesting design with the lowest baud being 2400.  That clearly puts it fairly well into the //e era probably, since before that there were a lot of commonly used serial devices that wouldn't go that high.

The idea of using a system clock divided down for baud rate isn't unique, Apple's 6850 based Communications card uses a similar concept, except that card is really intended for devices running at 110 or 300 baud.  There were mods people would do for those cards to run them at higher speeds.

There are a number of other cards like the CCS 7710 and the Mountain Hardware CPS that use the 6850, but most of those are a more conventional design using a baud rate generator chip and a crystal.

 

Offline
Last seen: 1 week 4 days ago
Joined: Aug 4 2015 - 14:30
Posts: 150
No signal

I have not much luck getting the card to work with ADTPro.

I have a Super Serial Card and a cable which works in the Apple IIe and a Apple II europlus.

But this card won't work.

I have measured the TxD line (pin 2) on the DB25 connector and the Tx Data pin 6 on the 6850.

It gives a signal of 300-500 millivolt when idle and up to 600 mV when I try do use DIR command in ADTPro, which is far too low afaik.

But the signals just looks like random noise.

Could the 6850 be faulty or cold the fault be somewhere else?

I'm not an expert in these things, so any advice?

Offline
Last seen: 6 hours 54 min ago
Joined: Jul 5 2018 - 09:44
Posts: 2536
The 6850 is not responsible

The 6850 is not responsible for generating the 12V line levels, it operates at 5V if I'm not mistaken, the 1488 and 1489 chips are the RS-232 line drivers.

 

Log in or register to post comments