Bill,

I agree with soooo much of what you said, thank you.?

As to scopes it was my understanding Tek had a patent on their circuit and didn’t license it. ?We had tek scope in the EM shop at Opto div. ?

Regards,

?

Stephen Hanselman

Datagate Systems, LLC

On Apr 25, 2020, at 13:43, Bill E <solartron@...> wrote:

?This topic has sure mutated. Can't let it die now, I'll jump in with a bunch of random babbling, some I've done before, but hey, can't remember which topics. Senility, I suppose.

I posted a few days ago how horrible Win-based equipment was. Just what was anyone thinking? Maybe HP did get paid off my MS. XP was pretty atrocious, massively buggy, security issues all over the place. Both HP and Tek did good embedded sw, then went lame. OTOH, could be part of that move to outsource development, driven by clueless bean-counters. I transitioned from all HW to mostly all SW over my career, always fighting the 'we can hire 5 3rd-world devs for what we pay you'. Ok, do that, hmm, nothing works. So, still pay a real dev plus those 5 useless hacks. Bean-counters apparently can't figure out that they're paying 2x one good dev for 0.9 total output, since all the code has to be dealt with (rewritten) by the good dev.

Next rant, sw bloat. I ended up doing a lot of sw for telecom, massive servers, all in Java. Talk about bloated code, gigs of open-source libs, some dragged in just to get one trivial method. And, modern pseudo-devs have no understanding of hardware at all, and just as little understanding of basic sw algorithms, Just blindly use whatever they stumble across, then wonder why we needed a server farm to run what should have been possible on one machine.

Moving on, those DS 'top-hat' pseudo-nvrams. Who ever thought those were a good idea? 10 yrs, dead. I'm suspecting bean-counters again, make sure to obsolete the equipment. There's a guy selling ferromagnetic rams to replace said DS horrors for Tek scopes. Time to start doing the same for our HP stuff. And yes, my latest acquisition, 53310A, appears to have one of those damn things in it.

Have I run out? Of course not.

On to C vs C++. I really never liked C++, and don't get me started on C#. My definition of C++, a pseudo-OO variant done by someone that really didn't understand C. Stroustrup is on my hate-list. I've written hundreds of thousands of lines of C, and still my favorite lang for embedded tiny stuff. I started with IBM 360 assembler (BALR sound familiar, waiting for responses to that), then Fortran, then PDP-8, PDP-11, Multics, Unix, Linux. Used Unix at Bell Labs early on. It is truly amazing how it has evolved.

Hmm, any more? Oh, wait, of course.....

Cheap instruments from China. All based on various US highly-integrated chips. Some of the stuff is useful. most is just junk. Amazing 'specs' in the ads. Any real test data? Of course not. Get some of them, Ok, yes, they do the basic operations, but, example, cheap signal generators, horrible purity and stability. About as much out-of-peak energy as is in the fundamental.

Finally, modern instruments of any kind.

What happened to the great engineering we used to have? Now it's all about design-to-cost. Selling points seem to be 'internet connected'. I'll stack my $300 5370 time-interval analyzer up against anything you can get today. GUI? Nope. Internet? Nope. Amazing? Yes. Same for my 54542, and still amazing even today, given what one costs vs new junk. BTW, friend of mine from college worked at HP instrument division, he's told me many a sad tale about the decline of HP. Blames one person, Carly.

In closing, maybe I can start another flame war. My 54542 is the first HP scope I'd ever considered.Everything else I now have in my home lab is HP, but Tek? scopes really dominated. I love my Tek 2465A, one of the last great analog scopes. My HP friend said they got tired of everyone complaining about the lame triggering performance of HP scopes, decided to do a 'Tek killer', and succeeded. True? Don't know. I do know that one of the early Tek digital scopes that I had used this what seems to me to be a total hack, linear CCD to capture the analog, then read it out slowly enough for the lame A/D to handle. And, I do know I just love my 54542. Well, until the already-fading CRT dies. At least it doesn't run Win. :)

Hope I've amused you,
Bill

Oh, and I want to put a plug in for some of the classic Brit engineering. My 2 Datron 4910 voltage refs used to be national-standard level. My Solartron 7081 8.5 digit meters rock (although it is sloooow at 8.5, and really isn't quite as good as a 3458A, but at a fraction of the price if you can find one. BTW, it didn't use the LTZ ref both the 3458 and my Datron 4910's used. It had a 'zero-temp-factor' Zener, selected, and then compensated more by a temp measurement, trimmed per-Zener to get the best zero TC. Love it.

