Topics

Grid Bias Adjustment on 475A


 

I've been trying to run through the calibration process for the 475A that had a blown Z-axis amp, and I've run into trouble right off the bat. I was unable to get a visible spot on screen at the +15 V intensity level as indicated in the service manual. I had to crank the intensity up nearly to maximum, and then the trim pot (R1375) for grid bias had to be cranked to one extreme to see anything on screen.

I tried trimming up the power supply, even though everything was within spec (everything was a little low, but not out of spec, so I trimmed the +50 V rail up to +50.18V, just shy of the upper range in the service manual. All the other low voltage rails now look spot on.

I tried the same process on two other 475s (non-A) and was easily able to get a visible spot at the +15 V intensity level, as well as adjust the grid bias for a visible spot across a wide range of the grid bias trim pot.

Next, I followed the diagnostic procedure in the service manual for the Z-axis circuit on page 4-18. I set up the scope as indicated, lifted on end of CR1343, and measured TP1364 to be +6.05 V (as indicated in the SM). When I press BEAM FIND the voltage jumps up to +29.3V, which is higher than the SM's indicated +25 V. Next I checked the voltage across VR1362, which is +6.23 V, which is the indicated value. Next I removed Q1362 and checked the voltage at the base of Q1352 which measured +1.34 V, and the voltage at the base of Q1344 which measured +0.67 V, both as indicated in the SM.

After that I lifted one end of CR1341 and measured the voltage on the collector of Q1338 while turning the INTENSITY knob through its full range of travel; the collector voltage measured from +7.6 V to -1.1 V (the later being a little short of the indicated -1.3 V from the SM). I also checked the voltage at the base of Q1332 which measured -1.75 V (as indicated). I then removed Q1332 and disconnected J1333 and J1334, and measured the collector voltage on Q1338 again as I turned INTENSITY through its range; again +7.6 V to -1.1 V

So I'm seeing a slightly higher than expected voltage when pressing BEAM FIND at TP1364, and a slightly narrower than expected range of voltage at the collector of Q1338, with the "top" of the range falling about 200 mV short of what the SM indicates. I'm not sure that any of this explains why I can't do the grid bias adjustment as indicated in the manual.

Just for good measure I also swapped the CRT from one of the 475s that I was able to successfully perform the grid bias adjustment on with the CRT in the 475A (they happened to be exactly the same part number, even though the SM for the 475 says that the CRT should be PN 154-0677-00, what I found in my donor was 154-0677-10, which is the correct part for a 475A), but the trouble did not follow the CRT.

I guess I could still check to see if the trim pot (R1375) appears to be in spec (if it isn't it would explain a lot), but I'm otherwise at my wits end.

Any help or advice would be appreciated.

-- Jeff Dutky


adesilva_1999@...
 

On Thu, Dec 31, 2020 at 04:36 PM, Jeff Dutky wrote:


I also checked the voltage at the base of Q1332 which measured -1.75 V (as
indicated).
The diagrm I have for 475A has the Q1332 base grounded. I am not sure about the readings you have mentioned. I have not gone through the whole manual but just looking at the schematic only.

Also, do you have the 150 volts on one leg of the R1375 trim pot?


adesilva_1999@...
 

Sorry! that should be 50 volts. Not 150 volts!


 

I have not examined the grid bias trim pot (R1375) yet, so I don't know what voltages it sees. Like I said, if it turns out that the trim pot has drifted out of spec that would explain a lot of what I'm seeing. I'll see what I can measure straight away.


 

Okay, I've examined R1375 (with the power off, and I'm not sure where to probe for anything other than that center tap on the pot, so I can't really answer whether it's seeing the full +50 V at one of the outer terminals) but the pot seems (to my uneducated eye) to be acting strangely. I was able to find a pad that is connected to the center tap of the pot, so I measured its resistance between center tap and ground, and between center tap and the +50 V rail, as I turn the pot through its range. It starts a effectively zero ohms (0.001 K ohm), goes up to about 6 K ohms, then drops down again to around 2 K ohms. The schematic says this is a 25 K ohm pot, so shouldn't I expect it to go from 0 to 25 K ohms?

