21 Nov
2007
21 Nov
'07
7:54 p.m.
All:
As you all know from previous posts, I?m having trouble with the drive
system on my IMSAI. While I wait to receive a memory board and some other
stuff from a friend of mine, Herb Johnson gave me some new energy in looking
at the existing iCOM Frugal Floppy system.
This system uses Pertec FD400 drives. One works perfectly, the other one
has consistent CRC errors as evidenced by the CRC light on the drive
cabinet, and BDOS errors reported by CP/M. I ensured that the spindle is
engaging the diskette hub, and I also cleaned the heads using a wet cleaning
diskette.
Is there any common failure mode on this model of drive that I should
start tracking down? I looked over the maintenance procedures for the drive
and many if the tests allude to a diagnostics program that can manually
control the spindle and head stepping. None of the iCOM manuals I have
provide diagnostics code for manually controlling the drives in this manner.
Before I start coding, does anyone have diagnostics code I can use? I
found some code in the SIGM archive (for the iCOM 3712 controller) but it
doesn?t seem to work properly with this controller.
Thanks.
Rich
--
Rich Cini
Collector of Classic Computers
Build Master and lead engineer, Altair32 Emulator
http://www.altair32.com
http://highgate.comm.sfu.ca/~rcini/classiccmp
21 Nov
21 Nov
10:47 p.m.
From: rcini at optonline.net> > All:> > As
you all know from previous posts, I?m having trouble with the drive> system on my
IMSAI. While I wait to receive a memory board and some other> stuff from a friend of
mine, Herb Johnson gave me some new energy in looking> at the existing iCOM Frugal
Floppy system.> > This system uses Pertec FD400 drives. One works perfectly, the
other one> has consistent CRC errors as evidenced by the CRC light on the drive>
cabinet, and BDOS errors reported by CP/M. I ensured that the spindle is> engaging the
diskette hub, and I also cleaned the heads using a wet cleaning> diskette.> > Is
there any common failure mode on this model of drive that I should> start tracking
down? I looked over the maintenance procedures for the drive> and many if the tests
allude to a diagnostics program that can manually> control the spindle and head
stepping. None of the iCOM manuals I have> provide diagnostics code for manually
controlling the drives in this manner.> > Before I start coding, does anyone have
diagnostics code I can use? I> found some code in the SIGM archive (for the iCOM 3712
controller) but it> doesn?t seem to work properly with this controller.>
Hi
As I've said before, if it is a belt drive, the surfaces have to
be vary smooth. The Flywheel end is especially sensistive to
all most invisible roughness. Feel it with your finger. You can sense
even small roughness.
Of course, check the speed with a strobe.
Dwight
_________________________________________________________________
Share life as it happens with the new Windows Live.Download today it's FREE!
http://www.windowslive.com/share.html?ocid=TXT_TAGLM_Wave2_sharelife_112007
10:56 p.m.
I thought about the speed. It's a direct-drive DC spindle motor, not belt
drive, in this model so I'll have to hunt around to see if there's a speed
control pot somewhere.
On 11/21/07 10:47 PM, "dwight elvey" <dkelvey at hotmail.com> wrote:
Rich
--
Rich Cini
Collector of Classic Computers
Build Master and lead engineer, Altair32 Emulator
http://www.altair32.com
http://highgate.comm.sfu.ca/~rcini/classiccmp
From: rcini at optonline.net> > All:>
> As you all know from previous posts, I?m
having trouble with the drive> system on my IMSAI. While I wait to receive a
memory board and some other> stuff from a friend of mine, Herb Johnson gave
me some new energy in looking> at the existing iCOM Frugal Floppy system.> >
This system uses Pertec FD400 drives. One works perfectly, the other one> has
consistent CRC errors as evidenced by the CRC light on the drive> cabinet,
and BDOS errors reported by CP/M. I ensured that the spindle is> engaging the
diskette hub, and I also cleaned the heads using a wet cleaning> diskette.> >
Is there any common failure mode on this model of drive that I should> start
tracking down? I looked over the maintenance procedures for the drive> and
many if the tests allude to a diagnostics program that can manually> control
the spindle and head stepping. None of the iCOM manuals I have> provide
diagnostics code for manually controlling the drives in this manner.> >
Before I start coding, does anyone have diagnostics code I can use? I> found
some code in the SIGM archive (for the iCOM 3712 controller) but it> doesn?t
seem to work properly with this controller.>
Hi
As I've said before, if it is a belt drive, the surfaces have to
be vary smooth. The Flywheel end is especially sensistive to
all most invisible roughness. Feel it with your finger. You can sense
even small roughness.
Of course, check the speed with a strobe.
Dwight
_________________________________________________________________
Share life as it happens with the new Windows Live.Download today it's FREE!
http://www.windowslive.com/share.html?ocid=TXT_TAGLM_Wave2_sharelife_112007 
11:08 p.m.
At 06:56 PM 11/21/2007, you wrote:
I thought about the speed. It's a direct-drive DC
spindle motor, not belt
drive, in this model so I'll have to hunt around to see if there's a speed
control pot somewhere.
The peak to peak voltage from the heads will tell if its a head
problem. There are still places selling spare parts for those drives.
22 Nov
22 Nov
7:58 a.m.
That's interesting -- who?
