Dim oscilloscope screen

[yugu]

Hi guys. I bought a $50 analog oscilloscope from ebay (I'm a fan of Dave  ). It's Instek GOS-635G. It works fine except that the screen is quite dim. When I turn the intensity knob up from the min position, the screen doesn't show anything before 70% position. From 70% to 90% position, the brightness keeps going up. And then it starts to drop if I continue to turn it up. So the maximum brightness happens when the knob is at 90%. But it's still very dim. I can barely see the trace.

Does anybody have any clue about what's wrong and how to fix it? I am wondering maybe there's something wrong with the potentiometer (it feels smooth though). But would like to hear what you guys have to say before I open it up 

Thanks!

[Stray Electron]

   Did you try taking the cover off and cleaning the front of the CRT and the back side of the protective glass?   You'd be surprised at how dirty they can get, particularly in an old scope.  I just bought a well optioned and working Lecroy 9374. I got it for $40 because it looked like the CRT was shot and was barely visible. But after cleaning it it's clearly visible at 64% brightness setting.

[yugu]

Thanks for the suggestion! Will clean it once I open it up.

[tautech]

Thanks for the suggestion! Will clean it once I open it up.

Be careful around the PDA lead on the CRT, it might bite you.    Even when off !
Best if it is dicharged with a screwdriver and a grounding strap.
Find a manual if you can and check HV voltages are to spec.

Be warned, they can be as high as 2.5 KV and a HV probe or capable MM are needed for this.
There is often high value resistive dividers in the CRT circuitry and the resistors can drift or go OC. Check these.

[PaulAm]

Some CRT scopes have a bias adjustment that can be used to set the brightness adjustment range, some don't.  If your scope has one of these, try to find the service manual for instructions on how to set it.  You may need a HV probe for your multimeter to set it correctly.

Sometimes the resistors in the focus/intensity string drift.  If you check these, make sure you have completely discharged the HV circuits before sticking any probes in there (and make sure it's unplugged, as well).

Something may have flaked out in the Z axis or blanking.  Something could be shot here restricting the intensity range.

It's possible the CRT could be bad, but normally you only see that if a scope has been left on 24/7 for extended periods.  That would be the last thing I would look at.

Do you have a service manual or schematic?

[yugu]

Thanks guys for the reply. I wasn't able to find the service manual.

Here is the internal. it looks quite clean. The potentiometer of the intensity knob is hiding under the crt. I have to remove the CRT to see it.

I didn't see the PDA lead of the CRT for discharging. could someone point it out in the following pictures?

Any thoughts?

[tautech]

As you have pointed out in the OP, the increase in brightness followed by dimming at max intensity does indeed indicate a problem.
Your CRT does not have a visible PDA lead, however use a grounded probe throughout the HV circuitry to discharge stuff around the tube.
PaulAm's description is accurate and I suggest you download some service manuals to get a "feel" for how the CRT HV is set up.
Also get the Tek pdf in the first post in the Repair board.

[yugu]

Thanks. I've been playing with it for some time today. So far what I know is

• There is nothing wrong with the intensity potentiometer. I've verified it with a multimeter.
• There is another intensity adjustment potentiometer on the pcb. However, it does not improve the maximum intensity.

I will dig into the HV circuitry when I got a chance. Seems the hypothesis is the HV for the CRT is not high enough, right? Without a service manual from Instek, how do I know the expected values?

[SeanB]

More likely resistors in the brightness circuit have gone high in value, and possibly a few capacitors in the power supply are failing and giving low power rails which means HT is low and the control range is low as well.

[yugu]

Yay found the circuitry for GOS-620FG. Though not the same model, they should be very similar. Really appreciate it if you could take a look to see which elements may go wrong. It will certainly take me a long time to understand the schematics.

[PaulAm]

Those aren't very useful.  The last page on the user manual has a block diagram; you really need to stuff in the boxes marked CRT  Circuit, Z-axis Amp and HV supply.  On the bright side, the 635 only uses a 2 KV HV supply.

[tautech]

Thanks. I've been playing with it for some time today. So far what I know is

• There is nothing wrong with the intensity potentiometer. I've verified it with a multimeter.
• There is another intensity adjustment potentiometer on the pcb. However, it does not improve the maximum intensity.

I will dig into the HV circuitry when I got a chance. Seems the hypothesis is the HV for the CRT is not high enough, right? Without a service manual from Instek, how do I know the expected values?

IIRC the trimmer is used to set minimum trace brightness while the main intensity is at a low setting.
Normally trace is adjusted to a "just visible" value.
Many adjustment procedures are done in a specified order to ensure the correct setup.

The HV adjustments will not be as critical as the rest of the scope circuitry, but you might strike it lucky and some values are on the PCB overlay as test point values.
There would be no harm in contacting Instek for a Service manual or failing that say a PSU and CRT schematic. Those should give the info you need for a repair. (Voltages, test points and component list)

[Alex Eisenhut]

Well I didn't want to post but that behavior sounds an awful lot like "double peaking". Usually means the tube is at end of life.

[David Hess]

The schematics do not show any PDA (post deflection acceleration) voltage connection and no PDA lead is visible in the photos so I assume the CRT lacks this which is common in slow oscilloscopes.  The documentation says that only the faster oscilloscopes in this series have PDA.  High PDA is necessary for a bright and sharp CRT and a good reason to prefer using fast oscilloscopes even slow signals.

If the horizontal and vertical deflection sensitivity is correct then the CRT cathode voltage is correct.  The manual says the cathode voltage is -2000 volts but from the looks of those schematics, it is lower.  Is the horizontal and vertical calibration correct on slow signals?

