Testing a Flyback Transformer (Leader LBO-302 Oscilloscope)

[bsfeechannel]

Don't lose hope. With the CRT disconnected, turn VR401 all the way up to +27V. This will cut Q406 off. The voltage on TP402 would be the next to zero. Now bring the scope up on the variac keeping an eye on R316.

With your sencore hooked up to TP401, and a meter to TP402, turn VR401 gradually until you see the waveform below.

[floobydust]

You disconnected the multiplier? Uh I wouldn't do it because the HV can shoot up and damage something. It really needs a load. I use old CCFL tube salvaged from LCD monitors with a ballast resistor. Or just a resistor if you have a few megs at a watt or so for short term use. Technically, the HV should have regulated due to feedback on R419, R420 etc.
The HV regulator Q406, Q402/Q403 I would check for shorted transistors which might supply too much base current to the flyback transistor Q401.

TP401 (on HV OSC board) tells you the health of the flyback oscillator, looking at current in Q401. If should be 1.7Vpk across R403 10R, so 0.17Apk. This will give some idea if R316 is getting too much. A small series light bulb i.e. nightlight 120VAC 7-10W might help if you need to run things for longer to make more measurements.

The only way I have killed HV multiplier is with arcs or partial discharges. The surge current of a 1,000pF cap dumping into a rectifier can usually kill it. That's why there are 22k resistors, to limit the current. Did it die, to only give 2V now?

I wouldn't blame the CRT just yet. It takes several seconds for the filaments to warm up, and during that "cold" time, the HV sees no load.

[Smoky]

Thanks for the help guys!

I want to get back into that transformer tonight because my first concern are the primary and secondary windings.

Let me do that first and check those resistors too.

What beats me up is that I don't own a transistor tester. I usually just take them off the board and do a basic diode test on them.

That's the next purchase for my toolbox

And on a side note, I'm not sure what the little neon lamps do on the main board but I noticed that they didn't exhibit any light or flickers either.

BSFeeChannel, I also have a restored Tektronix TDS420 that functions flawlessly since I rebuilt it last Fall. The scope does great with low voltages using the "zoom" feature.

[floobydust]

Try not to get struck on ideas, like it's the windings or something. If there is a crack or extra air-gap, poor fit loose screws in the fragile delicate ferrite core, it can cause problems like high current.
A waveform from TP301 will reveal a lot.

[Smoky]

BSFeeChannel, I can get no reading off of TP401. I tried every combination with CRT, w/o CRT, HV red wire from transformer connected, not connected, transformer connected to HV Oscillator board, not connected to HV Oscillator board, etc, etc...

And even using the DMM to read transistor Q401 on the HV Oscillator board I read maybe .83 volts on the base, .15 volts on the emitter, and .46 volts on the collector.

I was able to bring 27 volts to the base of Q406 too, just as requested, and still nothing. My scopes, as in multiples, could find something more emitting from my fingertips than TP401     I raised the Variac while turning VR401 too.

But this again, I think is because the rectifier circuit has been taken out and so things should be screwy?

After scratching my bald head, here's what I did after all of that testing madness:

I took a brand new rectifier module and installed it on the transformer. I hooked everything back up, the CRT, the Flyback transformer, the HV Oscillator board, everything but...

...the HV red wire. I attached it to the Sencore SC61.

I slowly brought up the Variac and watched the digital display on the Sencore.  The voltage reading climbed to about -1650 volts and it remained there for about 20 seconds, and that's when the Sencore display dropped to zero.

Does this indicate the current going into the rectifier circuit is above the threshold of the 2CL70 and DL800 diodes or resistors?

Anyway, this is fun stuff to me, and I want this to be the third oscilloscope that is fixed in my house this year

And to top it off, and really blows my mind, is that the original fuse on the back of the scope is still good!

And another observation while testing, resistor R404 on the HV Oscillator board, on the schematics it reads it as a 820 Ohm resistor, but what is installed is a 470 Ohm resistor.

[Smoky]

And here is my trivia question for tonight:

Can a 2N3739 NPN transistor replace a 2SC1059 (Q401) NPN transistor?



Also, reading a little more about transistors, I came across several articles about 2SC458's, and their failure rate. These would be used as Q402 and Q403 in the LBO-302. They say that they can fail in circuit but test good on a transistor tester.

Transistor BC549 was recommended as their replacements.

Is anyone familiar about that? And would they work as replacements in this scope?

Nothing like reading and modifying my posts in bed when I should be sleeping

[bsfeechannel]

And here is my trivia question for tonight:

Can a 2N3739 NPN transistor replace a 2SC1059 (Q401) NPN transistor?

Yes. According to these guys.

