开云体育

I have a Tek 453 that was given to me like 18 years ago. I used it to look at waveforms of guitar pedals I made and that sort of thing. It had sat unused for probably 10 years until I decided to recently start using it again. The "A Sweep Trig'd" indicator lamp was burned out and I replaced it with an LED instead. Was a very simple mod and it's been working great. The scope has been working OK for the most part, but it has a few problems that I have not been able to figure out. I'm hoping that someone here might know what might be the cause of these issues.

First one is the trace is unstable for about 4-5 minutes when turned on after not being on for a while. Hard to explain but it's kind of jumping around/vigorously spasming but staying in generally the same place (it's not jumping around all over the screen). With auto trigger on and no input, If I push the trace finder button, the bottom of the image is kinda going crazy. This goes away after it warms up for a few minutes and is quite stable. As far as I can tell the power supply is OK unless it's one of those big can caps under the B Sweep PCB. Is there another place in the circuitry that could cause that? When I first got the scope it was very stable when I turned it on.

Second one is that the right end of the sweep is a little unstable. In other words, the left end of the waveform is very stable but the right end is constantly shifting around a little, the end of the waveform is not stable and staying in one place. Doesn't matter if I'm using A or B Sweep modes.

Third one which is the most perplexing is that I have noise on the waveform that is like a cyclical fishbone pattern. This doesn't really show up until you use the Mag feature on the Horizontal Display switch and when in EXT HORIZ mode. That is where it is very annoying because you have this nice Lissajous waveform with a fuzzy halo behind it. With no input I get a dot with a line extending left and right from the dot about 1/8" each direction. I've been thru the entire Low Voltage Power Supply, Horizontal Amplifier, Vertical Amplifier, Vertical Switching, Tigger Pickoff, Trigger Preamp, B Trigger Generator, B Sweep Generator, A Trigger Generator and Z Axis Amplifier. I found a few transistors that were out of spec and replaced them with good ones off PCBs from a donor scope. I also checked the CRT Grid Bias, etc. Nothing I have done has rectified this. It sure does seem like noise coming from somewhere but I have no idea from where. Could it be the CRT itself? It seems to be working very well... it will get intensely bright, I can get a very sharp image, etc. This problem only shows up when in EXT HOR or MAG modes. What the heck could be causing this? Switching noise from a dodgy tunnel diode? Or a noisy transistor in the CRT HV Oscillator? I didn't find any voltages that were off in that circuit either.

Have you checked the LVPS power rails for excess ripple? You can have "good" voltages and still have excess ripple. This will drive the circuitry to do some crazy things.

--
Michael Lynch
Dardanelle, AR

After long storage times, old gear tends to get flaky due to mechanical and chemical deterioration in the switches, pots, for instance. A lot of problems can be cleared up by working over all of the controls through their ranges, a good number of times - you'll be able to tell as you go if things improve, and whether certain things need more exercise. Do this kind of thing before opening anything up. Chances are that a lot of issues will clear up. The ones that remain may be caused by the same things, but internally, so then look at the internal controls (like pots), connectors, and device sockets.

Ed

diyguitarist,

The two gentlemen, Michael L. and Ed B., are absolutely right. Others will have great advice as the post. I have the 453 and 4/A series I have restored. I did restuff the main filters, replaced the 1mfd axial decoupling with film and a couple of tantalums of higher capacitance to give higher filtering, used Nichicon UHE 220mfd 25V as decoupling for the two 12V supplies and higher capacitance for the 75V decoupling. I did replace carbons with 1% types and raised the wattage of some resistors. The triggering was adjusted as well as other circuits to get it right after resistor and condenser replacement. I did read that raising the resistors in the high voltage area to 1W should be done. I did this in mine. I did use 1% types as that is the tolerance I use. One thing I do suggest is to oil set screws and shafts. Experience taught me this. What you did to replace the trigger lamp with an LED is fine. I did this in one of mine with a 4300 ohm resistor to the white LED because the lamp was burned out. My LED is about 2,5mA operating and is bright enough. I got mine cleaned up and cleaned the crt because I wanted mine to look as nice as possible. If you want any further help, we are here. What I said I did is what I did to mine. You do what you want with yours.

