Zenith 9" color TV 1986
 

I got this and it worked fine the first time I turned it on, it wasn't hooked to anything so I just got snow and static - but it looked good. So I unplugged it and went to another room to plug it into a VCR, and now all it does is come on for a second or two - not long enough to get a picture - then shut itself off.

I'm not familiar with stuff this modern, but I suspect most likely problem is something like some sort of circuit breaker is tripping due to bad electrolytics in the power supply. But I can't figure out the chassis number to get a schematic, so I can actually locate stuff. Model Number = C0920U, main board = 9-468

there is a sam's?

https://samswebsite.com/en/photofact...x/model/CO920U

Thanks. It looks like that's an 'O' and not a '0' - that's probably why I couldn't come up with anything.

A very common issue with those sets were bad solder connections on the horizontal output transistor and the five legged voltage regulator STR mounted next to it on the right side heatsink.

Tried resoldering this stuff, wasn't it ----bad solder connections on the horizontal output transistor and the five legged voltage regulator STR mounted next to it on the right side heatsink

I have the sams now, got the whole folder, but there's no way to test for anything, to check voltages, to narrow down where the problem might be -- because the TV shuts itself off after a few seconds. And because everything's mounted on a board - I can't just disconnect wires to test capacitors, I actually have to remove capacitors to test them - and at that point you might as well replace them even if they are still good.

Start by replacing e-caps in the power supply? Or is there a better strategy? I'm looking at the power supply schematic - I really don't even see what it uses to shut itself off. I've always kind of wanted one of these, but this reminds me why I've always preferred tube stuff...

Everything worked fine the first time I turned it on. After that it always shut itself off after a few seconds -- never stays on long enough for me to hear it developing HV, no raster, but if I let it sit long enough in-between times trying to turn it on (hours not minutes) it will stay on for a second or two, long enough for me to hear static -- If I try to turn it on several times in a row, the time it stays on shortens to the point where it shuts it self off immediately - this might point to the behavior of a leaky cap in the power supply?

Run it on 90 V with a variac for a few secs at first. Be sure the HOT isnt getting hot. Sounds like HVSD either caused by bad load on HV / Hoz
OR power supply isnt regulating. Get clues first, otherwise it may
not work again.
Zeno

Heathkit version of the 9" Zenith set
 

Not sure if you're still interested in debugging this problem from last year, but I discovered this thread because I'm debugging a similar issue.

There are many variations of this Zenith set, with 3 major versions:
1. AC Only model
2. AC/DC model
3. AC Model with Space Command remote.
There are even more model numbers, some just for cabinet color, but there also appear to be revisions in the chassis between models. I have 2 versions of service info, and can see minor design revisions and layout changes.
This page has some info and a pointer to Sams data on some versions:
https://crtdatabase.com/crts/zenith/zenith-e0930s They don't even mention your model number.

This Zenith set was also sold in kit form by Heathkit, model GR-9009. I have one of those. It's the AC/DC version. Heathkit supplied their own schematic and service info, and I also downloaded the Sam's data for model D0930S from the above website. As I worked, I discovered differences. Some parts (resistors, etc.) have different numbers, and one transformer on my main board is rotated 90 degrees from the pictures in the Sam's. Heathkit provided their own board layouts that do match my set. The different revisions appear to be functionally very similar, but beware that what I describe may not match your set exactly.

My set has exhibited 2 problems:
1. The picture periodically fades to black (brightness drops).
2. Sometimes, when powered on, power only stays on for about 2 seconds, then drops. I've also had it spontaneously turn off by itself later, but only a couple of times.

These symptoms are without the space command remote control, which controls power differently. My set just has a mechanical on-off switch. My DC/AC inverter shouldn't matter, it's not used when running off AC.

Sams provides some debug tips, and Heathkit supplies others. Heathkit also supplied a helpful theory of operation.

First, note the set has 2 grounds. Parts of the chassis are hot (voltage present), so be careful working on it. You should use an isolation transformer for safety. The large metal parts on the left side of the set near the HV lead are on the AC ground (line voltage), other metal parts (tuner, antenna bracket, etc.) are on DC ground, which is separate from the AC ground. When measure voltages, measure with reference to the correct ground (DC ground for most of the low voltage areas).

First, line voltage is rectified and filtered to 150V (C3249, 470uF 200V zip tied to the heat sink under the CRT neck). This is then regulated down to 123 volts on CX3205 (both referenced to the AC ground). These may be live even if the rest of the set fails to come on. Sams points out that if the rest of the set is off, the voltage on CX3205 will be higher, because the regulator needs some load on it to regulate properly. Without that, it will be closer to the raw rectified and filtered line voltage. Lower voltages for the rest of the set are derived from the horizontal transformer that gets its power from here through the horizontal ouptut transistor. and transformers TX3205 and TX3204 (horizontal output, flyback, and low voltage power). This is essentially a switching power supply for the lower voltages. Without them, the set appears to be dead.

