10 May
2006
10 May
'06
11:40 p.m.
Okay, so, I'm back to working on my /34. I think I made significant
progress tonight.
What's going on is, I've narrowed it down to low AC levels coming off
the transformer. The printset and H7xx modules specify an incoming
AC of 20-30v (printset says specifically 28v.) All lines coming off
the transformer at P1 are reading a value of 10-15v.
So that leaves two possibilities. It's either the transformer, or
the AC input box.
I'd like to test the levels coming out of the input box, but I
wouldn't know where to go, as my understanding of electronics fades
fast at this time of the night when I've been up since 07:00.
Okay, so maybe I traced it back that far, it got more complicated,
and it's late and I'm feeling lazy and hoping someone will throw me a
bone here :) in any case, advice is always appreciated.
toodles
Julian
11 May
11 May
11:19 a.m.
Okay, so, I'm back to working on my /34. I think I made significant
progress tonight.
What's going on is, I've narrowed it down to low AC levels coming off
the transformer. The printset and H7xx modules specify an incoming
AC of 20-30v (printset says specifically 28v.) All lines coming off
the transformer at P1 are reading a value of 10-15v.
I am working from memory, but IIRC there's a connector with 8 black wires
that carries the transformer secodnary connections (4 windings, 2 wires
per winding). It goes into the power distribution PCB, and each winding
it then connected to one of the regulator 'bricks'.
Now, you should see 20-30V _between the 2 wires going to a particular
regulator_. Not between either wire and earth.
If you measure the voltage between a particular wire and earth (logic 0V
rail, case, whatever), then you'll see one of 2 things.
1) If there's no regulator connected to that winding, you'll measure 0V.
The winding is electrically isolated from the rest of the machine
2) If there is a regulator connected, then each wire will swing between
ground and the full secondary voltage at mains frequency, due to the
action of the bridge rectifier in the 'brick'. Now, what that will appear
at on your meter depends on the meter, but I could well believe some
would average it out to half-voltage.
So that's the first explanation, you're measuring the voltage between one
end of the secondary and ground, not the voltage across the secondary.
A second explation, if you are measuring the voltage across the
secondary, and it's half what you expect, is that the primary windings
are wired incorrectly. The power transgformer has 2 primary windings,
rated at 115V each. For 115V mains, you connect them in parallel (taking
care of the phase, of course!), for 230V mains you connect them in series
(ditto). The primary windings go to a 4 pin connector that plugs into the
front of the AC input box. There are 2 different AC input boxes, for 115V
or 230V mains which connect the pins of that connector appropriately --
there are other internal changes relating to the mains switching
contactor and its control supply. Is it possible that you have a 230V
unit plugged into a 115V mains socket?
So that leaves two possibilities. It's either the transformer, or
the AC input box.
I thing it's very unlikely to be the transformer. Shorted turns would
make it get hot and bothered very quickly.
-tony
12:18 p.m.
Tony, responses to your statements are below, but as a general question to
those who are helping me with this, is the H745 light supposed to come on if
it's just connected to the power distribution board? i.e. am I supposed to
se a light on with no load like the H7441s do?
Wow, OK, so what is it then? This really has me stumped.
Is it still possible I have an input box problem?
-----Original Message-----
From: cctalk-bounces at classiccmp.org
[mailto:cctalk-bounces at classiccmp.org] On Behalf Of Tony Duell
Sent: Thursday, May 11, 2006 1:20 PM
To: cctalk at classiccmp.org
Subject: Re: BA11-K low AC output levels
This could explain my reading. I was only measuring one wire at a time.
Also, when I measure each of the 8 wires on P1 in sequence, one is slightly
higher than the other, alternating between the 2. Secondly, when measuring
off the power distribution board, I get the same type of reading for each
regulator, one at 15-16v, one at 10-13v
Not that it matters, but the thing that got me started on this path is that
the H745 has two pins, who each say "20-30VAC"
Okay, so, I'm back to working on my /34. I think I made
significant
progress tonight.
What's going on is, I've narrowed it down to low AC levels
coming off
the transformer. The printset and H7xx modules
specify an
incoming AC
of 20-30v (printset says specifically 28v.) All
lines
coming off the
transformer at P1 are reading a value of 10-15v.
I am working from memory, but IIRC there's a connector with 8
black wires that carries the transformer secodnary
connections (4 windings, 2 wires per winding). It goes into
the power distribution PCB, and each winding it then
connected to one of the regulator 'bricks'.
Now, you should see 20-30V _between the 2 wires going to a
particular regulator_. Not between either wire and earth.
If you measure the voltage between a particular wire and
earth (logic 0V rail, case, whatever), then you'll see one of
2 things.
1) If there's no regulator connected to that winding, you'll
measure 0V.
The winding is electrically isolated from the rest of the machine
2) If there is a regulator connected, then each wire will
swing between ground and the full secondary voltage at mains
frequency, due to the action of the bridge rectifier in the
'brick'. Now, what that will appear at on your meter depends
on the meter, but I could well believe some would average it
out to half-voltage.
So that's the first explanation, you're measuring the voltage
between one end of the secondary and ground, not the voltage
across the secondary.
A second explation, if you are measuring the voltage across
the secondary, and it's half what you expect, is that the
primary windings are wired incorrectly. The power
transgformer has 2 primary windings, rated at 115V each. For
115V mains, you connect them in parallel (taking care of the
phase, of course!), for 230V mains you connect them in series
(ditto). The primary windings go to a 4 pin connector that
plugs into the front of the AC input box. There are 2
different AC input boxes, for 115V or 230V mains which
connect the pins of that connector appropriately -- there are
other internal changes relating to the mains switching
contactor and its control supply. Is it possible that you
have a 230V unit plugged into a 115V mains socket?
I highly doubt this as the AC input box is clearly marked 120V.
So that leaves two possibilities. It's either the
transformer, or the
AC input box.
I thing it's very unlikely to be the transformer. Shorted
turns would make it get hot and bothered very quickly.
-tony
4:34 p.m.
Tony, responses to your statements are below, but as a general question to
those who are helping me with this, is the H745 light supposed to come on if
it's just connected to the power distribution board? i.e. am I supposed to
se a light on with no load like the H7441s do?
Yes, you are. The lamp is simply connected across the output of the
regulator -- that is between the output line and ground. It'll light up
if the regulator is working.
Now, you
should see 20-30V _between the 2 wires going to a
particular regulator_. Not between either wire and earth.
This could explain my reading. I was only measuring one wire at a time. Not that it matters, but the thing that got me started
on this path is that
the H745 has two pins, who each say "20-30VAC"
Yes, it's somewhat confusing. It means you need to connect 20-30V AC
between those 2 pins. The printset should show how the transformer is
connected (note that there are no connections to the secondary windings
other than to these regulator bricks, there are no grounds, no taps
connected to ground, etc). And the prints of the regulator should justify
_why_ you connect it like that. From what I remember, those to pins go
to the AC terminals of a bridge rectifier, the output of which goes to a
large smoothing capacitor. And at least for the +5V brick, one side of
that capacitor is then grounded.
contactor and
its control supply. Is it possible that you
have a 230V unit plugged into a 115V mains socket?
I highly doubt this as the AC input box is clearly marked 120V. Wow, OK, so what is it then? This really has me
stumped.
Is it still possible I have an input box problem?
See above. I don't think there's anything wrong with the PSU at all. Just
that you're not measuring the voltage you think you're measuring.
-tony
5:12 p.m.
You're right. Somehow when I reassembled the power supply when I wasn't
testing it (just put it back together to move it out of the way)I put the
H745 in the WRONG SIDE. Therefore, I have it now hooked up to the battery
backup plug.
Duh, duh, and, more duh. :p
Needless to say I'm going to be starting all over when I get home.
Thanks for all your help in my chasing shadows.
Julian
-----Original Message-----
From: cctalk-bounces at classiccmp.org
[mailto:cctalk-bounces at classiccmp.org] On Behalf Of Tony Duell
Sent: Thursday, May 11, 2006 6:34 PM
To: cctalk at classiccmp.org
Subject: Re: BA11-K low AC output levels
wire at a time.
That might well explain it.
Try measuring the AC voltage between the 2 wires at each
regulator labelled '20-30V AC'. That is, connect one lead of
your voltmeter to one of the wires, the other lead to the other wire.
OK. Converting one to the other is possible, in that the
metal frame is the same apart from the markings on it. But
there are several internal wiring changes, I think the little
intenral transformer and the contactor are replaced, etc. I
don't think it's likely this has happened.
You could try looking at the wiring to the socket where the
transformer primary is connected (IIRC this is a 4 pin
connector on the front side of the input box). Does it look
as though the windings are in parallel (wires from the
contactor each go to 2 pins on that connector), or in series
(wires from the contactor go to 1 pin each, the remaing 2
pins are linked but not connected to anything else)? The
former is 115V, the later 230V.
Tony, responses to your statements are below, but as a general
question to those who are helping me with this, is the H745 light
supposed to come on if it's just connected to the power
distribution
board? i.e. am I supposed to se a light on with
no load
like the H7441s do?
Yes, you are. The lamp is simply connected across the output
of the regulator -- that is between the output line and
ground. It'll light up if the regulator is working.
Now, you
should see 20-30V _between the 2 wires going to a
particular regulator_. Not between either wire and earth.
This could explain my reading. I was only measuring one Not that it matters, but the thing that got me
started on
this path is
that the H745 has two pins, who each say
"20-30VAC"
Yes, it's somewhat confusing. It means you need to connect
20-30V AC between those 2 pins. The printset should show how
the transformer is connected (note that there are no
connections to the secondary windings other than to these
regulator bricks, there are no grounds, no taps connected to
ground, etc). And the prints of the regulator should justify
_why_ you connect it like that. From what I remember, those
to pins go to the AC terminals of a bridge rectifier, the
output of which goes to a large smoothing capacitor. And at
least for the +5V brick, one side of that capacitor is then grounded.
contactor
and its control supply. Is it possible that you have a
230V unit plugged into a 115V mains socket?
I highly doubt this as the AC input box is clearly marked 120V. Wow, OK, so what is it then? This really has me
stumped.
Is it still possible I have an input box problem?
See above. I don't think there's anything wrong with the PSU
at all. Just that you're not measuring the voltage you think
you're measuring.
-tony
6:44 p.m.
Disregard this, I didn't have it on the wrong side. I was viewing the
diagram wrong.
I forgot that the diagram says the H754/784 AC input connectors are white :p
-----Original Message-----
From: cctalk-bounces at classiccmp.org
[mailto:cctalk-bounces at classiccmp.org] On Behalf Of Julian Wolfe
Sent: Thursday, May 11, 2006 7:13 PM
To: 'General Discussion: On-Topic and Off-Topic Posts'
Subject: RE: BA11-K low AC output levels
You're right. Somehow when I reassembled the power supply
when I wasn't testing it (just put it back together to move
it out of the way)I put the
H745 in the WRONG SIDE. Therefore, I have it now hooked up
to the battery backup plug.
Duh, duh, and, more duh. :p
Needless to say I'm going to be starting all over when I get home.
Thanks for all your help in my chasing shadows.
Julian
OK. Converting one to the other is possible, in that the
metal frame is the same apart from the markings on it. But
there are several internal wiring changes, I think the little
intenral transformer and the contactor are replaced, etc. I
don't think it's likely this has happened.
You could try looking at the wiring to the socket where the
transformer primary is connected (IIRC this is a 4 pin
connector on the front side of the input box). Does it look
as though the windings are in parallel (wires from the
contactor each go to 2 pins on that connector), or in series
(wires from the contactor go to 1 pin each, the remaing 2
pins are linked but not connected to anything else)? The
former is 115V, the later 230V.
-----Original Message-----
From: cctalk-bounces at classiccmp.org
[mailto:cctalk-bounces at classiccmp.org] On Behalf Of Tony Duell
Sent: Thursday, May 11, 2006 6:34 PM
To: cctalk at classiccmp.org
Subject: Re: BA11-K low AC output levels
It'll light
Tony, responses to your statements are below, but as a general
question to those who are helping me with this, is the H745 light
supposed to come on if it's just connected to the power
distribution
board? i.e. am I supposed to se a light on with
no load
like the H7441s do?
Yes, you are. The lamp is simply connected across the output of the
regulator -- that is between the output line and ground. up if the regulator is working.
wire at a time.
That might well explain it.
Try measuring the AC voltage between the 2 wires at each regulator
labelled '20-30V AC'. That is, connect one lead of your
voltmeter to
Now, you
should see 20-30V _between the 2 wires going to a
particular regulator_. Not between either wire and earth.
This could explain my reading. I was only measuring one one of the wires, the other lead to the other
wire.
20-30V AC
Not that it matters, but the thing that got me
started on
this path is
that the H745 has two pins, who each say
"20-30VAC"
Yes, it's somewhat confusing. It means you need to connect between those 2 pins. The printset should show
how the
transformer is
connected (note that there are no connections to
the secondary
windings other than to these regulator bricks, there are no
grounds,
no taps connected to ground, etc). And the prints
of the regulator
should justify _why_ you connect it like that. From what I
remember,
those to pins go to the AC terminals of a bridge
rectifier,
the output
of which goes to a large smoothing capacitor. And
at least
for the +5V
brick, one side of that capacitor is then
grounded.
> > contactor and its control supply. Is it possible that
you have a
230V unit plugged into a 115V mains socket?
I highly doubt this as the AC input box is clearly marked 120V. Wow, OK, so what is it then? This really has me
stumped.
Is it still possible I have an input box problem?
See above. I don't think there's anything wrong with the PSU
at all. Just that you're not measuring the voltage you think
you're measuring.
-tony
12 May
12 May
4:56 p.m.
I forgot that the diagram says the H754/784 AC input
connectors are white :p
I am now trying to work out how the colour of the connector can have any
bearing on the voltage you measure!
[YEs, OK, I did once see measurements that showed that black plastic
boxes were more electrically leaky than the equivalent white ones --
presumably the black colouring contained carbon or something. But that
was working with voltages of 10's of kV, and currents of nanoamps. Not
applicable to the PDP11 PSU]
-tony
4:53 p.m.
You're right. Somehow when I reassembled the power supply when I wasn't
testing it (just put it back together to move it out of the way)I put the
H745 in the WRONG SIDE. Therefore, I have it now hooked up to the battery
backup plug.
I am not sure how this has any effect on the voltages you measure on the
transformer secondaries.
Can we go back a bit, please.
1) What is the problem with this machine -- it appears you think there's
a PSU problem. Which output(s) are defective?
2) What do you measure at the (transformer) inputs to the regulator
bricks. Rememebr a (working) voltmeter has two leads, so you are always
mesuring a voltage between 2 points. What are those points? When checking
the transformer output, you connect one voltmeter probe to each of the
transformer wires at the regulator brick -- not between either wire and
ground.
-tony
13 May
13 May
7:07 a.m.
What the problem with the machine, at least I believe, is, is that the
memory is not getting any power.
The light on the H745 is not coming on. I'm trying to figure out why.
-----Original Message-----
From: cctalk-bounces at classiccmp.org
[mailto:cctalk-bounces at classiccmp.org] On Behalf Of Tony Duell
Sent: Friday, May 12, 2006 6:54 PM
To: cctalk at classiccmp.org
Subject: Re: BA11-K low AC output levels
You're right. Somehow when I reassembled the power supply when I
wasn't testing it (just put it back together to move it out of the
way)I put the
H745 in the WRONG SIDE. Therefore, I have it now hooked up to the
battery backup plug.
I am not sure how this has any effect on the voltages you
measure on the transformer secondaries.
Can we go back a bit, please.
1) What is the problem with this machine -- it appears you
think there's a PSU problem. Which output(s) are defective?
2) What do you measure at the (transformer) inputs to the
regulator bricks. Rememebr a (working) voltmeter has two
leads, so you are always mesuring a voltage between 2 points.
What are those points? When checking the transformer output,
you connect one voltmeter probe to each of the transformer
wires at the regulator brick -- not between either wire and ground.
-tony
1:22 p.m.
What the problem with the machine, at least I believe, is, is that the
memory is not getting any power.
The light on the H745 is not coming on. I'm trying to figure out why.
OK, let's start there. Make sure the bulb isn't blown. It happens, all
too often.
Then check the AC input to that brick. Connect an AC voltmeter between
the 2 wires that come from the transformer (the ones labelled '20-30V
AC'. What do you read there?
Then connect a DC voltmeter between the 0V connection (or system ground)
and each of the output terminals in turn. What voltages do you read there?
-tony
6800
days inactive
6802
days old
9 comments
2 participants
participants (2)
-
ard＠p850ug1.demon.co.uk -
fireflyst＠earthlink.net