> The harness I have is not the same as either of the two that someone sent me
> pictures of. It is not a ribbon cable, it has discrete wires, of different
> lengths, its part number is 70-20450-01, rev C1 and there is a date of 9th
> Dec 1985 on it. The actual connectors though look like the connectors on the
> bad ribbon cable harness, I think they are MTA-156.
AFAIK this is a side issue. While the older pawer harness can overheat
_in use_, it will not cause the PSU to fail at switch-on. In other words,
if you have the older harness, you should reuild/replace it, but this is
not the cause of your problem/
>
> Some have suggested it blew because I had insufficient load. I had a TK50
> and an RD53 attached, would that no be sufficient. Furthermore I did not
As I said earlier, I'd not use thos 2 devices as dummy loads, they are simply
too valuale for that. Get soem 6V and 12V car bulbs...
While I beleive there are SMPSUs that fail if turned on with no/insufficient
load, I've yet to meet one. I've seen supplies that trip repeatedly if
given no load. I've seen supplies (including ones from DEC) where some
of the outputs give far too _low_ a voltage if the main output is not
suffiiently loaded. But not one that fails.
> actually switch the PSU on, just connected it to the mains, would the load
> connected to the PSU matter when the PSU is not actually switched on?
Hang on Are you saying the rocker switch on the front was not turned on?
In which case most of the PSU electronis was not powered. About the only
thing that could have failed in the way you descrie are the mains input
filter capacitors.
Have you opened up the PSU case yet? Can you see anything obviously
burnet or exploded? If so, what?
>
> There was also a suggestion that the PSU would have needed switching
> separately for 50Hz operation as well. The hardware manual I have for the
> machine tells me how to switch between 110 and 240, but does not say
> anything about switching it for frequency, so I suspect this was not an
> issue here.
It isn;t. There is no frequency adjsutment on the BA23 PSU.
>
> >From the various responses I think it would be unwise for me to attempt a
> repair. Is there anyone on this list who is in the UK who would be willing
> and able to fix this PSU, or who knows someone who might be?
Where are you in the UK?
-tony
Since it seems like Zilog is intent on committing suicide by getting rid
of its engineers, does anyone have any ideas on who will acquire the ruins
when it's all over?
--
David Griffith
dgriffi at cs.csubak.edu
A: Because it fouls the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?
8 volts and 16 volts is correct. Frankly, I would stay on LOW side (but not
below 7.5 volts). MANY systems burn up boards because they are towards the
high side (it was a huge issue for early SOL-20's). S-100 boards use linear
regulators, and many of them pushed those WAY too hard, to the point that
they got so hot they charred the board. Some notable candidates includes
the Tarbell double density controller card, the IMSAI SIO2-2 card and a
number of memory cards. At 9 volts, the linear regulator has to dissipate 4
volts x {whatever current}. A board like the Tarbell double density board
has one regulator for the entire board, about 60 ICs including some big LSI
chips (179x FDC & Intel 8257 DMA) and ONE linear regulator that gets
ridiculously hot at even 8 volts. The difference between 8 volts and 9
volts is 25% and makes a HUGE difference.
Barry Watzman
------------------------------
Message: 25
Date: Sat, 28 Feb 2009 10:03:10 -0500
From: "Andrew Lynch" <lynchaj at yahoo.com>
Subject: S-100 power supply voltage ranges
To: <cctalk at classiccmp.org>
Message-ID: <6FB56B9FDD5B49A78CDD0BB4DD3B4BF5 at andrewdesktop>
Content-Type: text/plain; charset="us-ascii"
Hi! What are acceptable voltage supply ranges for an S-100 power supply? I
am assuming 115VAC input. My intended application is a small home brew
S-100 backplane.
I have seen the S-100 voltage rails vary but I believe +15VDC, -15VDC, and
+9VDC would be sufficient. Herb's S-100 page lists the voltages as +18VDC,
-16VDC, and +8VDC but I have seen other voltages used.
Thanks and have a nice day!
Andrew Lynch
On Thu, 26 Feb 2009 14:18:59 -0700 Mike Loewen wrote:
> On Thu, 26 Feb 2009, Gordon JC Pearce wrote:
>
>> Didn't someone once work out that to store a single MP3 you'd need a
>> stack of 80-column cards ten miles high?
