Maker Pro
Maker Pro

HP 3310B Malfunction Generator

R

rickman

Jan 1, 1970
0
I bought this thing some 10 years ago used and it has worked ok, mainly
as an audio generator. I fired it up the other day and the upper ranges
seem to output a funky sine wave now. The 10 kHz and 100 kHz ranges
output a distorted waveform that does a direction reversal a little bit
after going negative of the midpoint. It almost looks like a rectified
sine wave, but the waveform is more than half a cycle. I don't see any
issues on the other waveforms.

I can play with the range switch and find that turning it slowly can
have an impact on the point of the waveform where the reversal occurs.
But it doesn't really feel like a switch problem. I would disassemble
the switch to see if there are mechanical issues, but it is one of those
multi-gang wafer switches with components mounted directly on it and
each gang is soldered to the board, a real nightmare to remove. This
thing was made to never break, not to be repaired.

I don't have any info on it. The circuit board has some dozens of
transistors, a couple of what are likely op amps (metal cans) and well
over a hundred passives. I don't know where to begin trying to fix
it... other than connectors and switches are the primary point of
failure. But even removing and reseating boards looks like a bear in
this thing.

Any suggestions on ways to repair this?

I saw the other thread on new units and had looked at some of the little
$8 boards on eBay. Funny that there isn't much in between the $8 boards
and the $400 boxes. I would have thought this is something that could
be done very inexpensively these days. I would use a PC audio output
but my signal is outside the 20 kHz upper limit of audio outputs.

Maybe I'll add a simple sig-gen to the prototype circuit I'm building.
Lots more than $8 of effort, but I'll know what I'm getting.

Rick
 
J

Jamie

Jan 1, 1970
0
rickman said:
I bought this thing some 10 years ago used and it has worked ok, mainly
as an audio generator. I fired it up the other day and the upper ranges
seem to output a funky sine wave now. The 10 kHz and 100 kHz ranges
output a distorted waveform that does a direction reversal a little bit
after going negative of the midpoint. It almost looks like a rectified
sine wave, but the waveform is more than half a cycle. I don't see any
issues on the other waveforms.

I can play with the range switch and find that turning it slowly can
have an impact on the point of the waveform where the reversal occurs.
But it doesn't really feel like a switch problem. I would disassemble
the switch to see if there are mechanical issues, but it is one of those
multi-gang wafer switches with components mounted directly on it and
each gang is soldered to the board, a real nightmare to remove. This
thing was made to never break, not to be repaired.

I don't have any info on it. The circuit board has some dozens of
transistors, a couple of what are likely op amps (metal cans) and well
over a hundred passives. I don't know where to begin trying to fix
it... other than connectors and switches are the primary point of
failure. But even removing and reseating boards looks like a bear in
this thing.

Any suggestions on ways to repair this?

I saw the other thread on new units and had looked at some of the little
$8 boards on eBay. Funny that there isn't much in between the $8 boards
and the $400 boxes. I would have thought this is something that could
be done very inexpensively these days. I would use a PC audio output
but my signal is outside the 20 kHz upper limit of audio outputs.

Maybe I'll add a simple sig-gen to the prototype circuit I'm building.
Lots more than $8 of effort, but I'll know what I'm getting.

Rick


Caps..and power supply.

Jamie
 
T

tm

Jan 1, 1970
0
rickman said:
I bought this thing some 10 years ago used and it has worked ok, mainly as
an audio generator. I fired it up the other day and the upper ranges seem
to output a funky sine wave now. The 10 kHz and 100 kHz ranges output a
distorted waveform that does a direction reversal a little bit after going
negative of the midpoint. It almost looks like a rectified sine wave, but
the waveform is more than half a cycle. I don't see any issues on the
other waveforms.

I can play with the range switch and find that turning it slowly can have
an impact on the point of the waveform where the reversal occurs. But it
doesn't really feel like a switch problem. I would disassemble the switch
to see if there are mechanical issues, but it is one of those multi-gang
wafer switches with components mounted directly on it and each gang is
soldered to the board, a real nightmare to remove. This thing was made to
never break, not to be repaired.

I don't have any info on it. The circuit board has some dozens of
transistors, a couple of what are likely op amps (metal cans) and well
over a hundred passives. I don't know where to begin trying to fix it...
other than connectors and switches are the primary point of failure. But
even removing and reseating boards looks like a bear in this thing.

Any suggestions on ways to repair this?

I saw the other thread on new units and had looked at some of the little
$8 boards on eBay. Funny that there isn't much in between the $8 boards
and the $400 boxes. I would have thought this is something that could be
done very inexpensively these days. I would use a PC audio output but my
signal is outside the 20 kHz upper limit of audio outputs.

Maybe I'll add a simple sig-gen to the prototype circuit I'm building.
Lots more than $8 of effort, but I'll know what I'm getting.

Rick

Service manual here:

http://www.home.agilent.com/agilent...o&pageMode=OV&pid=3310B:epsg:pro&cc=US&lc=eng

As always, check the power supplies first.

Then RTFM
 
T

Tim Williams

Jan 1, 1970
0
Not knowing the 3310 offhand, I thought for a moment you were talking
about a piece of test equipment which intentionally generated malfunctions
for testing purposes.

Hey, it's HP, who knows what kind of stuff they made!...

Tim
 
W

WoolyBully

Jan 1, 1970
0
Not knowing the 3310 offhand, I thought for a moment you were talking
about a piece of test equipment which intentionally generated malfunctions
for testing purposes.

Hey, it's HP, who knows what kind of stuff they made!...

That is kinda funny.

Except for the top post.
 
R

rickman

Jan 1, 1970
0
Not knowing the 3310 offhand, I thought for a moment you were talking
about a piece of test equipment which intentionally generated malfunctions
for testing purposes.

Hey, it's HP, who knows what kind of stuff they made!...


Lol!

I don't care if you top post, but in this case your sig made all of my
post disappear... even better, I didn't have to snip! I'll try top
posting in some of the idiot threads that are so popular here. That
should encourage others to trim a bit... lol

Rick
 
R

rickman

Jan 1, 1970
0
Operator and Service Manual for P 3310A/B is available here:
http://cp.literature.agilent.com/litweb/pdf/03310-90003.pdf

Thanks. I'll take a look at it. I guess I'm just intimidated by the
disassembly job alone.

It doesn't really look like a PSU issue to me. I would suspect the
switches, but when I work them it just doesn't feel like that's the
problem. The unit has tons of test points so that should make it easier
to see what is going on.

Rick
 
J

Jamie

Jan 1, 1970
0
rickman said:
Thanks. I'll take a look at it. I guess I'm just intimidated by the
disassembly job alone.

It doesn't really look like a PSU issue to me. I would suspect the
switches, but when I work them it just doesn't feel like that's the
problem. The unit has tons of test points so that should make it easier
to see what is going on.

Rick
I am not sure the age of that unit but I hope it's newer than the tube
era :) A LCR meter would serve you will, along with a DMM.

Oh, and lets us not for get to use a scope on the power supply for cap
ripples and possible linkage in selenium devices and maybe germanium.

Jamie
 
R

Robert Baer

Jan 1, 1970
0
Jamie said:
Caps..and power supply.

Jamie
I have an HP 3312A Function Generator, and the manual.
The looks are quite different, but the general operation should be
very similar.
Basically, it starts with a triangle / square generator,and the sine
is created with a DFG from the triangle.
So,your sine waveform problem is due to a bad diode or resistor in
the diode / resistor chain on the polarity side where you see the defect.
I could scan and send diagrams as need.
 
R

Robert Baer

Jan 1, 1970
0
Jamie said:
Caps..and power supply.

Jamie
I have an HP 3312A Function Generator, and the manual.
The looks are quite different, but the general operation should be
very similar.
Basically, it starts with a triangle / square generator,and the sine
is created with a DFG from the triangle.
So,your sine waveform problem is due to a bad diode or resistor in
the diode / resistor chain on the polarity side where you see the defect.
I could scan and send diagrams as need.
 
R

Robert Baer

Jan 1, 1970
0
Tim said:
Not knowing the 3310 offhand, I thought for a moment you were talking
about a piece of test equipment which intentionally generated malfunctions
for testing purposes.
Reminds me of the Missing Pulse Generator used in the Nike/Ajax
missle test van....
 
T

tm

Jan 1, 1970
0
Robert Baer said:
NOTE: The OP _only_ complained about the sine output, and mostly at
higher frequencies; definitely not PS oriented; prob is in one of the DFG
chains.