?

Hi,

I used Chipquick to remove a broken ADC from my TDS540B acquisition board and from a donor board.
Worked great.

I added Chipquick to all the joints and heated the whole chip using hot air to remove it.
Afterwards I cleaned the soldering pads using a normal iron and flux.

Putting the donor part back was done using a normal, small flat soldering iron.

Pictures can be found here:

/g/TekScopes/album?id=21745

Some companies have ISO or other certifications that require all test equipment (there are some exceptions if marked Not Subject To Periodic Calibration or similar) be periodically calibrated by a certified test/cal facility.? I once got some of my personal HP equipment calibrated for free because I had brought it into the R&D Lab where I worked.? They didn't have all the gear I needed for a project so I brought in what I needed from home.? During the product's development they started work on getting an ISO certification that required all equipment in the lab and production areas to have current calibration stickers.? They calibrated my gear knowing it was going back home with me at some point, a sweet deal!

Clay Scott

Some companies have ISO or other certifications that require all test equipment (there are some exceptions if marked Not Subject To Periodic Calibration or similar) be periodically calibrated by a certified test/cal facility.? I once got some of my personal HP equipment calibrated for free because I had brought it into the R&D Lab where I worked.? They didn't have all the gear I needed for a project so I brought in what I needed from home.? During the product's development they started work on getting an ISO certification that required all equipment in the lab and production areas to have current calibration stickers.? They calibrated my gear knowing it was going back home with me at some point, a sweet deal!

Clay Scott

WHAT’S LYSDEXIA?


On Sat, Apr 25, 2020 at 4:44 PM Jeremy Nichols <jn6wfo@...> wrote:

AND SPREAD THE WORD OF DOG.


On Sat, Apr 25, 2020 at 4:27 PM Bruce <bruce@...> wrote:

I am also dyslexic and there was a Fac ook post recently that I loved:

DYSLEXICS OF THE WORLD UNTIE

:-)


Quoting Paul Bicknell <paul@...>:

Sorry I didn't mean it to, and you have to put up with my bad spelling

as I

am dyslexic



All my mechanical engineering until after I left school came from my

father

and that was grate as my best subjects where science and metalwork



Paul

_____

From: [email protected]
[mailto:[email protected]] On Behalf Of Mike

Feher

Sent: 25 April 2020 21:36
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] Test Equipment For Sale



Wow, did this turn into a contest? I was doing that in the early 60's.
That's enough from my book. All yours -



Mike B. Feher, N4FS

89 Arnold Blvd.

Howell NJ 07731

848-245-9115



From: [email protected]
<[email protected]> On Behalf Of Paul Bicknell
Sent: Saturday, April 25, 2020 4:19 PM
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] Test Equipment For Sale



Hi



In 75 I was playing with gun diodes and Klystrons at the age of 17 only
test equipment where diode detectors on the end of a meter and an
oscilloscope I had rebuilt after someone had used it for spare parts



Paul

_____

From: [email protected]
[mailto:[email protected]] On Behalf Of Mike

Feher

Sent: 25 April 2020 20:29
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] Test Equipment For Sale



Well, in the early 70's, to be able to do this at home, was fun and
rewarding. From some heavy math to a final product. -

<>
Mike

Mike B. Feher, N4FS

89 Arnold Blvd.

Howell NJ 07731

848-245-9115



From: [email protected]
<[email protected]> On Behalf Of Paul Bicknell
Sent: Saturday, April 25, 2020 3:00 PM
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] Test Equipment For Sale



Hi



yes a lot of filter work can be verified using peak hold can be time
consuming on a adjusting complicated filter



Regards Paul

_____

From: [email protected]
[mailto:[email protected]] On Behalf Of Mike

Feher

Sent: 25 April 2020 19:33
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] Test Equipment For Sale



In my early days of designing filters, I used to use a broadband noise
generator with a 141T spectrum analyzer. Worked great in max hold after

a

few sweeps. Just had to make sure the total in-band energy did not

overload

the front end. Was even able to adjust the zeroes pretty accurately.

