Oscilloscope drawing lots of current and blowing fuses

[gkmaia]

I got an old analog oscilloscope that works perfectly well, all functions, rails, voltages are sharp and fine, but it slowly raises current till blows Fuse 501.

The power supply can be feed by an external 220v-20v transformer or directly with 12VDC.

When feed with DC I can monitor the current and it starts at 1.5amps and climbs all the way to 2.5amps after 20 to 30 min starts a hissing noise then blows fuse F501.

If I reduce the crt voltage from 1.5Kv to 1.35Kv it lasts much longer before blowing the fuse. But if I bring the crt voltage to 1.5Kv it blows a fuse in minutes.

I am suspecting the second transformer L501 has some winding short or issue but it also provides all voltages perfectly.

Is there a way to precisely diagnose if transformer is actually faulty?

[xavier60]

There could be a problem with the buck regulator gradually losing efficiency. If so, there might be a changing in the waveform to see.
It would be useful to rule out most of the output rails by looking for change in voltage drop across the filter chokes.

[gkmaia]

Thanks for you reply xavier.

I toke some measurements on all secondaries and the table bellow shows what I got.

My CRT main drops from 1500v to 1350v as the current consumption raises.

My 5v rail is a bit above at 6.5. And my 10v rails are uneven. But I guess the big issues is the voltage drop on the CRT main.

What you suggest next?

[L_Euler]

Try shooting some of those transistors with freeze spray and see if current drops.

[xavier60]

 The time taken for the current rise does seem to relate to something rising in temperature.
There is a lot happening with those voltages. Most of the changes correspond with the change in GND voltage.
The presence of the GND voltage isn't right. There could be a poor connection between the 2 grounds or excessive current flow that likely should not exist. This needs to be looked into. A current measurement can be taken between the 2 GND points.

The CRT Main voltage regulator uses +10V for its reference, so the change of CRT Main voltage could be just because of a change of the +10V.
Also check all bypass capacitors for excessive ripple voltage.
Is Q510 getting very hot?

What is IC 501?

[xavier60]

CRT Main voltage could be disabled by removing D516 and D517 to see what effect it has.

[shakalnokturn]

I'd follow xavier60's advice on this it may give you a slight clue.
The rare cases I've had with similar symptoms were related to the HV winding of the transformer, and with not much measurable effect other than excessive current draw.
One definite thing to watch is the transformer's temperature, if the transformer itself is the cause (weak insulation) the extra power consumed is consumed in the transformer itself.
I'd bet by the time your fuse blows (D512 and D517 removed or not...) your transformer is too hot to hold.

[gkmaia]

Had to build a bunch of jumper cables to be able to flip the power supply and have access to the transformer.

And yes I will follow xavier's advice and:

- check Q510 temperature
- check transformer temperature
- check everything else temperature wise around that block

A couple questions:
- best way to check what is leaking to ground?

In regards to disabling D516/D517 that block goes to wire 2 and 5 that goes to focus. CRT main 1.5k voltage comes from P502.

[xavier60]

- best way to check what is leaking to ground?

That could just be a poor connection. Check the resistance between the 2 grounds first with power off.

[jdragoset]

If you can disconnect both leads from the L 501 winding feeding the D 516/D 517 HV doubler with all else connected and the current does not increase with time, that would point to a .01 capacitor in the doubler cicuit breaking down, or other HV wiring/terminal leakage, look for carbon burn-through/tracks along all HV wire/terminal insulation.
If it were the diodes, I would not expect a gradual change.

[bob91343]

My initial diagnosis is bad electrolytic capacitors in the power supply.  You can qualitatively evaluate that by feeling the capacitors to see if they are warming.  A power supply capacitor should not generate palpable heat.

If they leak, they warm up and that increases leakage.  Thermal runaway.

[bsfeechannel]

What is the model of this oscilloscope?

[gkmaia]

Ok, some updates.

Did measure temperatures with infrared thermometer. The transformer L501 gets hot with time but it is tricky to capture the temperature. It only gives a high temperature if I aim the infrared in the winding gaps. Anywhere else temperature is normal 35-40ºC.

The voltage regulator IC 501 gets up to 60ºC and he is the one supplying current to L501 together with Q510. But Q510 does not heat up. 

Haven't checked the caps. But removed dozens of caps from this scope and they were all Elan and all still in pretty good condition. So not sure I can blame the caps.

[xavier60]

The transformer temperature doesn't look good. Next would be to run it with all loads disconnected if not already. Measure the current draw also.

[gkmaia]

ok, just to confirm I will disconnect all wires marked in the attached image.

It will draw way less current. But I imagine what you want to see is if after a few minutes the current draw start to raise?! Right?

[xavier60]

ok, just to confirm I will disconnect all wires marked in the attached image.

It will draw way less current. But I imagine what you want to see is if after a few minutes the current draw start to raise?! Right?