To provide low voltage power, a square wave is input to transformer TX3205. This drives the horizontal output transistor on and off, providing input to the main horizontal/flyback/power transformer TX3204.

Horizontal driver transformer TX3205 gets its input from two different sources, one for startup, and one for continued operation. This is described by Heathkit:

Horizontal Start-up Circuit:

Transistors Q3224 and Q3213, capacitors C3213 and C3224, and resistors R3212 , R3213, and R3215 are connected as an astable multivibrator. The output of this circuit, at the collector of Q3224, is connected through diode CR3216 to the primary winding of the horizontal driver transformer TX3205. When the TV set is turned on, the horizontal sweep transformer provides the isolated power supply voltages so the regular horizontal drive circuit can take over the function of the multivibrator. The start-up multivibrator is controlled by a two-transistor circuit consisting of transistors Q3211 and Q3210. This circuit detects the rising 123-volt DC power supply voltage at the time the TV set is turned on, which then starts the multivibrator. After a time delay, this circuit turns the multivibrator off.

Horizontal Drive:

The horizontal drive signal is provided at pin 37 of IC2376. This signal is a square wave with a 50% duty cycle at the horizontal scan frequency of 15.734 kHz. The positive portion of the square-wave signal switches on transistor Q3209, which inverts the signal. The positive portion of this inverted signal switches on transistor QX3202, which again inverts the square-wave signal. The drive signal is then applied to the base of horizontal driver transistor Q3206. As this transistor switches on or off, its collector voltage goes low or high at the horizontal scan rate. This varying voltage is transformer-coupled through TX3205 to produce a high-current / low-voltage driver signal to switch horizontal output transistor QX3208 on or off.

Got that?

Basically, when the set is running, the horizontal transformer has to run at exactly 15.734 kHz, synced to the receive signal. This is controlled by IC2376. But IC2376 runs on 18 volts provided by rectifying the voltage off the power supply winding on the horizontal output transformer. That voltage is present only when the horizontal output is switching. The set needs some means to get the 18 volt supply turned on in the beginning to get the horizontal oscillator in IC2376 going. That's the job of the startup circuit. Once the set is running, though, the startup circuit needs to get out of the way, because it's not providing a precise 15.734 kHz signal synced to the broadcast signal. If it kept providing input, the picture sync would be all messed up.

Once up and running, AC power for the set continues to flow in through the horizontal output transistor, which powers the low voltage supplies, plus the horizontal drive, plus the high voltage. That's why it's a big power transistor mounted on a heat sink, it's really passing all the power for the entire set. It's both horizontal output, and the power switching for the low voltage switching power supply.

So. If your set powers up briefly, the initial rectifiers and regulators must be working, and providing the first 150V, and regulated 123V power to the horizontal output transistor, and also to the startup circuit. The horizontal output transistor must also be working, as are the horizontal driver and output transformers. All that has to work to provide startup power, so the front panel LED display can come on, and you may hear sound from the speaker, and high voltage to the CRT may also come up. IC2376 should also be powered up and start providing the 15.734 kHz horizontal frequency.

Then, the startup circuit shuts down. If horizontal frequency from IC2376 is missing or doesn't make it through the 3 transistors and transformer to the horizontal output transistor, the horizontal output transformer will stop providing power to the rest of the set, and everything will appear to shut down, though the 150V and 123V from line voltage may still be present.

That's what I thought was happening with my set. I was able to scope the output of IC2376 during the short time power was on, and it looked correct: This shows the horizontal signal at the output IC3276 during the brief time the set is on. It closely matches the waveform shown by both Sams and Heathkit's documentation. I was also trying to check voltage with the meter in the background, but power dropped before it stabilized. I was able to capture this brief horizontal output using triggered sweep.

https://videokarma.org/attachment.ph...1&d=1774761996

The mystery is when it loses power, what happens first? Does IC2376 lose power, or does the horizontal output stop first? I note the scope frequency reading is a bit low, but maybe this is because I sampled right at startup? I wanted to probe the path after this to see how far the signal got, but I wasn't able to probe the transistors after this with the set assembled (the inverter for 12VDC operation sits on top of the chassis, making it even harder to reach to probe), so I took it apart, unsoldered Q3209, Q3202, and Q3206, the 3 transistors in the path, and a couple of capacitors. They all tested OK out of circuit, along with resistors and the TX3205 primary tested in-circuit. Everything after this needed to work for startup to work, so I assume it's good. I reassembled. I'm kind of stumped. It seems like it should work, but it still didn't, at least not at first.

