PG506 disassembly advice sought


Andy Warner
 

I have a PG506 (< B40000), which needs some switches cleaned, and I want to
replace the tantalums while I am at it. I've checked the service manual and
the archives and cannot find any advice on disassembly.

I'm seeking advice to get to the cam switches and both sides of any board
with tantalums.
--
Andy


Michael W. Lynch
 

Andy,

I have two PG506's that I have just recently fixed. Mine are both over S/N B040000, but the units are almost the same. As I recall, you can pull the A2 Period Board and the A5 Relay board and get to most of the main board Take a lot of pictures of the various connectors and cables. it is very easy to get the cable connectors swapped upside down, etc. The display board has cam switch contacts on the back and you pull the screws from the cams; (they stay attached to the AQ1 Main Board for now) then lift the A2 board out CAREFULLY, so as not to bend the cam contact blades on the back of the A2 board in the process. Once A2 and A5 are out of the way, you can get to both sides of the A1 Main Board and cam switches on the main board. After you do this 5 or 10 times, you get good at it. I seriously had one of mine apart at least 10 times. This is almost a complete disassembly because of the way that they have it all screwed together. Feel free to pick my few remaining brain cells if you get stuck.

Good lucj! .

--
Michael Lynch
Dardanelle, AR


Jean-Paul
 

Hello, beware of the very special precautions for cleaning and services of elastomeric switches for function and other rotary controls in PG506, even 99% isopropyl alcohol can damage them.

See the service manual for details.

Jon


Andy Warner
 

Michael,
Thanks for the advice, I see what you mean about it getting
easier with practice. I ended up needing to almost completely tear mine
down, the mode switches (S180A) on the A1 board needed a really thorough
cleaning and adjustment. I ended up removing the front panel in order to
remove the cam assembly from the A1 board.

Now I get to debug and repair the unit, so I am going at assume I’ll have
to dismantle it more than once as I work through the various problems
beyond crappy switches (the arcing from VR280 being the first and most
obvious failure mode.)

On Wed, Jun 2, 2021 at 21:51 Michael W. Lynch via groups.io <mlynch003=
yahoo.com@groups.io> wrote:

Andy,

I have two PG506's that I have just recently fixed. Mine are both over
S/N B040000, but the units are almost the same. As I recall, you can pull
the A2 Period Board and the A5 Relay board and get to most of the main
board Take a lot of pictures of the various connectors and cables. it is
very easy to get the cable connectors swapped upside down, etc. The
display board has cam switch contacts on the back and you pull the screws
from the cams; (they stay attached to the AQ1 Main Board for now) then lift
the A2 board out CAREFULLY, so as not to bend the cam contact blades on the
back of the A2 board in the process. Once A2 and A5 are out of the way,
you can get to both sides of the A1 Main Board and cam switches on the main
board. After you do this 5 or 10 times, you get good at it. I seriously
had one of mine apart at least 10 times. This is almost a complete
disassembly because of the way that they have it all screwed together.
Feel free to pick my few remaining brain cells if you get stuck.

Good lucj! .

--
Michael Lynch
Dardanelle, AR





--
Andy


Michael W. Lynch
 

Andy,

It took me 3 years (on and off again) to finally get mine fully repaired. The most difficult part of mine was figuring out the non- functional display. I have some notes on the display if you need them. Let me know if there is anything I can help you with.

Michael Lynch

From My I-Phone

mlynch003@yahoo.com

479-477-1115

On Jun 6, 2021, at 12:12 PM, Andy Warner <andyw@pobox.com> wrote:

Michael,
Thanks for the advice, I see what you mean about it getting
easier with practice. I ended up needing to almost completely tear mine
down, the mode switches (S180A) on the A1 board needed a really thorough
cleaning and adjustment. I ended up removing the front panel in order to
remove the cam assembly from the A1 board.

Now I get to debug and repair the unit, so I am going at assume I’ll have
to dismantle it more than once as I work through the various problems
beyond crappy switches (the arcing from VR280 being the first and most
obvious failure mode.)

On Wed, Jun 2, 2021 at 21:51 Michael W. Lynch via groups.io <mlynch003=
yahoo.com@groups.io> wrote:

Andy,

I have two PG506's that I have just recently fixed. Mine are both over
S/N B040000, but the units are almost the same. As I recall, you can pull
the A2 Period Board and the A5 Relay board and get to most of the main
board Take a lot of pictures of the various connectors and cables. it is
very easy to get the cable connectors swapped upside down, etc. The
display board has cam switch contacts on the back and you pull the screws
from the cams; (they stay attached to the AQ1 Main Board for now) then lift
the A2 board out CAREFULLY, so as not to bend the cam contact blades on the
back of the A2 board in the process. Once A2 and A5 are out of the way,
you can get to both sides of the A1 Main Board and cam switches on the main
board. After you do this 5 or 10 times, you get good at it. I seriously
had one of mine apart at least 10 times. This is almost a complete
disassembly because of the way that they have it all screwed together.
Feel free to pick my few remaining brain cells if you get stuck.