If it turns out that the pot is dead, I guess that's good, but I don't relish the idea of removing the main interface board to unsolder and replace this trim pot.


Dave Peterson
 

Trying to find R1375 & associated components on your earlier pictures, but I think from looking at my 465 that they're just out of frame.

Also using my 465 as proxy, it looks like the zener diode and 10k resistor node (don't have your schematics at hand for #'s) are accessible. I'm not advocating, and I wouldn't blame anyone for not being comfortable operating their scope with the shield off, but can you get a measurement off this output of the R1375 pot? Looks like it'd be possible to get a 10x scope probe on one of the leads of the cap maybe? Reassemble the shield and see if the value responds to turning the pot - again with the shield on. And this all depends on the 475A board layout, accessibility, clearance for the probe with shield, etc.

I'd be wary of trying to get a good R value in situ with the 50v supply being connected to the regulator divider feedback network.

Dave


 

R1375 is deep under the HV shield, but that's hardly my issue (I've been operating the scope without an HV shield for weeks). The real problem is that the grid bias pot is snuggled up close to what I believe to be the circuitry that generates the KV potentials, and I'm not too happy about poking around in there with a test lead (much less a scope probe) with the power on. I can hook the lead of a flying capacitor in the grid bias circuit, but it's separated from the center tap of the pot by another resistor (still not a huge problem, but it just adds to my analytical burden). The layout for the 465 looks a little different than the 475. It looks like the corresponding pot on the 465 is called R1480 "CRT BIAS" and is much further removed from the HV section than on the 475.

I am worried about getting a good reading in circuit, but am I wrong in thinking that any path from either ground or the +50 V rail should be fairly high impedance (like Megohms)?

Now that you mention it I notice that there's another component that could be dragging on the output of the grid bias pot, if it has failed that is: C1374 is a 1uF decoupling cap of almost identical make to the one that failed on my father's 475. If that were conducting significantly it would surely mess up the grid bias voltage.

-- Jeff Dutky.


adesilva_1999@...
 

Happy new year guys!

C 1374 is a possible culprit. You can always replace it and give it a try. If not, you could solder a couple of leads to the viper and the 50 volt ends of the pot and bring them away from the HV area to test. Since one end of the pot is grounded, you should be able to easily check the other end for 50 volts. If that is OK, then see if the viper has a voltage that is somewhere in between 0-50 volts. The issue is that it might be close to one end and the best test is to observe that lead while the power is on and turning it slightly with a plastic screw driver. If you decide to do that, make sure to mark the initial position as it is.


Jim Ford
 

LOL, I think you mean "wiper", not "viper"!  A viper is a kind of poisonous snake!  Such is the confused but fascinating mess that is the English language.   Jim Ford Sent from my Verizon, Samsung Galaxy smartphone

-------- Original message --------From: "adesilva_1999 via groups.io" <adesilva_1999=yahoo.com@groups.io> Date: 1/1/21 1:30 AM (GMT-08:00) To: TekScopes@groups.io Subject: Re: [TekScopes] Grid Bias Adjustment on 475A Happy new year guys!C 1374 is a possible culprit. You can always replace it and give it a try. If not, you could solder a couple of leads to the viper and the 50 volt ends of the pot and bring them away from the HV area to test. Since one end of the pot is grounded, you should be able to easily check the other end for 50 volts. If that is OK, then see if the viper has a voltage that is somewhere in between 0-50 volts. The issue is that it might be close to one end and the best test is to observe that lead while the power is on and turning it slightly with a plastic screw driver. If you decide to do that, make sure to mark the initial position as it is.


Michael W. Lynch
 

Happy New Year to all!

Resolve the condition and function of R1375 before you move on. What you describe above sounds suspicious to me.

After that, there are some other component to check as well.

Check VR1374 which is the 82V Zener, CR1373 and CR1371 (both 175V, 100mA fast switching diodes). These are well known to fail and cause extreme brightness, I suppose it would be possible for them to fail some other manner and cause a "dim" display as well?

C1373 is another known problem area.

Good Luck!