Since I

only built Elliptics, I knew were the poles and zeroes were. 73 - Mike



Mike B. Feher, N4FS

89 Arnold Blvd.

Howell NJ 07731

848-245-9115

No virus found in this message.
Checked by AVG - www.avg.com <>
Version: 2016.0.8048 / Virus Database: 4793/15886 - Release Date:

08/14/18

Internal Virus Database is out of date.

No virus found in this message.
Checked by AVG - www.avg.com <>
Version: 2016.0.8048 / Virus Database: 4793/15886 - Release Date:

08/14/18

Internal Virus Database is out of date.

--

Jeremy Nichols
Sent from my iPad 6.

--
Jeremy Nichols
Sent from my iPad 6.

Very good Sean,

?

Good luck on the meters.?

?

Dave

?

?

On Apr 25, 2020, at 13:43, Bill E <solartron@...> wrote:

?This topic has sure mutated. Can't let it die now, I'll jump in with a bunch of random babbling, some I've done before, but hey, can't remember which topics. Senility, I suppose.

I posted a few days ago how horrible Win-based equipment was. Just what was anyone thinking? Maybe HP did get paid off my MS. XP was pretty atrocious, massively buggy, security issues all over the place. Both HP and Tek did good embedded sw, then went lame. OTOH, could be part of that move to outsource development, driven by clueless bean-counters. I transitioned from all HW to mostly all SW over my career, always fighting the 'we can hire 5 3rd-world devs for what we pay you'. Ok, do that, hmm, nothing works. So, still pay a real dev plus those 5 useless hacks. Bean-counters apparently can't figure out that they're paying 2x one good dev for 0.9 total output, since all the code has to be dealt with (rewritten) by the good dev.

Next rant, sw bloat. I ended up doing a lot of sw for telecom, massive servers, all in Java. Talk about bloated code, gigs of open-source libs, some dragged in just to get one trivial method. And, modern pseudo-devs have no understanding of hardware at all, and just as little understanding of basic sw algorithms, Just blindly use whatever they stumble across, then wonder why we needed a server farm to run what should have been possible on one machine.

Moving on, those DS 'top-hat' pseudo-nvrams. Who ever thought those were a good idea? 10 yrs, dead. I'm suspecting bean-counters again, make sure to obsolete the equipment. There's a guy selling ferromagnetic rams to replace said DS horrors for Tek scopes. Time to start doing the same for our HP stuff. And yes, my latest acquisition, 53310A, appears to have one of those damn things in it.

Have I run out? Of course not.

On to C vs C++. I really never liked C++, and don't get me started on C#. My definition of C++, a pseudo-OO variant done by someone that really didn't understand C. Stroustrup is on my hate-list. I've written hundreds of thousands of lines of C, and still my favorite lang for embedded tiny stuff. I started with IBM 360 assembler (BALR sound familiar, waiting for responses to that), then Fortran, then PDP-8, PDP-11, Multics, Unix, Linux. Used Unix at Bell Labs early on. It is truly amazing how it has evolved.

Hmm, any more? Oh, wait, of course.....

Cheap instruments from China. All based on various US highly-integrated chips. Some of the stuff is useful. most is just junk. Amazing 'specs' in the ads. Any real test data? Of course not. Get some of them, Ok, yes, they do the basic operations, but, example, cheap signal generators, horrible purity and stability. About as much out-of-peak energy as is in the fundamental.

Finally, modern instruments of any kind.

