BK Precision 2120 trace issues

[vexatag]

Update: the resistors came in and I figured out the problem with my soldering (cheapo 63/37 solder doesn't leave a nice shiny joint ;__;), now I have the casing off but I don't know how to proceed with removing the pcb in question. Is there a "best" order for removing the connectors? How do I disconnect the switches from the front face?

[bdunham7]

Update: the resistors came in and I figured out the problem with my soldering (cheapo 63/37 solder doesn't leave a nice shiny joint ;__;), now I have the casing off but I don't know how to proceed with removing the pcb in question. Is there a "best" order for removing the connectors? How do I disconnect the switches from the front face?

Fake solder seems to be a thing now too.  I have some that came 'free' with some junk or another and it is like stirring mud.  As far as disassembly, take pictures for putting it back together, post them if you want.  I don't remember how the knobs and switches came off and the one I fixed was a little different than yours.  It can be tricky.  Can you do it without disassembling the front?

[vexatag]

Practically speaking, no. Between needing to desolder and solder the resistors from a very small vantage point without damaging any other components and the cleanup of the fresh component leads, it's a bit beyond my comfort zone. The square buttons are easy to remove however (pull on the metal shaft) and I know I somehow got the x position knob removed from the pot by accident, so it does seem possible.

[vexatag]

Ok, swapped the resistors out. TP260V is reading 265.5v, however the trace is still just a point. The trace is also even further to the left, I can get it about 1.5 cm away from the left edge in either X-Y mode or in x10 mode.

[bdunham7]

That's too bad--it was looking hopeful for a bit once you found those bad resistors.

So if you feed a signal (perhaps the probe calibrator) into the X-channel (in XY mode) can you get at least a small horizontal line that varies with the input voltage range setting?  And if you feed the same signal into the Y channel, can you get a vertical signal at all?  What else do you have that you can use to input a signal?

If you get a vertical signal at all and the X-axis responds even slightly, my first guess is that your next problem is within the horizontal final amp circuit.  You can start testing that circuit by verifying the +24 and +12V points are correct, and then depowering and measuring all of the resistors, starting with everything to the right of Q807.  Remember, high resistance = broken, low resistance = needs further thought.

[vexatag]

I uploaded a video of the channel testing to Youtube, testing on a cheap 1khz triangle wave generator in addition to the calibration point.

12v and 24v test points are reading ~12v and ~24v respectively. I'm about to test the resistors.

[bdunham7]

What happens at about 1 minute of your video is not clear to me--the dot splits up at an angle and I don't have a good explanation for that.  When you do channel 1, it appears to have a proper vertical response.  I'm going to have to think about that and I'm away for a bit, so test those resistors and don't give up yet.

If you don't find a problem, you should try turning XY mode off and look at it in the normal mode.  Figure out what moves the dot and what doesn't.

[vexatag]

Tested all of the resistors to the right of Q807 in the horizontal final amp circuit (aside from R828 which I can't for the life of me find on the PCB). All of them are within an ohm or so of spec, with the exception of R838 and R839 (measuring at 15.92k and 14.97k respectively).

When I replaced the resistors, for some reason the normal mode wasn't able to come on the display at all. I fiddled with... something and got it displaying, the results are the same as in XY mode. Additionally, the x position knob is now functioning similarly in XY mode and normal mode, giving ~5mm of horizontal movement at the center of display either direction. The directions of movement are different, however. Channel 1 XY mode clockwise is left, normal mode clockwise is right. Channel 2 clockwise is left for both modes.

[bdunham7]

OK, you need to stick with the normal mode for a while.  The normal mode will not normally show on the screen at all unless it is triggered or in AUTO mode.  So let's set this up for normal mode operation:

All six of the square pushbuttons in the 'out' position.  Coupling to DC/LF.  Source to CH1.  Connect the calibrator output to CH1 and set for 0.5 V/div for now.  Set the TIM/DIV to 1 ms/div, push in the center knob and make sure the XPOS/MAG knob button is pushed in.  Turn the TRIG LVL button to about the center and push it in.  Tell me what you have at that point.

