2 Feb
2014
2 Feb
'14
6:42 a.m.
As you may be aware I am trying to build an emulator for MFM disks (mainly
DEC RD5x disks) using an FPGA. Things have been going moderately well in
that I have managed to get the MicroVAX 2000 formatter (TEST 70) to
recognise the appropriate RD5x disk by counting cylinders.
However, now I am at the stage when I want the controller in the 2000 to
write something. I am trying to use the formatter to write data to the disk.
However, the WRITE GATE signal never gets asserted.
I cannot see what is stopping the 2000 from asserting WRITE GATE. I have
implemented the Index signal which looks right on the oscilloscope, and of
course all the signals for track selection. I am sending dummy data on READ
DATA (all logical 0s as far as I can be sure). I am as sure as I can be that
I have the right pin on the connector (32 according to the 2000's tech
manual).
This is the output from the 2000:
>> test 70
KA410-A RDRXfmt
VSfmt_QUE_unitno (0-2) ? 0
VSfmt_STS_Siz .......... RD31
VSfmt_QUE_SerNbr (0-999999999) ? 0
VSfmt_QUE_RUsure (DUA0 1/0) ? 1
VSfmt_STS_RdMbb ..........?
VSfmt_STS_FMTing .??
VSfmt_RES_ERR #3
84 FAIL
>>
Any ideas?
Thanks
Rob
2 Feb
2 Feb
12:20 p.m.
On 2014-02-02 15:42, Robert Jarratt wrote:
As you may be aware I am trying to build an emulator
for MFM disks (mainly
DEC RD5x disks) using an FPGA. Things have been going moderately well in
that I have managed to get the MicroVAX 2000 formatter (TEST 70) to
recognise the appropriate RD5x disk by counting cylinders.
However, now I am at the stage when I want the controller in the 2000 to
write something. I am trying to use the formatter to write data to the disk.
However, the WRITE GATE signal never gets asserted.
I cannot see what is stopping the 2000 from asserting WRITE GATE. I have
implemented the Index signal which looks right on the oscilloscope, and of
course all the signals for track selection. I am sending dummy data on READ
DATA (all logical 0s as far as I can be sure). I am as sure as I can be that
I have the right pin on the connector (32 according to the 2000's tech
manual).
Just the obvious, but:
seek_complete is low?
ready is low?
track 0 is low?
index you said is OK,
write_fault is low?
2:44 p.m.
-----Original Message-----
From: cctalk-bounces at classiccmp.org [mailto:cctalk-
bounces at classiccmp.org] On Behalf Of emanuel stiebler
Sent: 02 February 2014 20:21
To: General Discussion: On-Topic and Off-Topic Posts
Subject: Re: Trouble Getting MFM Controller to Write
On 2014-02-02 15:42, Robert Jarratt wrote:
the
As you may be aware I am trying to build an
emulator for MFM disks
(mainly DEC RD5x disks) using an FPGA. Things have been going
moderately well in that I have managed to get the MicroVAX 2000
formatter (TEST 70) to recognise the appropriate RD5x disk by counting
cylinders.
>
>
>
> However, now I am at the stage when I want the controller in the 2000
> to write something. I am trying to use the formatter to write data to disk.
I have checked all of these, but when I get back to this next weekend I will
check again. I don't think the track counting would work if most of those
were not right, but it is still worth another check.
Regards
Rob
However, the WRITE GATE signal never gets
asserted.
I cannot see what is stopping the 2000 from asserting WRITE GATE. I
have implemented the Index signal which looks right on the
oscilloscope, and of course all the signals for track selection. I am
sending dummy data on READ DATA (all logical 0s as far as I can be
sure). I am as sure as I can be that I have the right pin on the
connector (32 according to the 2000's tech manual).
Just the obvious, but:
seek_complete is low?
ready is low?
track 0 is low?
index you said is OK,
write_fault is low?
3 Feb
3 Feb
12:28 p.m.
I just realised, you said Write Fault is low. Do you mean logical low or
physical low? According to the docs logical low is +5V on the interface,
and that is what I believe I am sending, ie logical low.
Regards
Rob
On 2 February 2014 22:44, Robert Jarratt <robert.jarratt at ntlworld.com>wrote:
-----Original Message-----
From: cctalk-bounces at classiccmp.org [mailto:cctalk-
bounces at classiccmp.org] On Behalf Of emanuel stiebler
Sent: 02 February 2014 20:21
To: General Discussion: On-Topic and Off-Topic Posts
Subject: Re: Trouble Getting MFM Controller to Write
On 2014-02-02 15:42, Robert Jarratt wrote:
the
As you may be aware I am trying to build an
emulator for MFM disks
(mainly DEC RD5x disks) using an FPGA. Things have been going
moderately well in that I have managed to get the MicroVAX 2000
formatter (TEST 70) to recognise the appropriate RD5x disk by counting
cylinders.
>
>
>
> However, now I am at the stage when I want the controller in the 2000
> to write something. I am trying to use the formatter to write data to disk.
I have checked all of these, but when I get back to this next weekend I
will
check again. I don't think the track counting would work if most of those
were not right, but it is still worth another check.
Regards
Rob
However, the WRITE GATE signal never gets
asserted.
I cannot see what is stopping the 2000 from asserting WRITE GATE. I
have implemented the Index signal which looks right on the
oscilloscope, and of course all the signals for track selection. I am
sending dummy data on READ DATA (all logical 0s as far as I can be
sure). I am as sure as I can be that I have the right pin on the
connector (32 according to the 2000's tech manual).
Just the obvious, but:
seek_complete is low?
ready is low?
track 0 is low?
index you said is OK,
write_fault is low?
3979
days inactive
3980
days old
3 comments
2 participants
participants (2)
-
emu＠e-bbes.com -
robert.jarratt＠ntlworld.com