Re: How do you decipher HP serial numbers for date.
Hi All - Back about five years ago, someone who'd worked for HP posted an explanation of the post-1960 serial numbering system, which I then re-posted with my interpretation of it on another forum. Here is a copy of my forum post - the original text is at the start, and my understanding of it follows.
"Copied and pasted from a post to the HPAK mailing list explaining the serial numbering system; the original author worked for HP:
Having worked as a Production Line Engineer, Production Line Manager and Product Support Manager(Americas) for one HP Division, please let me clarify the process for Engineering Changes and how that affects Serial Prefixes. 1. The Basic Identity of an Instrument is the Model No. (e.g. HP 8510A,HP8510B,HP8510C). These Models are all independent of each other and each of the three examples would start with a Serial No.(Suffix) of 00101 2. The most Major Engineering Changes to an Instrument (e.g. Addition of a bunch of New Features or Upgrades) could be implemented into a New Model No. (e.g. change from "A" model to "B" model). In the case of changing to a New Model No. the Serial No.(Suffix) will start again at 00101. Short of a New Model No., all other changes would be handled through a "Production Change Order". 3. A Production Change Order may or may not trigger a New Serial Prefix. Typically, for instance, the change of a single component value would not cause a Serial Prefix change. I believe (but I'm not 100% on this), that would be decided by Production Engineer in conjunction with Product Support folks, who are responsible for Service Manuals. Major Engineering Changes would always trigger a New Prefix. 4. Instrument Manufacturing is done in Batches, called Production Runs. The Size of each Run is determined by Sales Volume. More Sales, Larger Runs. 5. A New Serial Prefix would be implemented for the Start of a New Production Run. Before issuing Work Orders for a New Production Run, I believe a review of Engineering Change Level would be done and the decision to introduce a New Prefix would be made or had already been made. That could be due to one Major change or sometimes a roll-up of a bunch of Minor ones. Keep in mind none of this is instantaneous, 'cos of lead times for (new)material to be provisioned. 6. In an Emergency (quite rare), it has been known for a Production Change to be implemented immediately, maybe 'cos of a Safety concern(I did it only once) or some major Defect being discovered. 7. I think by now it has been well established that the Serial Prefix first two digits are the number of Years since 1960 and the last two are the week number of the last Production change included in the Instrument. Sorry for being a bit Verbose, but that's how it works.
Paraphrasing my understanding of the above:
1: The model number and its letter suffix identifies the instrument type. Each is considered to be independent of the others for serial numbering purposes, and starts with a 00101 serial number suffix.
2: An expansion of (1) reiterating that the serial number resets with a new letter revision to an instrument when major changes are made. If it is only a rev change then the serial number will continue to increment sequentially from the last instrument made under the previous engineering rev.
3: Minor changes would not trigger a prefix update; more major ones (but those that were not enough to trigger a new model suffix revision) would trigger a prefix update.
4: When a design is released, they build a bunch at a time.
5: This one hung me up for a bit until I thought it through a bit more and paid attention to the wording. A new serial prefix would be implemented at the start of a new production run (note that it's 'would be', not 'will be' or 'is' - it only changes if the design is revised) Prior to the new run, they'd look everything over, and determine if a new prefix was needed; if it was not then the new run would get the old (but current to the design, as it's essentially unchanged) prefix.
To summarize my understanding of what 2-5 mean, for example, if the last unit of the run of the mythical HP 0000A Whatsis (introduced in the 32nd week of 1972) serial number 1232A00250 is finished and the design undergoes a major change to become an HP 0000B Whatsis in the second week of January, 1975, then the first new HP 0000B off the line will be serial number 1502A00101 (new prefix, reset serial). If, instead, the new iteration of the HP 0000A comes off the assembly line with a lesser engineering change (not enough to warrant it becoming a 'B' model), then the first unit in the new run will have a serial number of 1502A00251 (new prefix, incremented serial). If it's just a new run with no real changes, then the first off the line in January of 1975 would be serial number 1232A00251 (old prefix, incremented serial).
6 & 7: self explanatory.
I left the OP's name off because it was posted to the mailing list; if anyone wants to dig it was posted on 10/05/17."
-Pat
> ------ Original Message ------
> From: jn6wfo@...
> To: [email protected]
> Sent: Thursday, September 1, 2022 10:32 AM
> Subject: Re: [HP-Agilent-Keysight-equipment] How do you decipher HP serial numbers for date.
Actually, Matt¡¯s serial number ID (mmmmXnnnnn where X is the country of manufacture) applies to designs (redesigns) finalized in or after 1970. A=America (USA), G=Germany, J=Japan, U=United Kingdom, etc.
For 1960-1969, the serial number is of the form nnn-mmmm where the first part is the date code and the second part the series number. The first digit in the date code is added to 1960 to give the year. The second two digits are often said to be a week number but I¡¯ve seen numbers higher than 52 so I¡¯m unsure. For instruments with small unit runs, the serial number block is sometimes shared with other instruments.
For instruments before 1960, there¡¯s a simple serial number that has no date significance. Numbers range from 1 to 5 digits; comparing the various versions of the manuals and catalogs, if you can find them, can give a rough idea of the date.
The serial number consists of a 4-digit date code followed by a letter indicating the country of manufacture, followed by a sequence of digits that uniquely identifies the instrument.
In the 4-digit date code, the first two digits are the number of years since 1960 and the last two digits are the number of weeks into the year. This date code represents the date of design, not manufacture. It¡¯s a common misconception that this means build date. It doesn¡¯t.
For the country code, off memory, A is USA, J is Japan, M Malaysia. Probably others too. I think this indicates location of manufacture rather than design, but I could be wrong on that.
I think the 5-digit suffix is sequentially assigned based on build date, but I don¡¯t know.
So, to answer your question you can¡¯t really know when your instrument was built other than it was sometime after the date code. Previously, I expect HP would have been able to provide this information if you called in, but I doubt keysight would (or probably even could) tell you that anymore. But you could try.
FYI this question has been posed and answered a number of times before. A search on the HP-Agilent-Keysight group would yield the same info. The serial number format is also usually described at the beginning of instrument manuals.
Jeremy Nichols
6.
|
Re: How do you decipher HP serial numbers for date.
Actually, Matt¡¯s serial number ID (mmmmXnnnnn where X is the country of manufacture) applies to designs (redesigns) finalized in or after 1970.? A=America (USA), G=Germany, J=Japan, U=United Kingdom, etc.
For 1960-1969, the serial?number is of the form nnn-mmmm where the first part is the date code and the second part the series number. The first digit in the date code is added to 1960 to give the year. The second two digits are often said to be a week number but I¡¯ve seen numbers higher than 52 so I¡¯m unsure. For instruments with small unit runs, the serial number block is sometimes shared with other instruments.?
For instruments before 1960, there¡¯s a simple serial number that has no date significance. Numbers range from 1 to 5 digits; comparing the various versions of the manuals and catalogs, if you can find them, can give a rough idea of the date.?
The serial number consists of a 4-digit date code followed by a letter indicating the country of manufacture, followed by a sequence of digits that uniquely identifies the instrument.
In the 4-digit date code, the first two digits are the number of years since 1960 and the last two digits are the number of weeks into the year. This date code represents the date of design, not manufacture. It¡¯s a common misconception that this means build date. It doesn¡¯t.
For the country code, off memory, A is USA, J is Japan, M Malaysia. Probably others too. I think this indicates location of manufacture rather than design, but I could be wrong on that.
I think the 5-digit suffix is sequentially assigned based on build date, but I don¡¯t know.
So, to answer your?question you can¡¯t really know when your instrument was built other than it was sometime after the date code. Previously, I expect HP would have been able to provide this information if you called in, but I doubt keysight would (or probably even could) tell you that anymore. But you could try.
FYI this question has been posed and answered a number of times before. A search on the HP-Agilent-Keysight group would yield the same info. The serial number format is also usually described at the beginning of instrument manuals.
|
Re: Hp 8590a spectrum analyzer
Unfortunately I don't have another SA. I'll check the diodes before I replace and check the DB with corrections off. I'll take a picture and post it. Should I pose the cal data too??