Eventually, the set reverted to its other problem. Power stayed up, but brightness kept intermittently dropping, sometimes all the way to black. I tried to troubleshoot this. I checked voltages to IC2376, 12V and 175V to the video output board plugged on the CRT, and also checked the voltages from the black level and picture (brightness) control. All were correct and stable while the brightness was varying. Here's a couple of traces. This one is while it's working correctly:

https://videokarma.org/attachment.ph...1&d=1775277598

Blue is composite video going into IC2376, yellow is the IC2376 output that's driving the red gun of the CRT. (IC2376 is the chroma, luma, sync, vertical and horizontal processor.) Zero volts for both are at the bottom edge of the screen. You can see that the base of the sync pulse, and the base of the red drive (with blanking applied) line up. Blanking keeps the drive down during both the horizontal sync and back porch. When working correctly, the yellow (red drive) is higher than the corresponding composite input.
Here's a view when the picture has faded to near black:

https://videokarma.org/attachment.ph...1&d=1774762308

The red drive black level and blanking level is in exactly the same place, but the video during the scan line has fallen to about the same level as the composite input signal. It looks like it's lost amplification, or is being attenuated. But the voltage on the black level/brightness input to the IC is exactly the same (video doesn't pass through the black level/picture (brightness) controls, they simply provide an adjustable 1.8V to 2.2 constant bias voltage to the IC). Not varying at all. This tells me the problem must be inside IC2376.

IC2376 might be dropping the horizontal output at times. This might explain why power drops. But, I haven't been able to prove this. That's harder to capture on my scopes (1 digital, 1 analog). The cheap digital scope fails on this due to aliasing: If I decrease the sweep speed sufficiently to be able to watch the power drop, the horizontal signal isn't sampled fast enough. Aliasing causes it to disappear from the scope display. I can't tell if the oscillator is still running. Here I tried dropping power at a slow sweep speed:

https://videokarma.org/attachment.ph...1&d=1774762308

Blue is the 18V supply to IC2108. Yellow is the horizontal oscillator signal. You can't see the oscillation at all due to the 100ms/div sampling rate. I got the trace by turning the power switch off, because it wasn't exhibiting the power up drop problem at the time. Power seems to come back on, I think this is because of switch bounce in the mechanical on-off switch. I can see this because with the proper triggering mode, the scope holds the signal for display.

I also have an analog scope. That doesn't have the aliasing problem inherent in digital scopes. I can slow the sweep speed and the horizontal oscillator doesn't disappear, it paints the screen with solid green over the peak to peak signal range. I can set it to single sweep, and trigger on the power drop on the 2nd channel. But, there's no storage, so it's hard to see the transient. If I'm lucky enough to trigger on a power up power loss, it's only on the screen momentarily, then gone. I'll try this if the TV exhibits the power problem, but it may be hard to capture.

Edited to add: I think I have a better idea to capture this on my scope: I'll use the digital one, and set the horizontal rate to be able to see the horizontal signal from the IC, and connect the other channel to the 18V power to the IC. This is probably too fast to see any slope in the 18 V if it's dropping, but if I trigger on the horizontal in normal (not auto) mode, I think it will keep tracing until the horizontal stops. The other channel will show the power voltage at that time. If it's below 18V, the power must be dropping first, or if it's still 18V, the horizontal signal is failing first. I just need the set to cooperate. Right now, the power dropping problem isn't happening. Instead, I'm seeing the dimming picture problem.

Conclusion:

I think my IC2376 is bad. I'm pretty sure it's causing my intermittent dropping brightness problem, and it might also be intermittently dropping the horizontal out, which would cause power to drop. I haven't proven that, but since IC2376 is already suspect, and everything else checks out OK, that might be it.

Your set may also be powering up on the startup circuit but the horizontal oscillator may not keep it going. That could be IC2376, or QX3209, QX3202, or QX3206, or associated resistors and capacitors. If your channel display LED lights briefly, the rest is probably good. It gets 5V power that's derived from the 18V supply that also powers IC2376. The only difference you noted is the length of time power stays on varies? Mine seems pretty consistent at about 2 seconds whenever it's acting up.

I don't think the set even has a high voltage shutdown circuit (for over-voltage), and I don't think bringing it up slowly on a variac is going to work well due to the switching supply and startup circuit.

Zenith's and Heathkit's fix is to replace the entire module (main board). I haven't been able to locate one of them, or the IC yet. The only possibility may be a parts set.

This is taking much more time than I expected. I'm going this far only because I built this set new back in the 1980s, and I'd really like to see it working again.