[vexatag]

I wasn't able move the center of the time/div knob in or out, but running all the settings as you said gives me two points stacked vertically, about 8 mm away from each other.

[bdunham7]

OK, the TIM/DIV center knob turns instead of pushing, no big deal.

If you change the V/DIV one click each way does that 8mm become 4 mm one way and 15 the other (or something like that--precision is unimportant).

And then, if you change the SOURCE to EXT, pull the TRIG LVL out and center it--do the does stay or go away?

[vexatag]

If you change the V/DIV one click each way does that 8mm become 4 mm one way and 15 the other (or something like that--precision is unimportant).

Changing to 1 v/div halves the distance, 0.2 v/div roughly doubles it.

And then, if you change the SOURCE to EXT, pull the TRIG LVL out and center it--do the does stay or go away?

This makes absolutely zero difference.

[bdunham7]

I'm out of ideas for the moment, but don't throw it out the window yet.  Test the rest of the resistors in the horizontal final amp and maybe give it a break for a day or two...

One more thing--is the dot in the same area that is was before you replaced the defective resistors?

[vexatag]

One more thing--is the dot in the same area that is was before you replaced the defective resistors?

No, when I first got the scope the dot's position was strongly on the left, the x-pos knob moved the dot from the left (off screen) to a little over 1 cm away from the center. Now the entire range of the x-pos knob is within 1 cm to the left of the center line (the entire range fits inside one measured box) but it's much closer at the closest setting.

I'll check the rest of the resistors and report back.

[vexatag]

Checked the remaining resistors in the horizontal amp, here are the ones that were out of spec enough for me to notice:

• R823: 1.58k
• R825: 86.k
• R827: 9.2k
• R831: 4.57k
• R840: 1.55k

[bdunham7]

R825--is that a typo--is is 8.6K or 86K? 

You're getting to the point where one would typically use another oscilloscope to repair the oscilloscope, but since that's not possible, we'll keep trying with what you have.

If you don't already know how this circuit works, it helps to understand the theory a bit.  The horizontal final amp moves the beam by varying the voltage on two plates on either side, those are the outputs D1 and D2.  In normal mode, the horizontal sweep generator (elsewhere) would send a sawtooth-shaped sweep signal to the input marked FROM X-Y MODE IN.  When you switch to XY mode, the horizontal final amp just follows the X input.  The controls to the left of Q807 all control the overall fixed position of the beam.  Since you don't have another scope, we can try using very slow inputs and trying to see if they show up on your meter. 

Take the following measurements using your 1000VDC range, and do this again all hands-off by turning off, waiting, connecting and turning on.  Set XPOS to center, XMAGX10 off (pushed in) and XY MODE on (pushed in).

1.  DMM negative lead to ground, positive to the TO CRT D1 point.
2.  DMM neg to grd, pos to TO CRT D2 point.
3.  DMM neg to TO CRT D1, pos to TO CRT D2 (reading across the two plates)

Now, with the DMM still connected across the plates in the 3. configuration, adjust the XPOS control all the way counterclockwise, record the voltage, then all the way clockwise and record again.  Pull the knob out (turn on XMAGX10) and do it all again.

Next we can try generating the slowest sweep possible, by turning off XY MODE and XMAGX10 and setting the sweep speed to the slowest setting, which I think is .2S/div.  You can turn the center knob counterclockwise to slow it down a bit more.  See if there's any change on the plate voltage this way, and then try again with XMAGX10 on (out). 

Lastly, do the same as the last step, except connect the DMM neg to ground, pos to the left side of R834 or where it says FROM XYMODE IN.  You may have to turn the voltage setting down on the DMM, but start at 1000VDC and go down one range at a time (always a good practice on unknowns)  If you can see the sweep voltage ramping up over 2-3 seconds here but not on the plates, then we know you are in the right area.

[vexatag]

I don't see anything labeled TO CRT D1 or D2 on the pcb, but there is a plug, PG106, that seems to be where the D1 and D2 would be, is this a correct understanding?

Also I can technically use my computer's sound card as an oscilloscope but the cabling I'd be using is very sketchy (test probe into a bnc to rca adapter into an rca switcher into an rca to 3.5 mm stereo adapter into a very cheap 3.5 mm audio cable into my sound card).