--
Michael Lynch
Dardanelle, AR


Dave Peterson
 

Yes, C1374 is in a good place to explain a lot of what you're seeing. But schematics aren't board layout, and the TekWiki links for 475A don't seem to have the board layouts. So I don't know how accessible these components are. I was hoping you might be able to use the hook clip of a 10x probe to hook on that cap lead, power up, and adjust the grid bias just to see if it is adjusting that voltage up and down. And by how much. If you get to that point record the max, min, and how linear it is. Then you can see if those values make sense relative to the schematic. The online schematic has some voltages annotated, but doesn't seem to indicate exactly which nodes (+107 ?).
Looking at the 50v PS circuit there's an adjustable voltage divider that feeds back to the regulator "-" input. The online schematics are difficult to make out values, but even high k Rs in parallel are going to add to a much lower resistance (1/Rt = 1/R1 + 1/R2 + 1/Rn ...). There are also several PN junctions along the way that an ohm meter may well forward bias and contribute to "false" low R values while powered off and in situ. Testing for a dead short or complete open is one thing, but beyond that I wouldn't put too much stock in such a measurement. That's my personal take/preference/opinion. FWIW, YMMV, ... And there's no guarantee that the +50v at the grid bias is direct connected to the PS output (but it does say it goes to sheet 12).
Dave

On Thursday, December 31, 2020, 11:01:00 PM PST, Jeff Dutky <jeff.dutky@gmail.com> wrote:

R1375 is deep under the HV shield, but that's hardly my issue (I've been operating the scope without an HV shield for weeks). The real problem is that the grid bias pot is snuggled up close to what I believe to be the circuitry that generates the KV potentials, and I'm not too happy about poking around in there with a test lead (much less a scope probe) with the power on. I can hook the lead of a flying capacitor in the grid bias circuit, but it's separated from the center tap of the pot by another resistor (still not a huge problem, but it just adds to my analytical burden). The layout for the 465 looks a little different than the 475. It looks like the corresponding pot on the 465 is called R1480 "CRT BIAS" and is much further removed from the HV section than on the 475.

I am worried about getting a good reading in circuit, but am I wrong in thinking that any path from either ground or the +50 V rail should be fairly high impedance (like Megohms)?

Now that you mention it I notice that there's another component that could be dragging on the output of the grid bias pot, if it has failed that is: C1374 is a 1uF decoupling cap of almost identical make to the one that failed on my father's 475. If that were conducting significantly it would surely mess up the grid bias voltage.

-- Jeff Dutky.


Michael W. Lynch
 

On Fri, Jan 1, 2021 at 11:58 AM, Dave Peterson wrote:


But schematics aren't board layout, and the TekWiki links for 475A don't seem
to have the board layouts.
I happen to have the board Layouts for 475 and 475A and after a cursory look, they appear to be almost identical. They are, as someone else noted, completely different from the 465, but they do function in a similar manner. . Having worked on both, the 475 and 475A I can confirm that the layouts are for the most part the same. And there is very little room to work, as Jeff has already said. C1374 is not too bad to replace. VR1374 is a tough one, but doable. C1373 is also a PITA but again, doable. They just have packed a bunch of little components into a small area.

--
Michael Lynch
Dardanelle, AR


adesilva_1999@...
 

Of course! After a few drinks and too much to eat, at 1:30 AM while being a one finger typist, with the computer accepting anything I type, viper sounds like wiper anyway! I am glad I did not type Viagra. Imagine that!


Jim Ford
 

Good thing! :DLOL!   Happy New Year, everyone!     JimSent from my Verizon, Samsung Galaxy smartphone

-------- Original message --------From: "adesilva_1999 via groups.io" <adesilva_1999=yahoo.com@groups.io> Date: 1/1/21 12:03 PM (GMT-08:00) To: TekScopes@groups.io Subject: Re: [TekScopes] Grid Bias Adjustment on 475A Of course! After a few drinks and too much to eat, at 1:30 AM while being a one finger typist, with the computer accepting anything I type, viper sounds like wiper anyway! I am glad I did not type Viagra. Imagine that!


 

I lifted one end of C1374 and tested it with my DMM916. It acted much more like a capacitor than like a resistor: while measuring resistance I could see it charge up (the resistance increased over several seconds till it became open circuit) then I switch over to voltage measurement and watched it discharge from about 1.1 V down to 0 V over several seconds). Using the capacitance measure shows this is a 0.8 uF capacitor, which is in spec for a the 1 uF electrolytic shown in the schematic.

i checked CR1371 and CR1373 in circuit using the diode check mode of the 916 and they both measured 0.536 V forward voltage drop. They look good.