Good lucj! .

--
Michael Lynch
Dardanelle, AR





--
Andy




--
Michael Lynch
Dardanelle, AR


Andy Warner
 

Will do, my display is also non-functional (shows random digits, but
doesn’t seem to track the variable amplitude.)

These are funny little units, they have so many disparate functions crammed
in, and then reuse functional blocks between them (like the divider chain
for the variable period and the voltmeter.)

Here’s hoping it doesn’t take me 3 years too.

On Sun, Jun 6, 2021 at 15:50 Michael W. Lynch via groups.io <mlynch003=
yahoo.com@groups.io> wrote:

Andy,

It took me 3 years (on and off again) to finally get mine fully repaired.
The most difficult part of mine was figuring out the non- functional
display. I have some notes on the display if you need them. Let me know
if there is anything I can help you with.

Michael Lynch

From My I-Phone

mlynch003@yahoo.com

479-477-1115


On Jun 6, 2021, at 12:12 PM, Andy Warner <andyw@pobox.com> wrote:

Michael,
Thanks for the advice, I see what you mean about it getting
easier with practice. I ended up needing to almost completely tear mine
down, the mode switches (S180A) on the A1 board needed a really thorough
cleaning and adjustment. I ended up removing the front panel in order to
remove the cam assembly from the A1 board.

Now I get to debug and repair the unit, so I am going at assume I’ll have
to dismantle it more than once as I work through the various problems
beyond crappy switches (the arcing from VR280 being the first and most
obvious failure mode.)

On Wed, Jun 2, 2021 at 21:51 Michael W. Lynch via groups.io <mlynch003=
yahoo.com@groups.io> wrote:

Andy,

I have two PG506's that I have just recently fixed. Mine are both over
S/N B040000, but the units are almost the same. As I recall, you can
pull
the A2 Period Board and the A5 Relay board and get to most of the main
board Take a lot of pictures of the various connectors and cables. it
is
very easy to get the cable connectors swapped upside down, etc. The
display board has cam switch contacts on the back and you pull the
screws
from the cams; (they stay attached to the AQ1 Main Board for now) then
lift
the A2 board out CAREFULLY, so as not to bend the cam contact blades on
the
back of the A2 board in the process. Once A2 and A5 are out of the way,
you can get to both sides of the A1 Main Board and cam switches on the
main
board. After you do this 5 or 10 times, you get good at it. I
seriously
had one of mine apart at least 10 times. This is almost a complete
disassembly because of the way that they have it all screwed together.
Feel free to pick my few remaining brain cells if you get stuck.

Good lucj! .

--
Michael Lynch
Dardanelle, AR





--
Andy






--
Michael Lynch
Dardanelle, AR





--
Andy


Michael W. Lynch
 

Andy, When I started with this current addiction (Tek Scopes) about 4 years ago, I basically knew nothing about this part of the electronics field. The three year time frame was mostly me getting up to speed and to some level of competency so that I could understand what was going on inside the PG506. They are very complex, lots of functions crammed into a small box.

Michael Lynch

From My I-Phone

mlynch003@yahoo.com

479-477-1115

On Jun 6, 2021, at 4:10 PM, Andy Warner <andyw@pobox.com> wrote:

Will do, my display is also non-functional (shows random digits, but
doesn’t seem to track the variable amplitude.)

These are funny little units, they have so many disparate functions crammed
in, and then reuse functional blocks between them (like the divider chain
for the variable period and the voltmeter.)

Here’s hoping it doesn’t take me 3 years too.

On Sun, Jun 6, 2021 at 15:50 Michael W. Lynch via groups.io <mlynch003=
yahoo.com@groups.io> wrote:

Andy,

It took me 3 years (on and off again) to finally get mine fully repaired.
The most difficult part of mine was figuring out the non- functional
display. I have some notes on the display if you need them. Let me know
if there is anything I can help you with.

Michael Lynch

From My I-Phone

mlynch003@yahoo.com

479-477-1115


On Jun 6, 2021, at 12:12 PM, Andy Warner <andyw@pobox.com> wrote:
Michael,
Thanks for the advice, I see what you mean about it getting
easier with practice. I ended up needing to almost completely tear mine
down, the mode switches (S180A) on the A1 board needed a really thorough
cleaning and adjustment. I ended up removing the front panel in order to
remove the cam assembly from the A1 board.

Now I get to debug and repair the unit, so I am going at assume I’ll have
to dismantle it more than once as I work through the various problems
beyond crappy switches (the arcing from VR280 being the first and most
obvious failure mode.)

