Author Topic: EEVblog #757 - HP4263A LCR Meter Teardown  (Read 12002 times)

0 Members and 1 Guest are viewing this topic.

Online EEVblog

  • Administrator
  • *****
  • Posts: 29884
  • Country: au
    • EEVblog
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4045
  • Country: tw
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
Re: EEVblog #757 - HP4263A LCR Meter Teardown
« Reply #1 on: June 24, 2015, 11:02:09 am »
Great score Dave, I scored 4263B around a year ago, for those who interested in comparison, I had documented some repairs and teardown photos about B version of 4263 in this forum thread

Also we had some sucessful hacking for options around this thread.
Options are free now, but only if you buy new instrument from Agilent/Keysigh (which is not trivial, as these LCR's are EOL long ago).
You cannot get free option key for already used unit.  :palm:

Using my LCR with cheapie ebay LCR twizzer, which I had modify following genuine HP LCR connection scheme. Step-to-step tutorial I had in article about my 4263B.

Dave, can you get a photo of input connection area on analog PCB, like this angle:



I'd like to know resistance value near J101 port.

« Last Edit: June 24, 2015, 11:10:18 am by TiN »
YouTube | Chat room | Live-cam | Have documentation to share? Upload here! No size limit, firmware dumps, photos.
 

Offline VK3DRB

  • Super Contributor
  • ***
  • Posts: 1507
  • Country: au
Re: EEVblog #757 - HP4263A LCR Meter Teardown
« Reply #2 on: June 25, 2015, 01:16:11 pm »
I'm surprised you don't have an ESD safe attachment for your vacuum cleaner. Compressed air just blows the dust everywhere, much better to suck it up.

Agreed. Blowing dust around creates a horrible mess and compressed air cans cost $ if you don't own a real compressor and air gun. One trick I found was to use a small horse's hair paint brush (NOT nylon), and a trusty Dyson vacuum cleaner, even if it does have no carbon emissions. Use the paint brush to stir up the dust and suck it up with the vacuum cleaner. Sucking air is reputedly better than blowing air to minimise risk of ESD damage to components. Using this method, I give my PC's a good clean once per year to keep the insides looking and breathing like new. And there is no mess.

By the way, that HP LCR meter is nice meter to own. Quality capacitors and manufacturing so it is working like new after 20 years.
 

Online rsjsouza

  • Super Contributor
  • ***
  • Posts: 3454
  • Country: us
  • Eternally curious
    • Vbe - vídeo blog eletrônico
Re: EEVblog #757 - HP4263A LCR Meter Teardown
« Reply #3 on: June 25, 2015, 03:56:58 pm »
Interesting review; some lunatics have been tagging these above $1k on eBay, which is really not worth it.

I have been reading around on the internet that the DER DE5000 seems quite comparable to the Agilent 1733C while costs half. Is there any particular reason why the Agilent is better?
Vbe - vídeo blog eletrônico http://videos.vbeletronico.com

Oh, the "whys" of the datasheets... The information is there not to be an axiomatic truth, but instead each speck of data must be slowly inhaled while carefully performing a deep search inside oneself to find the true metaphysical sense...
 

Offline free_electron

  • Super Contributor
  • ***
  • Posts: 7234
  • Country: us
    • SiliconValleyGarage
Re: EEVblog #757 - HP4263A LCR Meter Teardown
« Reply #4 on: June 25, 2015, 06:40:47 pm »
BNC board to board connectors ? BNC ?? BEE ENNN EFFING CEEE ? DAVE ! they are SMB connectors !

As for the National semi part : that is an ULA Uncommitted LogiC array.  Basically a chip that has a bunch of logic gates Think of it as an FPGA that you mask program. THey make these wafers in mass quantitiy , stop the before metallisation. THen when an order comes in they take the half finshed wafers from stock , run the metallisation pattern ( basically the interconnects) and deliver a batch.

Doing it this way keeps the cost down , even for small production runs
1) the wafers are pre-made in large volume keeping the cost per wafer low
2) the customer only pays for the two metallisation masks that are unique to his design
3) since they have premade wafer sin stock they can spin this thing in single batches of 25 or 50 wafers at a time. so you don;t need mass orders.

AMI and LSI were the kings of this kind of design in the mid 80's bbut as complexity of designs increased this approach became too costly for larger designs and it died out. It is sometimes also called a a 'sea-of-gates' design

The package you have on the board sis the plastic brother of the PGA ( Pin grid array like they used on 486 cpu's. instead of the expensive ceramic package they use a circuit board as substrate and then simply overmold it with epoxy just like a normal chip.

the only Difference between A and B is the processor and display. the big motherboard is identical. They can measur ethe same. only the two line display can show you more data simultaneously, on the B you need to press a button to go to the next parameter.

the DC dC converter is the DC bias supply .

the Asahi Kasei is a A/D converter. it spits out a PCM bitstream. i have the datasheet for it.
« Last Edit: June 25, 2015, 06:43:48 pm by free_electron »
Professional Electron Wrangler.
Any comments, or points of view expressed, are my own and not endorsed , induced or compensated by my employer(s).
 

