[Repaired] EDC 522 DC Voltage Calibrator

[nidlaX]

Well good news, my de-soldering is less incompetent than I thought, the plating around the other three through-holes is still ok. Looks like I didn't clean it vigorously enough...

[jlmoon]

Hello,
I have been following this thread for a bit.  Might I suggest you measure the voltage at the junction of R334 and R335.  These are precision (individually selected per unit) 16.681K resistors.  Be real careful doing this because there could be - or + ~120Volts at this point.  This is the feedback reference from the output relay switching arrangement that sets the feedback loop up based on the front panel voltage range selection.  These instruments can be quite a trick to diagnose unless you understand the grand scheme of things and how they're supposed to function.  When you first power the unit up the display should go through a display diagnostics displaying the model number and then the IEEE-488 address.  After that it should reflect values from 0 - 9 based on what you have dialed on each decoder position.   I understand you're having trouble with the 5 Volt supply, have you corrected this yet?

JLM

Forgot to mention, when making this 16.681K junction measurement be sure to reference your negative meter lead to the junction of capacitors C416 and C417.  This is the +150V and -150V supply common.  WARNING !!! There is nearly 300 Volts potential in this area if across the two caps 416 & 417 .. so be really careful here.  This supply is the main function of the output stages driven by the differential amplifier created by Q301 and Q302.  Please take note this can be a dangerous unit to service if you're not familiar with this circuitry.  Best of Luck!

[nidlaX]

Hello,
I have been following this thread for a bit.  Might I suggest you measure the voltage at the junction of R334 and R335.  These are precision (individually selected per unit) 16.681K resistors.  Be real careful doing this because there could be - or + ~120Volts at this point.  This is the feedback reference from the output relay switching arrangement that sets the feedback loop up based on the front panel voltage range selection.  These instruments can be quite a trick to diagnose unless you understand the grand scheme of things and how they're supposed to function.  When you first power the unit up the display should go through a display diagnostics displaying the model number and then the IEEE-488 address.  After that it should reflect values from 0 - 9 based on what you have dialed on each decoder position.   I understand you're having trouble with the 5 Volt supply, have you corrected this yet?

JLM

Forgot to mention, when making this 16.681K junction measurement be sure to reference your negative meter lead to the junction of capacitors C416 and C417.  This is the +150V and -150V supply common.  WARNING !!! There is nearly 300 Volts potential in this area if across the two caps 416 & 417 .. so be really careful here.  This supply is the main function of the output stages driven by the differential amplifier created by Q301 and Q302.  Please take note this can be a dangerous unit to service if you're not familiar with this circuitry.  Best of Luck!

Thanks for the tips JLM. I had already measured the +-150V rails at the rectifier without knowing beforehand what the voltages were supposed to be.  As far as I can tell, that's the only functional rail at the moment. The 5V problem is still not fixed - I am concerned that the transformer might be damaged.

[jlmoon]

The 5V problem is still not fixed - I am concerned that the transformer might be damaged.

NidlaX,

I seriously doubt the windings are damaged, the windings used for the 5V generation are quite large in size making it pretty tough to burn up or short those, most of the time when severe loading is present on the secondaries of a transformer, the primary windings are usually the ones to melt or short down and start blowing fuses. 

The first thing I might suggest is to isolate that 5V supply down to just the bridge & filter caps and regulator.  If you do this be sure to remove and cap the windings that make the +- 150V so as not to bang the diff amp on the rails and blow up the output buffers due to improper DAC drive from the linear circuits.  This will save you lots of grief.  Actually I would recommend you isolate the HV supply until you get a good scaled output from the reference ladder and the DAC output op amp.  Once you get the 5V working and all the control systems then you can move forward into the power stuff that drives the output amp.  Does this make sense to you?


I have repaired and calibrated a few of these including JWallings 521. 
I think I can point you in the right direction if you give me some readings in various places.  Let me know.


JLM

[nidlaX]

The 5V problem is still not fixed - I am concerned that the transformer might be damaged.

I seriously doubt the windings are damaged, the windings used for the 5V generation are quite large in size making it pretty tough to burn up or short those, most of the time when severe loading is present on the secondaries of a transformer, the primary windings are usually the ones to melt or short down and start blowing fuses. 

Agreed, but the physical evidence is present that melting has occurred on at least one of the windings, unless it's just a manufacturing defect.

I'll take your advice and isolate the HV supply. Should I simply disconnect the transformer taps to the board and set them aside?

[jlmoon]

Absolutely Yes.  Cap them up so not to spoil your day.

[dacman]

2.6 Vdc is not enough to energize the relays.  This would result in no output and no feedback.

What is the line voltage switch set to?

What is the voltage at (what would be) the input to the LM309K if the 5V connector is disconnected?  How much current can it source?  What is the voltage across the HPCL-3700 IC pins 1 & 4?

