Author Topic: BK Precision 2120 trace issues  (Read 13576 times)

0 Members and 1 Guest are viewing this topic.

Offline vexatagTopic starter

  • Regular Contributor
  • *
  • Posts: 103
  • Country: us
Re: BK Precision 2120 trace issues
« Reply #75 on: July 27, 2020, 02:26:48 am »
It did not. What am I connecting with the bodge wire? L801 B directly to R838 B?
 

Offline bdunham7

  • Super Contributor
  • ***
  • Posts: 7691
  • Country: us
Re: BK Precision 2120 trace issues
« Reply #76 on: July 27, 2020, 02:34:21 am »
This certainly explains some of the inconsistencies I was seeing in the measurements.

That looks ugly.  It will take some scraping!  And at some point perhaps you should replace both of those resistor, but for now you need to see if anything else is causing this or the scope was just used/abused a lot.
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline bdunham7

  • Super Contributor
  • ***
  • Posts: 7691
  • Country: us
Re: BK Precision 2120 trace issues
« Reply #77 on: July 27, 2020, 02:37:44 am »
You tried resoldering?  I see at least two solder joints that look toasted.  The left side of R839 is not looking good.  If you can't get a connection and just want a bodge for testing now, then yes just replace the trace with a wire, from that burnt side of R838 to the left side (near the resistor symbol) connection of L801.

A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline bdunham7

  • Super Contributor
  • ***
  • Posts: 7691
  • Country: us
Re: BK Precision 2120 trace issues
« Reply #78 on: July 27, 2020, 02:38:19 am »
It did not. What am I connecting with the bodge wire? L801 B directly to R838 B?

Yes.
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline vexatagTopic starter

  • Regular Contributor
  • *
  • Posts: 103
  • Country: us
Re: BK Precision 2120 trace issues
« Reply #79 on: July 27, 2020, 04:14:40 am »
The pads for R838 have completely lifted off the PCB, explains why resoldering didn't fix the problem. I guess my next step is to wire in two bodges, one to L801 and one to D2?
 

Offline bdunham7

  • Super Contributor
  • ***
  • Posts: 7691
  • Country: us
Re: BK Precision 2120 trace issues
« Reply #80 on: July 27, 2020, 04:31:43 am »
Yes, and if you eventually replace the resistor, you can just use the long leads to go through the holes and over to the point that the trace used to go to.  If there's any danger of touching something else, the small clear teflon tubing can be put over the wires.
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline vexatagTopic starter

  • Regular Contributor
  • *
  • Posts: 103
  • Country: us
Re: BK Precision 2120 trace issues
« Reply #81 on: July 27, 2020, 05:20:51 am »
Bodge installed, not pretty but functional. The horizontal deflection still isn't happening but L801 to D2 is reading at 15.9k ohm (same value as R838 in isolation). What should I test next?
 

Offline bdunham7

  • Super Contributor
  • ***
  • Posts: 7691
  • Country: us
Re: BK Precision 2120 trace issues
« Reply #82 on: July 27, 2020, 04:09:10 pm »
I'm a little worried that the resistor burning up may have been caused by another problem, so don't leave the scope powered on too long at a time. 

I want you to go back to the step where you used one multimeter as an input to move the dot vertically and then changed to CH2/X to stimulate the horizontal circuit.  I had you give me a bunch of voltages measured at the transistors.  I didn't explain much, but you should understand what I was looking for to help you go forward here.  What I was looking for was both for the signal to change in response to the X input from your DMM--which it did in the appropriate areas--and for the B to E voltage difference to stay at approximately 0.6 volts--which it did.  The first thing to look for in transistor troubleshooting is for the base-to-emitter voltage to be something different than about 0.6V, like 0V or over 1V.  So you can rerun those tests at some point.

But first, lets make sure that there isn't a short somewhere dragging down that D2 line and burning up the resistor.  Check the voltages at D1 and D2, then disconnect PG106 (after a power-down, of course) and check them again.  If there is a big difference with PG106 disconnected, especially if D2 is near zero when connected but pops up a lot after disconnecting PG106, then stop and turn the scope off.  The CRT might be damaged. 