The way that the service manual tells you to check the speed is the pulse
width on a specific test point on the PCB. I was just going to use a laser
tach on the spindle...but I have to get a tach first :-)
On 11/21/07 11:08 PM, "Grant Stockly" <grant at stockly.com> wrote:
At 06:56 PM 11/21/2007, you wrote:
Rich
--
Rich Cini
Collector of Classic Computers
Build Master and lead engineer, Altair32 Emulator
http://www.altair32.com
http://highgate.comm.sfu.ca/~rcini/classiccmp
I thought about the speed. It's a
direct-drive DC spindle motor, not belt
drive, in this model so I'll have to hunt around to see if there's a speed
control pot somewhere.
The peak to peak voltage from the heads will tell if its a head
problem. There are still places selling spare parts for those drives.
12:13 p.m.
On Thursday 22 November 2007 12:58:17 Richard A. Cini wrote:
That's interesting -- who?
The way that the service manual tells you to check the speed is the pulse
width on a specific test point on the PCB. I was just going to use a laser
tach on the spindle...but I have to get a tach first :-)
High-brightness LED and a precise signal generator (or possibly an IO pin on
some sort of microcontroller).
Gordon
3:17 p.m.
On Thursday 22 November 2007 12:58:17 Richard A. Cini wrote:
That's interesting -- who?
The way that the service manual tells you to check the speed is the pulse
width on a specific test point on the PCB. I was just going to use a laser
tach on the spindle...but I have to get a tach first :-)
I'm coming in late on this - sorry if it's been discussed
Is there some reason to not check the spindle speed in software (watch for
index pulses)?
Many/most drives have 50Hz/60Hz zebra discs on the flywheel, for using a
flickering fluorescent light.
4:12 p.m.
On 22 Nov 2007 at 12:17, Fred Cisin wrote:
Is there some reason to not check the spindle speed in
software (watch for
index pulses)?
I gather that the system that these are hooked to is an S100 box that
may not have any sort of timer facility in it. To be sure, it would
be possible to hook at least the index link to a PC parallel port for
checking--but that would require writing code for the wretched thing.
Many/most drives have 50Hz/60Hz zebra discs on the
flywheel, for using a
flickering fluorescent light.
Indeed--and for 360 RPM, one can use a 45 RPM phono strobe disc just
fine (360/45 = 8.00). A 33 1/3 phono strobe disc works for 300 RPM
drives (300/33.333... = 9.00..). Strobe discs are available free for
the download at sites such as
http://www.extremephono.com/free_turntable_strobe_disk.htm
I prefer a neon bulb or LED for strobe checks--the strobe bars stand
out more clearly. Just be sure to find a disc that matches your
mains frequency (50 or 60 Hz).
Cheers,
Chuck
10:58 p.m.
Chuck/Tony/Dwight/der Mouse:
This is exactly why I love you guys...
I forgot about the neon bulb and all of that. It's been years, but I had
a linear tracking turntable as a teenager and it had a built-in neon-based
strobe.
The Postscript code is great. I also have a TM100 in the shop and I
downloaded the turntable strobe disk.
Something to do this weekend :-)
Thanks again all, and Happy Thanksgiving.
Rich
On 11/22/07 4:12 PM, "Chuck Guzis" <cclist at sydex.com> wrote:
On 22 Nov 2007 at 12:17, Fred Cisin wrote:
Rich
--
Rich Cini
Collector of Classic Computers
Build Master and lead engineer, Altair32 Emulator
http://www.altair32.com
http://highgate.comm.sfu.ca/~rcini/classiccmp
Is there some reason to not check the spindle
speed in software (watch for
index pulses)?
I gather that the system that these are hooked to is an S100 box that
may not have any sort of timer facility in it. To be sure, it would
be possible to hook at least the index link to a PC parallel port for
checking--but that would require writing code for the wretched thing.
Many/most drives have 50Hz/60Hz zebra discs on
the flywheel, for using a
flickering fluorescent light.
Indeed--and for 360 RPM, one can use a 45 RPM phono strobe disc just
fine (360/45 = 8.00). A 33 1/3 phono strobe disc works for 300 RPM
drives (300/33.333... = 9.00..). Strobe discs are available free for
the download at sites such as
http://www.extremephono.com/free_turntable_strobe_disk.htm
I prefer a neon bulb or LED for strobe checks--the strobe bars stand
out more clearly. Just be sure to find a disc that matches your
mains frequency (50 or 60 Hz).
Cheers,
Chuck
6:14 p.m.
That's interesting -- who?
The way that the service manual tells you to check the speed is the pulse
width on a specific test point on the PCB. I was just going to use a laser
tach on the spindle...but I have to get a tach first :-)
You can of course determine the spindle speed by looking at the timing of
the index pulses (one per revolution on a soft-sectored disk, so 5Hz for
a 300 rpm disk, etc).
But there is one ;gotcha' in doing this. Don't simply hook up a frequency
counter and measure the freqwency. If you have a 1-second gate time,
you'll get 5 pulses through for quite a wide range of speeds. It's just
not accurate enough. It's better to use the period maode of your counter
and see how many (say) 1MHz clock pulses you can count between index
pulses from the drive.
Alternatively make a stroboscope disk (either copy the one off a TM100
pulley, or work out howm many segments you need and write a trivial bit
of postscript to draw it), stick that on the spindle pulley or
direct-drive motor rotor and view it under amins-powered lamp.
-tony
6086
days inactive
6087
days old
9 comments
7 participants
participants (7)
-
ard＠p850ug1.demon.co.uk -
cclist＠sydex.com -
cisin＠xenosoft.com -
dkelvey＠hotmail.com -
gordonjcp＠gjcp.net -
grant＠stockly.com -
rcini＠optonline.net