RE: [N8VEM: 2481] Re: Disk I/O IDE Session Log

["Andrew Lynch" <lyn...@yahoo.com>]

Hi Rich!  This is getting better by the moment!  I just noticed that pin 27
on the XT-IDE project (IORDY) has a 10K ohm pull down resistor on it.  It is
a wonder anyone gets an IDE interface to work at all!

It appears the IORDY is an output from the IDE hard drive so I cannot see
why that would be pulled low on the controller side.  

Thanks and have a nice day!

Andrew Lynch



> -----Original Message-----
> From: n8...@googlegroups.com [mailto:n8...@googlegroups.com] On Behalf Of
> Richard A. Cini
> Sent: Tuesday, February 10, 2009 7:13 PM
> To: N8VEM-Post
> Subject: [N8VEM: 2481] Re: Disk I/O IDE Session Log
> 
> 
> All:
> 
>     Andrew sent me a few links to Web sites used as ³inspiration² for the
> Disk-I/O interface. The information is inconsistent (again, the IDE/ATA
> "standard" is so flexible), but basically here¹s what I¹ve concluded
> today:
> 
>     If the host controller isn¹t using DMA, DMARQ (an output signal from
> the
> drive) should be left floating (this is R2).
> 
[AJL>] 

Agree.  Pulling low an IDE hard drive output makes no sense in retrospect
even though some IDE interface designs incorporate it anyway.


>     /DMACK is an output from the host to the drive. If DMA isn¹t being
> used,
> it should be pulled-up. The current Disk-I/O schematic shows it as being
> pulled down.
> 

[AJL>] 

Agree.  Easy fix to cut the GND trace leading to R3 and add a jumper to VCC.

>     For Cable Select (CSEL), it appears that the drive usually provides
> the
> 10k pull-up. If not using Cable Select, make sure to use the jumper on the
> drive to select Master so that it's pulled-down.
> 

[AJL>] 

Agree.

>     So, for testing purposes (tonight hopefully), I plan on pulling the
> bottom legs (closest to the IDE connector) of R1 (CSEL pull-up) and R2
> (DMARQ pull-up), and R3 and connecting the free leg of R3 to Vcc (to
> create
> a pull-up rather than a pull-down). People¹s experience with their drives
> working with all resistors removed is consistent with what I outlined
> above
> with the exception of the pull-up on /DMACK.
> 

[AJL>] 
Good luck my friend!  Good hunting!

Thanks and have a nice day!

Andrew Lynch

PS, I fear we will be going through this all over again with the 8 bit IDE
project on the VCF so save your notes!  There are all sorts of squirrelly
stuff on the XT-IDE design as well.  I am glad to have insisted on building
a prototype for them or it probably would have ended badly!  That's no
guarantee but it's better than the guess game that the IDE "standard" is
turning out to be!  :-)

> 
> Rich
> 
> --
> Rich Cini
> Collector of Classic Computers
> Build Master and lead engineer, Altair32 Emulator
> http://www.altair32.com
> http://www.classiccmp.org/cini
> 
> 
> 
>