[bdunham7]

I'm not sure.  Look at the schematic and see if you can relate that plug to R838 and R839.  The top of R838 and the bottom of R839 are alternate points to measure from.

[vexatag]

I'm definitely kicking myself for not taking high-quality photos when I had the pcb underside easily accessible 

PG106 has four wires, tan, tan, orange, and black. R838 has continuity with the left tan pin, R389 has continuity with the right tan pin.

[bdunham7]

There you go.  Measure away.

[vexatag]

Oh and R825 was 8.6k

[vexatag]

Measurements!

Checking to ground:

• Left pin: 825 mV
• Right pin: 38.2 V

Neg on left, pos on right:

• X-pos neutral: 146 V
• X-pos clockwise: 73 V
• X-pos counterclockwise: 220 V
• X10 mode neutral: 115 V
• X10 mode clockwise: 1.3 V
• X10 mode counterclockwise: 260 V

Slowing the sweep:

• X-pos neutral: 141 V
• X-pos clockwise: 145 V
• X-pos counterclockwise: range between 136 V and 138 V
• X10 mode neutral: 141 V
• X10 mode clockwise: 1.3 V
• X10 mode counterclockwise: range between 136 V and 138 V

Measuring the left leg of R834 (when viewed from the front of the scope) at 0.5mS/div in XY mode gave about 1.6 V through the entire x range and X10 mode, but at 0.2S/div in normal mode gave -1.42 V at neutral and -2.32 V when the xpos knob was rotated fully clockwise.

[bdunham7]

Measuring the left leg of R834 (when viewed from the front of the scope) at 0.5mS/div in XY mode gave about 1.6 V through the entire x range and X10 mode, but at 0.2S/div in normal mode gave -1.42 V at neutral and -2.32 V when the xpos knob was rotated fully clockwise.

I meant the left side on the schematic, IOW the side that is not connected to the base of Q808.  It would help if you had a signal generator that could feed in a very slow ramp or triangle wave of about 1 volt and 0.1 Hz, but the sweep generator in the 0.2S/div mode will maybe do.  What I'm looking for is a smooth change from low to high that repeats itself.  This may be difficult to spot depending on how fast your meter updates.

[vexatag]

Oops, my bad. Doing the proper test (checked the leg labeled B for continuity with either side of R834, then hooked the positive lead to the leg without continuity and the negative lead to the chassis), it started at about -15 mV but has been falling over about three or four minutes. I wish I could have a visual of this data, I understand now why they say you need a scope to fix a scope.

[bdunham7]

Oops, my bad. Doing the proper test (checked the leg labeled B for continuity with either side of R834, then hooked the positive lead to the leg without continuity and the negative lead to the chassis), it started at about -15 mV but has been falling over about three or four minutes. I wish I could have a visual of this data, I understand now why they say you need a scope to fix a scope.

Yes, another scope is the best way!  What did you measure R834 at?  It is only a 68R resistor so I would not expect a continuity meter to show open. 

So what we are trying to do is trace signals.  If you look up at the Trig & Sweep section, you'll see that S605 is your XY MODE switch.  In normal mode, the sweep generator sends a ramp signal to the horizonal final amp and that is amplified and sent to the horizontal deflection plates causing the beam to sweep from left to right.  In XY mode, the CH2/X signal is fed instead, causing the beam to move according to the signal.  Since neither of these functions work and they are at least partly independent of each other, I was going on the assumption that your problem lies somewhere in the horizontal final amp.  Now we should take a step back and verify as many of those assumptions as we can, and tracing a signal into R834 is on very important point.  The other would be verifying as best we can that the CRT is responding to the horizontal deflection voltage.  So, let's try a few things.

First, unplug PG106 and power up the scope.  Where is the dot?

Second, tell me what equipment you do have, other than your DMM.  Power supplies, electronic parts such as potentiometers, batteries, wire, solder, breadboards, etc etc.  We need to rig up a signal we can trace through the scope.  Even a battery and a 10K potentiometer might work.