Offline simingx

  • Contributor
  • Posts: 32
  • Country: sg
Re: EEVblog #757 - HP4263A LCR Meter Teardown
« Reply #5 on: June 26, 2015, 12:57:45 am »
Dave, did you ever get around to taking apart the 4260A I sent you, a while back? Would be interesting to compare how far they've come since then...
 

Offline HighVoltage

  • Super Contributor
  • ***
  • Posts: 4259
  • Country: de
Re: EEVblog #757 - HP4263A LCR Meter Teardown
« Reply #6 on: June 27, 2015, 11:38:37 am »
Interesting tear down and explanation, thanks.
When it came to good quality LCR meters, for some reasons, I always stayed with the Philips / Fluke PM6303A and PM6306 with a much much easier user interface but with the same principle of operation.

All the Fluke PM630x came with a little chart, explaining the operation in one nice picture.
Here is a small but helpful excerpt of the chart, showing the different angles of reactance.

The Original Philips handbook also has an amazingly good explanation of operation with all the math inside.




 
There are 3 kinds of people in this world, those who can count and those who can not.
 

Offline TheAmmoniacal

  • Supporter
  • ****
  • Posts: 1150
  • Country: no
  • Cut me some slack, I'm a biochemist!
Re: EEVblog #757 - HP4263A LCR Meter Teardown
« Reply #7 on: June 28, 2015, 08:28:20 am »
(...)
I have been reading around on the internet that the DER DE5000 seems quite comparable to the Agilent 1733C while costs half. Is there any particular reason why the Agilent is better?

A video recommendation for Dave, do a quick LCR-meter shootout - a side-by-side comparison of the DE-5000 and U1733C, and other handheld LCR meters you own. Just get some F/H/ESR measurements on each meter and compare with the HP4263A measurement (take this value as reference?).

I would highly appreciate it!

I collect [corporate] mugs.
MTBF ~ 700.000 h
 

Offline PTR_1275

  • Frequent Contributor
  • **
  • Posts: 496
  • Country: au
Re: EEVblog #757 - HP4263A LCR Meter Teardown
« Reply #8 on: June 28, 2015, 09:13:38 am »
A shoot out would be fantastic. Even throw some of those $20 eBay lcr testers.

I'm in the market for a better lcr meter and Dave's videos have persuaded me one way or the other on several of my purchases.
 

Online EEVblog

  • Administrator
  • *****
  • Posts: 29884
  • Country: au
    • EEVblog
Re: EEVblog #757 - HP4263A LCR Meter Teardown
« Reply #9 on: June 28, 2015, 09:58:42 am »
Only LCR meters I have left I think are the ("genuine") DE-5000 from IET, and the U1733C
The BK Precision 879 and Extech ones I gave away.

Not sure if the old DE-5000 from IET is representative of the current ebay cheapies?
 

Offline TheAmmoniacal

  • Supporter
  • ****
  • Posts: 1150
  • Country: no
  • Cut me some slack, I'm a biochemist!
Re: EEVblog #757 - HP4263A LCR Meter Teardown
« Reply #10 on: June 28, 2015, 07:15:19 pm »
I don't actually know the differences between the "genuine" IET DE-5000 and the cheaper DER EE DE-5000. I thought they were the same, both based on the same Cyrustek chipset (ES51919/ES51920), just one being a rebadge of the other. IET Labs might have their own calibration procedures?
I collect [corporate] mugs.
MTBF ~ 700.000 h
 

Offline RussellB

  • Newbie
  • Posts: 1
Re: EEVblog #757 - HP4263A LCR Meter Teardown
« Reply #11 on: July 06, 2015, 02:45:31 am »
I'm surprised you don't have an ESD safe attachment for your vacuum cleaner. Compressed air just blows the dust everywhere, much better to suck it up.

Agreed. Blowing dust around creates a horrible mess and compressed air cans cost $ if you don't own a real compressor and air gun. One trick I found was to use a small horse's hair paint brush (NOT nylon), and a trusty Dyson vacuum cleaner, even if it does have no carbon emissions. Use the paint brush to stir up the dust and suck it up with the vacuum cleaner. Sucking air is reputedly better than blowing air to minimise risk of ESD damage to components. Using this method, I give my PC's a good clean once per year to keep the insides looking and breathing like new. And there is no mess.

By the way, that HP LCR meter is nice meter to own. Quality capacitors and manufacturing so it is working like new after 20 years.

Yes, I agree also. Vacuuming is the way to go (with a natural bristle brush). Using a HEPA filter in the vacuum cleaner's exhaust isn't a bad idea, either. There is no telling what is in the dust from the labs from whence the test equipment came. I have a couple of 'scopes that came from a smelter and there is some pretty gritty stuff in there. I'm thinking I may even do the water wash on the 'scopes as per maintenance manual-- maybe not.
Speaking of carbon emissions, I learned how to vacuum by cleaning powdered ink from xerographic duplicators. It would be a mess of a different magnitude with compressed air!
BTW, I have a DER DE5000 I'm quite happy with.
I really enjoyed that video. Thanks!
« Last Edit: July 07, 2015, 10:52:35 pm by RussellB »
 

Offline SeanB

  • Super Contributor
  • ***
  • Posts: 15117
  • Country: za
Re: EEVblog #757 - HP4263A LCR Meter Teardown
« Reply #12 on: July 06, 2015, 06:57:17 pm »
I have 2 vacuum cleaners at work in my office. A regular one for dust and stuff and the "dirty" cylinder one with the HEPA filters for fine powders. I cleaned one after a few years of operation and basically took it apart and washed the entire unit, motor and all, to get the fine room dust out.  Lubricated it and it lives again.
 