Thank you for your help !
|
Re: How do you decipher HP serial numbers for date.
Matt, That applies to equipment made after 1960. ?To a certain time when serial numbers changed again.? For equipment pre 1960 it¡¯s not that easy.? The 400D I inquired about recently was mid 1950¡¯s.? That was determined by researching when equipment was introduced and various publication with dates. The ultimate determination on age will be further narrowed down by observing date codes on components.? So it can be more complex based on vintage.?
toggle quoted message
Show quoted text
On Sep 1, 2022, at 5:52 AM, Matt Huszagh <huszaghmatt@...> wrote:
? The serial number consists of a 4-digit date code followed by a letter indicating the country of manufacture, followed by a sequence of digits that uniquely identifies the instrument.
In the 4-digit date code, the first two digits are the number of years since 1960 and the last two digits are the number of weeks into the year. This date code represents the date of design, not manufacture. It¡¯s a common misconception that this means build date. It doesn¡¯t.
For the country code, off memory, A is USA, J is Japan, M Malaysia. Probably others too. I think this indicates location of manufacture rather than design, but I could be wrong on that.
I think the 5-digit suffix is sequentially assigned based on build date, but I don¡¯t know.
So, to answer your?question you can¡¯t really know when your instrument was built other than it was sometime after the date code. Previously, I expect HP would have been able to provide this information if you called in, but I doubt keysight would (or probably even could) tell you that anymore. But you could try.
FYI this question has been posed and answered a number of times before. A search on the HP-Agilent-Keysight group would yield the same info. The serial number format is also usually described at the beginning of instrument manuals.
|
Re: How do you decipher HP serial numbers for date.
The serial number consists of a 4-digit date code followed by a letter indicating the country of manufacture, followed by a sequence of digits that uniquely identifies the instrument.
In the 4-digit date code, the first two digits are the number of years since 1960 and the last two digits are the number of weeks into the year. This date code represents the date of design, not manufacture. It¡¯s a common misconception that this means build date. It doesn¡¯t.
For the country code, off memory, A is USA, J is Japan, M Malaysia. Probably others too. I think this indicates location of manufacture rather than design, but I could be wrong on that.
I think the 5-digit suffix is sequentially assigned based on build date, but I don¡¯t know.
So, to answer your?question you can¡¯t really know when your instrument was built other than it was sometime after the date code. Previously, I expect HP would have been able to provide this information if you called in, but I doubt keysight would (or probably even could) tell you that anymore. But you could try.
FYI this question has been posed and answered a number of times before. A search on the HP-Agilent-Keysight group would yield the same info. The serial number format is also usually described at the beginning of instrument manuals.
|
Re: How do you decipher HP serial numbers for date.
What is your serial number?? There are several ways to decode the number based on the number. ?I recently started a similar thread and was helped out by the group knowledge here.?
Dave
toggle quoted message
Show quoted text
On Sep 1, 2022, at 4:54 AM, Mikek <amdx@...> wrote:
?I recently bought an HP 3400A, and want to get a build date and also find the proper schematic for repair.
?Is there a site that tells how to decipher the serial number?
?I thought I saw this discussed recently, but not sure wher.
????????????????????????????????????????? hanks, Mikek
|
Hello friend!
ok, i will call you in private, thank you very much! Em qui., 1 de set. de 2022 08:40, piecolav via <piecolav= [email protected]> escreveu:
Hi imaurente,
at
you will find a series of manuals related to your analyzer, including the complete one for the power supply.
Also I have the CLIP manual of the HP8560E, but I can't find the link from where I downloaded it. If you send me a PM message, I can send it to you.
Greetings.
|
How do you decipher HP serial numbers for date.
I recently bought an HP 3400A, and want to get a build date and also find the proper schematic for repair.
?Is there a site that tells how to decipher the serial number?
?I thought I saw this discussed recently, but not sure wher.
????????????????????????????????????????? hanks, Mikek
|
Hi imaurente,
at
you will find a series of manuals related to your analyzer, including the complete one for the power supply.
