The CPU has the 14 slot main backplane and a 9 slot expansion backplane,
connected together with the M9202 jumper card pair. When I first opened the
top the M9202 jumper card was plugged into slots 12 and 14 of the main
backplane. I assume it is really supposed to plug into slot 14 of the main
backplane and slot 1 of the expansion backplane to bridge the two so I moved
it there. The M9302 terminator was in slot 9 of the expansion backplane.
I believe all row D connectors except the empty 10,11,12 memory slots either
had a card installed or had a stubby little card grant card installed. I
didn't pull any of them to see what they were (G727, G7270, G7273?)
> BTW - how many slots / backplanes are curently installed? How many free
>slots? How many BUs Grant cards are in open slots? It is also possible
>that you will need access to the underside of the machine at some time (to
>attach/remove NPR jumpers) so don't mount it where that would be
>difficult, until you have the System stable and configured the way you
>want it.
>
I did pull the power cord panel off of the back of the power supply and it
looks pretty easy to swap the existing power cable with one with a more
common plug so that's what I'll do after I manage to track down a
replacement M7095 and get to the point of trying to power up the CPU.
> Not too bad - get a 15A plug from any of the Home DIY places or a big
>hardware store. Take off the (should be attached with various screws and
>clamps) existing 20A plug and attach the 15A one. Remember that the the
>Black (hot) wire goes to the Brass terminal, the White (neut) wire goes to
>the Silver terminal, and the Green (gnd) wire goes to the green (or black)
>Ground terminal.
>
> Your CPU, configured as it is now, will draw maybe 6 or 7 amps from the
>wall at 120V, so you're good to go.
_________________________________________________________________
MSN 8 helps eliminate e-mail viruses. Get 2 months FREE*.
http://join.msn.com/?page=features/virus
On Dec 10, 22:36, Philip Pemberton wrote:
> pete(a)dunnington.u-net.com wrote:
> > On things like that, I often don't even try to rescue any suspect ICs
> > or even passives, just cut them off close to the PCB with a very fine
> > pair of sidecutters, and then desolder the stub of pin.
> That's what I've been doing. I've been desoldering using "Soder-Wick" - I
> got a small reel of it for 90p.
You'd be much better off using a nice BIG solder sucker. Wick is really
good for removing excess solder, say from a "generously" reworked SMT
device, but not ideal for common-or-garden ICs.
> In order to
> limit damage to the ICs due to heat, I gave them a quick blast of freeze
> spray and then soldered them.
I hate to say it, but that's not a good idea. IC's don't mind heat if it's
evenly applied (think infra-red reflow oven) but thermal shock can break
seals etc.
> > How adept are you
> > with a soldering iron?
> Six years of experience, three with a Maplin elcheapo, three with my
Antex
> XS25 (25 watt).
Well, at least it's not one of those nasty Antex 15W irons, but I prefer a
50W temperature-controlled iron. The risk with a small iron is that the
tip cools down as you put it on the work, and the lower-wattage irons can't
heat it up very quickly again. Result is that the joint isn't really hot
enough.
> > Did you get the 2114s and Z80 I sent you? They should have arrived
> > this morning.
> Yup, they arrived this morning. Fitted them and the random, static
garbage
> has been replaced with scrolling garbage. Grr...
Well, it's *some* improvement :-)
> As for your offer to have a look at it, I might just take you up on it.
I've
> got a schematic for it (drawn by Bodo Wenzel) for a clone of the Ace,
some
> bits are different, 90% of it is the same though.
OK. I'll mail you my address and phone number off-list, and maybe we can
get together.
--
Pete Peter Turnbull
Network Manager
University of York
Hi,
I've just bid on some Hitachi HN462532G EPROMs on Ebay - item numbers
1943283712 1943523215 and 1943750626 (I've bid on the first two - feel free
to outbid me, just leave me with at least one pack of three). Can someone
please confirm that these are pin-compatible with the TI TMS2532? I found a
pinout on R. Steve Walz's webpage (www.armory.com/~rstevew), but I'd like
confirmation from someone who's attempted to use the Hitachi chips as
replacements for the TIs.
Thanks.
--
Phil.
philpem(a)dsl.pipex.com
http://www.philpem.dsl.pipex.com/
Well, I can officially stop complaining. I've just scored a software
engineering position with a local natural gas production company. This will
have three effects with respect to the ClassicCmp list:
* I will no longer gripe about lack of funds in nearly every post. :-)
* My collection will experience growth and an increase in quality, which
hopefully will enable me to do things like software and documentation
archiving, hardware testing and restoration, and of course, playing with
shiny new toys.
* You should see me start to offer free stuff to list members soon.
Yes, today is a good day.
--
Jeffrey Sharp
Wow!
I think we've all had a wack at poor Phil.
Hi Phil
I think the most important thing is to save the board
over saving parts. If you think you've lifted a trace that is a
thru hole( and these are two sided boards ), take a piece
of wire wrap wire through the hole and solder it to the
topside of the board and it goes through the hole.
Don't solder it in the hole. Make sure that the component
lead will fit through. When you put the part in,
fold the piece of wire down so even if you heat it
enough to melt the solder on the top, it won't
rotate.
If the top trace is clearly visible with the lead
in there, you can solder the lead from the top. Use
an Ohm meter to verify that you've made connections
to all the right places.
Dwight
Hi all,
Well, it looks like I've finally destroyed the Ace. I've just spent the
past hour trying to desolder the dead RAMs and buffers from the Ace's main
PCB. Unfortunately it looks like the board was designed to self-destruct
when anyone tried to repair it.
The pads appear to have been designed to peel off on the application of
heat, they're less than 5 mils around the hole (what do you think that
means?) and they don't even seem to be through-hole plated. The tin plating
was applied straight on top of oxidised copper - I've had to retin some pads
and tracks courtesy of that major screwup.
In my opinion, the Jupiter Ace is one of the most appallingly-made
machines I've ever tried to repair. Heck, the Commodore 64 was bad, but at
least the pads were easy enough to desolder. It looks to me like Jupiter
Cantab's PCB designer was either *VERY* inexperienced or just wanted to make
sure that no-one could fix an Ace if it failed. It's also beginning to look
like the ROMs are stuffed, but that shouldn't be too hard to sort out --
I've just bid on some 2532 EPROMs on eBay from someone in Austria.
Does anyone know how I could rescue this machine? It looks like the RAMs
are definetly fried, along with some of the logic as well. Font RAM and
Video RAM are still not being loaded on startup so the output of the video
generator is still 100% noise, however it *is* changing when the machine is
powered off and then back on again. I'm shotgunning all the RAMs (there's
only six of them) and the bus muxes.
Has anyone here either repaired one of these machines or got a spare Ace
to sell me? I've got a proper PSU now, with only one connector (the jack
plug the Ace uses), so I can say with near absolute certainty that the same
mistake will not occur again.
Later.
--
Phil.
philpem(a)dsl.pipex.com
http://www.philpem.dsl.pipex.com/
I totally agree.
Remember the concept of the $99 computer. You dont sell a computer for $99
if it costs
you $150. I suspect that commercially, the cost breakdown probably went
something like this:
$99.00 Final RRP (Margin + Freight + Wholesale price)
$30.00 Sales Margin (30%)
$ 5.00 Distribution freight
$65.00 Wholesale price
$65.00 Wholesale price (Margin + freight + Sinclair sell price)
$ 9.75 Wholesale Margin (15%)
$ 2.00 Wholesale freight
$53.25 Sinclair Sell Price
$53.25 Sinclair Sell price (Manufacture cost + R&D Recovery + Sinclair
Margin)
$10.65 Sinclair Margin (20%)
$10.65 R&D Recovery (20%)
$31.95 Manufacture cost
$31.95 Manufacture Cost (System + Plugpack + Lead + Manual Printing)
$16.00 Component Parts
$ 1.35 Membrane Keyboard
$ 3.00 Modulator
$ 1.60 Case
$ 3.00 Plug Pack
$ 1.00 Video Cables
$ 1.00 PCB
$ 2.00 Assembly Charges
$ 2.00 Manual Printing
$ 1.00 Packaging (foam etc)
These figures have been fudged to fit, but you can see that at $99 there is
basically *no* rooe to provide any type of quality.
Just my thoughts, don't flame me too hard!
Doug Jackson
Director, Managed Security Services
Citadel Securix
+61 (0)2 6290 9011 (Ph)
+61 (0)2 6262 6152 (Fax)
+61 (0)414 986 878 (Mobile)
Web: <www.citadel.com.au>
Offices in Melbourne, Sydney, Canberra, Hong Kong, Boston
> -----Original Message-----
> From: J.C.Wren [mailto:jcwren@jcwren.com]
> Sent: Wednesday, December 11, 2002 7:19 AM
> To: cctalk(a)classiccmp.org
> Subject: RE: One more screwup with the Ace...
>
>
> Phil,
>
> Most likely the board was designed for low cost. If there is no
> soldermask, larger annular rings increase the chance of solder bridges
> during manufacturing. But more likely, smaller annular rings
> were used to
> reduce the amount of solder needed, thus lowering the cost.
> I don't believe
> they Ace nor the ZX81 nor any of that genre of computers was
> designed to be
> repaired. They were designed to be produced as cheaply as possible.
>
> --John
>
> -----Original Message-----
> From: cctalk-admin(a)classiccmp.org
[mailto:cctalk-admin@classiccmp.org]On
Behalf Of Philip Pemberton
Sent: Tuesday, December 10, 2002 15:04
To: cctalk(a)classiccmp.org
Subject: One more screwup with the Ace...
Hi all,
Well, it looks like I've finally destroyed the Ace. I've just spent the
past hour trying to desolder the dead RAMs and buffers from the Ace's main
PCB. Unfortunately it looks like the board was designed to self-destruct
when anyone tried to repair it.
The pads appear to have been designed to peel off on the application of
heat, they're less than 5 mils around the hole (what do you think that
means?) and they don't even seem to be through-hole plated. The tin plating
was applied straight on top of oxidised copper - I've had to retin some pads
and tracks courtesy of that major screwup.
In my opinion, the Jupiter Ace is one of the most appallingly-made
machines I've ever tried to repair. Heck, the Commodore 64 was bad, but at
least the pads were easy enough to desolder. It looks to me like Jupiter
Cantab's PCB designer was either *VERY* inexperienced or just wanted to make
sure that no-one could fix an Ace if it failed. It's also beginning to look
like the ROMs are stuffed, but that shouldn't be too hard to sort out --
I've just bid on some 2532 EPROMs on eBay from someone in Austria.
Does anyone know how I could rescue this machine? It looks like the RAMs
are definetly fried, along with some of the logic as well. Font RAM and
Video RAM are still not being loaded on startup so the output of the video
generator is still 100% noise, however it *is* changing when the machine is
powered off and then back on again. I'm shotgunning all the RAMs (there's
only six of them) and the bus muxes.
Has anyone here either repaired one of these machines or got a spare Ace
to sell me? I've got a proper PSU now, with only one connector (the jack
plug the Ace uses), so I can say with near absolute certainty that the same
mistake will not occur again.
Later.
--
Phil.
philpem(a)dsl.pipex.com
http://www.philpem.dsl.pipex.com/
CAUTION - The information in this message may be of a privileged or confidential nature intended only for the use of the addressee or someone authorised to receive the addressee's e-mail. If you are not the intended recipient, you are hereby notified that any dissemination, distribution or copying of this e-mail is strictly prohibited. If you have received this email in error please notify postmaster(a)citadel.com.au. Any views expressed in this message are those of the individual
sender and may not necessarily reflect the views of Citadel Securix.
Feel free to visit the Citadel Securix website! Click below.
http://www.citadel.com.au
On Dec 10, 23:14, Philip Pemberton wrote:
> Dwight K. Elvey wrote:
> > If you use a solder sucker, don't use a small one.
> > Use one with as big a bore and stroke as you can find.
> Hmm... I've got an Antex (www.antex.com) "PRO-DESOLD" black desolder pump
in
> my toolbox. I take it that would be better than a roll of Soder-Wick
(sic)?
Depends how big it is; the bigger the better (within reason). Practise on
a board you can afford to live without, and see. The syringe-style sucker
I keep in my kit in the car, and the one I keep most handy on the bench,
are about 1" diameter and about 9" long. I have a much smaller one
somewhere, but it's so long since I last used it, I couldn't tell you
where.
--
Pete Peter Turnbull
Network Manager
University of York
Hi Phil
If you know a part is bad and you want to remove it
>from the bard with the least damage, clip the leads
close to the part and then remove the leads, individually,
with a pair of tweezers/iron. Always use a temperature
controlled iron set to the point that a clean tip
will melt the solder of the joint in about 10 to 15 seconds.
Even really cheap PC board can be reworked if one
uses some care in methods. I don't recommend using solder
wick. Many do well with it but it tends to get stuck
to traces if not overly heated and will lift traces.
If you use a solder sucker, don't use a small one.
Use one with as big a bore and stroke as you can find.
Dwight
>From: "Philip Pemberton" <philpem(a)dsl.pipex.com>
>Hi all,
> Well, it looks like I've finally destroyed the Ace. I've just spent the
>past hour trying to desolder the dead RAMs and buffers from the Ace's main
>PCB. Unfortunately it looks like the board was designed to self-destruct
>when anyone tried to repair it.
> The pads appear to have been designed to peel off on the application of
>heat, they're less than 5 mils around the hole (what do you think that
>means?) and they don't even seem to be through-hole plated. The tin plating
>was applied straight on top of oxidised copper - I've had to retin some
pads
>and tracks courtesy of that major screwup.
> In my opinion, the Jupiter Ace is one of the most appallingly-made
>machines I've ever tried to repair. Heck, the Commodore 64 was bad, but at
>least the pads were easy enough to desolder. It looks to me like Jupiter
>Cantab's PCB designer was either *VERY* inexperienced or just wanted to
make
>sure that no-one could fix an Ace if it failed. It's also beginning to look
>like the ROMs are stuffed, but that shouldn't be too hard to sort out --
>I've just bid on some 2532 EPROMs on eBay from someone in Austria.
> Does anyone know how I could rescue this machine? It looks like the
RAMs
>are definetly fried, along with some of the logic as well. Font RAM and
>Video RAM are still not being loaded on startup so the output of the video
>generator is still 100% noise, however it *is* changing when the machine is
>powered off and then back on again. I'm shotgunning all the RAMs (there's
>only six of them) and the bus muxes.
> Has anyone here either repaired one of these machines or got a spare
Ace
>to sell me? I've got a proper PSU now, with only one connector (the jack
>plug the Ace uses), so I can say with near absolute certainty that the same
>mistake will not occur again.
>
>Later.
>--
>Phil.
>philpem(a)dsl.pipex.com
>http://www.philpem.dsl.pipex.com/
>
>
I just found one of these in a load of scrap. It appears to be similar to the HP 9000 200 and 300 series computers and uses the same DIO cards but also has a built in CRT and floppy and hard disk drives. It's designed to be rack mounted and it's supposed to have a fold down keybaord but the keyboard is missing on this one. Fortunately the unit uses a HP-HIL keyboard socket so I can plus in a regular HP_HIL keyboard. Has anyone here used one of these? Are they supposed to show a list of installed memory and interfaces when they power up like the 9000 200s do? I'm not getting any display on the CRT but I can see the sweep and retraace lines on the CRT when I turn the brightness all the way up so the CRT is working. The hard drive light illuminates for several seconds then the floppy drive accesses for a split second so the system seems to be working but still don't get any display on the screen.
Joe