So you would not recommend first checking the power supplies on a 40+ year
piece of test equipment before doing any follow-up trouble shooting? Even
when it is a well known fact that electrolytic capacitors that old are often
found degraded.

Oh well, whatever.
 
R

rickman

Jan 1, 1970
0
So you would not recommend first checking the power supplies on a 40+
year piece of test equipment before doing any follow-up trouble
shooting? Even when it is a well known fact that electrolytic capacitors
that old are often found degraded.

Oh well, whatever.

Hey, that is one of the "rules of thumb" for repairing any circuit fault
in my opinion. First check the supplies, then if a digital circuit,
check the clocks. They are fast, easy checks to make.

I think Robert is saying he has some insight into the failure based on
the symptoms, so maybe the PSU can be skipped this time. I will be
looking at the manual I download and if it has good info on the test
points, I'll be checking the power supplies.

Thanks,

Rick
 
R

rickman

Jan 1, 1970
0
Reminds me of the Missing Pulse Generator used in the Nike/Ajax missle
test van....

Oh, how long has it been missing?

lol

Rick
 
T

Tim Williams

Jan 1, 1970
0
rickman said:
I don't care if you top post, but in this case your sig made all of my
post disappear... even better, I didn't have to snip! I'll try top
posting in some of the idiot threads that are so popular here. That
should encourage others to trim a bit... lol

That's the spirit! ;-)

Plus it's nice if you're just slapping something on which isn't very
relevant to quoted material. Saves time, makes it easy to read.

Tim
 
R

Robert Baer

Jan 1, 1970
0
tm said:
So you would not recommend first checking the power supplies on a 40+
year piece of test equipment before doing any follow-up trouble
shooting? Even when it is a well known fact that electrolytic capacitors
that old are often found degraded.

Oh well, whatever.
I did NOT say that; i mentioned where the problem was most likely to
be found.
Yes, it is possible that there may be one or two degraded
'lytics..even the old wet electrolytics lasted 20 years at best and even
then many could be recovered by adding electrolyte.
On old tube equipment, powering them up with a variac at zero and s l
o w l y raising the line voltage to full will allow most of degraded
'lytics to re-form and perform adequately at worst.
 
R

rickman

Jan 1, 1970
0
Listen to Maynard Philbrick at your own risk.

Ok, I'll bite, who is Maynard Philbrick? I thought the name was
familiar, but googling it doesn't provide anything useful, mainly
genealogy pages. I can't even seem to exclude them they are so
pervasive. Linking the name to HP doesn't do it either.

The name does sound familiar. Is that a name here?

HP used germanium transistors in the '60s designs, like everyone
else, then moved to silicon. They weren't penny pinching, they were
cutting edge designs.

You should always verify the power supply is in spec on test
equipment before doing anything else. Slight errors are compounded, and
can cause you all kinds of headaches including a dead piece of
equipment. Start with the +/-25 volt outputs are right and have no
ripple. I see no Germanium transistors listed. I see are bipolar
silicon, and a few FET. The parts list is in the middle of the manual,
the manual is dated 1973, and has a lot of design changes in the back
and sorted by serial number groups. Unless you are very familiar with
troubleshooting, you may need help.

I've mainly worked on digital stuff including state of the art (at the
time of course) floating point array processors. Analog is not my
forte, but I'm comfortable with it.

It seems I am spending more time getting a new lab setup than actually
working on anything. I cleared out some old appliances and now have
room for an 8 foot workbench which I plan to build. But before I do
that I am turning my hand router and hand saw into table tools to help
with the workbench construction. The test gear is the same way, I need
this to fix that and before I can use that I need this other thing...
I've let things slide too long.

Wasn't it Lincoln who said, "Give me six hours to chop down a tree and I
will spend the first four sharpening the axe"?

I'd prefer to use a chain saw...

Rick
 
T

tm

Jan 1, 1970
0
Robert Baer said:
I did NOT say that; i mentioned where the problem was most likely to be
found.
Yes, it is possible that there may be one or two degraded 'lytics..even
the old wet electrolytics lasted 20 years at best and even then many could
be recovered by adding electrolyte.
On old tube equipment, powering them up with a variac at zero and s l o
w l y raising the line voltage to full will allow most of degraded 'lytics
to re-form and perform adequately at worst.

OK then. Sorry for the comment. I miss-understood what you were implying.
And agree with your assessment.
 
Top