On 4/30/18 10:20 AM, Eric Christopherson via cctalk wrote:
On Fri, Apr 27, 2018 at 5:40 PM, Alan Perry via cctalk <
[email protected]> wrote:
I keep them all. Not counting the bad ones in the SS1 and 2, I have 7.
I can send them to you. I don't mind pick up the shipping costs for
something small like that. But the $70 that is it going to cost to ship the
SS20 to its new home is another matter.
alan
On 4/27/18 3:33 PM, systems_glitch wrote:
You can always send me the dead modules and I'll rebuild them
(GlitchWorks == me, my wife sometimes helps with assembly). Whatever you
do, don't throw out the dead NVRAMs -- I'll buy them or pay for you to ship
them or whatever, they're not making more and they're the only solution
that's 100% compatible.
Yeah, the "still works but pukes errors" is the typical symptom of the
newer, slightly incompatible 48T02s in Sun machines. I don't recall if mine
kept accurate time with the newer modules.
Thanks,
Jonathan
On Fri, Apr 27, 2018 at 6:28 PM, Alan Perry via cctalk <
[email protected] <mailto:[email protected]>> wrote:
The ones from Mouser work well enough in every system that I have
used them in. I still get the IDPROM corrupt message on boot on
some systems, but it holds the MAC and the systems boot without
intervention.
I tried to repair a few and botched most of them. I know that I
should be using the GlitchWorks stuff, but it has been easier to
just buy something that I can plug in.
alan
On 4/27/18 3:15 PM, systems_glitch via cctalk wrote:
Don't get the new MK48T02/MK48T08s from Mouser et al, they're
not fully
compatible. They will retain NVRAM but the clock part is
different and
you'll get an error on that (system won't autoboot). Rebuild
your old
NVRAM! I made up some little boards to make the repair cleaner
and faster
to do (I had about 50 NVRAMs to repair):
http://www.glitchwrks.com/2017/08/01/gw-48t02-1
<http://www.glitchwrks.com/2017/08/01/gw-48t02-1>
There are other guides for tacking on a coin cell holder
without cutting
off the entire top encapsulation, but if you do that, it may
not fit under
SBus cards if you're doing it on a system that puts SBus slots
over the
NVRAM.
Thanks,
Jonathan
On Fri, Apr 27, 2018 at 6:03 PM, Glen Slick via cctalk <
[email protected] <mailto:[email protected]>> wrote:
On Fri, Apr 27, 2018 at 1:55 PM, Alan Perry via cctalk
<[email protected] <mailto:[email protected]>> wrote:
- SPARCstation 1. Chassis is intact. It has a bad
IDPROM; aside from that
it passes onboard diagnostics. It has 12M memory, no
HDD now, and a 3.5"
floppy drive. It has no SBus cards. Aside from the
IDPROM, it doesn't
have any issues (but I haven't run an OS on it yet).
Like the SS2, it
needs a bath. A small portion of the plastic cover
over the rear of the
case is broken off.
What are these "actual parts expenses"? IDPROMs are
around $25 on Mouser.
SCSI HDDs start around $70 shipped on eBay and SCSI2SD
are $60 plus
shipping to me plus the SD price. Given the price of
25 year old HDDs
with
a stated service life of 5 years (according to one
spec sheet that I
read),
SCSI2SD looks pretty attractive.
When you say IDPROM, is that a Dallas built-in battery
NVRAM type of
thing? I have an SS1 with a dead NVRAM thing. Are the
currently
available versions of those new at Mouser fully
compatible? Those are
one of those things that the new versions aren't always fully
compatible with the old versions for some systems, even
though they
are supposed to be.
My SS1 is also in the Seattle area. If there is much
demand for those
it's probably one of those systems I'll never get around
to doing
anything with it myself. I also have a 4/110. Those seem
to be a lot
less common, and maybe more collectible.
Last year some time I replaced the timekeeper in my first SPARC with one I
ordered off eBay. It's been a while since I tried it in that machine* but,
last I knew, it was slow at keeping time but time was still moving forward,
and it hadn't lost its MAC or hostid. Would that be caused by it being one
of the newer not-100%-compatible ones? Alternately, I read somewhere that
timekeeper/NVRAM/IDPROM chips need to either have power available to them
most of the time or need to have the computer powered up regularly (not
sure which it was) -- is there any truth to either of those?
* I did try the "new" timekeeper in a newer SPARC acquisition of mine a
week or so ago, because I knew the new workstation's battery was bad.
Unfortunately the computer failed to give any monitor signal when I did
that, and I haven't had a chance to hook it up via serial to see what's
working and what isn't. I wonder if that could be another failure mode of
those chips, or if it just means the workstation's not quite working (I was
told that it had just been booted recently using its original IDPROM, but
defaults had to be set at boot; I also haven't gotten a chance to try the
original IDPROM in it to see what would happen.)
Jonathan the GlitchWorks guy has better experience with this than I do,
but I have new M48T02s/M48T08s from Mouser in all of the systems that I
have had to replace them in. That would be in SPARCstation 1+, 5, 20,
IPC, IPX, Classic, and LX. They have worked good enough for me, but I
only occasionally power them on (which is why I was happy that VCF gave
me an excuse use them and network them all together).
I retrofitted an external battery to the one in my Axil 320, but never
found a satisfactory way to fix the battery in place. After I finish
clearing out the SPARCs that came from Pete's place, I will be giving
the Axil some attention and may go ahead and GlitchWorks it (since I am
out of M48T08s to just grab out if the tube).
alan