On Wed, Jun 2, 2021 at 21:51 Michael W. Lynch via groups.io <mlynch003=
yahoo.com@groups.io> wrote:

Andy,

I have two PG506's that I have just recently fixed. Mine are both over
S/N B040000, but the units are almost the same. As I recall, you can
pull
the A2 Period Board and the A5 Relay board and get to most of the main
board Take a lot of pictures of the various connectors and cables. it
is
very easy to get the cable connectors swapped upside down, etc. The
display board has cam switch contacts on the back and you pull the
screws
from the cams; (they stay attached to the AQ1 Main Board for now) then
lift
the A2 board out CAREFULLY, so as not to bend the cam contact blades on
the
back of the A2 board in the process. Once A2 and A5 are out of the way,
you can get to both sides of the A1 Main Board and cam switches on the
main
board. After you do this 5 or 10 times, you get good at it. I
seriously
had one of mine apart at least 10 times. This is almost a complete
disassembly because of the way that they have it all screwed together.
Feel free to pick my few remaining brain cells if you get stuck.

Good lucj! .

--
Michael Lynch
Dardanelle, AR





--
Andy






--
Michael Lynch
Dardanelle, AR





--
Andy




--
Michael Lynch
Dardanelle, AR


Harvey White
 

Looking at the schematic, you have a bunch of transistors making a square wave, which is fed to pin 14 of a 7490, U665.  Period cal sets the fixed frequency, and when going to variable, period var changes the frequency.  This waveform is fed through the 7490's (hooked up as divide by 5 then divide by 2, always giving a symmetric output.  The range switch picks off the frequency from a divider chain.  The Div100 and Div1000 (U666 and U667 feed latches, which directly drive the displays.  Looks like the latches are enabled a variable time after the start of the count, capturing and freezing a count that goes from 0 to 99.

So:

1) you should have a fixed frequency at the input (pin 14) of U665.  In variable mode, this frequency should change.

2) you should have outputs from each of the 7490's at pin 11, which should be 1/10 the frequency at the chip's pin 14.

3) at pins 4 and 13 of either latch, you should see a pulse mostly low, high for a bit, then low again.  The outputs of the latch should change when that pulse is high, and be stable when low.  It's a latch (not clocked, transparent when enable is high), not a register (clocked on an edge).


If the outputs of the 7490's are bad, then and the clock input is ok, then they're likely bad.  If the output from Q565 (sheet 3) is not a good waveform, that messes up the display.

if the latch outputs don't freeze when the enable input is low (4&13), then they're bad.

The display is disabled (blanked) in the non-variable position, but the counters and latches still work.

The enables are pin 13 of U673 and U675.  The signal should be stable and not change unless the var setting is used.


The reading of the display is dependent on the enable pulse for the latches.  If that's messy, no telling what you get.

In the fixed mode, I don't think that the oscillator frequency will vary at all, so you should get a more or less constant value from the latches, likely a 0 for both displays.

Hope this helps a bit, unless you've figured this out already. If this isn't right, it ought to be fairly close.

Harvey

On 6/6/2021 4:47 PM, Michael W. Lynch via groups.io wrote:
Andy,

It took me 3 years (on and off again) to finally get mine fully repaired. The most difficult part of mine was figuring out the non- functional display. I have some notes on the display if you need them. Let me know if there is anything I can help you with.

Michael Lynch

From My I-Phone
mlynch003@yahoo.com

479-477-1115


On Jun 6, 2021, at 12:12 PM, Andy Warner <andyw@pobox.com> wrote:

Michael,
Thanks for the advice, I see what you mean about it getting
easier with practice. I ended up needing to almost completely tear mine
down, the mode switches (S180A) on the A1 board needed a really thorough
cleaning and adjustment. I ended up removing the front panel in order to
remove the cam assembly from the A1 board.

Now I get to debug and repair the unit, so I am going at assume I’ll have
to dismantle it more than once as I work through the various problems
beyond crappy switches (the arcing from VR280 being the first and most
obvious failure mode.)

On Wed, Jun 2, 2021 at 21:51 Michael W. Lynch via groups.io <mlynch003=
yahoo.com@groups.io> wrote:

Andy,

I have two PG506's that I have just recently fixed. Mine are both over
S/N B040000, but the units are almost the same. As I recall, you can pull
the A2 Period Board and the A5 Relay board and get to most of the main
board Take a lot of pictures of the various connectors and cables. it is
very easy to get the cable connectors swapped upside down, etc. The
display board has cam switch contacts on the back and you pull the screws
from the cams; (they stay attached to the AQ1 Main Board for now) then lift
the A2 board out CAREFULLY, so as not to bend the cam contact blades on the
back of the A2 board in the process. Once A2 and A5 are out of the way,
you can get to both sides of the A1 Main Board and cam switches on the main
board. After you do this 5 or 10 times, you get good at it. I seriously
had one of mine apart at least 10 times. This is almost a complete
disassembly because of the way that they have it all screwed together.
Feel free to pick my few remaining brain cells if you get stuck.