Mark

On Mon, Apr 4, 2022 at 01:25 PM, Michael W. Lynch wrote:

Have you checked the LVPS power rails for excess ripple? You can have "good"
voltages and still have excess ripple. This will drive the circuitry to do
some crazy things.

--
Michael Lynch
Dardanelle, AR

Yeah, I have 10mV on the +12V & -12V supplies. I still need to check the 75V supply one more time, but as I recall it was nothing significant.

On Mon, Apr 4, 2022 at 05:05 PM, Ed Breya wrote:

After long storage times, old gear tends to get flaky due to mechanical and
chemical deterioration in the switches, pots, for instance. A lot of problems
can be cleared up by working over all of the controls through their ranges, a
good number of times - you'll be able to tell as you go if things improve, and
whether certain things need more exercise. Do this kind of thing before
opening anything up. Chances are that a lot of issues will clear up. The ones
that remain may be caused by the same things, but internally, so then look at
the internal controls (like pots), connectors, and device sockets.

Ed

Yeah I've went thru all of the controls & switches. None of them have a problem as far as I can tell. I did notice that if I put the Horizontal Display switch between DELAYED SWEEP (B) and EXT HORIZ that I get a dot without the line, but there is no horizontal deflection and it won't display a waveform. I've been trying to figure out what that may indicate but I'm still stumped.

On Mon, Apr 4, 2022 at 07:23 PM, Mark Vincent wrote:

diyguitarist,

The two gentlemen, Michael L. and Ed B., are absolutely right. Others will
have great advice as the post. I have the 453 and 4/A series I have restored.
I did restuff the main filters, replaced the 1mfd axial decoupling with film
and a couple of tantalums of higher capacitance to give higher filtering, used
Nichicon UHE 220mfd 25V as decoupling for the two 12V supplies and higher
capacitance for the 75V decoupling. I did replace carbons with 1% types and
raised the wattage of some resistors. The triggering was adjusted as well as
other circuits to get it right after resistor and condenser replacement. I did
read that raising the resistors in the high voltage area to 1W should be done.
I did this in mine. I did use 1% types as that is the tolerance I use. One
thing I do suggest is to oil set screws and shafts. Experience taught me this.
What you did to replace the trigger lamp with an LED is fine. I did this in
one of mine with a 4300 ohm resistor to the white LED because the lamp was
burned out. My LED is about 2,5mA operating and is bright enough. I got mine
cleaned up and cleaned the crt because I wanted mine to look as nice as
possible. If you want any further help, we are here. What I said I did is what
I did to mine. You do what you want with yours.

Mark

Since it's a unit from 1969 (serial #37389) I have considered committing re-cappicide. I really don't like the design of the whole can capacitor arrangement. It appears to me that you have to dismantle half the scope in order to replace those. That's not a road that I really want to go down if I can help it.

I have made progress on one of the issues. The instability on start up, that was due to the wrong transistor being used for Q1189. The transistor that was in there apparently could not handle the voltage it was seeing (the data for it sheet said it was rated for only 75V). Manual says that a 2N5184 is a replacement for Tektronix 151-0250-00 but I don't have any of those lying around but I did have an NTE equivalent (NTE 128) from my donor scope, so I put it in there to see what happened. Now it's stable from the second I turn it on! Interesting because Q1189 is really only there for short circuit protection, but I guess because it's connected to the regulated 75V and unregulated 150V supplies, it must mess with all of the circuits connected to those power supplies until it warms up.

The weird fishbone thing is still going on when in MAG and EXT HORIZ modes though. I tried to get a good video of it but my iphone won't sync well and the video is terrible but here is a 10 second video showing two pictures of it. It travels cyclically from left to right along the waveform. Don't worry, the CRT isn't as bad as it looks, its just my makeshift CRT cover made from a cassette tape that my iphone can't get a good picture of either. I need to get one of those original smoke grey CRT covers but they seem to be non-existent items on ebay. :-(

UPDATE: I have looked at basically every circuit in the scope at this point. I have suspected that maybe there is an oscillation problem somewhere and even made a "snooping loop" as described in this Tektronix document but haven't found anything conclusive (I think).

