Re: Help required finding a Tek 466 power rail short


Harvey White
 

Ok, interleaved:

On 3/8/2022 10:32 PM, James55 wrote:
Hi Harvey.

As the scope was essentially dead with only 0.6v at the test points on the 15v, 5v, -8v and -15v rails with 38 and 43 volts on the 140 and 65v test points respectively with no lights, fan, or any signs of life, I disconnected the five bridge rectifiers by lifting one of the AC input wires on each one. I did this because I wanted to understand to what extent the PSU problem lay.
And there's the first problem.  I'm using the 468 power supply (ought to be close) as a reference.  Bridge rectifiers can be used two ways, one is a full wave bridge.  The +5 volt supply (in the 468) has a transformer winding without a center tap.  Isolating it (and my schematic shows a fuse in series with the transformer lead to the bridge.  Removing that fuse ought to be easier than unsoldering a leg of the bridge.  Since one side (- side) of the bridge is grounded, there is only one bulk voltage.  Disconnecting that supply lead will work and kill the +5v supply.  (the 468 has a substantial amount of 5 volt logic in it, so has a heavy 5 volt supply.  Similar situation with the -8v, (although it's flipped a bit), +15 volt (although the fuse is on the output, so harder to shut off)

Now the 150 volt supply is a bit odder.  It's got a center tap on the transformer.  The two filter capacitors are effectively in series, but what you have is two stacked 75 volt supplies. There's a regulator on the 75 volt supply that produces the +55, and one that takes the +150 unregulated voltage and produces the +110.  To kill those supplies, you must disconnect both AC inputs to the bridge.



The point of ground reference was the one of the 'Gnd' test points.
+
I wanted to see if there was any point where (something near) the correct voltages were present.

With the Bridges out of circuit, the scope finally showed some resonable voltages (if a little high) between the 65 and 140 volt Tp's and the Gnd test point. This voltage was coming directly from the transformer and is the 'reference' voltage we discussed the other day.
Actually, that is not the reference voltage.  From the +55 volts there's a resistor to a zener diode.  That diode is used as the reference voltage for the "error" amplifiers to adjust the power supplies.  Please review series regulated supplies a bit more. The "error" amplifier takes the difference between the reference voltage and the divided down output voltage.  That difference produces an error signal that controls the pass transistors to allow more or less current through (it's a big voltage divider, bulk supply, pass transistor, load).  That regulates the +55.  The +55 volts is used as a reference for all other supplies.  Get it right, and they're on the right voltage.



At this point there was zero voltage on the output of any of the rectified rails

Following that, I placed the lifted AC input wire back into its hole on the 15v rectifier, then turned the scope on, whilst observing the voltage between the 65v tp and Gnd, followed by the 140v test point and finally the 15v test point. Seeing as none of these voltage readings dropped, even at all, I soldered the AC wire back into the circuit.
I then repeated the procedure with the 5v rail, testing first, before resoldering the supply wire back into the circuit.
This was repeated with each of the remaing bridge rectifiers.

I did things this way as I had spent so much time, searching, locating and replacing various components on four of the boards, that when the scope was still absolutely 'dead' after powering it up, If I could not have found any life I would have walked away and given up.
At least now, I have one clear direction and area of circuitry to investigate.

Your input has been invaluable and I hope I can help you out in return someday?

If I am honest, I don't quite understand 'why' the scope was 100% lifeless with all rails and boards connected, yet full of life when each rail was reconnected in turn, but I am not going to complain.
Didn't you find and replace some capacitors on another board?

Here's what I'd do, and I strongly suggest that you do this and understand why you're doing this.  It can make a difference in how you repair the system.

Look at the power supply schematics.  If there's a fuse that can be removed in the secondary of the transformer to the bridge, do so for all supplies.  If there's a fuse in the output, remove it except for the +55/+110 supply.  The +55 supply is critical.  (in some scopes, it's the -50 volt supply).  The 15 volt supply will have to have a line lifted on the bridge.

Disconnect the main power supply from the rest of the scope.

Turning it on, and looking only at the +55 volts, what do you measure?

Look at the 110 volts, is that ok?

Fix those supplies, ignore the rest, they aren't on.

Add in other supplies one by one.  See what they read and can be adjusted to.

With the supplies adjusted properly, plug in other boards one by one, measuring the voltages at the supplies.  Any supply that doesn't work properly indicates a potential short on that supply line on the board you just plugged in.


Troubleshooting a supply like this is a big decision tree.  You go different places depending on the results of the previous test.  Regardless of the level of experience you have in troubleshooting, you're going to do the same kind of thing, perhaps faster, perhaps not.  The logic behind this whole thing is fairly constant in the world of power supplies, although as the supply architecture varies, so do the steps and what you must do. The logic, though, is what you started to do, but I think didn't complete.  Isolate and diagnose.


A design aid/troubleshooting aid is to print out the board layout, then color highlight the parts belonging to a circuit. You will find that it makes life easier in locating components, especially when the board is loaded with parts.

Annotate a copy of the schematic with the voltages you find, especially if they're the wrong ones.  Compare those voltages against the schematic.  If they're wrong, why?


Harvey



We are getting there.


Thanks to all who have helped.


James






Join TekScopes@groups.io to automatically receive all group messages.