What kind of signal are you using as a test and at what time/div setting?  If the repetition rate is low, then fast time/div settings are going to look dim.

Make sure the x10 horizontal magnification is turned off and the hold off is set to minimum.

When the trigger is set to automatic and the timebase is slow, the trace should be nice and bright.

[yugu]

thanks guys!

yeah it sounds similar to the symptoms of "double peeking". I hope it's not because of an old tube   

To David: Good point. I test it with a 4v 1kHz sinusoid using 1ms/div and 1v/div. The measurement of the time looked correct to me. I didn't pay attention to the value of the peak-to-peak voltage, though. Let me verify it tonight. Do you mean if the vertical axis is way off, then there is still hope for the tube?

Also I will measure the voltage when I have a HV probe.

[David Hess]

yeah it sounds similar to the symptoms of "double peeking". I hope it's not because of an old tube   

It is sometimes difficult to distinguish double peaking from blooming.

Another possibility is that the CRT has become gassy.  If this is the case, just operating it for hours to days with the intensity turned down but not off may restore operation.

To David: Good point. I test it with a 4v 1kHz sinusoid using 1ms/div and 1v/div. The measurement of the time looked correct to me. I didn't pay attention to the value of the peak-to-peak voltage, though. Let me verify it tonight. Do you mean if the vertical axis is way off, then there is still hope for the tube?

If the cathode voltage is low which will cause a dim CRT, then the horizontal and vertical calibration will be off as well.  Low voltage will increase the sensitivity; if the cathode voltage is one half of what it should be, then the deflection sensitivity will be doubled.

So if the horizontal and vertical deflection is much higher than they are suppose to be, it indicates that the cathode voltage is low (closer to ground) which will also result in a dimmer CRT.

A more likely problem is that the repetition rate of the test signal is low.  Your 1 kHz test signal should be bright at 50uS/div and slower but progressively dimmer at faster sweep speeds.  The x10 horizontal magnification function will also dim the trace to 1/10th of its normal brightness.

Also I will measure the voltage when I have a HV probe.

If the horizontal and vertical calibration are correct, then you do not need to bother measuring the high negative cathode voltage because it is correct.

[yugu]

The explanation of the voltage makes perfect sense to me now.

I will double check the sensitivity when I got a chance. Thanks!

[yugu]

The sensitivities of both horizontal and vertical axis are correct. I measured the voltage of the tube pins (referenced to ground). They are as follows

H=-2.2kv
H=-2.2kv
K(G1)=-2.2kv
G1(k)=-2.2kv
P1=-1.65kv

p2=77.7v
G2=96.5v

Attached is the photo of the pins. Do those voltages look normal to you? I will purchase a replacement tube if these data is enough to give it a death sentence.

[tautech]

Those measurements are referenced to ground right?
Measure voltage between H & H, that should be the heater filament voltage.

[yugu]

Those measurements are referenced to ground right?

yep they are referenced to ground.

Measure voltage between H & H, that should be the heater filament voltage.

The voltage between H&H is AC 6.7v

Here are the parameters for the tube found from internet:
Type: KX-S-6  15SJ118Y14
Screen Size: 6"
Voltage(V): 6.3
Current(A): 0.2
4th Anode Voltage (V): --
3th Anode Voltage (V): 2000±100
2th Anode Voltage (V): 250~450
1th Anode Voltage (V): 2000
Cutoff Voltage (V): —35~-85
Horizontal: 19~21.6
Vertical: 10.8~13.2

[yugu]

If understand it correctly, the intensity control is adjusting the voltage between G1 and K. I checked the brightness of the screen on different voltages. Here is the result(turning the knob from min to max):

V(G1)-V(K) = 68      brightness = 0
V(G1)-V(K) = 62      maximum brightness (very dim)
V(G1)-V(K) = 44.5   brightness ~= 0.1 * maximum brightness

G1 is control grid and G2 is screen grid, right? what are P1 and P2?

[tautech]

Did you get the TEK PDF for troubleshooting CRO's?
Study from P35 on

[VK5RC]

If it isn't a circuit problem and a spare tube can't be found for reasonable $,  there are some guys who have had a little success in getting a bit of extra life out of the tube.  KE5FX, not for the faint hearted! http://www.ke5fx.com/crt.html

[yugu]

Thanks. I am not experienced in schematics/circuits. So I wanted to directly measure the voltages of the tube, to see if the tube is probably biased. If the voltages are good, and the tube is still too dim, then I will try to replace it (or zap it  )

Did I miss anything in this approach?

If understand it correctly, the intensity control is adjusting the voltage between G1 and K. I checked the brightness of the screen on different voltages. Here is the result(turning the knob from min to max):

V(G1)-V(K) = 68      brightness = 0
V(G1)-V(K) = 62      maximum brightness (very dim)
V(G1)-V(K) = 44.5   brightness ~= 0.1 * maximum brightness

Here V(G1) is higher than V(K). I believe it's because the labels on the PCB is misleading. They probably use K(G1) to indicate the pin could be either K or G1 in different models. And in my case, I should read the labels inside the parentheis...

G1 is control grid and G2 is screen grid, right? what are P1 and P2?

[tautech]

Thanks. I am not experienced in schematics/circuits. So I wanted to directly measure the voltages of the tube, to see if the tube is probably biased. If the voltages are good, and the tube is still too dim, then I will try to replace it (or zap it  )

Did I miss anything in this approach?

I have never need to Zap the CRT's of any of the old CRO's I have fixed, but I've always had the schematics.
IME its all about faulty componentry, correct voltages and setup.

Maybe there is somebody out there that can take some measurements for you of critical test points and PSU voltages/ripple etc.
Start a new thread requesting such with the model and make in the title and/or edit this thread and add the same.