475A : excessive jitter on delayed sweep


Alberto I2PHD
 

I fixed my 475A by replacing five tantalum caps with quality electrolytics. So far so good.
But now a new problem has surfaced. Let's see if I am able to explain it correctly, with the aid of a couple photos...

I send a 1 kHz square wave signal to the 'scope, set the A sweep to 50 us/div, triggering on the negative slope of the square signal.
Doing so, I see on the screen only the lower half of the cycle (50 us/div * 10 div = 500 us, half cycle).
Then I set the B sweep at 1 us/div, "Start after the delay" (important...)

Pressing "A intensified" I see this, which is what is to be expected :

http://i2phd.org/public/A-intensified.jpg

Now I press "B delayed", and what I obtain is this, BUT ONLY AFTER A COLD START OF THE 'SCOPE :

http://i2phd.org/public/B-delayed.jpg

As long as the 'scope is cold, the image is very stable. After about ten minutes, when the 'scope has had time to warm up, the display is jittering badly, going left and right, even beyond the borders of the screen...

It is clearly a problem related to temperature... Has anybody words of wisdom about which components could cause, when warning up, this problem ?

TIA

Alberto








--
/*73 Alberto I2PHD*
Credo Ut Intelligam/


Reed Dickinson
 

Hi Alberto:

What you see on the display is probably correct.  With your scope cold, press the A push button and turn the A TIME/DIV knob such that two or more cycles are shown.  Rotate the B TIME/DIV control two or three positions past the A setting and press the A INTEN push button.  Adjust the intensity to show the highlighted portion of the A trace, rotate the DELAY TIME POSITION (DTP) control and watch the intensified portion move.  Using the DTP control position the starting portion of the highlighted A trace right on the rising edge, now press the ALT push button, if all is working OK you should see the A trace start then continue at the rising edge of the A trace but at the B rate.  Now press the B DEL'YD button, you should see the portion of the highlighted A trace swept at the B rate, you can move the display using the DTP control.

As to the 'hot jitters', with the scope hot, carefully measure each power supply with a good DC VOM, if you see no variation use another known good scope to view the various supplies, if you don't have another scope switch the VOM to AC and, observing polarity, connect a 10uF, 200VDC cap in series with the positive lead of the VOM, you are looking for low frequency variations in the supplies and the cap is used to block the DC component.  Once you find a supply that is varying the VOM reading, troubleshoot it and continue.  I have found that the 741 op-amp IC in the power supply is frequently at fault, a cheap fix!

Good luck!

Reed Dickinson

press the

On 4/15/2019 3:06 PM, Alberto I2PHD wrote:
I fixed my 475A by replacing five tantalum caps with quality electrolytics. So far so good.
But now a new problem has surfaced. Let's see if I am able to explain it correctly, with the aid of a couple photos...

I send a 1 kHz square wave signal to the 'scope, set the A sweep to 50 us/div, triggering on the negative slope of the square signal.
Doing so, I see on the screen only the lower half of the cycle (50 us/div * 10 div = 500 us, half cycle).
Then I set the B sweep at 1 us/div, "Start after the delay" (important...)

Pressing "A intensified" I see this, which is what is to be expected :

http://i2phd.org/public/A-intensified.jpg

Now I press "B delayed", and what I obtain is this, BUT ONLY AFTER A COLD START OF THE 'SCOPE :

http://i2phd.org/public/B-delayed.jpg

As long as the 'scope is cold, the image is very stable. After about ten minutes, when the 'scope has had time to warm up, the display is jittering badly, going left and right, even beyond the borders of the screen...

It is clearly a problem related to temperature... Has anybody words of wisdom about which components could cause, when warning up, this problem ?

TIA

Alberto








Alberto I2PHD
 

On 2019-04-16 1:08, Reed Dickinson wrote:
Hi Alberto:

What you see on the display is probably correct.  With your scope cold,
press the A push button and turn the A TIME/DIV knob such that two or
more cycles are shown.  Rotate the B TIME/DIV control two or three
positions past the A setting and press the A INTEN push button.  Adjust
the intensity to show the highlighted portion of the A trace, rotate the
DELAY TIME POSITION (DTP) control and watch the intensified portion
move.  Using the DTP control position the starting portion of the
highlighted A trace right on the rising edge, now press the ALT push
button, if all is working OK you should see the A trace start then
continue at the rising edge of the A trace but at the B rate.  Now press
the B DEL'YD button, you should see the portion of the highlighted A
trace swept at the B rate, you can move the display using the DTP control.
Hi Reed,

   thanks for your answer. Yes, that is exactly what I did to obtain what is shown in my photo Nr. 2.
All normal and all well understood.  But what I expected to see was a stable rising edge of the signal, as in the photo, taken with the 'scope cold, just switched on.
The problem is that when the instruments warms up, that rising edge is no more stable, the trace moves left and right, in a random fashion.
Clearly a problem related to the warming of the 'scope...