I measured the trim pot (R1375) while C1373, C1374, and VR1374 were lifted. The trim cap has a range of travel of about 280 degrees:

0 deg = 0 ohms
45 deg = 3.1 K ohms
90 deg = 4.8 K ohms
135 deg = 7.0 K ohms
180 deg = 6.75 ohms
225 deg = 4.3 K ohms
270 deg = 2.7 K ohms

The specified value of R1375 is 25 K ohms, and I expect that it should range smoothly from 0 ohms to 25 K ohms.

I also verified continuity between R1375 and both the ground and +50 V rails.

I checked C1373 (the cap in parallel with VR1374) and it measured as 96.5 nF (it's specced as 0.1 uF on the schematic), so that looks good.

VR1374 is not shorted (I didn't check that it's breakdown voltage was 82 V, as I sadly have no way to do that).

This really seems to point to the trim pot as the culprit, which means I need to get the main interface board out. Joy.

-- Jeff Dutky


Tom Lee
 

If it is indeed the trimpot, that is an unusual failure mode. I can recall seeing something similar only once, and it was due to a small sliver of solder that managed to bridge the pads connected to the outer two terminals.

Out of curiosity, what is the resistance you measure between the two non-wiper terminals?

--Tom

--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
http://www-smirc.stanford.edu

On 1/2/2021 00:32, Jeff Dutky wrote:
I lifted one end of C1374 and tested it with my DMM916. It acted much more like a capacitor than like a resistor: while measuring resistance I could see it charge up (the resistance increased over several seconds till it became open circuit) then I switch over to voltage measurement and watched it discharge from about 1.1 V down to 0 V over several seconds). Using the capacitance measure shows this is a 0.8 uF capacitor, which is in spec for a the 1 uF electrolytic shown in the schematic.

i checked CR1371 and CR1373 in circuit using the diode check mode of the 916 and they both measured 0.536 V forward voltage drop. They look good.

I measured the trim pot (R1375) while C1373, C1374, and VR1374 were lifted. The trim cap has a range of travel of about 280 degrees:

0 deg = 0 ohms
45 deg = 3.1 K ohms
90 deg = 4.8 K ohms
135 deg = 7.0 K ohms
180 deg = 6.75 ohms
225 deg = 4.3 K ohms
270 deg = 2.7 K ohms

The specified value of R1375 is 25 K ohms, and I expect that it should range smoothly from 0 ohms to 25 K ohms.

I also verified continuity between R1375 and both the ground and +50 V rails.

I checked C1373 (the cap in parallel with VR1374) and it measured as 96.5 nF (it's specced as 0.1 uF on the schematic), so that looks good.

VR1374 is not shorted (I didn't check that it's breakdown voltage was 82 V, as I sadly have no way to do that).

This really seems to point to the trim pot as the culprit, which means I need to get the main interface board out. Joy.

-- Jeff Dutky




 

Tom,

I measure 2.6 K ohms between the two non-wiper terminals, but the pot is still in circuit, so this is really measuring everything that bridges the +50 V and ground planes in parallel, so is this a meaningful measurement?

It seems like I've eliminated every other possible suspects, and I don't have any other explanation for the measurements I've gotten for the pot resistance (but what the heck do I know? Am I wrong about what I expect to measure on a 25 K ohm pot?).

But I've come this far, and this scope was never meant to be anything other than a parts mule, so taking it down to the studs seems like the logical next step.

-- Jeff Dutky


Tom Lee
 

Hi Jeff,

I misunderstood -- I thought that all of your pot measurements were done with at least 2 terminals isolated. If you did not isolate the terminals, your measurements don't really point to anything specific.

To understand why, draw a schematic of the pot (two resistors in series, with their junction brought out as the wiper connection). Now represent the (still-connected) other stuff as resistances that connect each terminal of the pot to the other two terminals of the pot. If you stare at that for a second, you'll realize that an ohmmeter will be measuring things besides the pot itself. You'd only be able to indict the pot for sure if you measured higher than 25k. Measurements that are below 25k would be expected, as all that other stuff is in parallel with the pot.