If there isn't a big change and you have significant positive voltage on D1 and D2 with PG106 connected (say at least 50 volts on each) then you want to start testing at each point both with and without the input from the DMM.  Test this way at each point on each component on the schematic going all the way around from L801/802 to Q807--that's L801, R838, Q808, R835, Q807 on the D2 side and you can figure it out for the D1 side.
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline vexatagTopic starter

  • Regular Contributor
  • *
  • Posts: 103
  • Country: us
Re: BK Precision 2120 trace issues
« Reply #83 on: July 27, 2020, 08:26:26 pm »
Right-o, D1 connected is 192 V, D1 disconnected is 193 V. D2 connected is 101 V, D2 disconnected is 102 V. I'll test the other points in a bit but wanted to give that little update.
 

Offline vexatagTopic starter

  • Regular Contributor
  • *
  • Posts: 103
  • Country: us
Re: BK Precision 2120 trace issues
« Reply #84 on: July 27, 2020, 11:32:39 pm »
Here's the underside of the PCB by the way, I took this just before I soldered the bodge wires in.
 

Offline vexatagTopic starter

  • Regular Contributor
  • *
  • Posts: 103
  • Country: us
Re: BK Precision 2120 trace issues
« Reply #85 on: July 28, 2020, 12:16:06 am »
Ok, measurements have been made. All measurements were made with the negative lead on either D1 or D2 and the positive lead on the component being tested.

From D1:
L802: 71 V in either state
R839: 71 V in either state
Q809 base: -193 V in either state
Q809 emitter: -256 V in either state
R836: -193 V in either state

From D2:
L801: 162 V in either state
R838: 162 V in either state
Q808 base: -190 V in either state
Q808 emitter: -104 V in either state
R835: -104 V in either state.

The only component I wasn't able to test properly was Q807, it's small and sits too close to the pcb for me to clip a test lead to it. I suppose the next step is to test at the resistors directly adjacent to the base and emitter?
 

Offline bdunham7

  • Super Contributor
  • ***
  • Posts: 7691
  • Country: us
Re: BK Precision 2120 trace issues
« Reply #86 on: July 28, 2020, 12:41:50 am »
Sorry, I think I wasn't clear.  That helps, but it is confusing for me to try and calculate what I need.  Can you do this again with the negative DMM lead to chassis ground and the positive at the test point?  Are you in XY mode and have verified that you can 'move the dot' in the Y channel again before switching to the X channel at the same V/DIV?  Something isn't working that was working before.  Don't worry about Q807 at this point if it is too hard to get a safe measurement there.

EDIT:  Check at the base of Q808 first and verify that it changes with and without an input.  If it does not, stop right there.

EDIT2:  Check both sides of R835 and R836.  One side of each is connected to the other but I'm not sure it is obvious which just by looking at the board.
« Last Edit: July 28, 2020, 12:56:12 am by bdunham7 »
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline vexatagTopic starter

  • Regular Contributor
  • *
  • Posts: 103
  • Country: us
Re: BK Precision 2120 trace issues
« Reply #87 on: July 28, 2020, 05:08:34 am »
Having bnc leads connected to both channels is tripping the system up, but that's an issue to troubleshoot later.

I wasn't testing in XY mode, there's a lot of new concepts for me and things are escaping my notice. Y and X channels both have full motion from the position knobs. Additionally, the X position knob is now going off the screen both left and right.

Testing in XY mode with the BNC lead plugged into CH1/Y, DMM1 negative lead on chassis ground:
Q808 base goes from 1.527 V without a load and 1.532 V with a load. (it's hard to be specific, the exact reading can change just by moving the bnc cable around (I guess this is why you don't get the cheap cables?))
Q808 emitter: .961 V to .955 V
L801: 265 V either way
L802: 265 V either way
R835 (side not connected to R836): .960 V to .969 V
R835 (side connected to R836): -2.30 V to -2.29 V
R836 (side connected to R835): -2.30 V to -2.30 V
R836 (side not connected to R835): .488 to .488
Q809 emitter: .514 V to .514 V
Q809 base: 1.08 to 1.08
 

Offline bdunham7

  • Super Contributor
  • ***
  • Posts: 7691
  • Country: us
Re: BK Precision 2120 trace issues
« Reply #88 on: July 28, 2020, 05:22:27 am »
So go back to normal mode and see if you get a sweep with the trigger level knob centered and pushed in.  If not, try switching the COUPLING to 'LINE' and you may have to switch source to EXT as well--but try both ways.  Make sure all 4 buttons below there are out.  Set the timebase to one of the slowest settings.
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline bdunham7

  • Super Contributor
  • ***
  • Posts: 7691
  • Country: us
Re: BK Precision 2120 trace issues
« Reply #89 on: July 28, 2020, 05:42:29 am »

I wasn't testing in XY mode, there's a lot of new concepts for me and things are escaping my notice. Y and X channels both have full motion from the position knobs. Additionally, the X position knob is now going off the screen both left and right.