>
> Assuming a 6MB MP3 file and 40 bytes per card:
>
> 6 * 1048576 = 6291456 bytes
> / 40 bytes per card = 157286.4 (157285) cards
> * .178mm card thickness = 27996.73 mm high
> / 25.4mm per inch = 1102.233465 inches
> / 12 = 91.85278871 (91.85) feet
>
>
> Mike Loewen mloewen at cpumagic.scol.pa.us
> Old Technology http://sturgeon.css.psu.edu/~mloewen/Oldtech/
Umm if you store binary ... that is 80 bytes per card .. 46 feet per file.
Now how many dec tapes is that?
Errr, ISTR that a standard IBM, etc. punched card was 0.0065" thick rather than the 0.007" implied by the 0.178mm, BICBW. Also, since there are 12 rows per column, the most efficient packing technique could store 120 bytes in the up to 960 "bits" (hole positions) in the card.
Using those numbers, I came up with the following:
6,291,456 bytes / 120 = 52,428.8 cards (OK, 52,429)
52,429 * 0.0065 = 340.7885 inches or about 28.4 feet per file.
What an improvement. ;) (Is anyone else bored today?)
I also seem to recall that Univac or some other computer manufacturer actually used a 90 (?) column card for even better packing density. I'll leave that improved calculation to someone who actually remembers how many columns those cards were blessed with.
Later,
Charlie Carothers
--
My email address is csquared3 at tx dot rr dot com
Qbus -11s are poor candidates as they are hard to stall. That said you can display
the dynamic address and data. This does meand latching the multiplexed address and
data and providing led drivers for that. Also note as Chuck has said very little
of the CPU state is brought to the bus so you cannot see 11/34 or 11/70 style
info. To be able to write to memory or incrementially display a location is far
harder as the CPU is continiously active and hard to stop and restart without
some code.
A switch flipping blinkin lights front pannel for an 11 is not simething I've missed
and I"ve run 11/34s and 11/70s. In both those cases they were never used or needed.
Allison
>
>Subject: Re: compact pdp11 and front panel
> From: Dave McGuire <mcguire at neurotica.com>
> Date: Mon, 02 Mar 2009 12:11:44 -0500
> To: "General Discussion: On-Topic Posts Only" <cctech at classiccmp.org>
> Cc: cctalk at classiccmp.org
>
>On Mar 2, 2009, at 4:33 AM, David Griffith wrote:
>> Does anyone know anything about shoehorning a front panel with lots
>> of lights and switches onto a compact pdp11 (say, a /73 or /83)?
>
> I don't know anything about it specifically, but just from
>thinking about it...I think the best you'd probably be able to do is
>address and data LEDs, plus RUN, and maybe a few others...not much
>more (i.e., no address spaces, processor states, etc) because the
>signals aren't brought out to the pins on the J11 chip.
>
> But if that'd be enough, you could probably do it by using some
>buffers to drive the LEDs, and doing DMA cycles onto the bus for the
>switches. I'd probably take a whack at it if I had some time.
>
> -Dave
>
>--
>Dave McGuire
>Port Charlotte, FL
Does anyone know anything about shoehorning a front panel with lots of
lights and switches onto a compact pdp11 (say, a /73 or /83)?
--
David Griffith
dgriffi at cs.csubak.edu
A: Because it fouls the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?
On Thu, February 26, 2009 8:42 pm, Pete Turnbull wrote:
>>> UC07's are QBUS cards which have a single SCSI port and support
>>
>> Uh, UC07s are UNIBUS, not QBUS. I've got a UC08, the dual-SCSI port
>> version of that.
>
> No, they're definitely QBus, as is UC08. They're both quad-height
> cards, at least the ones I've come across, but I believe there was a
> later dual-height version with the same designation. I don't have one
> but they're listed in lots of places and I know they work in microVaxen
> as well as PDP-11s. UC17 and UC18 are Unibus, I believe.
This is correct; the UC17/UC18 are Unibus. I've got a UC17 here. I've
never seen a UC07, but docs say it is Qbus.
-Dave
--
Dave McGuire
Port Charlotte, FL
At 8:46 AM -0600 3/1/09, Dan wrote:
>0x105 here, damn you guys make me feel old, I'm 41
>and grew up with the old systems and move to unix later too...
0x8A, 46 years old. Sigh. Wasted years....at least I understand the scoring.
--
- Mark 210-379-4635
-----------------------------------------------------------------------
Large Asteroids headed toward planets
inhabited by beings that don't have
technology adequate to stop them:
Think of it as Evolution in Fast-Forward.