Yes, and the transformer temperature should rise if it has shorted turns. Also check the continuity of the HV winding.
What is the current draw?

[gkmaia]

I had a bit of a go. No load at all. Temperature of all components including transformer at 35-40ºC to 20min and current at 0.240A all time.

First 15 min current was between 0.238 and 0.240.

Last 5 min current was fast moving between a 0.225 and 0.255. But temperature was stable.

I could try for a bit longer tomorrow. As it is late already.

[xavier60]

As well as checking the resistance of the HV winding, check for leakage to GND, preferably with an insulation tester.

 
Also check for leakage to the winding that connects to P502.

[duak]

Are the temperatures of the main switching transistors Q511 & Q512 about the same as each other?  If one has become leaky at high temperatures or its beta falls off, it may unbalance the transformer primary currents and cause heating.  I wonder if these transistors are germanium rather than silicon?  If so, they are far more prone to aging.  The diodes D510 & D511 could also be leaky or high resistance.  I would compare the voltages and waveforms on the collectors and bases of those two transistors.

[bsfeechannel]

Yes, and the transformer temperature should rise if it has shorted turns. Also check the continuity of the HV winding.
What is the current draw?

A single shorted turn in a transformer always leads to a catastrophic failure. All other windings will be seen by their respective circuits as short circuits.

If the power supply is capable of producing the nominal voltages, there are no shorted turns.

The overheating of the transformer may indicate some overload.

Since the OP didn't deign to declare the model of the scope, we cannot determine its age. Nor have a more comprehensive view of the schematic. But since he said its old and analog, I'll suppose its more than 30 years old. In that case, the usual suspects are capacitors. All of them. Especially the high voltage ones. Not only the electrolytic. This followed by transistors and diodes. Resistors should be checked for drifted values.

[gkmaia]

Yes, and the transformer temperature should rise if it has shorted turns. Also check the continuity of the HV winding.
What is the current draw?

A single shorted turn in a transformer always leads to a catastrophic failure. All other windings will be seen by their respective circuits as short circuits.

If the power supply is capable of producing the nominal voltages, there are no shorted turns.

The overheating of the transformer may indicate some overload.

Since the OP didn't deign to declare the model of the scope, we cannot determine its age. Nor have a more comprehensive view of the schematic. But since he said its old and analog, I'll suppose its more than 30 years old. In that case, the usual suspects are capacitors. All of them. Especially the high voltage ones. Not only the electrolytic. This followed by transistors and diodes. Resistors should be checked for drifted values.

It is a Trio CS1352. I was trying to find a way to upload the manual I bought as it is not available for free. Just having the model wouldn't be of any help as you would not be able to find an SM to download.

SM is here:
https://elektrotanya.com/trio_cs1352_sm_en.pdf/download.html

All bypass capacitors in the power supply board have been replaced except by a couple I do not have the values here. Otherwise all new. See image.

Are the temperatures of the main switching transistors Q511 & Q512 about the same as each other?  If one has become leaky at high temperatures or its beta falls off, it may unbalance the transformer primary currents and cause heating.  I wonder if these transistors are germanium rather than silicon?  If so, they are far more prone to aging.  The diodes D510 & D511 could also be leaky or high resistance.  I would compare the voltages and waveforms on the collectors and bases of those two transistors.

Have a look at the image attached you will see they are the same. They are silicon. I will try to put them on my curve tracer.

As well as checking the resistance of the HV winding, check for leakage to GND, preferably with an insulation tester.
Also check for leakage to the winding that connects to P502.

I think we can exclude the ground issue. My scope was not properly set to lower voltages. With it set the ground is in the order of 1-5 mV.
I will check the resistance of the HV winding.

[xavier60]

I should be calling it "EHT" winding.
If there is a problem with the transformer, it will likely involve the EHT winding. It could be partly open or arcing over to an adjacent winding.

This winding was troublesome in oscilloscopes that had the EHT winding on the mains transformer. The DC imposed on it by the doubler would cause accelerated corrosion.

[gkmaia]

There you go.

Got the transistors on a curve tracer see images attached. They seem fine.

Also tested the resistance of the EHT winding and turned out to be 0.136 Ohms. I believe there is something wrong with it (see image).

To get 0.136 ohms on a wire with a cross section of 0.15mm (~35G) would need just 15 cm of wire. That is probably 5 turns. But lets say the wire is 0.25mm thick that would need 40cm in length to build such resistance.

[xavier60]

Also tested the resistance of the EHT winding and turned out to be 0.136 Ohms. I believe there is something wrong with it (see image).

With a resistance that low, it would have been producing very little voltage.  Are you certain that the correct winding is being checked?
Be certain to check for possible leakage to other windings.

[gkmaia]

Pretty sure that is the one. See the attached schematics.  I could try measuring it with another multimeter, but the Fluke is precise. And I do agree with you, that length of wire would not be able to generate that voltage. Quite confusing...