Offline jeffg

  • Newbie
  • Posts: 2
  • Country: us
Re: EEVblog #757 - HP4263A LCR Meter Teardown
« Reply #13 on: July 01, 2016, 08:38:05 am »
A couple of things:

All HP/Agilent/Keysight LCR/Impedance analyzers are made at the same factory in Kobe, Japan at "Kobe Instrument Division".  Basically any model starting with 42xx or 43xx is a KID product, specifically so-called "Product Line 36" or PL36, and is some type of impedance measurement device.  HP ran out of 4 and 5 digit numbers in the late 1990s so now it's E49xx for LCR meters now.  So because this is a 4263A, it's KID and its an LCR meter.  (Yes, pretty much ALL HP/Ag/Ks instrument can be broken down by their part numbers)

Both KID and HP/Ag/Ks' other Japanese division, HSTD (Hachioji System Test Division) were originally part of a joint venture between HP and Yokogawa formed in the 1960s, which was called "YHP" or Yokogawa-Hewlett-Packard.  This joint venture was 100% bought out by HP in the mid-1980s.  Occasionally you see old YHP logos but it's been absolutely disconnected from Yokogawa for decades.  Yokogawa still has its own instrument product lines but there is no longer any interaction between them and HP/Ag/Ks. 

HP got into this market because of the growing Japanese electronics industry in the 1960s but because they were not Japanese they were not accepted.  Japanese companies buy from Japanese companies first and tenth.  So HP got a Japanese partner, Yokogawa to build Japanese legitimacy in the local Japanese market - it took 20 years to stand on their own with the help of Yokogawa.

The reason for the linear supply is for the instrument side - you need low noise for most instrumentation so linear is the way to go.  Switcher noise is problematic.  The switcher is for the digital which is more forgiving and isolated from the sensitive analog front-end.

The handler interface is specially for testing surface mount devices (capacitors and inductors) using package part handlers which feed raw SMDs in and output tapes of parts that have passed spec and binned by performance - the LCR meter can automatically trigger binning by things like Q, ESR or value.  So those reels of parts you use likely were tested with a KID LCR meter. 

All the derived units like capacitance, inductance and other values like ESR, Q, etc. are "dumb algebraic derived" and depend on getting the perfectly corrected raw magnitude-phase value first which is where the Open-Short-Load come in - these are for correcting the fixturing that holds your DUT and getting that to "zero".   The delta Ref is only for "icing on the cake" final correction where you have only one type of uncorrected offset i.e. only capacitance or only inductance or only resistance but never combinations of these - for these you need to use Open-Short correction because removing it requires a 2-port (4 distinct complex values) rather than a 1-port correction (1 distinct complex value).   

The UI on all Japanese HP/Ag/Ks instruments suck - it's just how it is.  Software is NOT their thing - building hardware is.  The best UIs come from either Keysight Colorado Springs (where the scopes come from and which are the very best) or from Sonoma county (where spectrum analyzers and network analyzers come from).  These are also the biggest and most profitable divisions so they can afford to invest in UI (which is not cheap).  HP/Ag/Ks have always pushed profit-and-loss responsibility down to divisions/PLs and they also decide what to invest in for their product designs as part of this, so you see a far amount of variability across the entire corporation; but you avoid bureaucracy because decision are taken locally in a small group of people who know the product, technology and market best.

These are still great LCR meters because of both the accuracy and the measurement ranging, which far exceeds most lower cost units.  The main reason is the methodology of measurement which is a "auto-balancing bridge" method.  Most low-end LCR meters use a a RC/RL time constant method which is not amenable to manufacturing test but is good enough for incidental hobby and technician use.  One thing you can only do with an bridge method is apply variable AC and DC bias while measuring the impedance which is important for ceramic capacitors and ferrite core inductors as these vary.  Also you can measure far lower and high values better (most PL36 LCR meters can measure down into the fF range and up to F ranges).  This makes even older units golden.  I still see production test lines using 4284A and 4263A LCR meters even thought these have been obsoleted 2 decades (most recently a few months back).

I used to be the field applications engineer for PL36 (as well as others including PL1H which are products from HSTD and others).  If you have questions on these, especially measurement theory about LCR meters or any other instruments generally, feel free to ask.  I was trained on every instrument in the old HP catalog.

Also a good reference for LCR meters and impedance measurement is the Impedance Measurement Handbook which is written by KID.

http://cp.literature.agilent.com/litweb/pdf/5950-3000.pdf
 
The following users thanked this post: oPossum, materialsguy


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf