Re: HP-8753E Which EPROM Holds CAL Constants?
The calibration along with serial number info reside in a XICOR EEPROM , if I remember correctly it was a 2864 , this EEPROM is soldered on the board unless somehow you got a hold of one of my special boards I modified back in 2001
?
toggle quoted message
Show quoted text
From: HP-Agilent-Keysight-equipment@groups.io <HP-Agilent-Keysight-equipment@groups.io>
On Behalf Of Rich Miller via groups.io
Sent: Wednesday, March 9, 2022 12:06 PM
To: HP-Agilent-Keysight-equipment@groups.io
Subject: Re: [HP-Agilent-Keysight-equipment] HP-8753E Which EPROM Holds CAL Constants?
?
Hi Marcus:
I am referring to the service calibrations not the kit constants. I am able to write the calibration constants, and retrieve cal kit, and probe tables. I am referring to Calibrations such as Source, and Sampler Definitions associated with
test starting around 44.?
Which EEPROM are you referring too? There is U402, and U408. One is a Xicor, and the other is made by AMD.?
I understand your suggestion regarding the subject of the post I chose to posts. They way I view this is, how would one search in group.io for such content. I suppose I could add more context to the subject line, but that too could be confusing.
I know in the past I have searched for content, and missed something because it was posted in an unrelated top level subject.
Rich
On Mar 9, 2022, at 11:46 AM, Marcus Gustafsson <mankan@...> wrote:
?Rich,
what kind of cal constants are you referring to?
-
Calibration kit constants as well as power probe tables?
-
Calibration data for the 8753 HW?
I think the answers are.
-
NVRAM, i.e. the battery backed DRAM. The internal drive data is also here.
-
EEPROM
A friendly tip: It might be easier to follow your efforts and provide help, if you avoided starting new threads with lacking context each time you have a new question. Although I see a point of starting a new thread for a new question.
Best regards
Marcus
|
Re: HP-8753E Which EPROM Holds CAL Constants?
Hi Marcus:
I am referring to the service calibrations not the kit constants. I am able to write the calibration constants, and retrieve cal kit, and probe tables. I am referring to Calibrations such as Source, and Sampler Definitions associated with test starting around 44.?
Which EEPROM are you referring too? There is U402, and U408. One is a Xicor, and the other is made by AMD.?
I understand your suggestion regarding the subject of the post I chose to posts. They way I view this is, how would one search in group.io for such content. I suppose I could add more context to the subject line, but that too could be confusing. I know in the past I have searched for content, and missed something because it was posted in an unrelated top level subject.
toggle quoted message
Show quoted text
On Mar 9, 2022, at 11:46 AM, Marcus Gustafsson <mankan@...> wrote:
?Rich, what kind of cal constants are you referring to?
- Calibration kit constants as well as power probe tables?
- Calibration data for the 8753 HW?
I think the answers are.
- NVRAM, i.e. the battery backed DRAM. The internal drive data is also here.
- EEPROM
A friendly tip: It might be easier to follow your efforts and provide help, if you avoided starting new threads with lacking context each time you have a new question. Although I see a point of starting a new thread for a new question. Best regards Marcus
|
Re: HP-8753E Which EPROM Holds CAL Constants?
Rich, what kind of cal constants are you referring to?
- Calibration kit constants as well as power probe tables?
- Calibration data for the 8753 HW?
I think the answers are.
- NVRAM, i.e. the battery backed DRAM. The internal drive data is also here.
- EEPROM
A friendly tip: It might be easier to follow your efforts and provide help, if you avoided starting new threads with lacking context each time you have a new question. Although I see a point of starting a new thread for a new question. Best regards Marcus
|
Re: A homemade diode power sensor for HP meters
Understood ,
Generally if I were to pursue a sensor design I would design a USB sensor rather than pursuing a replacement for a existing sensor , there are small mixed signal Processors with USB such as CYPRESS PSOC that can be used , for the sensing
element I would use analog devices detector mmic which are cheap and readily available or for higher frequency a self balancing bridge design using thermistors or a thermal sensor design
?
Back when I was a teen getting started with microwave ham radio I build my first powermeter using a thermal bridge ,? the design used a 50ohm chip termination with a thermistor attached using thermal conductive glue , the thermistor was
put in a bridge circuit and used a DMM module for readout , the beauty of this design was that you could do a rough calibration using a DMM and DC power supply
?
toggle quoted message
Show quoted text
From: HP-Agilent-Keysight-equipment@groups.io <HP-Agilent-Keysight-equipment@groups.io>
On Behalf Of n8zmTWH via groups.io
Sent: Wednesday, March 9, 2022 7:58 AM
To: HP-Agilent-Keysight-equipment@groups.io
Subject: Re: [HP-Agilent-Keysight-equipment] A homemade diode power sensor for HP meters
?
I didn’t think you were. My ‘you’ was just a general case.
?
?
Oh I was not talking about myself, I have calibrated power sensors and meters this was just for people to consider?
On Mar 9, 2022, at 00:07, n8zmTWH via groups.io <n8zmTWH@...> wrote:
?
A couple of thoughts on that…
?
It assumes you don’t have friends with a suitable sensor, or are unwilling to pay a commercial cal house to do it for you.
?
Also, if you have a decent SA and sig gen that you have reasonable faith in, those can be used, Yes, they aren’t quite as good as having another sensor, but for hobby purposes,
you could get within a half dB with care and decent gear. If you have both the SA and SG, you now have two reference instruments, and if they are in agreement then you can be fairly confident of your cal. Most SG’s provide output level indications, and SA
markers are more accurate than eyeballs.
?
It should be noted that power meters are the most accurate devices for power measurements, if traceable to NIST or equivalent. Typically less than .2 dB uncertainty. SG’s are
calibrated for output power with a power meter. The 8566 is +/- 1 dB or more over its frequency range (from memory). More modern SA’s, such as the PSA and PXA series, can almost match a power meter for uncertainty, however.
?
Building a sensor as a challenge is a good exercise even if you have a real one. You could keep the real one on the shelf as a reference and use the homebrewed unit for you daily
testing.
?
BTW, I am trying to revive an really old HP sensor from the 60’s that used 4 AGX 0.01 amp fuses as the sense and reference elements. So far they appear to be unobtanium. Even
the fuse makers don’t list them anymore and I haven’t yet found anything that looks to be equivalent. Just finding a 0.01 amp fuse in any style is a tough one.
?
?
the problem you run into when developing any substitute sensor ,? which is calibration , in order for a sensor to be of value you have to calibrate it which usually is done by comparing it to a calibrated sensor but obviously if you have
a calibrated sensor there was no need to design a sensor?
|
Re: Scoping the Power Rails [8566B]
The “ref unlock, oven” messages are normal when the unit has not been plugged in for a while. It takes about 5 min. for the OCXO oven to come up to temp., there is no accurate or no output from the OCXO until that happens. If the 8566 is plugged in and switched off at the front panel it is in standby mode and the oven stays powered on.
The “battery” indicator means that the 3.6V LiOn battery on the A15 brd needs replacement. “ YTO” unlock means that the 8566 cannot phase lock the YTO tuning over it full range.
Troubleshooting this from the front panel:
a) set the start/stop frequencies 2GHz/6.2GHz at a slow sweep (ie. 2-10 sec.). Check to see if the “YTO” goes out at any point during the sweep. The YIG is actually sweeping from about 2GHz - 6.2GHz for these settings and can be checked at the front panel YTO front panel port.
b) if you see the “YTO” go off at the start or end of the sweep, then flip the SA on its side, remove the bottom cover of the 85660, find A19 and A20 brds. and the adjustments on these brds. labeled 2 and 6.2. Gently push on these adjustment screws 2 for the start of sweep, 6.2 for the end of sweep. Over time there have been issues with these potentiometers. If these are the issue then if you have the ETE to adjust the frequency tune alignment on an 8566, you may be able to rotate these 20 turn pots end to end a few time and clean the crud off the internal wiper of the pot.
The blue Bourns pots were a little worse in reliability than the black pots, but after 30+ years we all get a little tired and cranky.
c) if the pots show no sensitivity to mechanical touch then suspect the DAC’s on these two brds., particularly if the YTO unlock occurs in the somewhere in the middle of the sweep or goes on and off in the middle of the sweep.
d) if the “YTO” unlock is sweep time related then you can check the A16 brd. where the sweep and sweep times are generated. Check the sweep ramp on this brd with an oscilloscope. For the settings in a) it should be a 0Vdc to +10Vdc smooth ramp, if it is not a smooth ramp then there is a sweep ramp DAC on this brd. that has been a problem.
e) if the sweep is too short or too long by greater than 10%, the IF section has the A8 brd. that sets the EOS (end of sweep) for the digital/analog sweep. The adjustments and DAC on this brd have also been known to cause problems.
f) for any DAC troubleshooting there is a KS function that will exercise all of the DAC’s in the 8566, in this mode you can use the up arrow to exercise the DAC’s upward (0,1,2,4,8,16 - 255) and use the down arrow to exercise the DAC’s downward (255 -
15,7,3,1,0). You can use this to check for missing or wrong bits from the DAC’s.
To leave this mode use the IP.
Don Bitters
|
Hi there, mates!
I am thinking on buy a 8562A for my lab and I ask for advice. It has option 026, so 9 kHz to 26.5 GHz. Is this a good unit in general terms??
I wonder how can I see the unit is ok in a few steps at the time of check it for buy.
My 8590L has a CAL key in front panel and makes an auto calibration procedure using the 300 MHz built-in source, so itself can say 'hey, is all Ok'. But the 8562A has no CAL key. I review the manual and don't found that 'autocalibration' function. Maybe is easier than I think, but you maybe can tell me how to do that.
At the moment I don't know about more options installed or modules in the back. I observed that these old units has modules with different options instaled in the back. My 8590L doesn't have that kind of module or dock for it, ninety models are newer versions that doesn't need external memory? Maybe those modules are hackeable to get more functions...
All pieces of advice or recommendations are welcome.
Best regards,
Ruben
|
Re: A homemade diode power sensor for HP meters
I didn’t think you were. My ‘you’ was just a general case. ?
toggle quoted message
Show quoted text
From: HP-Agilent-Keysight-equipment@groups.io <HP-Agilent-Keysight-equipment@groups.io> On Behalf Of Lothar baier
Sent: Wednesday, March 09, 2022 2:03 AM
To: HP-Agilent-Keysight-equipment@groups.io
Subject: Re: [HP-Agilent-Keysight-equipment] A homemade diode power sensor for HP meters? Oh I was not talking about myself, I have calibrated power sensors and meters this was just for people to consider?
On Mar 9, 2022, at 00:07, n8zmTWH via groups.io <n8zmTWH@...> wrote:
? A couple of thoughts on that… ? It assumes you don’t have friends with a suitable sensor, or are unwilling to pay a commercial cal house to do it for you. ? Also, if you have a decent SA and sig gen that you have reasonable faith in, those can be used, Yes, they aren’t quite as good as having another sensor, but for hobby purposes, you could get within a half dB with care and decent gear. If you have both the SA and SG, you now have two reference instruments, and if they are in agreement then you can be fairly confident of your cal. Most SG’s provide output level indications, and SA markers are more accurate than eyeballs. ? It should be noted that power meters are the most accurate devices for power measurements, if traceable to NIST or equivalent. Typically less than .2 dB uncertainty. SG’s are calibrated for output power with a power meter. The 8566 is +/- 1 dB or more over its frequency range (from memory). More modern SA’s, such as the PSA and PXA series, can almost match a power meter for uncertainty, however. ? Building a sensor as a challenge is a good exercise even if you have a real one. You could keep the real one on the shelf as a reference and use the homebrewed unit for you daily testing. ? BTW, I am trying to revive an really old HP sensor from the 60’s that used 4 AGX 0.01 amp fuses as the sense and reference elements. So far they appear to be unobtanium. Even the fuse makers don’t list them anymore and I haven’t yet found anything that looks to be equivalent. Just finding a 0.01 amp fuse in any style is a tough one. ? ? the problem you run into when developing any substitute sensor ,? which is calibration , in order for a sensor to be of value you have to calibrate it which usually is done by comparing it to a calibrated sensor but obviously if you have a calibrated sensor there was no need to design a sensor?
|
HP-8753E Which EPROM Holds CAL Constants?
As I troubleshoot the problem of writing CAL Constants to my 8753E it occurred to me those constants may not be stored in the main EEPROM U402. On the CPU Board it appears there are a variety of memory devices which can be written to. There is U408, which comes from the factory as a socketed PLCC. I have reviewed the contents of this EEPROM, and it appears to have data related to selecting disk drives in it. Then there seems to be a series of Dallas NVRAM. I know in older Tektronix products CAL Data was stored in Dallas NVRAM (because I replaced a ton of them). Does anyone know where the constants are stored specifically in the 8753 series? I may probing a chip with my logic analyzer that does not even store cal data.
Rich
|
Re: Scoping the Power Rails [8566B]
Okay - I have a quick update.
I tested the output from the front panel 100Mhz CAL out BNC socket and it shows 100Mhz spot-on on my 1Ghz counter and a nice clean stable sine wave on my 350Mhz Tek analog scope. I'm seeing 200mV p-p into 50 ohms.
Manually tuning the center frequency across a range from min to 13Ghz, the YTO Unlock error was present throughout that whole range.
I just managed to squeeze those two checks in this morning and will be around again later this evening if anyone has any further thoughts.
|
Re: Agilent 8714ES missing graticule
Hi Maurit Great idea, thank you for the tips. Any info for the type number and place to purchase ? Kind regards Kurt ? Fra: HP-Agilent-Keysight-equipment@groups.io <HP-Agilent-Keysight-equipment@groups.io> P? vegne af maurit via groups.io
Sendt: 9. marts 2022 12:12
Til: HP-Agilent-Keysight-equipment@groups.io
Emne: Re: [HP-Agilent-Keysight-equipment] Agilent 8714ES missing graticule ? I recommend you replace the CRT with a VGA LCD display.
It's very simple. You will solve the problem and have a beautiful VNA that you won't replace for a long time.
|
Re: Agilent 8714ES missing graticule
Hi Kurt,
I recommend you replace the CRT with a VGA LCD display.
It's very simple.
You will solve the problem and have a beautiful VNA that you won't replace for a long time. 
|
Re: Scoping the Power Rails [8566B]
@Bruce
No problem! The thread is so long now it's getting confusing and even more so for newcomers. So I'm considering starting a separate one off from scratch as soon as I reach a suitable juncture for doing so.
Thanks for the clarification on the strobing/stating. I'll do the manual scan as soon as I can fit it in.
Cheers,
J.
|
Re: HP 5334A Oscillator Support PCB
Hi, a very good description by Harvey! However, I would like to add that there is a major difference between
LF grounding and RF grounding but it is a different story...
G?ran
|
Re: A homemade diode power sensor for HP meters
Oh I was not talking about myself, I have calibrated power sensors and meters this was just for people to consider?
toggle quoted message
Show quoted text
On Mar 9, 2022, at 00:07, n8zmTWH via groups.io <n8zmTWH@...> wrote:
?
A couple of thoughts on that…
?
It assumes you don’t have friends with a suitable sensor, or are unwilling to pay a commercial cal house to do it for you.
?
Also, if you have a decent SA and sig gen that you have reasonable faith in, those can be used, Yes, they aren’t quite as good as having another sensor, but for hobby purposes,
you could get within a half dB with care and decent gear. If you have both the SA and SG, you now have two reference instruments, and if they are in agreement then you can be fairly confident of your cal. Most SG’s provide output level indications, and SA
markers are more accurate than eyeballs.
?
It should be noted that power meters are the most accurate devices for power measurements, if traceable to NIST or equivalent. Typically less than .2 dB uncertainty. SG’s are
calibrated for output power with a power meter. The 8566 is +/- 1 dB or more over its frequency range (from memory). More modern SA’s, such as the PSA and PXA series, can almost match a power meter for uncertainty, however.
?
Building a sensor as a challenge is a good exercise even if you have a real one. You could keep the real one on the shelf as a reference and use the homebrewed unit for you daily
testing.
?
BTW, I am trying to revive an really old HP sensor from the 60’s that used 4 AGX 0.01 amp fuses as the sense and reference elements. So far they appear to be unobtanium. Even
the fuse makers don’t list them anymore and I haven’t yet found anything that looks to be equivalent. Just finding a 0.01 amp fuse in any style is a tough one.
?
From: HP-Agilent-Keysight-equipment@groups.io <HP-Agilent-Keysight-equipment@groups.io>
On Behalf Of Lothar baier
Sent: Wednesday, March 09, 2022 12:37 AM
To: HP-Agilent-Keysight-equipment@groups.io
Subject: Re: [HP-Agilent-Keysight-equipment] A homemade diode power sensor for HP meters
?
the problem you run into when developing any substitute sensor ,? which is calibration , in order for a sensor to be of value you have to calibrate it which usually is done by comparing it to a calibrated sensor but obviously if you have
a calibrated sensor there was no need to design a sensor?
|
Re: A homemade diode power sensor for HP meters
A couple of thoughts on that… ? It assumes you don’t have friends with a suitable sensor, or are unwilling to pay a commercial cal house to do it for you. ? Also, if you have a decent SA and sig gen that you have reasonable faith in, those can be used, Yes, they aren’t quite as good as having another sensor, but for hobby purposes, you could get within a half dB with care and decent gear. If you have both the SA and SG, you now have two reference instruments, and if they are in agreement then you can be fairly confident of your cal. Most SG’s provide output level indications, and SA markers are more accurate than eyeballs. ? It should be noted that power meters are the most accurate devices for power measurements, if traceable to NIST or equivalent. Typically less than .2 dB uncertainty. SG’s are calibrated for output power with a power meter. The 8566 is +/- 1 dB or more over its frequency range (from memory). More modern SA’s, such as the PSA and PXA series, can almost match a power meter for uncertainty, however. ? Building a sensor as a challenge is a good exercise even if you have a real one. You could keep the real one on the shelf as a reference and use the homebrewed unit for you daily testing. ? BTW, I am trying to revive an really old HP sensor from the 60’s that used 4 AGX 0.01 amp fuses as the sense and reference elements. So far they appear to be unobtanium. Even the fuse makers don’t list them anymore and I haven’t yet found anything that looks to be equivalent. Just finding a 0.01 amp fuse in any style is a tough one. ?
toggle quoted message
Show quoted text
From: HP-Agilent-Keysight-equipment@groups.io <HP-Agilent-Keysight-equipment@groups.io> On Behalf Of Lothar baier
Sent: Wednesday, March 09, 2022 12:37 AM
To: HP-Agilent-Keysight-equipment@groups.io
Subject: Re: [HP-Agilent-Keysight-equipment] A homemade diode power sensor for HP meters? the problem you run into when developing any substitute sensor ,? which is calibration , in order for a sensor to be of value you have to calibrate it which usually is done by comparing it to a calibrated sensor but obviously if you have a calibrated sensor there was no need to design a sensor?
|
Re: A homemade diode power sensor for HP meters
the problem you run into when developing any substitute sensor ,? which is calibration , in order for a sensor to be of value you have to calibrate it which usually is done by comparing it to a calibrated sensor but obviously if you have a calibrated sensor there was no need to design a sensor?
|
HP 3324A - sweep menu frustrations
Hi all, I've had this minty 3324A for a pretty long time now, and I just stopped trying to use it at some point due to frustrations with setting up sweeps on it. Hopeful some good people here would take pity on my wits and assist me out of these heartburny woods. Thank you in advance for that.? Let's say I start the instrument up and, after the self tests ("set default"), it goes into 1kHz sine WaveForm. I'd then press Sweep and start trying to set it up to do one single sweep - let's say from 10MHz to 20MHz on a 5s sweep time.?
- I press "Down Arrow" to switch from SINGLE to CONT. Select.
- "Retrace Time" comes into the screen, I go "5s," "Right Arrow."
- Goes to MARKER, and I assume it wants either something like "1MHz" (so it'd put a marker at 11MHz, 12MHz, etc. till 19MHz). No go here. I can't even see anything coming out of this step on my HP54522A FFT set to look at the entire sweep.?
- I've been able to go to the next setting here (not this afternoon, apparently), but it'd not want to go to START-FREQ. It usually skips directly to STOP-FREQ. [5 minute intense fiddling pass by] I just managed to get into the START-FREQ option and set it. But I really can't grasp what I did, and what is the expected sequencing in doing my selections and getting on to the next step. Every time I managed to set it up in some single step way I am not exactly sure what I did.??
The other very strange default option is that is seems to assume I want "1 of 7" sweeps. Has this been maybe customized by the prior owner? Otherwise, why not one beautiful single sparkly sweep, like most folks probably want? It seems "3325A mode" means dropping all prior settings upon restarting (can anyone think of a more cryptic description for just doing that?...). I don't think this one is set to default to that setting.? If I do manage to start a sweep, I'd be leaving the screen flashing on some variable as it probably wants me to set another six sweeps (or "intervals?..."). Not sure how I can get the screen to be static (being that I'm done) and I resigned myself to a flashing screen.?? This HP unit seems to be one those from times when the quality of the manuals took quite a fall - no "component level service manuals" available any longer (wink wink, Bruce... ;)), and the operating and programming manual is just not very good. The actual menu system of these units seems to be a pretty contorted mess that's not intuitive in any of its parts, and much less so in whole.? Sorry for spilling my guts a bit here. Many thanks for your input in advance.? Radu.?
|
Re: HP 5334A Oscillator Support PCB
regarding grounds:
you're dealing with what's called circulating currents, or ground
currents.
Let's say that you have an analog circuit, and that its ground is
common (on the board) with a high power circuit.
as the high power circuit operates, the analog ground goes up and
down, and could cause any number of problems with the analog
circuit.
The trick is to make sure that the ground currents from the high
power circuit don't flow through the analog ground, keeping the
analog ground steady (and quiet).
A way to ensure this is to connect the analog ground (separate!)
and high power ground (separate!) and digital ground( separate!)
at only one point, which is where power enters the board.?
Look up the concept of "star grounding", for example.
the grounds are the same, in that they're all ground.? They are
not the same in that they are NOT the same ground.
Hope that helps a bit, especially the lookup topic.
(NOTE: grounding is not a simple topic, and there are many
different opinions).
Harvey
On 3/8/2022 6:11 PM, Tony wrote:
toggle quoted message
Show quoted text
Hi,
I have an HP 5334A frequency counter and I want to install a
10MHz oven quartz oscillator to improve the accuracy. Although
it’s not hard to find the oscillator, getting the support PCB is
proving tricky. So I thought I would make one as it’s not a
complex circuit or layout. Spacing of mounting holes etc is a
bit more tricky - ?but I mocked up a prototype with cardboard to
get rough sizing etc.
My question is about circuit grounds. The circuit (attached) has
several separate supply grounds and my question is about the
note 3 regarding “instrument common” and “digital circuit
return”. They are distinguished on the schematic but appear to
be connected on the PCB (see attached pic). For example C1 and
C2 are clearly connected to edge connector pin 4 although they
are different “grounds” according to the note.
could anybody advise me on what the best thing to do here is
please? Are they the same?
thanks
Tonys
?
|
Re: Scoping the Power Rails [8566B]
Re: voltage jumping around - not necessarily, the sweep actually stats and stobs between bands. Doing the manual scan is probably a good idea - sorry about the confusion on the LED. Cheers! Bruce Quoting Jinxie <paul666@...>:
@Bruce
I believe it was Ozan who posted the image showing an LED in the stack unlit - certainly not me anyway.
I don't think we can attribute this to scanning/sweeping as that would be predictable and regular, no? What I'm seeing is completely random, like a child fooling around with a light switch. The intervals between breakdowns are completely arbitrary. I do like your final suggestion, though; must try that tomorrow...
|
Re: Scoping the Power Rails [8566B]
Gonna have to leave it for now as it's once again getting really late here now. Thanks for all the help, guys, I'll be back tomorrow....
|
RubenRubio
Jinxie
Kurt Poulsen
maurit
G?ran Krusell