Repairing bad power supply for Tektronix 212 vintage analog oscilloscope, need h

[SittingBear]

Hello EEVBlog Repair Forums,
The issue for my oscilloscope is the 0.4A fuse blows the instant I connect to AC mains. The physical power switch is set to "OFF". There is no obvious visible damage. All diodes and bridge rectifiers are OK. Forward voltage of 0.55V measured with multimeter across diodes and bridge rectifier. Outer 2 bridge pins are 1.1V. No capacitors read 0-Ohms nor do I get a low ohm-reading from Positive output to Negative outputs; typically greater than 1Meg-ohms in either polarity with multimeter across positive and negative supply rails and AC mains input.

Here is the

of the power supply circuit. I also took the liberty of photographing the circuit board in question. It is completely disconnected from all others as to avoid partial damages of the digital components in the repair process - the problem persists even when disconnected.

I've been skimming through the Tektronix Oscilloscope Troubleshooting Guide to find a few possible issues, though I have yet to find the root of the problem.

Any help is greatly appreciated!

[Vito_R]

Recheck all rectifier diodes makes sure they are also not leaky when you reverse the leads.  If you get a reading desolder one leg and test again.  If you are sure then the next thing I would check are all those disc capacitors and tantalum caps known to sometimes short when then go bad, and anything across the mains that could short out your circuit before the switch.

[Vito_R]

Looking at your pics again I would also clean off those terminal pins on the board labeled P4  P6. Lot's of corrosion and looks like some kind of arcing or short occured between 2 terminals? Doesn't hurt to clean it all off and recheck.

[WarSim]

If the power switch is off all it can be is the filter, snubber or bridge.  The arc over point could defiantly be a symptom.  Given the arc is between P4-2 and P6-1.  I would have to suspect the filter and snubber.  Both a meant to control voltage spikes at P6-1. 


Sent from my iPad using Tapatalk

[tautech]

Parts of the PCB are a mess as far as mains is concerned.
Firstly it needs a good clean on the top side.
The green/blue around some connections and pads is corrosion ie. salts that are very conductive at mains potential.
The arcing could be a result of back EMF from the common mode choke when the fuse blows.
Make up a "dim bulb tester" to limit the energy on power on.
Remove the bridge and double check it for correct operation, maybe with a small AC wallwart.
I can't see a reason why it could not be powered with bridge removed for testing.
Other suspects are shorted caps and a shorted common mode choke.

When the repair is completed it would be wise to use a PCB lacquer.

I also note a few Tantalum bead caps that give me the heebee geebees although they are after the switch.

[SittingBear]

If the power switch is off all it can be is the filter, snubber or bridge.  The arc over point could defiantly be a symptom.  Given the arc is between P4-2 and P6-1.  I would have to suspect the filter and snubber.  Both a meant to control voltage spikes at P6-1. 


Sent from my iPad using Tapatalk

What is the best way to remove corrosion from the pins? Should I test for only continuity across components to see if they're bad?

[WarSim]

The cheapest way is allot of running water, then IPA wash, then remove with a fibreglass pen.  Repeat rinsing and washing as required when using fibreglass pen.   Once you get rid of all the corrosion replace missing traces.  And through hole plating. 
If you have to remove allot of contaminated PCB you will need to replace it with high temp epoxy.  3M sells EpoxyPatch for allot of money but it is basically pigmented high temp epoxy. 
Applica Traces and through hole grommets can be overly expensive.  For non professional jobs I have been experimenting with alternatives.  My best hope is using copper tape. 
Once affixed and prepared tin the new traces.  If you do not have tinning solution then tin with solder.  But the temp is critical.  Too hot and the traces will lift.  Too low and you do not get bonding. 
Next over coat the traces with PCB lacquer, then clean the pads.  Or cover with high temp epoxy.  If you don't have a hot air tool and you need to use a heat gun you may need to remove more components. 
Once dry install components, clean then lacquer again if you used lacquer previously.   
This is the basics of board repair.  There are many more steps but these are the ones I foresee you will require.

Testing continuity test will only test for conduction it will not test it ability to handle the specified current load. 
In reality the only real way to test a component is a curve tracer.  But a component tester is the most common accepted method.  IE an LCR meter.  Continuity across anything but a resistor is no guaranty that the component is good.  With out a component tester or curve tracer you will need to find the part through symptoms and processes.  Such as using lower AC input, change the bridge to a resistive load, applying DC to the AC input to name a few. 


Sent from my iPad using Tapatalk

[Vito_R]

If the power switch is off all it can be is the filter, snubber or bridge.  The arc over point could defiantly be a symptom.  Given the arc is between P4-2 and P6-1.  I would have to suspect the filter and snubber.  Both a meant to control voltage spikes at P6-1. 


Sent from my iPad using Tapatalk

What is the best way to remove corrosion from the pins? Should I test for only continuity across components to see if they're bad?

I usually just scrub everything off with a small brush and alcohol.  You can recoat the pcb with lacquer at the end when your repairs are done and the board is cleaned up.  If there are any damaged traces repair them.  To test those components and specifically the caps I highly recommend using LCR and ESR meters.  Desolder each component and properly test them to be 100% sure it is functioning and in spec.  There's not much to check so just do it one at a time.

[Vito_R]

Did you desolder and test your bridge again for forward and reverse voltages? Anything leaky?

[SittingBear]

I haven't had the chance to work on the board yet. I will tonight when I get home from university. I do not have an LCR meter nor an ESR meter. Granted, to do a reliable test I'd have to remove the element from the circuit to isolate it from any loads. The best I can do is re-test the bridge and the rectifiers along with simply testing resistance shorts across elements. I'm fairly convinced the fault is simply arcing across the pins and will try it out after fixing the corrosion. I do not have PCB laquer available to me.

I'll most likely go at it with some fine and small steel wool with acetone. This should be strong enough to rid the corrosion, but shouldn't further damage or expose the traces due to not using coarse material.

[WarSim]

You can use a constant current power supply or a cheeper powersupply and a constant current load to test caps for value and leakage.  Same for inductors but less accurately.  An LCR meter (most are also an ESR meter) I just a specific use tool to achieve the same thing more accurately. 
You just need to make sure the caps are in the ballpark so a calibrated special tool is not strictly required. 


Sent from my iPad using Tapatalk