[floobydust]

The mysterious 20 second delay, is it due to the CRT warming up and then presenting a load?
I know the anode connection (red wire) is off, but the other CRT grids are there and can be a load. I only have the LBO-301 schematic and it's clipped off, missing the CRT portion, might be the pdf viewer in Firefox. You can get unwanted screen-grid current when the anode is not catching electrons.
Or something else heats up after 20 seconds.

My rules- don't run it open-loop (no voltage reg feedback), don't run it not connected to any load. Just to avoid damaging the insulation on something with HV being too high.

You could isolate it by putting a dummy load on the (flyback) CRT filament winding 75AKB1 CRT 6.3VAC 600mA, so a 10R load. Or winding a few turns on the core and using that to give it some load, some work to do.

These one transistor oscillators can be finicky. This is a strange repair where the oscillator would not start: https://www.eevblog.com/forum/repair/tek-model-464-repair/msg1704728/#msg1704728

P.S. Found the press release for LBO-302 in March 1974 $700 ($3,640 today). LBO-301 was 1971.

[Smoky]

Thank you guys!

FloobyDust, the best I could do about the CRT schematic was to take several pictures of it. You'll need to piece them together.

Over the next few weeks, I'm going to acquire new transistors and resistors. Some I have, some are obscure and would need some alternative. But before any new parts are installed, I'm going to check all of the resistor values on the boards, do tests on diodes, etc.

But what adds frustration is finding different value resistors and capacitors on the schematic than what's installed.

I can only imagine, at the factory, there were points on each board where voltages and currents needed to be, they pass or fail. The schematics show voltages, but when components don't match the schematics, then what?

So when I read -1650 volts, that was without a load. The tolerance on the schematic is +/- 100 volts. I'm sure the CRT would have brought it down and the VR would fine tune that number.

Is it possible to bring those test point voltages to service manual specs then work the voltages back up when the loads are applied?

Anyway, this project will be kept on the little bench that it rests on for however long.

...and if you have any parts or boards to this machine, I'll buy 'em

[floobydust]

I'll bet there's nothing wrong with the HV power supply now. Just a theory in my head I'm liking.
The problem is the CRT is overloading it, once warmed up. This is the 20 second delay.
Vacuum tubes are normally full-on devices (once the cathode is warmed up) like a depletion-mode FET. They need a negative-bias voltage from cathode wrt control grid to lessen or cutoff current. This is what the intensity control is doing, as well as blanking pulse.
The control-grid CRT pin 7 or maybe screen-grid could be acting as the anode, due to the absence of grid-bias or the control grid is stuck near 0V.

The only hole in my theory is the HV DC should not go so low to 2V, the CRT could take it down to 100V min. as a guess.
Can you test Q405. Shorted C408 would light up the two NE-2 neon lights.
I will stare at the schematics more and see if I am out to lunch or this is a possibility.

[bsfeechannel]

I can only imagine, at the factory, there were points on each board where voltages and currents needed to be, they pass or fail. The schematics show voltages, but when components don't match the schematics, then what?

So when I read -1650 volts, that was without a load. The tolerance on the schematic is +/- 100 volts. I'm sure the CRT would have brought it down and the VR would fine tune that number.

The CRT is irrelevant for the HV supply because the supply is regulated. And we regulate a power supply to make its output voltage independent of the load.  Besides, the CRT is "absent" anyway for the first half a minute, every time you turn the scope on. So disconnect it. Preferably remove it from the scope and keep it in a safe place. It is one variable less. And an unobtainium one.

Now, instead of changing out every component, let's troubleshoot. It's fun, I promise.

When you bring 27 volts to the base of Q406, you cut it off. The oscillator should stop. Q401 should be also cut off and R316 should be as cool as a cucumber in this condition. If it is not, there's something fishy going on. So let's investigate.

With Q406 cut off, you should read around zero volts on its collector, which is connected to the base of Q403.

Q403, and Q402 will also be cut off. This means that you should read next to zero volts at the base of Q401.

Now, please, turn VR401 to 27V, bring the scope up until bearable by R316. Then, please, measure these voltages and report them to us.

Q406 Vb, Vc
Q403 Ve, Vc
Q402 Vb, Ve
Q401 Vb, Vc, Ve

This is what I expect to see

Q406 Vb = 27V, Vc = 0V (or less)
Q403 Ve = 0V, Vc = 27V
Q402 Vb = 0V, Ve = 0V
Q401 Vb = 0, Vc = whatever is there on TP-301, Ve = 0V

The aim here is to see if there's something wrong with the reference Q406, the control circuit Q403 and Q402, and the oscillator Q401. If nothing's wrong, rule them out. If there's something strange somewhere, watch it under a magnifying lens.

