TRS-80 Model 4 help requested

From: Tony Duell <ard_at_p850ug1.demon.co.uk>
Date: Sat Mar 1 16:48:00 2003

> On Fri, 28 Feb 2003, Tony Duell wrote:
>
> > > On Thu, 27 Feb 2003, Tony Duell wrote:
> > >
> > > > > Aparently, the Drive 0: is bad. Replacing it with the 1: or an
> > > >
> > > > Does the spindle motor run on this drive when you try to boot from it?
> > > > Does the LED come on?
> > >
> > > Sorry, guess I should have been more specific. The (drive that was
> > > 0:) drive spins, and the LED comes on, but the machine still says "Cass?".
> >
> > OK...
> >
> > What test equipment do you have available? Multimeter? Logic probe? Logic
> > analyser? 'scope?
>
> "Yes." I've actually got all of those. Don't particularly like using my

OK, excellent!

> DMM to observe changing signals, though... I need to find/build myself a
> decent analog meter sometime.

Yes, I find an analogue VOM to be more useful than a digital one for most
tests. It's rarely necessary to have a precise measurement of a voltage
(or whatever), it's much more useful to be able to see fluctuations.

However, the first tests probably will use the DMM. You should check the
output voltages of both PSUs in the machine. One can be checked at the
CPU board J4 connector (+5V, ground, +12V, -12V). The other at a disk
drive power connector (standard pinout).

Assuming the power supply voltages are OK, now try moving the drive head
(on drive 0) towards the spindle (off track 0) with the machine off. Put
in a disk and turn on. Does the head restore to track 0? If so, then the
CPU board is talking to the disk controller, and the latter is doing
something.

The drives have the standard pinout :
All odd # pins : gorund
8 : Index
10 : Select 0
12 : Select 1
16 : Motor On
18 : Step Direction
20 : Step Pulse
22 : Write Data
24 : Write Gate
26 : Track 00
28 : Write Protect
30 : Read Data
32 : Side Select
(Others not used on the controller)

With the machine trying to boot, look at the Index signal with the logic
probe. Is it pulsing? If you can, measure the frequency (should be 5Hz,
corresponding to 300 rpm).

Grab the read data signal with the logi analyser. You're not trying to
decode the data, but does it look reasonable?

Now look at the disk controller board. There should be a 4MHz clock on
TP18. Also check for 1 MHz clock on the outputs of U24b ('74).

Lok at U4 ('244). The outputs of sections (a) and (b) are the read and
write lines to the disk controller -- and are gated on the CPU board with
the appropriate drive select (CS/ on the disk controller chip is
grounded). Do these go low? If not, you may have CPU board problems.

OK so far? Check the output of U15a ('123), which is the (slightly
processed) read data signal from the drive. Is that toggling?

Look at TP11 too. That's the read clock output of the WD1691 chip. Is
that clcoking (and on-frequency)?

Look at the DRQ signal (output of U16f ('367)). Is that changing state?
If not, then for some reason the disk controller chip doesn't think it's
transfering data. Could be a problem with the 1793 (I've had to change a
number of these), or CPU board problems.

>
> > I can probably suggest signals to look at to find out exactly what's
> > going on...
>
> That would be most appreciated.
>
> > > The other drive (which I had replaced it with) attempts to read from the
> > > disk, and either displays "Diskette?" or nothing if there's no disk in the
> > > drive or a disk, respectively.
> >
> > Of course at this point you don't _know_ that the disk controller or CPU
> > boards are working correctly...
>
> That's what I'm betting the problem is.. it'd be nice to be able to rule
> out the drive controller and CPU board.

Which CPU board do you have? I only have information on the older one,
which has 2 40 pin chips (the Z80 CPU and a 6845 CRT controller, maybe
with a strange number, next to each other) and a load of small PALs and
TTL on it. I beleive later boards have a larger custom gate array chip on
them.

[Terminatior]

> > Are you sure? It normally goes in a DIL socket near the interface cable.
> > Is there a totally empty socket on both drives?
>
> There's a SIP socket pretty much 'on top of' the I/O connector. I don't

If it _is_ a SIP, then these are not the normal drives for an M4 (this is
an M4, right, not an M4P or M4D). Doesn't mean they can't work (Radio
Shack used standard drives, and my M4 has very non-stnadard drives (2 40
cylinder, 2 80 cylinder, all DS, all internal!). The only problem will be
that I may not have service data on the drives.

> see anything else (DIP shaped) on the board that isn't a 'normal' black
> plastic IC. I'll check again, but unless it's mounted inside the drive...

No, it's on the logic board.

>
> > > What resistance is it? I could probably try making one out of spare
> >
> > 150 Ohms, but I can't remember if the resistors are separate or if they
> > have a common connection to the highest-numbered pin on the package. I
> > think the former.
>
> OK, I think I have one of those laying around, with a common pin. If the
> RAM swapping doesn't help, then I'll finding a terminator pack to stick in
> there. Oh wait, I have one right here from a SCSI drive. Yes, testing it
> with my handy-dandy crappy-DVM, it's 148ohms. I might try this first.

Watch out. SCSI termiantors are not simple resistor-to-Vcc networks.
They're potential dividers (220 Ohms to Vcc, 330 Ohms to ground
normally). Not really what you want here.

-tony
Received on Sat Mar 01 2003 - 16:48:00 GMT