The Jasmin was a floppy disk controller and 3" floppy disk drive for the Oric. It was released in 1984 by T.R.A.N., a French company. It supported up to 4 drives and featured an operating system, named TDOS and later FTDOS. In 1985, T.R.A.N. released the Jasmin II that was, I believe, the same thing but with a double-sided 3" drive.
Here is a schematic diagram of the controller board (based on the WD1770 FDC). It does not come from T.R.A.N., but I believe it to be accurate. It was done from a Jasmin I.
Component list :
|
|
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
A question mark in the left column means that there was no mark on the board corresponding to that component. A question mark in the right column means that I could not decipher the marking on the component
This is another schematic diagram of the same controller board, straight from T.R.A.N., reprinted in Théoric no. 22 (1986-07). It includes a short technical explanation in French. English translation follows :
The Jasmin controller board
Exclusive property of T.R.A.N., the schematics of the Jasmin II controller are reproduced here for information only, for the benefit of Jasmin owners.
As one can see, the controller board is built around a custom chip (C16). The drive is handled by a Western Digital FDC 1770 (C13), a kind of microprocessor designed specifically for that purpose. The custom chip handles I/O, address decoding and switching to overlay RAM.
An EPROM 2716 (C17) contains the boot program. To fully understand how it all works, it is necessary to have a some knowledge of assembly language and of the programming of the FDC 1770. If there is enough demand, we will publish an article on the subject...
Let's take a look at the role of the other chips.
To access the overlay RAM, it is necessary to assert the !MAP signal (pin 1 of the extension connector). The timing is rather tight. !MAP must come 80 to 100 ns before the raising edge of PHI2 and remain active for 250 ns. That is the job of the monostables in C10. If an improper timing is used, the display is affected.
Everybody has heard about the problems caused by PHI2, the generally very noisy clock signal of the Oric. This signal is reconditioned in the Jasmin by C6, a 4069 CMOS gate connected in a somewhat unusual fashion. From the start, that simple gate has given the Jasmin the edge over the Microdisc (the latter prevents from loading cassettes when connected to the bus, which the Jasmin doesn't).
The most heavily used address signals, A0, A1, A2 and A3, are also routed through a 4069 gate, for impedance matching.
Thanks to these precautions, the Jasmin board functions reliably in spite of the shortcomings of the extension bus.
Finally, C14, a 1-to-8 multiplexer, handles the 8 outputs :
- 4 drive select registers (0 to 3)
- the !ROMDIS signal
- the RAM overlay access register
- the side select register
- the FDC 1770 reset signal
Obviously, the board is a sound design, based on mature technology.
I don't know who wrote the article. It's not signed.
The component list for the power supply of the Jasmin 1 :
|
|
|
|
||||||||||||||||||||||||||||||||||||||||||||||
http://www.teaser.fr/~amajorel/jasmin/, last modified AYM 2007-03-08.