[Smoky]

Here's what I read from the transistors with everything connected and Q406 having +27 volts at the base:

Q406

B +27.1
E +26.9
C -16.74

Q403

B  -.13
E +.38
C +14.69

Q402

B +.318
E  -.48
C  +14.69

Q401

B -.55
E +.98
C +189.5

TP 301 +185.8

Measurements were taken twice.

Everything is connected BSFeeChannel because I just took readings with everything connected after FloobyDust made some comments. So FloobyDust, my reply to you is next...

[Smoky]

FloobyDust, after you made your comments, I connected everything back up on the scope.

I wanted to go through all of the tests points with everything active. I toggled through the Variac to come up with these results:

*Test point TP402 -1056 volts

*Test point TP301 +136  (it should be +150, it's now low)

*Test point TP302 +118.6 (it should be +130, it's now low)

HV Red wire entering main board -1118 volts.

The unregulated test points TP301 and TP302 are now low, before, these test points were ~30 volts high.

I think things are improving.

...and here is a shot of the scope from test point TP401 on the HV Oscillator board:



Q401

B +.7
E +1.29
C +130

So something is drawing-down the voltage to the transformer.

No adjustment to VR401 will raise the test point TP402 any higher than -1056 (it should be -1500v at this test point).

BSFeeChannel, things are strange

[bsfeechannel]

Here's what I read from the transistors with everything connected and Q406 having +27 volts at the base:

Q406

B +27.1
E +26.9
C -16.74

Q403

B  -.13
E +.38
C +14.69

Q402

B +.318
E  -.48
C  +14.69

Q401

B -.55
E +.98
C +189.5

TP 301 +185.8

Measurements were taken twice.

Wonderful. Now we're getting somewhere. I forgot that R402 and R429 form a voltage divider, so the voltage at the collector of Q403 and Q402 should be 14.7V, which, your measurements show, are spot-on.

There's a discrepancy is the voltage at the collector of Q406, which your measurements show -16.74V and the base of Q403, which is at -0.13V. They should be exactly at the same voltage, since they are directly connected.

Another problem is that the emitter of Q402 is negative, -0.48V. There's no path for negative voltage there. Same thing with Q401, which has a negative voltage at its base, -0.55V, and a positive voltage at its emitter (+0.98V).

Without changing anything, see if you have some waveform at TP-401 and report to us what you see, please. After you check that, you're going to investigate the suspect voltages.

EDIT: Since you edited your post and showed TP-401 with an oscillation, now it is time to investigate why the collector of Q406 and the base of Q403 do not have the same voltage. Please do a visual inspection on the board. See if there is any cracked solder joint, raised tracks, or shorts. Measure the resistance between the two terminals.

Everything is connected BSFeeChannel because I just took readings with everything connected after FloobyDust made some comments. So FloobyDust, my reply to you is next...

That's OK. It won't change our investigation. Leave it the way it is.

[Smoky]

I apologize, I didn't realize that I had to wait for the numbers to climb when I took the initial voltage tests. Some of the transistors took longer than others to top-off.

So I decorated resistor R316 like a Christmas Tree with heat sinks and waited for the voltage numbers to come up and stabilize on the transistors.

I didn't do this in one swoop, so slight deviations can appear due to bringing the Variac up several times. It's not a digital one.

Q406

B +27.1
E +26.9
C -16.74

Q403

B  -16.74
E  -8.8
C +14.69

Q402

B  -8.8
E  -.6
C  +14.69

Q401

B  -.7
E +1.0
C +189.9

*TP 301 +149.9

TP301 is almost spot on at +150

TP302 is also close to spot on at 130.9

[bsfeechannel]

I apologize, I didn't realize that I had to wait for the numbers to climb when I took the initial voltage tests. Some of the transistors took longer than others to top-off.

No worries.

So I decorated resistor R316 like a Christmas Tree with heat sinks and waited for the voltage numbers to come up and stabilize on the transistors.

R316 shouldn't be overheating. Without changing a thing, measure the amplitude and frequency of the waveform at TP-401, please.

I didn't do this in one swoop, so slight deviations can appear due to bringing the Variac up several times. It's not a digital one.

Q406

B +27.1
E +26.9
C -16.74

Q403

B  -16.74
E  -8.8
C +14.69

Q402

B  -8.8
E  -.6
C  +14.69

Q401

B  -.7
E +1.0
C +189.9

*TP 301 +149.9

TP301 is almost spot on at +150

TP302 is also close to spot on at 130.9

The base-emitter junctions of Q402 and Q403 are in reverse breakdown. This means a current is flowing from their emitters to their bases, not the other way around. The only DC path for that current to flow would be through the base of Q401. But that current would have to flow from its emitter to its base, which would cut it off. You'd have no oscillation, no negative voltage on TP-402 and consequently no negative voltage at the base of Q402.