Good lucj! .

--
Michael Lynch
Dardanelle, AR





--
Andy





Andy Warner
 

Yeah, the use of the counters to display the ramp times from the comparator
is linked with the mode switch, ensuring a fixed clock when the counter
chain is being used for the error display. Clever, but intertwined,
especially if switches cannot be relied upon.

Now I have the switches mostly doing the right thing, I am just going to
start going through it, functional unit by functional unit.

On Sun, Jun 6, 2021, 17:21 Harvey White <madyn@dragonworks.info> wrote:

Looking at the schematic, you have a bunch of transistors making a
square wave, which is fed to pin 14 of a 7490, U665. Period cal sets
the fixed frequency, and when going to variable, period var changes the
frequency. This waveform is fed through the 7490's (hooked up as divide
by 5 then divide by 2, always giving a symmetric output. The range
switch picks off the frequency from a divider chain. The Div100 and
Div1000 (U666 and U667 feed latches, which directly drive the displays.
Looks like the latches are enabled a variable time after the start of
the count, capturing and freezing a count that goes from 0 to 99.

So:

1) you should have a fixed frequency at the input (pin 14) of U665. In
variable mode, this frequency should change.

2) you should have outputs from each of the 7490's at pin 11, which
should be 1/10 the frequency at the chip's pin 14.

3) at pins 4 and 13 of either latch, you should see a pulse mostly low,
high for a bit, then low again. The outputs of the latch should change
when that pulse is high, and be stable when low. It's a latch (not
clocked, transparent when enable is high), not a register (clocked on an
edge).


If the outputs of the 7490's are bad, then and the clock input is ok,
then they're likely bad. If the output from Q565 (sheet 3) is not a
good waveform, that messes up the display.

if the latch outputs don't freeze when the enable input is low (4&13),
then they're bad.

The display is disabled (blanked) in the non-variable position, but the
counters and latches still work.

The enables are pin 13 of U673 and U675. The signal should be stable
and not change unless the var setting is used.


The reading of the display is dependent on the enable pulse for the
latches. If that's messy, no telling what you get.

In the fixed mode, I don't think that the oscillator frequency will vary
at all, so you should get a more or less constant value from the
latches, likely a 0 for both displays.

Hope this helps a bit, unless you've figured this out already. If this
isn't right, it ought to be fairly close.

Harvey



On 6/6/2021 4:47 PM, Michael W. Lynch via groups.io wrote:
Andy,

It took me 3 years (on and off again) to finally get mine fully
repaired. The most difficult part of mine was figuring out the non-
functional display. I have some notes on the display if you need them.
Let me know if there is anything I can help you with.

Michael Lynch

From My I-Phone
mlynch003@yahoo.com

479-477-1115


On Jun 6, 2021, at 12:12 PM, Andy Warner <andyw@pobox.com> wrote:

Michael,
Thanks for the advice, I see what you mean about it getting
easier with practice. I ended up needing to almost completely tear mine
down, the mode switches (S180A) on the A1 board needed a really thorough
cleaning and adjustment. I ended up removing the front panel in order to
remove the cam assembly from the A1 board.

Now I get to debug and repair the unit, so I am going at assume I’ll
have
to dismantle it more than once as I work through the various problems
beyond crappy switches (the arcing from VR280 being the first and most
obvious failure mode.)

On Wed, Jun 2, 2021 at 21:51 Michael W. Lynch via groups.io
<mlynch003=
yahoo.com@groups.io> wrote:

Andy,

I have two PG506's that I have just recently fixed. Mine are both over
S/N B040000, but the units are almost the same. As I recall, you can
pull
the A2 Period Board and the A5 Relay board and get to most of the main
board Take a lot of pictures of the various connectors and cables. it
is
very easy to get the cable connectors swapped upside down, etc. The
display board has cam switch contacts on the back and you pull the
screws
from the cams; (they stay attached to the AQ1 Main Board for now) then
lift
the A2 board out CAREFULLY, so as not to bend the cam contact blades
on the
back of the A2 board in the process. Once A2 and A5 are out of the
way,
you can get to both sides of the A1 Main Board and cam switches on the
main
board. After you do this 5 or 10 times, you get good at it. I
seriously
had one of mine apart at least 10 times. This is almost a complete
disassembly because of the way that they have it all screwed together.
Feel free to pick my few remaining brain cells if you get stuck.

Good lucj! .

--
Michael Lynch
Dardanelle, AR





--
Andy