What happened to the great engineering we used to have? Now it's all about design-to-cost. Selling points seem to be 'internet connected'. I'll stack my $300 5370 time-interval analyzer up against anything you can get today. GUI? Nope. Internet? Nope. Amazing? Yes. Same for my 54542, and still amazing even today, given what one costs vs new junk. BTW, friend of mine from college worked at HP instrument division, he's told me many a sad tale about the decline of HP. Blames one person, Carly.

In closing, maybe I can start another flame war. My 54542 is the first HP scope I'd ever considered.Everything else I now have in my home lab is HP, but Tek? scopes really dominated. I love my Tek 2465A, one of the last great analog scopes. My HP friend said they got tired of everyone complaining about the lame triggering performance of HP scopes, decided to do a 'Tek killer', and succeeded. True? Don't know. I do know that one of the early Tek digital scopes that I had used this what seems to me to be a total hack, linear CCD to capture the analog, then read it out slowly enough for the lame A/D to handle. And, I do know I just love my 54542. Well, until the already-fading CRT dies. At least it doesn't run Win. :)

Hope I've amused you,
Bill

Oh, and I want to put a plug in for some of the classic Brit engineering. My 2 Datron 4910 voltage refs used to be national-standard level. My Solartron 7081 8.5 digit meters rock (although it is sloooow at 8.5, and really isn't quite as good as a 3458A, but at a fraction of the price if you can find one. BTW, it didn't use the LTZ ref both the 3458 and my Datron 4910's used. It had a 'zero-temp-factor' Zener, selected, and then compensated more by a temp measurement, trimmed per-Zener to get the best zero TC. Love it.

On Apr 25, 2020, at 15:36, "ghnatiuk@..." <ghnatiuk@...> wrote:

?Concerning the HP3458A 8-1/2 Digit Multimeter: A week ago a group member contacted me regarding his problematic HP3458A. The issue he brought up is important enough that I thought it would be best to address in Group chat rather than continue with a private conversation in case others face this same problem and to share a bit of development history with Group regarding the 3458A. The initial conversation is copied below to bring everyone up to speed and my reply will follow: ------------------------------------------------------------------------------------------------ Date: Sat, 18 Apr 2020 13:36:20 -0400 Hi there George, this may be a tall order but I hope you can help me out. My beloved 3458A has succumbed to the "ADC drift" failure of the A3 board. I've verified this as per the instructions in the related Service Note. I believe the solution for this is to just replace the A3 board, but I don't know if there's a better approach. Can you help? Good luck with your move, and stay safe during these weird times. --------------------------------------------------------------------------------------------------- On 4/18/20 2:35 PM, ghnatiuk@... wrote: Sorry but I do not remember assembly numbers. Give me the part number 03458-665xx or function. I never remember the A1,2,3 designations for any meter. We never used them in the lab -- that was a marketing designation given in the manuals AFTER the meters were designed so I never got to know what they were. LOL George ---------------------------------------------------------------------------------------------------- Date: Sat, 18 Apr 2020 15:12:10 -0400 Ohhh ok, sorry. I have the same problem sometimes with the PDP-11 minicomputers that I work with a lot at the museum. Their boards have "module numbers" and "option names", and I never remember the former! It looks like the A3 board is 03458-66503, called "A/D Converter and Inguard Logic". Another part number, 03458-69503, appears to pertain to the same board, but a "rebuilt" one. ---------------------------------------------------------------------------------------------------- On 4/18/20 5:30 PM, ghnatiuk@... wrote: That will be a tough one. The A/D board rarely ever fails. Not many spares around. I will look to see what I can do for you. Hate to see a 3458A ( code name: Sentry) go belly up. Is this meter a production unit with serial number? Need to identify vintage. George ------------------------------------------------------------------------------------------------- Date: Sat, 18 Apr 2020 17:44:57 -0400 Sentry, I didn't know that was its code name! Neat! Actually the A/D board has a high failure rate later in life; there's a lot of info about it online. I think that this wasn't the case early on, but later it was. It's discussed in Service Note 18. The converter boards don't outright "die", but they start showing high short term drift. My 3458A's serial number is 2823A 26542. Yeah I was very upset when I noticed the drift, then performed the test in that Service Note and verified that I had this problem. It's like a cancer rolling through older 3458As. -------------------------------------------------------------------------------------------------- Sat, Apr 18, 2020 04:59 PM Regarding the 3458A. I have been told by the cal lab techs at Keysight that the older first production 3458A can be much better that the current ones - more stable. This is due to newer instrument having different boards than what were first designed due to the need to revise critical parts that have become obsolete and discontinued from the vendor that went into the 3458a. It became necessary to redesign some of the boards from the original. I seriously have not followed the meter much since the mid 90s. We really did not dream the meter would still be in production today 32 years later and purchased a "life time" buy of critical parts back then that have since been depleted. If you could forward me a copy of that service note 18 so I could look it over, I would appreciate that. Thanks in advance, George ----------------------------------------------------------------------------------------------- OK, so here we are. I have a copy of the Service Note 18A/B and discussed the note with the author, who is a good friend of mine and worked in the lab with me. Together we sat over lunch and discussed the issues and filled in the missing pieces of information each of us had given that the note was written 13 years ago. There are two problems concerning drift in the 3458A and they are not related. The first issue has to do with the long term drift due to the reference which is an LTZ1000 along with associated support parts to get it at the FLAT part of its drift curve. The serial numbers in the service note: US28031400 / US28032927 most likely have to do with a manufacturing problem in the US that occurred when the instrument was being transferred to Malaysia for production. This was an embarrassing incident for Agilent. As sometimes happens during these transfers, not all goes according to plan. During this time, the references where being burned-in and put into sorting bins from the burn-in rack and were to be selected on a first-in , first-out basis as needed in production but, manufacturing was not following strict protocols regarding reference selection and this was not caught during the transfer until later. Many references sat in the bins a very long time and reverted back to a pre-aged state. The suspect instruments (about 1500 units) have been flagged in this service note and when they come into Loveland for calibration, they are subject to testing to see if their references are up to spec and the instrument powered on an appropriate period before calibration to insure the reference has low drift. That is what the following statement in the service note is addressing: If the reference assemblies are powered down for an extended period of time (such as being stored in a stock bin) the references revert back toward their “pre-aged state”. As a result the initial drift rate for the DCV reference may be high enough that the DCV function may be “out of specification” prior to the first calibration/adjustment interval (90 days or 1 year). Some of our references used in recent production have been stored in a stock bin long enough to exhibit this drift problem. The error associated with this long-term drift issue will cause a gain error for all voltage, resistance and current measurements. This error is expected to be less than 15 ppm of the reading. This tendency for references to "revert" back toward their pre-aged state when powered down for extended periods of time can be avoided by keeping your instrument continually powered up. There are other good reasons for never turning off your instrument that I address below. For those of you that are curious, the high stable reference option is just a selection from the references being burned-in over time at the production facility. The 3458A reference is powered up in a test rack with 100s and continuously monitored for drift over a period of 3 months or so. The lowest drift units are selected for the high stability option and monitored longer. It is the same reference as used in the standard 3458A just selected for low drift. There is a phenomena with respect to the reference that has to do with a power on/off offset that can occur with the reference that is not addressed in this service note. This problem has never been understood by engineering at either HP/Agilent/Keysight nor at Linerar Technology (now Analog Devices), who manufactures the LTZ1000. Most references are well behaved but some are not and the problem is independent of aging. Eventhough an LTZ1000 reference may be long term burned-in under severe conditions to stress it, it may take a SET when instrument power is cycled. That is, the reference voltage could change a bit (OFFSET) from the last power up state. There is no way to determine which references will shift as far as we know. It just happens. Curiously enough, the older production units from Loveland do not seem to have this problem. It took quite awhile for this problem to be identified for most customers always kept their 3458s continuouly powered up and the reference never shifts as long as it is powered up. The problem does not manifest itself until the instrument is power cycled. So all the more reason to keep your instrument powered up even when you are not using it. This is actually true for any instrument that uses voltage references of any type - keep the instrument always powered up - do not turn it off. A voltage reference will get better the longer it is powered up -- it becomes more stable - less drift over time - they settle down with time. Fortunately, this power cycle offset NEVER shifts the 3458A out of spec. It only becomes a problem in applications where you are trying to get the most out of the instrument with its 24 hour spec and need to be well within the 24 hour error window.