That's very good news.

Quote
Testing in XY mode with the BNC lead plugged into CH1/Y, DMM1 negative lead on chassis ground:

Oops!  You were supposed to check that you could deflect the dot vertically with CH1/Y but do the measurements on CH2/X.  But you don't need to redo everything.  The scope may be working at this point.  If not, just redo this test and look at the base and emitter of Q808.  Everything else looks good and the fact that you can get full control with the XPOS knob proves out most of the rest of the horizontal final amp.

A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline vexatagTopic starter

  • Regular Contributor
  • *
  • Posts: 103
  • Country: us
Re: BK Precision 2120 trace issues
« Reply #90 on: July 28, 2020, 06:25:18 am »
Q808 emitter: 0.94 V to -5.59 V
Q808 base: 5.61 V to 4.25 V
 

Offline bdunham7

  • Super Contributor
  • ***
  • Posts: 7691
  • Country: us
Re: BK Precision 2120 trace issues
« Reply #91 on: July 28, 2020, 03:29:34 pm »
If you go back a few pages to your reply #61, you were able to get about a 5 volt swing on Q808.

Quote
At the base, voltage when the test load was connected was -4.37 V, voltage when the test load disconnected was 0.92 V. At the emitter, voltage when the test load was connected was -5.01 V, voltage when the test load disconnected was 0.28 V.

Those numbers looked OK for Q808, but the latest ones don't.  So either there's an issue with your input method or something else is going on.  So check everything as best you can--make sure you can move the Y channel about half a screen like before, then move the BNC to the X channel.  I'm not sure what you meant by two BNCs 'tripping the system', but that doesn't sound good!

Next, the fact that you have full X-position control is very good because it shows the CRT and a lot of other stuff is working.  So we are down to Q808 and its input signal.  I'd like you to see if you can get a proper signal change on Q808 like we had before.  Try on both sides of R834 to make sure you aren't losing it across that resistor.  Also, perhaps first, give me the voltages on D1 and D2 (neg-to-ground, pos to test point) three different ways--with the XPOS set so the dot is on the left edge of the screen, centered, and on the right edge.

This scope has obviously been worked on before and one possibility that we have to consider is that a previous unsuccessful repair attempt may have included some adjustments to the trim pots that are now going to need to be corrected.
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline vexatagTopic starter

  • Regular Contributor
  • *
  • Posts: 103
  • Country: us
Re: BK Precision 2120 trace issues
« Reply #92 on: July 28, 2020, 07:15:36 pm »
Here's a Youtube video to what I mean by "tripping the system up". I needed something to test with and these were the cheapest leads that shipped from the USA (since international shipping is a bit of a mess right now). I'm assuming there's very little shielding, if any, and the actual probe clips have either bent out of shape or snapped off (it's hard to tell which), so I'll definitely be replacing them once the scope is able to generate a solid line across the screen when unplugged.

Testing without a test load in XY mode on channel 2 at 0.5 V/div (so I know where the center is, it's a 45 degree line at lower levels)
D1, dot left: 198 V
D1, dot center: 142 V
D1, dot right: 89 V

D2, dot left: 98 V
D2, dot center: 155 V
D2, dot right: 210 V

Testing on channel 2 at 10 mV/div, using my second multimeter in ohms mode as a test load
R834 (side not connected to Q808): 1.4 V to -4.6 V
R834 (side connected to Q808 base): 1.6 V to -4.3 V
Q808 B: 1.58 V to -4.16 V
Q808 E: 0.98 V to -4.40 V
 

Offline bdunham7

  • Super Contributor
  • ***
  • Posts: 7691
  • Country: us
Re: BK Precision 2120 trace issues
« Reply #93 on: July 28, 2020, 07:57:17 pm »
Which leads did you order?  Link? They don't look like they are necessarily junk, but maybe that is an issue for later.  It looks like you have an actual XY pattern going on there, but something is off.  What are those cables connected to on the other end????

