Forums :: Resources :: Features :: Photo Gallery :: Vintage Radio Shows :: Archives :: Books
Support This Site: Contributors :: Advertise


It is currently Mar Wed 20, 2019 8:05 am


All times are UTC [ DST ]





Post New Topic Post Reply  [ 10 posts ] 
Author Message
 Post subject: HP 8552B IF Section Help Requested
PostPosted: Feb Sat 02, 2019 10:55 am 
Member
User avatar

Joined: Nov Mon 02, 2009 7:01 am
Posts: 3200
Location: Lincoln City, OR
Greetings to the Forum:

I am working on my HP 8552B IF plug-in for my 141T spectrum analyzer. The original symptom was that the gain calibration could not be accomplished and the base line would be all over the screen intermittently. It now seems to have settled down a bit; the base line still won't hold still with changes in the log reference level, but the excursions seem to be less than 10 dB. However, it now will not display any signals at all when in the 2 dB log mode; 10 dB log and linear seem to work OK, subject to the calibration issue mentioned above. The manual can be downloaded from Keysight at the following URL:

https://www.keysight.com/main/techSuppo ... =FR&lc=fre

I have been looking at signals present on pin 7 of U2 (which is also the collector of Q1) on the Deflection Amplifier Assembly, A7 Board. This can be found on page 8-47 of the paper manual, which is a big fold-out. I haven't downloaded the pdf from Keysight to see exactly where to find it in there.

Here's what the signal looks like on the spectrum analyzer at 10 dB log (source internal calibrator):

Attachment:
Spec anal 10 db log.JPG
Spec anal 10 db log.JPG [ 44.29 KiB | Viewed 349 times ]


Here's what pin 7 of U2 looks like with 10 dB log selected; .2 volts / div; DC coupled; 0 volts on the center line.

Attachment:
U2 Pin 7 two tenths volt per cm 10 dB log.JPG
U2 Pin 7 two tenths volt per cm 10 dB log.JPG [ 48.45 KiB | Viewed 349 times ]


Here it is again at 1 volt per division; DC coupled; 0 volts on the center line.

Attachment:
U2 Pin 7 1 volt per cm 10 db log.JPG
U2 Pin 7 1 volt per cm 10 db log.JPG [ 49.67 KiB | Viewed 349 times ]


Here is what is present at the same point (U2 Pin 7) with 2 dB log selected; DC coupled; 1 volt per division; 0 volts on the center line:

Attachment:
U2 Pin7 1 volt per cm 2 dB log.JPG
U2 Pin7 1 volt per cm 2 dB log.JPG [ 52.67 KiB | Viewed 349 times ]


And here is what the spectrum analyzer display shows with 2 dB log selected:

Attachment:
spec anal 2 db log.JPG
spec anal 2 db log.JPG [ 45.15 KiB | Viewed 349 times ]


Relay K1 is energized by selecting 2 dB log; this is supposed to change the gain of the op-amp formed by U1 and associated circuitry from unity gain to X5. It is supposed to generate an offset as well so as to position the trace at the correct point on the display for the current log reference and attenuator settings. However, it is driving the signal to almost 3 volts positive which cuts off the following stage.

I'm not good enough on circuit analysis to figure out how U1 should behave and whether or not this magnitude of a voltage change should or should not cut off U2. I suspect that since it only seems to take a volt of signal at U2 pin 7 to accomplish full deflection of the spectrum analyzer display that a three volt change is unreasonable.... but I don't know for sure. Assuming that 3 volts DC is too much, where should I look to find this problem? Bear in mind that the amplifier "almost" works right at 10 dB log.

Any suggestions welcome.

Thanks,

_________________
Jim T.
KB6GM


Top
 Profile  
 
 Post subject: Re: HP 8552B IF Section Help Requested
PostPosted: Feb Sat 02, 2019 9:32 pm 
Member

Joined: Jul Wed 22, 2009 3:07 pm
Posts: 757
Hi Jim,

I have not dug into the details of the circuit, and your problem (I probably will), but before you do anything do check the all the supply voltages in the main frame!

