HP Agilent Spectrum Analyzer 8595E CAL AMPTD error

[charlyd]

hello after i got my 8595E in a few days ago, and did some test i got  a NBW gain failed error in CAL AMPTD.    my optionlist is 004,041,103,130.

i guess NBW is related to option 130  what can i do to fix this?  ( last night the cal. AMPTD. was ok  one time and 15 min later failed.)
unit gives no errors at startup and no freq. uncal or something



my SG stops at 2100Mhz and apart from what i see on youtube from an other 8595E Sa,  is that my BAND 1 looks bad/dead.   is there a way todo a real test of band 1  2.75GHz - 6.5GHz 
i read on this forum to use my 5GHZ wireless and connect.. is that a good test?

[charlyd]

For the second part of my question the YIG or YTF driver. i found a troubleshoot part in the manual for the YTF driver. chapter 4 verify the YTF driver.

Are there simple measurement checks i can do, on or arround the YIG. to see if these are what they should be? as a diagnose.

[Miti]

The display looks OK.

[charlyd]

hi miti, thanks for your response. i think in the middle picture i miss some harmonics..about 2 div.  amptd is overall slightly low.

[analogRF]

hello after i got my 8595E in a few days ago, and did some test i got  a NBW gain failed error in CAL AMPTD.    my optionlist is 004,041,103,130.

i guess NBW is related to option 130  what can i do to fix this?  ( last night the cal. AMPTD. was ok  one time and 15 min later failed.)
unit gives no errors at startup and no freq. uncal or something



my SG stops at 2100Mhz and apart from what i see on youtube from an other 8595E Sa,  is that my BAND 1 looks bad/dead.   is there a way todo a real test of band 1  2.75GHz - 6.5GHz 
i read on this forum to use my 5GHZ wireless and connect.. is that a good test?

the picture does not show how your band 1 is. It looks normal. The harmonics of cal signal above 2.9GHz are very small.
you calibration values dont look ok. see the service manual, there you can see a sample of normal cal constants.

if the calibration fails intermittently (passes sometimes and fails sometimes) it is likely that your attenuator has bad contacts. I have had that problem in 8596E and 8595E, both cases it was the attenuator which is repairable  the amplitude is way too low in that calibration picture.
try to play the attenuator contacts several times and then run the cal again
However, it can also be something wrong in the RBW filters...

your LO is fine, dont waste time on troubleshooting the YTO and specially not the YTO driver. If there is a hole in the YIG sweep, you can search for it with a small span and moving the center freq, if YTO amplitude goes down at some point you will get freq uncal or unlock error. I am pretty sure our YTO is fine...

"amplitude out of spec for >3GHz" can mean a lot of things, it does not necessarily mean SYTF is bad. You need to first observe what the "out of spec" is.

[charlyd]

this is the pictures after i [push] preset. yes it is low all over the bands from what i found on youtube..  see pictures attached to this post.

ok my attenuator have bad contacts you say, that can be a good check.. you are talking about A3A5 i assume..
the attenuator which is repairable.. 

yes indead the ampltd is very low if i compare again...

so is it that you try to tell me that if the YTO or driver fails /is wrong, i should get one of these errors:   freq uncal or unlock error.
and confidence test is also succesfull.

i dont get these that s true.!

@analogRF you say : try to play the attenuator contacts several times and then run the cal again .  did you open the attenuator itself? if yes what is there to do maybe?
                              or do i understand you wrong?

i was looking in the manuals i have but where do i get the cals examples where you talk about above  you said : you calibration values dont look ok. see the service manual, there you can see a sample of normal cal constants

[analogRF]

the amplitude is not low all over the band. The 300MHz is right at -20dB as it should be. the first few harmonics are also good
the small differences are due to calibration. your cal constants dont look normal, compare with sample values in the service manual.

it does seem your high band might be dead or very off tuned. still it does not necessary mean the SYTF driver is bad, in fact most of the time
it is the SYTF itself that is faulty or it could be out of cal. You do need a sig gen that goes beyond 3GHz, even a cheap ebay synthesizer board would help. Just playing with the SYTF driver when you dont have any means to check signals in that band is waste of time


still I would suggest you first make sure the attenuator is ok and consistent and then try to fix the issue of AMP CAL first (that should be fixed
even if the high band is bad). Just play with the attenuator many times when you are at 300MHz center freq and cal signal connected. If the amplitude remains constant by changing attenuator, it means it is working ok. Search the forum for attenuator repair, also there are videos on youtube. If you are very careful you can open the attenuator and clean the contacts and change the tiny rubber o-rings that get stiff or even break apart by aging.

[analogRF]

no, I didnt mean you should open the attenuator right away! First just check the contacts and see if it has any problem. I meant play with the contacts from the front panel up-down keys and see if the signal amplitude remains constant. if it doesnt then attenuator has issues

have you run the service routines in chapter 4? That should be your starting point. They test quite a bit of the hardware including the YTF driver. If one of those service routines fails then you can follow the troubleshooting.

sample/default CAL data is in the service manual, somewhere towards the end of the manual, maybe chapter 13 or 14 ...

also some of the basic adjustments in chapter 2 can be done without any sig gen. I think high freq sig gen was only needed for freq response adjustment , everything else can be done quite easily. But first you need to find out the problem with CAL AMP failure. Do the service routines first

[Miti]

The easiest way to check the attenuator is to apply a 0dbm within the frequency range of your SG, span about 10MHz, place a peak marker then play with the attenuation. The peak marker reading should stay at or very close to 0dbm. If it changes between iterations or it bounces, you need screwdrivers.

[charlyd]

maybe it is good to say i dont have a SG  which goes into band 1.   i have one on e the bench wiating for this SA to be working ;-)
and i have a working SG HP 8656B ( upto 990Mhz ). so i can do some test.  this having said…

i addend the sample above under the #1 pictures. what does it say?
i have option 130 installed. i read somewhere that a quick test -to see if it is related to #130- is pulling the board a do a new FREQ. followed by a AMPTD cal. this way i make my troubleshoot area smaller.

is that right, or am destroying crucial settings?

for option 130 there should be a foldout 11-1.  to see how it interacts with the unit. couldn t find it 
i will go through chapter 4 tomorrow.

@Miti thanks testing tips for the attenuator. i love them, which can make much progression with little effort.

[charlyd]

hi from what if found it looks like my 08591-60055 is broken as many people have this issue. a new card is on its way..

BTW while making this unit to be a future workbench gair. does anybody have firmware updates. ( if i am right we are talking about the content of 4 eproms ) 
and for the 8595e the latest version i found was fw.980615 can anybody upload these?

[Miti]

hi from what if found it looks like my 08591-60055 is broken as many people have this issue. a new card is on its way..

How did you conclude that? That is an option card, you can remove it and the SA should still work , is just that the RBW can’t go below 1kHz if I remember. You may be rushing things.

Edit:

BTW while making this unit to be a future workbench gair. does anybody have firmware updates. ( if i am right we are talking about the content of 4 eproms ) 
and for the 8595e the latest version i found was fw.980615 can anybody upload these?

What version is your FW? My 8594E has 980615, the 8591E has 9803xx and still works well. I don't know if it's worth upgrading.

[analogRF]

totally agree with Miti here

[MarkL]

On the firmware question...

  https://www.eevblog.com/forum/testgear/hacking-old-hp-spectrum-analyzers/msg2287293/#msg2287293

[Miti]

@MarkL

I got somewhat intrigued by this FW upgrade thing and I compared what I have, with the binaries in that collection. My 8594E has 980615 but 8591E has 940315 so it could take an upgrade. The processor board is 08590-60342 and it has two AM27C020 EPROMs installed. If I look at the Rev. L 98.06.15, it has 4 hex files inside. Does it mean that, if I want to upgrade to 980615, I have to program/ populate all  four EPROMS?
Do they preserve the calibration map or it needs a full calibration after the upgrade?
Do I get any benefits from doing that or it’s just a waste of time?

[MarkL]

@MarkL

I got somewhat intrigued by this FW upgrade thing and I compared what I have, with the binaries in that collection. My 8594E has 980615 but 8591E has 940315 so it could take an upgrade. The processor board is 08590-60342 and it has two AM27C020 EPROMs installed. If I look at the Rev. L 98.06.15, it has 4 hex files inside. Does it mean that, if I want to upgrade to 980615, I have to program/ populate all  four EPROMS?

Yes, you need four EPROMs starting with version 95.02.21.

I also have a 8591E with an old version processor board, 08590-60356.  It currently has four 95.09.14 EPROMs installed.

Do they preserve the calibration map or it needs a full calibration after the upgrade?

That's an excellent question but I don't know.  I did an upgrade on a 8595E and I don't recall that it wiped out the cal data, but there were other problems with that unit.

I'd highly recommend you follow the "fill-in-the-chart" cal backup procedure in the service manual before attempting to move to a different version.  (You should do the backup anyway, as a matter of good care and feeding if you haven't already done so.)

Do I get any benefits from doing that or it’s just a waste of time?

Some of the fixes in that above linked post/document have some fairly concerning bugs that were resolved.  Like, "CAL data loss, DLP erasure, blank-screen lockup, etc."  Unfortunately, it's really not enough information to know anything about the bugs or the conditions that cause them.

I personally like running the latest available on my TE.  Shame on me if I stumble across a bug that's already been fixed and then it ends up wasting even more of my time by interrupting my work AND doing the upgrade.

I wouldn't make it emergency.  Order some EPROMs and do it when you have some spare time.  I would keep the old EPROMs handy so it's easy to go back if something goes sideways with the "new" EPROMs.  They'll likely be pulls.

[Miti]

Thanks Mark!

Yes, that was the first thing I did when I got it, I saved the calibration data. But is the flatness data and the attenuator corrections all I need to save? It is not too much to restore if needed, but I see much more data there.
Also, I quote from the revision history: "Intro of 2X memory (4 2-megabit EEPROMS)". Do you know what EEPROMs are used? 27C020 is an EPROM not EEPROM.

[MarkL]

Thanks Mark!

Yes, that was the first thing I did when I got it, I saved the calibration data. But is the flatness data and the attenuator corrections all I need to save? It is not too much to restore if needed, but I see much more data there.
Also, I quote from the revision history: "Intro of 2X memory (4 2-megabit EEPROMS)". Do you know what EEPROMs are used? 27C020 is an EPROM not EEPROM.

The 8591E needs timebase, flatness, and attenuator to be backed up.  Once this data is restored, you can run the self-cal procedures described in the section "Instrument Calibration after Reloading the Correction Constants" to re-compute everything else.

The 27C020 is certainly an EPROM.  "EEPROM" in that document is a typo.  I've done the same thing many times myself.  I hardly ever deal with EPROMs anymore and my fingers are always typing "EEPROM" instead.

  https://octopart.com/search?q=AM27C020&currency=USD&specs=0

[charlyd]

for the Original Eproms do we talk about the 120ns or the 150ns

btw assuming each model has it own eprom version, what is the latest version for the 8595E ? 

or are all software versions of series 859X compatible for all unit of the 859X series so version from the 8594E can be used in the 8595E? would sound really strange for me.   

[MarkL]

for the Original Eproms do we talk about the 120ns or the 150ns

btw assuming each model has it own eprom version, what is the latest version for the 8595E ? 

or are all software versions of series 859X compatible for all unit of the 859X series so version from the 8594E can be used in the 8595E? would sound really strange for me.

The ones I pulled out of my 8595E when doing the upgrade are 150ns (08590-80216 and 08590-80217, both AM27C020-150DC).  So, 150ns or faster will work.  I didn't analyze the processor read cycle time, so I don't know if anything slower, like the 200ns version, would work.

At some point, HP moved from 1Mb to 2Mb EPROMs.  If you have a model with 1Mb EPROMs and you want to upgrade to the latest version for 2Mb EPROMs, I think you have to adjust some jumpers to accommodate the larger EPROMs.  The jumpers are shown in the CLIP schematic, but there's no table saying how they should be configured.  Maybe it's buried in a document somewhere, or it could ultimately be worked out from the schematic.

The EPROMs are the same for all the 859xE models, but there are some subtleties for older models described here:

  https://www.keysight.com/main/editorial.jspx?cc=US&lc=eng&ckey=458427&nid=-11143.0.00&id=458427

[charlyd]

Hi MarkL great post thanks  ok in my 8594E it is fw. 98 06 15  can i connect that with a letter still L or M ?... but dont see the logo so no M i assume ;-)
my 94E has the Tadinran Battery boarding  my 95E has the yellow CR2477 bat. on the ram board.  i think i will add the jumpers and more info if i see differences.

So markL your experience is that i don t need todo other settings then rerun the internal well known Freq. & Amptd Cal.  after i swapped out  the eproms

[MarkL]

Hi MarkL great post thanks  ok in my 8594E it is fw. 98 06 15  can i connect that with a letter still L or M ?... but dont see the logo so no M i assume ;-)
my 94E has the Tadinran Battery boarding  my 95E has the yellow CR2477 bat. on the ram board.  i think i will add the jumpers and more info if i see differences.

So markL your experience is that i don t need todo other settings then rerun the internal well known Freq. & Amptd Cal.  after i swapped out  the eproms

My recollection is that cal constants (flatness, etc.)  were preserved across an upgrade, but ALWAYS do a backup.

As far as that document is concerned, M only adds the Agilent logo.  But you say "no logo"?  You should see the "hp" logo in the upper left.  If not, your display might not be adjusted correctly.

[charlyd]

aaah in the upper left corner yes  i think so yes..   i thought   when powering on a big logo but ok i understand.  thanks!

[Miti]

I don’t think that’s the logo they are talking about. My 8591 with 84 version still shows the logo.

[charlyd]

So maybe still a big HP BOOT logo. 

ok today i did some diagnosing about the CAL AMPTD failed.  i removed option board 130 and placed jumper W1 on the log ampl. and then did the cals again.  and yes passed....CHECK!

+-+ YIG Problems +-+

but still my amplitude is a bit low and gone after 2.9 GHz .. and it think it has todo with YIG problems. So time for real testing But how can i test that without have a SG above 2.1 GHz.  Does YTF cal in the unit say anything.

Are there some tips , things you need to know/do to find out the band 1 is ok or not ok

+-+  option 130 board +-+

i understand there is not much info or schematich on the Option board 130.

But who has debugged the board? any points to start maybe. 

i see some testpoints on the option 130 board (TP) is it handy to get an extension card for it or is there not really anything to repair on that board?

Who jumps in...