Do you have a 5V power supply that can source about 1.5 Adc that can be used to power the 5V circuit?

[jlmoon]

2.6 Vdc is not enough to energize the relays.  This would result in no output and no feedback.

Nor would 2.6 volts be enough to properly run any of that digital logic.  Would be handy to remove output lead of that 309K and feed with a 5 V current regulated supply to determine if there is some excess loading going on, which would explain the heating of that bridge rectifier pin on pcboard and limiting the output on that 309K.  Is that regulator hot to the touch?

[nidlaX]

2.6 Vdc is not enough to energize the relays.  This would result in no output and no feedback.

Nor would 2.6 volts be enough to properly run any of that digital logic.  Would be handy to remove output lead of that 309K and feed with a 5 V current regulated supply to determine if there is some excess loading going on, which would explain the heating of that bridge rectifier pin on pcboard and limiting the output on that 309K.  Is that regulator hot to the touch?

When I installed the new rectifier, I was getting 0V from the rectifier output. That is currently the issue that I'm focusing on.

I thought about powering the regulator externally, but I was worried about damaging something.

[jlmoon]

If you have 0 V I would isolate supply and if output is normal then look for a shorted tantlum cap or component downstream.

[nidlaX]

I desoldered C420 and it seems to be ok. However, I still measure ~0 Ohms across transformer taps 10 and 11.

[dacman]

Do you mean that there is no output and the fuse is still good?

Are the +/- 15 VDC supplies OK (at two each 7815 and 7915 regulators).

In the third set of pictures there is a white connector with a green circle imposed on it.  If this connector is pulled, the 309K would be removed from the circuit.  The pins are standard 0.093” pins.

The 5V supply transformer windings should have low Ohms due to high current output.

[nidlaX]

Do you mean that there is no output and the fuse is still good?

Are the +/- 15 VDC supplies OK (at two each 7815 and 7915 regulators).

In the third set of pictures there is a white connector with a green circle imposed on it.  If this connector is pulled, the 309K would be removed from the circuit.  The pins are standard 0.093” pins.

The 5V supply transformer windings should have low Ohms due to high current output.

I wish I knew more about the specifications of the transformer. Suppose I have disconnected the rectifier D402 for the +11V DC output to the +5V LM309: should I still measure an AC voltage across taps 11 and 12?

The mains fuse is fine, I have not measured the +/- 15V supplies yet. I will test the LM309K by itself.

[dacman]

Taps 11 and 12 or 10 and 11?  (I have access to a 522 manual at my day job, but I don't access eevblog at work.  The manual would need to be available on-line.)

There should be AC voltage across this open transformer winding; I would guess 8 VRMS or more.  There should be AC voltage across it, even if the rectifier is installed.

If that unit has a separate Line Voltage Selector switch (115V/230V) I would want to know that it was good.  I would Ohm it out, or at least exercise it a few times to try and knock off any possible oxidation.

Jlmoon is giving good advice.  I wouldn't want 150V feeding back into the power amplifier.  I also would not want the power transistors going into saturation, which could happen whether or not the relays are energized.

[nidlaX]

I've revisited this repair over the past few days. Today, I fixed the 5V power supply problem with a new bridge rectifier. I drilled through the carbonized area of the PCB and soldered a bodge wire directly from the positive leg of the rectifier to the corresponding capacitor. The unit powers and goes through the self test.

Now, I have some weird problems with the output. If I set negative polarity, I can get output on all the ranges for all possible values most of the time. (Sometimes I get an overload when I am increasing the voltage) If I set positive polarity, I can only get output up to about 0.5 V on the 100 V range or about 0.2 V on the 10 V range. Anything else would trigger an overload, and I would continue to get overload until I back the set value down to near 0. At first, I thought this could be a front panel switch issue, so I sprayed each switch out with deoxit and actuated them many times. This does not seem to solve the root problem, however. I get similar results if I attempt to program the output over GPIB, but for some reason I can only reach 0.2 V on the 100 V range when setting remotely as opposed to the 0.5 V I can obtain when setting locally.

Are there any suggestions for how I should proceed to troubleshoot this?

[nidlaX]

Some more progress on this: the output at Z303 appears to become increasingly unstable at positive voltages. I see a half-sinewave like ripple when I increased output to >= 0.5 V on the 100 V range. FB and INP appear stable relative to SG. I swapped Z303 to little effect - the instability still appears eventually as I increase the voltage. The output will hold at any negative value, but overload will trip if I try to increase the voltage, so it seems that there is an issue any time the non-inverting input is greater than the inverting input.

Time to keep digging.

[dacman]

The output of the DAC (Z101) should be the same with the polarity switch set to either + or -.  Polarity gets inverted by FETs Q311 through Q314 (between Z101 and Z303, two FETs turn on and the other two turn off, inverting the input to Z303).