So I can not find anything wrong other than two test points that seem to be backwards from the schematic. They are in the A Trigger Generator, test points TP485 & TP494.
Having all of the scope controls set per Page 1 of the manual schematics, TP485 should be +12.5V and measures -0.32V & TP494 should be -0.98V and it measures +11.42V. These are almost exactly opposite of the schematic. Is there is a mistake on the schematic or do I have a legitimate problem here? This strange problem does definitely seem to be somehow related to the sweep, so I imagine it's more likely I have problem than the schematic being wrong. Interestingly, the snooping loop detects some activity in this same area and nowhere else except for the Z Axis PCB, where there is an oscillator for the CRT HV supply. However, if the problem is with A Trigger circuit, and if A & B sweep is grounded in Ext Horiz mode, then how can I possibly be getting this fuzzy halo around the waveform in X-Y mode? This is what confuses me.

Additionally, I did notice one other interesting thing: Although I cannot get just a dot in Ext Horiz mode (I get a dot with an 1/8" line extending horizontally in both directions, about 1/4" total), I can get a beautifully round one in A & B sweep modes using the Norm Trigger setting with no input, if I turn the intensity up enough and move the dot into view with the Horizontal Position control.

Not sure what I should looking at here... I'm used to working with positive only voltages and a ground. Any suggestions?

I checked one of my 453s that is above 20,000. The TP voltages in the schematic are right. Check the transistors and diodes in that area. The 3,3mfd condenser could be bad. I have had ones with high ESR I replaced with Nichicon ULD type. R473, R674 and R483 I raised to 1 and 2W respectively. Yours could be high in resistance. The originals are 1/2W carbons. The tunnel diode will be good if you can get a signal to trigger.

The trace problems you describe sound like the astig and geometry(?) need adjusting.

Mark

On Tue, Apr 5, 2022 at 07:18 AM, diyguitarist wrote:

I did notice that if I put the Horizontal
Display switch between DELAYED SWEEP (B) and EXT HORIZ that I get a dot
without the line, but there is no horizontal deflection and it won't display a
waveform. I've been trying to figure out what that may indicate but I'm still
stumped.

After latest fixes is above behavior still there? When the switch is in between nothing is connected to horizontal amplifier. If you are seeing a dot in when switch is in between the issue is before horizontal amplifier. EXT HORIZ amplifier (sheet <10>) only has +12V and -12V supplies. -12V transformer is shared with the HV oscillator, HV oscillator might be coupling.

I see that you already checked but you may want to double check the ripple on all the supplies, especially +75V, +12V and -12V, with a scope in AC coupled mode. Note that ripple spec is 2mVpp.

C937 on sheet <16>, C1142 and C1151 on sheet <13> could be bad if there is ripple on -12V.

Other possibility is sharing noisy equipment on the same power strip, check the scope in another room in another power outlet.

Ozan

On Thu, Apr 21, 2022 at 09:50 PM, diyguitarist wrote:

, TP485 should be +12.5V and measures -0.32V & TP494 should be -0.98V and it
measures +11.42V.

1. Those voltages are correct if you measure when the sweep is trigggered. They revert to the schematic values with a free-runing trace.
2. On my 453 (c 1970) the ripple & noise on the +12V and -12V rails, measured in the B trigger area (R667 and R636) relative to the local ground is in both cases around 3mV p-p.
3. There are two X-Y modes, not sure which you're using?
4. In both X-Y modes, the horizontal x10 Mag switch should not do anything as it is shorted out by the wafer 2R switch (shown on the X amp schematic 13).
John

On Thu, Apr 21, 2022 at 06:23 PM, Mark Vincent wrote:

I checked one of my 453s that is above 20,000. The TP voltages in the
schematic are right. Check the transistors and diodes in that area. The 3,3mfd
condenser could be bad. I have had ones with high ESR I replaced with Nichicon
ULD type. R473, R674 and R483 I raised to 1 and 2W respectively. Yours could
be high in resistance. The originals are 1/2W carbons. The tunnel diode will
be good if you can get a signal to trigger.