I recently got a tracking generator for the 110MHz RF-section that I have in my system, and it also came with a "free SA" (the whole set was cheap). That SA worked for all about 3 minutes for me, after that nothing on the screen.
Some of the PSs were shorted out, but the ones that worked were as much as 20% out of spec, giving 80V instead of 100V.
The resistors that were dividing down the output voltage for the error amplifiers (2W metal film resistors), a number of those were totally out of spec. Those resistors must have been creeping out of spec for quite some time, once the PSs were up to standards the vertical and horizontal positions and gains needed quite a bit of adjustments. They must have continuously corrected with those before the whole thing quit.

EDIT:
Looked at the schematics quickly.
The K1 relay also adds offset to the circuit originating from the -100V supply when in 2dB mode.
Does the R35 trimpot brings back the signal to the screen?
The schematics that I have has a scribbled DC voltage for the U3 opamp +input (pin 3), and it is around 0V. So it seems you should have around 0V at the collector of Q1.
Is that what you get when in 10dB mode?

These are just initial thoughts, so I might be wrong too...

Regards, Peter


Top
 Profile  
 
 Post subject: Re: HP 8552B IF Section Help Requested
PostPosted: Feb Sat 02, 2019 11:53 pm 
Member
User avatar

Joined: Oct Sun 15, 2006 12:57 pm
Posts: 6184
Location: Liberty, Missouri
You mentioned the base line moving around and problems with the 2db range. In my experience mechanical contacts, which this unit has in excess, are the most common problem with any kind of "moving target" be it amplitude, time, or any other "selected parameter".

Before you proceed further I recommend you frantically exercise each and every switch and pot at least ten times, from one extreme to the other. Although certainly no expert, I've worked on several 141s and associated plug-ins, and dirty contacts were always a big, if not the biggest contributor to intermittent and "mystery" performance issues.

_________________
Pics, Bold & underlined text, are usually links.

https://mikeharrison.smugmug.com/

Mike


Top
 Profile  
 
 Post subject: Re: HP 8552B IF Section Help Requested
PostPosted: Feb Sun 03, 2019 6:53 am 
Member
User avatar

Joined: Nov Mon 02, 2009 7:01 am
Posts: 3200
Location: Lincoln City, OR
Greetings to Peter, Mike and the Forum:

Thanks for the replies. I appreciate your time. For Peter:

The voltage on Q1's collector in the 10 dB log mode is essentially zero volts with the signal (spectral peaks) negative-going to around around - 1/2 volt for the largest magnitude peak. This is shown in the second waveform photo that I posted; the scope was DC coupled for this image at .2 volts per division.

Neither the 2dB log gain pot (R37) nor the 2 dB log offset pot (R35) have any effect whatsoever on the output in the 2dB log position. The forth waveform that I posted, labeled " U2 Pin7 1 volt per cm 2 dB log.JPG" is the voltage at the collector of Q1 in the 2 dB log position. (U2 Pin 7 and Q1 collector are tied together.) The transistor inside U2 that is connected to pins 6,7 and 8 of U2 is described in the service note as being connected as a diode to "raise the signal level 0.7 volt". In the 10 dB log position, the video signal passes through this device and appears on pin 8 of U2. In the 2 dB log position, no signal passes through this device; pin 8 is a flat line. I do not recall the DC voltage of this flat line; I will have to go back and measure it. However, it would seem that taking the emitter of this internal transistor in U2 3 volts positive could well be sufficient to cut off the emitter-base junction and thus completely remove video from passing any further.

The fact that R35 (offset) has no effect should have rung some bells last night; I will have to go look for the presence of the -100 volts and see if there is some problem with the feedback divider circuit.

For Mike: Yes... the Log reference Level switch is flaky in one position. I will try to get in and clean or at least exercise the controls and switches. Unfortunately, I have not been as fortunate as you in the problems that crop up in my 141T. I have had to replace a transistor in the sweep trigger circuitry. Also, I discovered that if you have a failure of one of the lamps that light up the Log Reference Level fine control (X or +) that the 2dB log relay control no longer works.... the transistor is biased through those lamps. I have also had to replace at least two electrolytic caps, one on a board somewhere (I don't recall where any more) and one in the power supply in the 141T main frame.

Each of these problems has led me a merry chase, so I am becoming disenchanted with the maintenance man-hours per flight hour. So, anything that can help me to zero in on the problem more quickly is greatly appreciated... so thanks again for your input, as well as Peter's and anyone else who wants to chime in.

73,


Each of these problems

_________________
Jim T.
KB6GM


Top
 Profile  
 
 Post subject: Re: HP 8552B IF Section Help Requested
PostPosted: Feb Fri 08, 2019 6:47 pm 
Member
User avatar

Joined: Nov Mon 02, 2009 7:01 am
Posts: 3200
Location: Lincoln City, OR
Greetings to the Forum:

It has been so cold here lately that I just can't bring myself to venture out into my unheated garage. I did for a bit and wrote down some voltage measurements around U2. I think this chip might have a problem. I am starting to think that the 3 volt change on the input to U2 at pin 7 might be the legitimate offset voltage for 2 dB log. If so, the input diode connected transistor ( pins 6,7 and 8 ) needs a bit more voltage on its base and collector to be able to pass the video. This voltage is set by a pull-up resistor and a pair of transistors in U2. The conduction of these transistors is established by the voltage applied by the relay driver transistor that controls the 2dB / 10 dB log switching of U1. The voltages on these transistors do not look right to me. I would appreciate comments. Here are the voltages as measured in 2 dB log (failed) and 10 db log (working) conditions:

Pin ..... 2dB ....... 10 dB
4 ......... .73 ........ -6.9
3 ......... 0 ........... .21
2 ......... .69 ........ .62
1 ......... .69 ........ .62

I am guessing that the transistor represented by pins 1,2 and 3 of U2 is shorted. Any thoughts?

Thanks,

_________________
Jim T.
KB6GM


Top
 Profile  
 
 Post subject: Re: HP 8552B IF Section Help Requested
PostPosted: Feb Sat 09, 2019 3:51 pm 
Member

Joined: Jul Wed 22, 2009 3:07 pm
Posts: 757
Hi Jim,

I have the sneaking suspicion that that IC (U2, and the transistors in it) are fine in your circuit!
But I will get to that later!

The first question is that how is that 2dB mode supposed to work?
Looking through the manual briefly I did not find any reference to it, neither "performance check and adjustment" instructions to it.

When in 2dB mode, there is an additional 5-times amplification in the vertical display, after the log conversion. It means that it is very easy to move any signal off the top of the screen, basically over driving it. That is what the "2dB offset" adjustment pot is supposed to address.
But that is only a set-up adjustment (a trimmer pot) and not an operational control!
So it seems that the 2dB mode is just a "magnifying glass" when the signal is quite small.
That would suggest that the 2dB offset adjustment is for adjusting the base line when there is no input signal. But any higher signal (8 graticules, 8 X 2dB = 16dB above the bottom level) would be off the top of the screen.

Contrast to this offset adjustment the 2dB mode offset adjustment in the W7ZOI SA:
http://w7zoi.net/sa-stuff.html
They produced a circuit modification for 2dB display, and in that circuit the offset adjustment are two pots, coarse and fine, and those are operational controls. That means that a signal at any level in the 2dB display mode can be brought back to the screen. Of course absolute level calibration would be off!

Interestingly, when looking at my 141, when displaying the -30dBm cal signal, and switching to 2dB display mode, the signal disappears at the bottom of the screen!
I can bring back the tip of the pip with that 12dB fine IF-gain adjustment pot on the 8552B!
I do not think that is right!
It is possible that the 2dB offset was (mis-)adjusted when the 100V PSs were off in my SA!

Going back to your circuit and why I do think that it is OK.
First of all, what you call pin3, I think it is pin 3 at U3, and that is pin 10 at U2!
That point is the input to the vertical pre-amplifier, and I think it should be at 0V in both cases when there is no input signal!
Your readings are close to that!
I wonder if you could make it the same with the 2dB offset adjustment pot?

I also plugged that part of the circuit into the simulator and it gave roughly the same voltages.
But you have not listed every voltages on the peripheries of that circuit (I made assumptions there), and I have not followed those through the pages of the circuit diagrams.

Regards, Peter


Top
 Profile  
 
 Post subject: Re: HP 8552B IF Section Help Requested
PostPosted: Feb Tue 12, 2019 5:28 pm 
Member

Joined: Jul Wed 22, 2009 3:07 pm
Posts: 757
Hi Jim,

I did check the 2dB display mode in the manual of another SA (Polarad 632A) where it is spelled out in details.
It means magnification when the signal display is at the top graticule!
Knowing this there is a hint of this in the specs of the 8552B:
The specs for the "Absolute Amplitude Calibration Range:" reads "... or 2dB/div expand below LOG reference."

Based on this the adjustment of the 2dB offset pot would be as follows:
Bring the pip of a signal to the top reference level in the 10dB/div mode.
Switch over to 2dB/div mode and the top of the pip should be still at the top reference level, or if not, bring it to there with the 2dB offset pot.

My SA nearly works this way, it could use a bit of adjustment.
Unfortunately, it does not look like that the adjustment is brought out to the front panel.
I guess I would have to find the location of that pot first, but then I do not have the "service connection cables and cards" to service the modules out of the main frame.

Hope this helps.

Regards, Peter


Top
 Profile  
 
 Post subject: Re: HP 8552B IF Section Help Requested
PostPosted: Feb Thu 14, 2019 1:17 am 
Member

Joined: Feb Sun 17, 2008 11:36 pm
Posts: 1970
Location: Dayton, Ohio
It has been more than 20 years since I worked on a h/p 8552, so I don't know if this will help; the incandescent lamps for the + and -, (iirc), must be working, to supply a bias for the Log/Linear circuits. The lamps for the "LOG REF LEVEL" lamps may also have to work. Can't remember for sure.

It's been too long ago for me to be more specific.

Good luck,

Charlie


Top
 Profile  
 
 Post subject: Re: HP 8552B IF Section Help Requested
PostPosted: Feb Fri 22, 2019 5:40 am 
Member
User avatar

Joined: Nov Mon 02, 2009 7:01 am
Posts: 3200
Location: Lincoln City, OR
Greetings to Mike and the Forum:

I suspect that your comment about switches needing exercising was the ticket. Other than this, I can only resort to the note we used to use all the time when I worked for the phone company: "Came clear while testing."

The original problem was that the unit could not be calibrated. When I ran into the strange 2 dB log behavior, I assumed that this might be a short cut to finding the non-calibratable problem. Well... while poking around at it last night, I was suddenly able to see the 2 dB log signals off the top of the screen, whereas before, they were not visible at all. Since I had twidled the 2 dB log cal and offset pots while attempting to observe the behavior of the circuit, I now had to put those back. It turns out that the "offset" pot is supposed to adjusted so as to cause a signal set to the log reference level in the 10 dB log position to also be at the log reference level when switched to 2 dB log. The gain is set to cause the signal to move down 5 divisions when the log reference level is increased by 10 dB.

In practice, since both pots affect the gain considerably, I had more luck setting the offset with the gain pot and the gain with the offset pot once I got close. Anyway, it is now working again. Whether it will still be working when the next Rickreal swap meet comes around in October is another question entirely.

A note for Charlie: I had already run into the lamp problem a few years back. You are correct... as I recall, the base current for the relay driver transistor that switches between 2 dB log and 10 dB log is supplied through one of these lamps... anyway they both have to be working in order for the circuit to operate correctly. I believe the lamps in question are for the fine reference control and are "x" and "+" respectively.

Anyway, it is working again and I now have some bench space. Thanks to all who responded. Sorry I can't be more specific about what I did to fix it (if anything) but that's the way it goes sometimes.

Regards,

_________________
Jim T.
KB6GM


Top
 Profile  
 
 Post subject: Re: HP 8552B IF Section Help Requested
PostPosted: Feb Fri 22, 2019 1:05 pm 
Member
User avatar

Joined: Oct Sun 15, 2006 12:57 pm
Posts: 6184
Location: Liberty, Missouri
Quote:
... Sorry I can't be more specific about what I did to fix it (if anything) but that's the way it goes sometimes....
You're absolutely right, just smile and take the "win".

Glad to see you got it running.

_________________
Pics, Bold & underlined text, are usually links.

https://mikeharrison.smugmug.com/

Mike


Top
 Profile  
 
Post New Topic Post Reply  [ 10 posts ] 

All times are UTC [ DST ]


Who is online

Users browsing this forum: Bob_Go and 6 guests



Search for:
Jump to:  


























Privacy Policy :: Powered by phpBB