So we need to discover what is providing the path for the reverse current at the base-emitter junctions of Q402 and Q403. If C405 or C404 be leaky, we have a problem. So let's check them for leakage.

[Smoky]

Thank you for your generous help BSFeeChannel!

And yes, R316 gets very hot even with +27 volts on the base of Q406.

I connected a TDS420 to TP401 for the measurements with the Variac turned up:



I will update again tonight after I test the capacitors and transistors.

*C404 tests good.

*C405 tests good.

[floobydust]

Um 197Hz     so the osc. stops and sits there and cooks? I see AC mains ripple as the only modulation there in that strange TDS420 scope trace.

I'm a little confused- when it's stalled, at TP401 you are getting steady -200mV (inverted?) with around ~80mVpk of AC?
I can't make out the units on the Tek222A snapshot, I'm assuming now that's when it's working, running at say 30kHz.
C403 is suspect, test at voltage on the Sencore, and the 150V rail filter cap 50uF 250V C302 is already small but the 150V rail should have less ripple at 20mA (-200mV/10R).

[Smoky]

HV Oscillator board:



C402 and C403:



I had to verify the accuracy of the Sencore Z Meter with my reference caps to confirm what I found and it is spot on:



1.1pf on one cap, .3pf on the other! These caps should measure .01uf or 10,000pf.

FLoobyDust, is this what was taken out when the -1650 volts went to zero?

If so, what other collateral damage could there be?

*I just attached the HV red and the black ground wire coming from the Flyback transformer to the Z meter, it reads only 2180pf. I tend to believe a capacitor was taken out in the rectifier module.

[bsfeechannel]

Thank you for your generous help BSFeeChannel!

My pleasure.

And yes, R316 gets very hot even with +27 volts on the base of Q406.

Certainly, something is drawing too much current for this resitor to overheat. Please measure again the voltage on Q401 and then on both leads of R316 like in the picture below (I'm using lbo-301's schematic).

I connected a TDS420 to TP401 for the measurements with the Variac turned up:

As floobydust pointed out, this screen capture is confusing. It is not possible to conciliate it with the measurement you took with the meter +0.98V.



The frequency is ridiculously low (around 200Hz), when it should be in the range of 30 to 40KHz.

Can you please clarify what is in the screen?

I will update again tonight after I test the capacitors and transistors.

*C404 tests good.

*C405 tests good.

So, please, turn the scope off. Wait for the capacitors to discharge, then measure the resistance as indicated in the figure below. Both ways, I mean, since the ohmmeter has polarity. We need to know if at that point there is a path for a current to flow both ways.



After that, test Q401 and C405.

[floobydust]

What I think is going on now, the oscillator is just weak and stops, leaving Q401 on and R316 cooks?

C403, C402, C404 are critical for the flyback oscillator tuning and gain. If C403 is open, that is a problem. I would say their failure is not from anything going on here, for a wound film cap to go open, they must have corrosion inside and died of old age. I'm really glad we found something.
If you replace C403, C402 that will change everything I think, as far as it staying running and not cooking up R316, unless the multiplier has croaked.

The second issue would be what happened to the new HV multiplier? Is it alive or did it go open?
What part number are the 1,000pF caps, they look smallish and can they take 4kV? With no load they would see more.

[Smoky]

Here's what I have so far:

Q401 tests good in circuit (diode test).

C404 and C405 test good with no leakage.

Resistance across C404 and C405 Is over-limit (OL) nothing.

FloobyDust, those 1000pf caps in the rectifier circuit were and are 3kv

[Smoky]

BSFeeChannel, you can see things are still apart.

Should the film caps C402 and C403 be replaced before the scope is fired up again?

I'm searching now for a 600+ volts .01 caps as I type.

As it stands at the moment, the CRT is out because it shrouded access to Q406, Q403, Q402, C405, and C401.

[Smoky]

Here's something else I noticed.

Q402 and Q403 should be 2SC458 B or C's by the schematic.

What's installed are two 2SC979's.

[floobydust]

Should the film caps C402 and C403 be replaced before the scope is fired up again?
I'm searching now for a 600+ volts .01 caps as I type.

I would not run it with open capacitors, they also tame the leakage inductance of the flyback transformer and we don't want to make Tesla coil here.
Stay with a metal film cap 0.01uF 630V +/-20% (service manual) like Kemet R73 PP with long leads Digikey 399-12529-ND. The same part is in a few other locations in the scope, to keep in mind if it is a high failure rate item.