The good news is you will know a lot more about oscilloscopes by the time we're done here.  The bad news is I think your scope is a beaten, cobbled wreck that is going to fight us all the way to the end.  No matter, I've fixed worse.  Sometimes I think the most important quality in a repair tech is sheer stubbornness. 

So, if you start with the dot on the left, nothing connected, put it in the non-XY normal mode (button out, right?) and set the timebase to 0.1S/Div, all 4 pushbuttons out, CH1 to 10mv/DIV with DC coupling, and the trigger centered and pushed in, what does it do?  And if you give it an input with your DMM in ohms just like the test before, does it trigger? (Does the dot move horizontally at all?)  If not, how about if you select trigger coupling "LINE"?

If not, then there's two separate things needed to track this down.

First, the inputs.  Something is not right. It looks like the polarity is inverted from what I was expecting, but that alone shouldn't be a serious problem. Still, take the one you're using as the source (the Aneng I presume) and connect it directly to the other DMM, pos-to-pos, neg-to-neg and test the output voltage of the ohms function, including whether or not it is a negative reading.  Then continuity-check your BNC cable to make sure you know which clip is connected to the center pin of the BNC connector.  I want  you to make sure in future tests that the positive source voltage is connected to the center pin--and if your voltage was negative in the first step, that means the negative lead of the DMM.  Then, can you verify that when connected like this to the scope, the DMM reads approximately 1Megohm when connected to either channel?

Second, it looks like we are getting some signal through to Q808, but I've no idea what the actual signal level should be.  So go back to XY mode, set the dot in the center to start and do the same measurements as the last step of last time, only on Q809 and at the base, emitter and the other side of R836.  Don't worry about R832 unless it is super easy to get to.  Then we'll compare the voltage swing at the base that it takes for Q809 to deflect the dot all the way from one side to the other to see if the 5-volt or so swing we had at Q808 should be enough to see(in which case Q808 is bad) or if the signal needs to be a lot more (in which case the horizontal final amp is fixed and we move on...).





« Last Edit: July 28, 2020, 08:02:58 pm by bdunham7 »
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline vexatagTopic starter

  • Regular Contributor
  • *
  • Posts: 103
  • Country: us
Re: BK Precision 2120 trace issues
« Reply #94 on: July 28, 2020, 10:06:53 pm »
Which leads did you order?  Link? They don't look like they are necessarily junk, but maybe that is an issue for later.  It looks like you have an actual XY pattern going on there, but something is off.  What are those cables connected to on the other end????

These bad boys are hooked up to... nothing at all. All the changes in the display are from me moving the cables around.


So, if you start with the dot on the left, nothing connected, put it in the non-XY normal mode (button out, right?) and set the timebase to 0.1S/Div, all 4 pushbuttons out, CH1 to 10mv/DIV with DC coupling, and the trigger centered and pushed in, what does it do?  And if you give it an input with your DMM in ohms just like the test before, does it trigger? (Does the dot move horizontally at all?)  If not, how about if you select trigger coupling "LINE"?

With settings dialed as specified and nothing connected (no lead hooked up) it's just a dot. With a lead hooked up to CH1/Y, it's a ~20mm vertical line. With one lead of the multimeter hooked up to the test lead, the line goes from top to bottom, growing thinner and dimmer. With both leads of the multimeter hooked up, it is impossible to notice. Trigger coupling "LINE" mode has the same results.

If not, then there's two separate things needed to track this down.

First, the inputs.  Something is not right. It looks like the polarity is inverted from what I was expecting, but that alone shouldn't be a serious problem. Still, take the one you're using as the source (the Aneng I presume) and connect it directly to the other DMM, pos-to-pos, neg-to-neg and test the output voltage of the ohms function, including whether or not it is a negative reading.  Then continuity-check your BNC cable to make sure you know which clip is connected to the center pin of the BNC connector.  I want  you to make sure in future tests that the positive source voltage is connected to the center pin--and if your voltage was negative in the first step, that means the negative lead of the DMM.  Then, can you verify that when connected like this to the scope, the DMM reads approximately 1Megohm when connected to either channel?

My source is the Aneng, hooking it to the other meter pos-pos and neg-neg gives 0.499 V. Checking the bnc, the red jack is the center and the black jack is the outset shield. I've been connecting neg-pos on my measurements with the bnc so far, should I connect pos-pos in the future? And yes, both channels are around 1 Megaohm.