I will check, as you suggested, the voltages. There is a suspect tantalum capacitor, which I did not replace, namely C933, which is in the path that brings working voltage to the Delay Time Position control.... just as good measure, I will replace that also...

Thanks and regards

Alberto


Alberto I2PHD
 

On 2019-04-16 16:35, Alberto I2PHD wrote:

There is a suspect tantalum capacitor, which I did not replace, namely
C933, which is in the path that brings working voltage to the Delay Time Position control.... just as good measure, I
will replace that also...
Problem fixed. I replaced that suspect capacitor with a 4.7uF, 100V electrolytic, and now the delay time after which the B sweep starts is rock stable...

Now I have a suspect, which maybe somebody could confirm or deny... look at this photo of the replaced cap :

http://www.i2phd.org/public/tantalum2.2uF.jpg

Should the mark indicate the negative side of the cap, then it was soldered in reverse, with the marked side connected to +14V....
If instead it indicates the positive side, then no manufacturing errors....  I am not much familiar with the markings of the tantalum caps....

In any case, now my 475A is perfectly working, and this is what counts... :-)


--
/*73 Alberto I2PHD*
Keyboard not found : Press F1 to continue/


 

On Wed, Apr 17, 2019 at 01:23 PM, Alberto I2PHD wrote:


If instead it indicates the positive side,
It does indeed. No assembly error. You would have noticed or would notice soon if incorrectly mounted...

Raymond


 

99.44 % sure the mark on a Ta cap is the positive terminal.  If it is in backwards it will say so loudly.

Regards

On 4/17/2019 7:23 AM, Alberto I2PHD wrote:
On 2019-04-16 16:35, Alberto I2PHD wrote:

There is a suspect tantalum capacitor, which I did not replace, namely
C933, which is in the path that brings working voltage to the Delay Time Position control.... just as good measure, I
will replace that also...
Problem fixed. I replaced that suspect capacitor with a 4.7uF, 100V electrolytic, and now the delay time after which the B sweep starts is rock stable...

Now I have a suspect, which maybe somebody could confirm or deny... look at this photo of the replaced cap :

http://www.i2phd.org/public/tantalum2.2uF.jpg

Should the mark indicate the negative side of the cap, then it was soldered in reverse, with the marked side connected to +14V....
If instead it indicates the positive side, then no manufacturing errors....  I am not much familiar with the markings of the tantalum caps....

In any case, now my 475A is perfectly working, and this is what counts... :-)


Alberto I2PHD
 

On 2019-04-17 23:27, Tom Miller wrote:

99.44 % sure
You have a very well calibrated way to assess the level of your certainty... :-)
Congratulations !

--
/*73 Alberto I2PHD*
Keyboard Not Found : Press F1 to continue/


Jim Cotton
 

Tektronix tended to have a square pad on circuit boards on the positive side for the capacitors... also square pad on banded end of diodes...

Jim n8qoh

On ‎Wednesday‎, ‎April‎ ‎17‎, ‎2019‎ ‎07‎:‎23‎:‎35‎ ‎AM‎ ‎EDT, Alberto I2PHD <i2phd@weaksignals.com> wrote:

On 2019-04-16 16:35, Alberto I2PHD wrote:

There is a suspect tantalum capacitor, which I did not replace, namely
C933, which is in the path that brings working voltage to the Delay Time Position control.... just as good measure, I
will replace that also...
Problem fixed. I replaced that suspect capacitor with a 4.7uF, 100V electrolytic, and now the delay time after which the
B sweep starts is rock stable...

Now I have a suspect, which maybe somebody could confirm or deny... look at this photo of the replaced cap :

http://www.i2phd.org/public/tantalum2.2uF.jpg

Should the mark indicate the negative side of the cap, then it was soldered in reverse, with the marked side connected
to +14V....
If instead it indicates the positive side, then no manufacturing errors....  I am not much familiar with the markings of
the tantalum caps....

In any case, now my 475A is perfectly working, and this is what counts... :-)


--
/*73 Alberto I2PHD*
Keyboard not found : Press F1 to continue/