Check the output transistors (Q306, Q308) and drivers (Q305, Q307), R311 and R315 (3.3 Ohm resistors) and the other output stage resistors, and the MJE350 transistors (Q301 to Q304).  Also, check the diodes (D305 to D309) and make sure the +/- 150 Vdc rails are ok.

[nidlaX]

The output of the DAC (Z101) should be the same with the polarity switch set to either + or -.  Polarity gets inverted by FETs Q311 through Q314 (between Z101 and Z303, two FETs turn on and the other two turn off, inverting the input to Z303).

Yes, the DAC output and polarity switching were both checked early on.

Check the output transistors (Q306, Q308) and drivers (Q305, Q307), R311 and R315 (3.3 Ohm resistors) and the other output stage resistors, and the MJE350 transistors (Q301 to Q304).  Also, check the diodes (D305 to D309) and make sure the +/- 150 Vdc rails are ok.

Thanks, I'm continuing to go through the output amplifier stage.

[dacman]

There will be a little ripple on the output of Z303, even if it is working well.  The ripple on the +/- 150 Vdc power supplies is higher than the output noise specification, and the way this is reduced is that Z303 follows the ripple (which is on top of the feedback) and takes most of it out.  FB will look stable but Z303-6 will not.

[nidlaX]

There will be a little ripple on the output of Z303, even if it is working well.  The ripple on the +/- 150 Vdc power supplies is higher than the output noise specification, and the way this is reduced is that Z303 follows the ripple (which is on top of the feedback) and takes most of it out.  FB will look stable but Z303-6 will not.

How much should I expect to be normal? Should it be on the order of the ripple of +/- 150 V vs power_gnd?

[nidlaX]

I don't know why I keep triggering myself by not checking the simple things first.

When probing Q301 through Q304 today, I noticed that the symptoms were "improving" depending on which transistor I was hooked onto. Turning the unit around revealed solder joints that were basically falling off to light pressure on the transistor legs. Given how bad the soldering job is throughout my unit, I should've known to triple check all the suspicious looking joints. 5 minutes of reflow later and I have stable output on all ranges again.

I could say repair completed, but I know deep down that I should reflow just about every single joint on the unit because they all look like shit. Thanks EDC or whoever worked on this unit last...

[Samogon]

Hello,
I have been following this thread for a bit.  Might I suggest you measure the voltage at the junction of R334 and R335.  These are precision (individually selected per unit) 16.681K resistors.  Be real careful doing this because there could be - or + ~120Volts at this point.  This is the feedback reference from the output relay switching arrangement that sets the feedback loop up based on the front panel voltage range selection.  These instruments can be quite a trick to diagnose unless you understand the grand scheme of things and how they're supposed to function.  When you first power the unit up the display should go through a display diagnostics displaying the model number and then the IEEE-488 address.  After that it should reflect values from 0 - 9 based on what you have dialed on each decoder position.   I understand you're having trouble with the 5 Volt supply, have you corrected this yet?

JLM

Forgot to mention, when making this 16.681K junction measurement be sure to reference your negative meter lead to the junction of capacitors C416 and C417.  This is the +150V and -150V supply common.  WARNING !!! There is nearly 300 Volts potential in this area if across the two caps 416 & 417 .. so be really careful here.  This supply is the main function of the output stages driven by the differential amplifier created by Q301 and Q302.  Please take note this can be a dangerous unit to service if you're not familiar with this circuitry.  Best of Luck!

Good point jlmoon !
I have 520a with "overload" turning on output +/-23V. On the junction between R334 and R335 measures few volts.   
Also noticed that at power supply +/- 150 volts experience voltage drops to +/- 130 volts depending on what polarity output set on control panel.
I assume something in Output Amplifier has failed and drags down PSU voltage.
Still searching for it. If you have ideas i would appreciate your input.

[dacman]

Check the output transistors and associated components (drivers, resistors, and diodes).  The MJE350s (and associated circuitry) can also fail.  I'd be curious to know what Series the unit is (on the back panel).  There is a resistor on the other side of the rail from the MJE350s.  It was originally a 13 kOhm 2 Watt carbon (it could short and take out the MJE350s).  It was revised to a 10 kOhm 5 Watt WW.  The two 22 kOhm 2 Watt bleeders in the power supply were revised to 47 kOhm bleeders (they can short and blow fuses).

[Samogon]

Serial is 14022
I did check power transistors they are easily detached, drivers need to desolder to check, will do tomorrow.
Resistor is 13k and two others 47k.
There is charring on MJE350 and resistors also charing on +/- 150 rectifier output.

[jondaddio]

Hi Jwalling,

Any chance I could get a copy of your scanned schematic and manual for the Khron-Hite 522 DC Voltage Calibrator?

TIA,

Jon