So, you can't conclude that the pot is dead. Indeed, I would say that your pot is most likely just fine. Its resistance is varying as you rotate the wiper, and the values are below 25k. That behavior is inconsistent with the standard failure modes of a pot.

--Tom

--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
http://www-smirc.stanford.edu

On 1/2/2021 01:14, Jeff Dutky wrote:
Tom,

I measure 2.6 K ohms between the two non-wiper terminals, but the pot is still in circuit, so this is really measuring everything that bridges the +50 V and ground planes in parallel, so is this a meaningful measurement?

It seems like I've eliminated every other possible suspects, and I don't have any other explanation for the measurements I've gotten for the pot resistance (but what the heck do I know? Am I wrong about what I expect to measure on a 25 K ohm pot?).

But I've come this far, and this scope was never meant to be anything other than a parts mule, so taking it down to the studs seems like the logical next step.

-- Jeff Dutky




 

Tom,

Thanks for setting me straight. Yes, when I draw the simple schematic it's pretty obvious what's happening, and why the measured resistance of the pot has a humped profile (when you reach the middle of the range you get current flowing preferentially through the other leg). Damn. I was really hoping that I had this licked (even though it meant pulling the scope completely apart).

So I'm still at a loss to explain what's happening here. I guess I'm going to have to put everything back together, power it up, and probe around the live circuit with a scope, and compare against the sample traces in the schematic.

Tally-ho! I guess?

-- Jeff Dutky


Michael W. Lynch
 

On Sat, Jan 2, 2021 at 02:32 AM, Jeff Dutky wrote:


VR1374 is not shorted (I didn't check that it's breakdown voltage was 82 V, as
I sadly have no way to do that).
Jeff,

Can you measure the voltage drop across the diode? That is partially how I discovered the "bad" VR1374 on my 475. You should measure 82V across the diode. Here is my conversation with Chuck Harris regarding the problem:


ME: I must admit, I do not fully understand the DC Restorer. I have been reading these and other posts on the subject.

CHUCK: Have you checked that VR1374 has 82V across it? That 82V puts the grid
bias control into the range where the grid is in the cutoff region for
the CRT.

ME: YES! Testing across VR1374 I get about 66VDC. This seems at odds with your statement quoted above. This is with the + lead on the band side and the - lead on the opposite end of the diode. Reversing the leads gives a negative voltage of similar potential. So I have "Low" voltage and something like you are describing below, correct?

CHUCK: High voltage zener's and the capacitors that parallel them often start
to break down at too low of a voltage... they become leaky... and that
would put your CRT grid into the blindingly bright range.

ME: I was expecting 82Volts but perhaps I am not measuring this correctly? From your statement, 60-66VDC would put me in the "blindingly bright range", correct?

ME: CR1373 and VR1374 check "good" with a conventional diode check, using a DVM. Both show infinity one direction and .576 VDC in the other.

Like I stated previously, the scope works OK for the first 15-20 cold minutes of operation. After that it all goes nuts."

END OF QUOTED CONVERSATION.

MORE FROM SAME THREAD (after testing):

ME: "I may have answered my own question. After writing that post, I thought that I might actually be following or understanding Chuck's explanation and that I might have a bad 82V Zener or a bad capacitor. Since I had a parts scope (the one that the CRT came from) I went out and pulled C1373 and VR1374 from that unit and installed them into my "malfunctional" unit. I am testing now, but the issue seems to be resolved. In addition, I put the unknown or "Bad" VR1374 into my Type576 and 577 curve tracers. In both cases, the component would show a normal Zener curve, then suddenly break down into a crazy looking curve (looked almost like a tunnel diode curve). Not sure what was going on, however, both the 576 and 577 showed the same results."

END OF QUOTED CONVERSATION

This scenario above may or may not be relevant. But if a "Low" voltage (less than 82V) is "blindingly bright", then does it not make sense that "High" voltage (more than 82V) would cause the display to dim? I can be way wrong, but you see my line of thinking?

Good luck!
--
Michael Lynch
Dardanelle, AR