Sorry I didn't mean it to, and you have to put up with my bad spelling as I
am dyslexic



All my mechanical engineering until after I left school came from my father
and that was grate as my best subjects where science and metalwork



Paul

_____

From: [email protected]
[mailto:[email protected]] On Behalf Of Mike Feher
Sent: 25 April 2020 21:36
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] Test Equipment For Sale



Wow, did this turn into a contest? I was doing that in the early 60's.
That's enough from my book. All yours -



Mike B. Feher, N4FS

89 Arnold Blvd.

Howell NJ 07731

848-245-9115



From: [email protected]
<[email protected]> On Behalf Of Paul Bicknell
Sent: Saturday, April 25, 2020 4:19 PM
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] Test Equipment For Sale



Hi



In 75 I was playing with gun diodes and Klystrons at the age of 17 only
test equipment where diode detectors on the end of a meter and an
oscilloscope I had rebuilt after someone had used it for spare parts



Paul

_____

From: [email protected]
[mailto:[email protected]] On Behalf Of Mike Feher
Sent: 25 April 2020 20:29
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] Test Equipment For Sale



Well, in the early 70's, to be able to do this at home, was fun and
rewarding. From some heavy math to a final product. - Mike



Mike B. Feher, N4FS

89 Arnold Blvd.

Howell NJ 07731

848-245-9115



From: [email protected]
<[email protected]> On Behalf Of Paul Bicknell
Sent: Saturday, April 25, 2020 3:00 PM
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] Test Equipment For Sale



Hi



yes a lot of filter work can be verified using peak hold can be time
consuming on a adjusting complicated filter



Regards Paul

_____

From: [email protected]
[mailto:[email protected]] On Behalf Of Mike Feher
Sent: 25 April 2020 19:33
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] Test Equipment For Sale



In my early days of designing filters, I used to use a broadband noise
generator with a 141T spectrum analyzer. Worked great in max hold after a
few sweeps. Just had to make sure the total in-band energy did not overload
the front end. Was even able to adjust the zeroes pretty accurately. Since I
only built Elliptics, I knew were the poles and zeroes were. 73 - Mike



Mike B. Feher, N4FS

89 Arnold Blvd.

Howell NJ 07731

848-245-9115

No virus found in this message.
Checked by AVG - www.avg.com <>
Version: 2016.0.8048 / Virus Database: 4793/15886 - Release Date: 08/14/18
Internal Virus Database is out of date.

No virus found in this message.
Checked by AVG - www.avg.com <>
Version: 2016.0.8048 / Virus Database: 4793/15886 - Release Date: 08/14/18
Internal Virus Database is out of date.

On Apr 25, 2020, at 4:09 PM, "[email protected]" <[email protected]> wrote:

?Hope you have a speedy recovery!

Appreciate the clarification. HP left enough slack in those option 002 cables to slide the rear panel assembly back far enough to get the fan module out. I have successfully extracted the bad fan and will now examine it to see if maybe it can be cleaned and put back into service. Otherwise, time to open the other meter.

Sean

On Sat, Apr 25, 2020 at 03:26 PM, Dave Smith W6TE wrote:

I’m in my hospital rm after having knee surgery yesterday. You’re right, four corner screws no eight.?

?

If you can’t move the back panel enough to get the fan in and out, you’ll need to unplug the sensor cables from the input board. They should be labeled but, if not, just note where they go. You should be able to move the back panel Enough without completely pulling the sensor leads back.?

Dave

I’m in my hospital rm after having knee surgery yesterday. You’re right, four corner screws no eight.?

?

If you can’t move the back panel enough to get the fan in and out, you’ll need to unplug the sensor cables from the input board. They should be labeled but, if not, just note where they go. You should be able to move the back panel Enough without completely pulling the sensor leads back.?

Dave