Also I have the CLIP manual of the HP8560E, but I can't find the link from where I downloaded it. If you send me a PM message, I can send it to you.
Greetings.
|
Ed, you're absolutely right. It's the same thing I thought when I started getting interested in buying one of these YTOs.
Especially the 856X series SAs are still today pretty good tools. They have a low noise floor, wide dynamics, low phase noise and digital resolution up to 1Hz. And surelythere are tens of thousands of instruments still in operation around the world. But, nevertheless, the number of YIG oscillators still seems to be too large.
But the biggest problem, I believe, is the high price required to repair them.
Even Keysight, whom I contacted, hasn't sold any in a while, but continues to repair these YTOs by shipping the complete instrument. Sum required? About 3000 euros!
They are probably shipped to the US for repair, I read in another forum that they would be built by the Microsource of Santa Rosa, and realigned on the instrument, but still it seems like an overpriced.
I understand that they are professional instruments, and I don't even want to think about how much a new instrument of similar characteristics costs today, but still they look like giants with feet of clay ...
|
Re: Hp 8590a spectrum analyzer
Let me try that on mine as well!
|
Re: Hp 8590a spectrum analyzer
There is a chance, of course, that something else is wrong if you see a lower level. The calibration output can be off, some filters or gain stages can be misaligned and so on. But if you have another spectrum analyzer capable of 2.04 GHz and a function generator, it should be easy to test the first converter separately as shown in the video.
On Wednesday, August 31, 2022 at 06:11:20 PM PDT, Richard Cook <richardcook331@...> wrote:
Thanks I ordered 5 of them just to have them on hand.?
|
Re: Hp 8590a spectrum analyzer
Thanks I ordered 5 of them just to have them on hand.?
|
Hello guys,
I have an 8560E analyzer, it is turning on but it shows error 335 on the screen, according to the service manual it refers to the sampling oscillator, I went deeper in the research and detected that the 28V voltage was absent on the A2, I did a brief analysis at the source, where the TP304 had no voltage. I ask, has anyone had this problem? Does anyone have here in the documentation group that the components for replacement can be identified? I appreciate the help of friends! thanks.
|
Re: Hp 8590a spectrum analyzer
Yes, but those are obsolete. These seem to be the replacements nowadays:
On Wednesday, August 31, 2022 at 03:24:13 PM PDT, Richard Cook <richardcook331@...> wrote:
Thanks Alex , I'll try that when I get the power supply re installed tomorrow.? I was wondering how to check that. What's the correct part number for that diode pack?. Hsms 2822 or something like that.? Thanks.?
On Wed, Aug 31, 2022, 18:02 Alexander Saydakov via <sandy_saydakov= [email protected]> wrote:
Connect the input to the calibration output and turn the corrections off. Is the main tone about -20 dBm? if it is significantly lower (say, -30) then most probably the mixer diodes in the first converter are bad. The corrections try to compensate, but it is too much to compensate. As I understand, the corrections must be within a fraction of dB typically.
On Wednesday, August 31, 2022 at 02:51:02 AM PDT, Richard Cook < richardcook331@...> wrote:
You are most right. It may be better to leave it alone. But I know the power Supplies are hard to come by so I think replacing the caps with high grade ones and safety caps isn't a bad idea. Like I said I never have delt with repair of one of these animals so I may just keep in mind the noise floor and use it.?
|
Re: Hp 8590a spectrum analyzer
Thanks Alex , I'll try that when I get the power supply re installed tomorrow.? I was wondering how to check that. What's the correct part number for that diode pack?. Hsms 2822 or something like that.? Thanks.?
On Wed, Aug 31, 2022, 18:02 Alexander Saydakov via <sandy_saydakov= [email protected]> wrote:
Connect the input to the calibration output and turn the corrections off. Is the main tone about -20 dBm? if it is significantly lower (say, -30) then most probably the mixer diodes in the first converter are bad. The corrections try to compensate, but it is too much to compensate. As I understand, the corrections must be within a fraction of dB typically.
On Wednesday, August 31, 2022 at 02:51:02 AM PDT, Richard Cook < richardcook331@...> wrote:
You are most right. It may be better to leave it alone. But I know the power Supplies are hard to come by so I think replacing the caps with high grade ones and safety caps isn't a bad idea. Like I said I never have delt with repair of one of these animals so I may just keep in mind the noise floor and use it.?
|
Re: Hp 8590a spectrum analyzer
Connect the input to the calibration output and turn the corrections off. Is the main tone about -20 dBm? if it is significantly lower (say, -30) then most probably the mixer diodes in the first converter are bad. The corrections try to compensate, but it is too much to compensate. As I understand, the corrections must be within a fraction of dB typically.
On Wednesday, August 31, 2022 at 02:51:02 AM PDT, Richard Cook <richardcook331@...> wrote:
You are most right. It may be better to leave it alone. But I know the power Supplies are hard to come by so I think replacing the caps with high grade ones and safety caps isn't a bad idea. Like I said I never have delt with repair of one of these animals so I may just keep in mind the noise floor and use it.?
|
I was looking for info on this YTO, and stumbled upon this ebay listing, which I found to be very informative, and disturbing:
https://www.ebay.com/itm/HP-AGILENT-YTO-YIG-REPAIR-5086-7903-8593E-8594e-8595e-8596e-8592L-8593em/252216883196
This has a bunch of info on these YTOs. What I find disturbing is that it appears that a whole family (856X and 859X) of HP portable SAs may have a serious reliability issue - big enough to support a cottage industry repairing their YTOs. Maybe the the actual failure percentage is tiny, and there are just a lots and lots of units out there, so no big deal, but it seems like an awful lot of failures to me. I don't have any of these models, so it doesn't affect me, but I wonder what the deal is. My experience with lots of old-school YIG devices is that they last virtually forever, unless abused.
Ed
|
Thank you for the information, John! I tried to increase the tuning coil current up to 200 mA but to no avail.
Then, I also wanted to look at the second YTO under the microscope, the one extracted from the HP8560E and I discovered to my surprise that the wire connected to the PIN of the 5 VDC power supply was interrupted.
With a lot of patience and using the finest soldering iron tip I had, I pre-wetted the PIN. Then using a thinned toothpick I bent the wire until it touched the PIN. And I lightly rested the tip of the soldering iron on it for a moment.
I checked and the soldering was successful. I did well? I used as little tin as possible to avoid tin bubbles that could out-scale the YTO.
Unfortunately my microscope has a maximum magnification of 57X. It would take one with at least double the magnification.
I connected the YTO to the power supplies by crossing my fingers and to my surprise I found that it worked!
The peak power, about 13.5 dBm, I get it with a tune current of 90 mA.
Instead I noticed that the absorbed currents, with the YTO running, have varied a lot, that is:
+5 VDC, 73 mA
+15 VDC, 20 mA
-15 VDC, 7 mA
I reinstalled the YTO in its SA and checked it was working properly. I kept it running for several hours doing all the tests possible, all passed positively.
At this point the hypotheses for the breakdown of the second YTO have narrowed down to only two:
1- internally the YTO of the HP8560E, although similar in structure, differed slightly from that of the HP8563E. Question: Were there different versions made? Or is the YTO mounted on the HP8560E not the 5086-7906? So why did it break?
I also found that this YTO has been repaired. Internally there are several scratches ...
2- The +5 VDC voltage of the HP8563E has some problems. I checked it and it is stable and precise but, I doubted that some tantalum capacitor (there are several on that line) is interrupted and the power on extra current, not absorbed by the capacitor, may have caused it to burn the two YTOs.
Could mine be a valid hypothesis?
Sorry if I'm insisting on the subject, but I just want to try to understand and learn and thank you all for your cooperation!
|
HP8561E and HP8341A as tracking source?
Hi All!
Tell me please if it possible to use HP8341A as tracking generator for HP8561E and how to interconnect them (HP-IB, BNC...)?
What steps of configuration in menus?
Best regards,
Roman.
|
Cubdriver
Jeremy Nichols
imaurente .
Anders