Second, it looks like we are getting some signal through to Q808, but I've no idea what the actual signal level should be.  So go back to XY mode, set the dot in the center to start and do the same measurements as the last step of last time, only on Q809 and at the base, emitter and the other side of R836.  Don't worry about R832 unless it is super easy to get to.  Then we'll compare the voltage swing at the base that it takes for Q809 to deflect the dot all the way from one side to the other to see if the 5-volt or so swing we had at Q808 should be enough to see(in which case Q808 is bad) or if the signal needs to be a lot more (in which case the horizontal final amp is fixed and we move on...).

Test of changes when adding a load:
Q809 base: 1.06 V to 1.08 V
Q809 emitter: 0.508 V to 3.34 V
R836 (Q808 side): -2.27 V to 3.46 V

Deflection test at base of Q809:
Left: 3.85 V
Center: 1.09 V
Right: -1.80 V
 

Offline bdunham7

  • Super Contributor
  • ***
  • Posts: 7691
  • Country: us
Re: BK Precision 2120 trace issues
« Reply #95 on: July 29, 2020, 12:46:44 am »
Yes, use the pos lead of the Aneng to the red clip on your BNC.  In your video, I think you are just picking up noise, you have it set at 10mV/div.  So now that we know that 5 volts is enough of  change at the base of Q809 to deflect from one side to the other, lets see what we get when we use a proper input.  You have about 0.5 volts, so if we set the CH2/X input to 100mV/div (I think that is 0.1V/div on your scope) then your meter's output should deflect the dot 5 divisions (if it were working). 

So do that.  Set it to 0.1V/div on CH2/X, use XY mode as before and test at Q808 base, emitter, D2, D1 and the side of R835 that is not connected to Q808's emitter.  Then we can compare the base voltage change and the D1-D2 change with what Q809 produces.
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline vexatagTopic starter

  • Regular Contributor
  • *
  • Posts: 103
  • Country: us
Re: BK Precision 2120 trace issues
« Reply #96 on: July 29, 2020, 03:42:15 am »
I figured they're probably just poorly shielded cables, I'll pick some decent ones up eventually.

As for the measurements (set as you outlined, connecting pos-pos on the test load):
Q808 emitter: 0.96 V to 1.55 V
Q808 base: 1.53 V to 2.14 V
D1: 143 V to 135 V
D2: 154 V to 161 V
R835 (Q809 side): 0.96 V to 1.55 V
 

Offline bdunham7

  • Super Contributor
  • ***
  • Posts: 7691
  • Country: us
Re: BK Precision 2120 trace issues
« Reply #97 on: July 29, 2020, 04:31:16 am »
OK, does the dot move when you do this?  How far? 

If it does move, the go down to 50mV/div and do the same test again, including recording all the same voltages, then go down to 20mv/div and do it again, then 10, etc until you get either get a difference between load and no load of at least 5 volts on the base of Q808 or you get the dot to move at least halfway, preferably all the way, across the screen (you can recenter as needed to give it room to move).

A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline vexatagTopic starter

  • Regular Contributor
  • *
  • Posts: 103
  • Country: us
Re: BK Precision 2120 trace issues
« Reply #98 on: July 29, 2020, 04:49:41 am »
In XY mode, the dot is not a dot but a 45 degree angle line as I mentioned here

Testing without a test load in XY mode on channel 2 at 0.5 V/div (so I know where the center is, it's a 45 degree line at lower levels)

When I apply the voltage, the line shifts position up and to the right about 1 CM, in addition to getting larger and fainter. I don't know if this changes anything in what I should be looking for.
 

Offline bdunham7

  • Super Contributor
  • ***
  • Posts: 7691
  • Country: us
Re: BK Precision 2120 trace issues
« Reply #99 on: July 29, 2020, 04:55:35 am »
The 45 degree line is what happens when you feed the same signal to both channels in XY mode.  There are noise and signal issues here that we haven't gotten to yet--I saw your videos and something is....wierd.  But for now I want to see whether the horizontal final amp is fixed before moving on.  So just to confirm, if you move the 'dot' or whatever it is with the XPOS knob, you apparently can get it all the way to the left or right.  Does it change shape or anything when you are moving it with the XPOS knob?
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf