14 Nov
2010
14 Nov
'10
2:47 p.m.
I have a critical ( late 80's microprocessor controlled ) piece of test gear
no longer supported. The power supply is dead and after sawing through it's
exoskeleton I was meet with a large epoxy block the size of the original
box. I don't intend to fix said power supply but rather to figure out what
voltages it provided and use a modern switcher. I know I need +12 and +5 as
there is a generic floppy drive and the logic / processor uses +5. Any ideas
how to strip the epoxy other than digging through it with implements of mass
destruction ? Any and all help is appreciated.
Best regards, Steven
14 Nov
14 Nov
3:24 p.m.
I have a critical ( late 80's microprocessor controlled ) piece of test gear
no longer supported. The power supply is dead and after sawing through it's
exoskeleton I was meet with a large epoxy block the size of the original
box. I don't intend to fix said power supply but rather to figure out what
voltages it provided and use a modern switcher. I know I need +12 and +5 as
there is a generic floppy drive and the logic / processor uses +5. Any ideas
how to strip the epoxy other than digging through it with implements of mass
destruction ? Any and all help is appreciated.
Itmight be easier to work out what the PSU needs to be than to dismantle
the origianl one.
Firstly, what is the instrument (People here have some amazing stuff, and
ity's possible somebody else has the same unit and can measure the output
voltages.
If not, then how many output wires are there? Presumably there are 3 for
the ground, +5V and +12V supplies. Is there a -12V (e.g. for an RS232
port) that you could identify?
What about the analogue side? This may well need a floating supply (that
is, one with a semarate ground rail, not the same as the logic supply
ground). Op=amps (if there are any) are likely to use something like a
+/-15V supply.
-tony
5:28 p.m.
On Sun, Nov 14, 2010 at 12:24 PM, Tony Duell <ard at p850ug1.demon.co.uk>wrote:
Tony,
It is a B & K Model 2102 Audio Analyzer. I'm certain it does have a negative
supply for the RS232 ( no Maxim chip back then ). P/S has two " Molex " type
connectors; one 9 pin male and one 12 pin female. It has a ridiculously
complex inter-connect so it will run at any " reasonable " input
voltage user selectable ( I don't care about the input voltage(s).
Definitely OP-Amps so I'm checking Voltage ratings on the bypass caps for a
clue on the analog plus and minus voltages. I hate digging thru epoxy. It's
like trying to recover peanuts from elephant poop. Thanks for the ideas
Tony.
Best regards, Steven
I have a critical ( late 80's microprocessor controlled ) piece of test
gear
no longer supported. The power supply is dead and
after sawing through
it's
exoskeleton I was meet with a large epoxy block
the size of the original
box. I don't intend to fix said power supply but rather to figure out
what
voltages it provided and use a modern switcher. I
know I need +12 and +5
as
there is a generic floppy drive and the logic /
processor uses +5. Any
ideas
how to strip the epoxy other than digging through
it with implements of
mass
destruction ? Any and all help is appreciated.
Itmight be easier to work out what the PSU needs to be than to dismantle
the origianl one.
Firstly, what is the instrument (People here have some amazing stuff, and
ity's possible somebody else has the same unit and can measure the output
voltages.
If not, then how many output wires are there? Presumably there are 3 for
the ground, +5V and +12V supplies. Is there a -12V (e.g. for an RS232
port) that you could identify?
What about the analogue side? This may well need a floating supply (that
is, one with a semarate ground rail, not the same as the logic supply
ground). Op=amps (if there are any) are likely to use something like a
+/-15V supply.
-tony
15 Nov
15 Nov
4:14 p.m.
It is a B & K Model 2102 Audio Analyzer. I'm
certain it does have a negative
supply for the RS232 ( no Maxim chip back then ). P/S has two " Molex " type
From your first message, I didn't even know it had
an RS232 interface...
It would be possible to get the voltages for that in some other way (e.g.
using a diode/capacitor multiplier driven from a buffered logic-level
clock signal), but I think it's unlikely. Having suitabrl outputs from
the main PSU makes more sense.
connectors; one 9 pin male and one 12 pin female. It
has a ridiculously
complex inter-connect so it will run at any " reasonable " input
voltage user selectable ( I don't care about the input voltage(s).
Definitely OP-Amps so I'm checking Voltage ratings on the bypass caps for a
clue on the analog plus and minus voltages. I hate digging thru epoxy. It's
You should also check the data sheets on any unuual/complex analogue ICs
to see if they have particular supply requirements. Most op-amps have a
fairly wide supply range, other things may not...
-tony
8:54 p.m.
Thanks Alexandre, I had forgotten about banana oil. Looks like 8 oz is about
$120 USD !!
Meanwhile, output pins tied together;
1 2 3 ( probably Gnd or +5 )
4 9 ( probably +12 or -12 )
5 6 ( probably +12 or -12 )
7 8 12 ( probably Gnd or +5 )
10 ( OP Amp power ? )
11 ( OP Amp power ? )
I only have the power supply here right now ( not the main unit ). When I
have the unit here I can buzz out and find which pair goes to Gnd and logic
+5, etc.
Best regards, Steven
On Sun, Nov 14, 2010 at 3:18 PM, Alexandre Souza - Listas <
pu1bzz.listas at gmail.com> wrote:
how to strip the epoxy other than digging through it
with implements of
mass
destruction ? Any and all help is appreciated.
Look for banana's oil...
9:19 p.m.
Sodium ethoxide is the thing to use for epoxy. Horribly caustic but
easy to make by *SLOWLY* dissolving lye pellets in denatured ethyl
alcohol (let the solution cool before adding the next pellet!).
http://en.wikipedia.org/wiki/Sodium_ethoxide
An alternative is to soak the epoxy block in a bucket of "aircraft
stripper" (available at most auto parts stores) for a couple of
weeks. Makes the epoxy rubbery and easy to carve away. However, it
also dissolves paint and plastic.
--Chuck
4:19 p.m.
Thanks Alexandre, I had forgotten about banana oil. Looks like 8 oz is about
$120 USD !!
My worry about this, or any other chemical that attacks epoxy is what
woudl it do to component packaging (some of which is epoxy-based). No, I
am not suggestign you would want to use the components again, but suyrely
you'd want to be able to read the markings on them.
Meanwhile, output pins tied together;
1 2 3 ( probably Gnd or +5 )
4 9 ( probably +12 or -12 )
5 6 ( probably +12 or -12 )
7 8 12 ( probably Gnd or +5 )
10 ( OP Amp power ? )
11 ( OP Amp power ? )
I assume you've identified linked pins using a test meter or simuilar. My
concern is that this PSU is defective, and something like a shorted
rectifier or output capacitor could make pins appear to be shorted
together when in fact they shouldn't be. Best to check when you get the
main unit.
-tony
14 Nov
14 Nov
9:28 p.m.
If you don't like DIY "bucket" type chemistry, there are commercial
solvents:
http://www.dynaloy.com/Products/epoxy_products.html
--Chuck
5010
days inactive
5011
days old
13 comments
6 participants
participants (6)
-
ard＠p850ug1.demon.co.uk -
cclist＠sydex.com -
cisin＠xenosoft.com -
ethan.dicks＠gmail.com -
pu1bzz.listas＠gmail.com -
scanning.cc＠gmail.com