The trace problems you describe sound like the astig and geometry(?) need
adjusting.

Mark

Thanks for doing that. I looked at that whole area and I didn't find any bad components. I have also suspected that maybe C485 (3.3uF) was bad. It tests as open with ESR meter and ha a very funny looking curve when looking at it with an "octopus" curve tracer. I replaced it but none of those voltages changed. Apparently, per a post after your, the voltages I measured are correct IF IT IS TRIGGERED. The schematics make it appear like you have to use the conditions on Page 1 of the schematics, but I guess I did not interpret that correctly.

Anyway, the astig and geometry are fine. When I'm using the scope in any other mode besides Mag (x10) and Ext. Horiz. the traces look pretty good except that in certain positions of the Time/Div switch you can see a little noise on the waveform, but it's still very usable. You can also see it with the beam finder too. Still have not found where it's coming from.

On Thu, Apr 21, 2022 at 06:41 PM, Ozan wrote:

On Tue, Apr 5, 2022 at 07:18 AM, diyguitarist wrote:

I did notice that if I put the Horizontal
Display switch between DELAYED SWEEP (B) and EXT HORIZ that I get a dot
without the line, but there is no horizontal deflection and it won't display

a

waveform. I've been trying to figure out what that may indicate but I'm

still

stumped.

After latest fixes is above behavior still there? When the switch is in
between nothing is connected to horizontal amplifier. If you are seeing a dot
in when switch is in between the issue is before horizontal amplifier. EXT
HORIZ amplifier (sheet <10>) only has +12V and -12V supplies. -12V transformer
is shared with the HV oscillator, HV oscillator might be coupling.

I see that you already checked but you may want to double check the ripple on
all the supplies, especially +75V, +12V and -12V, with a scope in AC coupled
mode. Note that ripple spec is 2mVpp.

C937 on sheet <16>, C1142 and C1151 on sheet <13> could be bad if there is
ripple on -12V.

Other possibility is sharing noisy equipment on the same power strip, check
the scope in another room in another power outlet.

Ozan

I don't think it's a PS ripple issue. I had a thought that maybe C937 was bad but it is OK. All of those LV PS caps test good with ESR meter and even though I measure around 10mV of ripple I don't think that could be causing this problem. I'm generally only looking at stuff in audio frequency range and not using it for very precise measurements where a little ripple would be a big issue. I just want to be able to use Ext Horiz mode and not have the fuzzy halo around it.

Don't think it's noise from other equipment as I have a second time capsule 453 I picked up a few weeks ago that works perfectly in same environment. This seems to be a pulse that is coming from somewhere. In x10 mode, with certain settings of the Time/Div switch, you can watch if go in cycles. It almost goes away, then grows to a max width which is about the same width as my line with the dot in X-Y mode (like a side view of Saturn) and then it diminishes to almost nothing. Then it does it again. So it's a cyclical repeating pattern, going from one side of the CRT to the other. I want to say that it's diode noise but I am not sure what that would look like in an oscilloscope.

If it were HV oscillator coupling, what would be the mechanism by which that is happening? Also, although fuse F937 is not blowing, what would happen if that 500pF "door knob capacitor (C953) was a little leaky? Could that cause something like this noise to happen?

Do you have second scope? if so, you can identify the nature of the problem by putting the 453 into X-Y mode with your quarter-inch "dot", then looking at the voltage on the X plates with the second scope. If it looks like the "noise" on either +12 or -12 supplies, you know where to start looking, If it's high-frequency coherent, there's only the EHT oscilator in the 453 which runs up around 30kHz. I assume you have the lid in place on the EHT section? C937 is the main decoupling for Q930 and is located behind the rear panel across a couple of the old-style ceramic tag strips. NB you've not said which X-Y mode you're using?
John

On Mon, Apr 25, 2022 at 04:04 AM, John Sykes wrote:

On Thu, Apr 21, 2022 at 09:50 PM, diyguitarist wrote:

, TP485 should be +12.5V and measures -0.32V & TP494 should be -0.98V and it
measures +11.42V.

1. Those voltages are correct if you measure when the sweep is trigggered.
They revert to the schematic values with a free-runing trace.
2. On my 453 (c 1970) the ripple & noise on the +12V and -12V rails, measured
in the B trigger area (R667 and R636) relative to the local ground is in both
cases around 3mV p-p.
3. There are two X-Y modes, not sure which you're using?
4. In both X-Y modes, the horizontal x10 Mag switch should not do anything as
it is shorted out by the wafer 2R switch (shown on the X amp schematic 13).
John

OK, well that explains why the waveforms at TP485 & TP494 look right but the voltages were "wrong". The schematic doesn't really make that clear about how to measure that. Was under impression that it was values a given on Page 1 of the fold out schematics. (I have a PDF that can be a little hard to read where the blue parts are).

The X-Y mode I am using using is feeding signal to the Ext. Trigger / Ext Horiz. jack and signal to Channel 1 jack. It doesn't matter how the B source or coupling switches are set, still have the same problem.

I checked all of the connections on Horiz Display switch when in Ext Horiz mode, and found nothing wrong with the switch - all connections are good. That is correct, the X10 switch does not do anything when using Ext Horiz mode, but per the Circuit Descriptions: "In the EXT HORIZ position of the HORIZ DISPLAY switch, the magnifier is connected into the circuit so the horizontal gain is correct for external horizontal operation regardless of the setting of the MAG switch." So the two things seem to be related to each other here, and these are the two functions on the scope where it's not working properly. On a second time capsule 453 I picked up a few weeks ago it works perfectly using it same way.

On Mon, Apr 25, 2022 at 07:59 AM, John Sykes wrote:

Do you have second scope? if so, you can identify the nature of the problem
by putting the 453 into X-Y mode with your quarter-inch "dot", then looking at
the voltage on the X plates with the second scope. If it looks like the
"noise" on either +12 or -12 supplies, you know where to start looking, If
it's high-frequency coherent, there's only the EHT oscilator in the 453 which
runs up around 30kHz. I assume you have the lid in place on the EHT section?
C937 is the main decoupling for Q930 and is located behind the rear panel
across a couple of the old-style ceramic tag strips. NB you've not said which
X-Y mode you're using?
John

I do have a second scope. I could give that a try. I looked at C937 yesterday, tested good with ESR meter. Not familiar with the term "EHT" but I have back cover on, the shield on the HV section is also on and is making a good connection to ground. Only covers off are the top and bottom covers for taking measurements. Still does it when everything is all buttoned up too.

Looking at the X plate signal will give a clue to what's going on and reduce the time spent looking for the culprit.
In the mode you're using, the X signal is travelling through the B trigger gen board (on [10]), so you could also have a look to see if the output from there (on pin CJ) is clean.
John

On Mon, Apr 25, 2022 at 07:49 AM, diyguitarist wrote:

If it were HV oscillator coupling, what would be the mechanism by which that
is happening?

The mechanism would be: HV supply feeding into F937 comes from the same secondary winding of the transformer as +12V. You can see the connection across C1142 in sheet <17>. A possibility is one of the caps I mentioned in the previous message (C937 on sheet <16>, C1142 and C1151 on sheet <13>) could be bad, not filtering the HV ripple on the supply, and it is leaking into +12V supply.

Also, although fuse F937 is not blowing, what would happen if
that 500pF "door knob capacitor (C953) was a little leaky? Could that cause
something like this noise to happen?

If your previous observation, pushing display switch between DELAYED SWEEP (B) and EXT HORIZ giving you a clean dot, is still correct the coupling has to be before the horizontal amplifier. We expect some issue downstream, like C953, should still be there even when input to horizontal amp is disconnected (switch at not fully engaged to any input).

If you increase/decrease intensity of the trace, does it change the coupling signal pattern?

Is +75V clean? Any ripple on +75V will show up in +12V and -12V.

Does 10mV ripple has the same burst like pattern, or do you see 2x line frequency ?

Ozan

If the Delayed Sweep switch is not in EXT HORIZ then the x10 mag will not engage. It's when the X amp is in magnified mode that he sees the noise.