Author Topic: #2000 Post : Teardown and study of Fluke 5700A Calibrator [56K warning!]  (Read 26414 times)

0 Members and 1 Guest are viewing this topic.

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4206
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
Post 2000!

Fluke 5700A of course!



Offtopic intro

If you may wonder, where did I disappear and why no posts last week, now you know why. I wanted to cook and publish this thread first to celebrate my 2000th (O-o) post here. But before that I'd like to have small retrospective. My first post here was tear-down/repair of Tek CSA7404 thread bit over 3 years ago, and I really did not expected that to become so popular, as even broken these Windows-based old Tektronix scopes still often fetch pretty penny. Then Keithley 2001 repair thread was started, covering about 7 years of my repair shenanigans, all stated with first bench-top DMM, bought on eBay for $200 USD. Rest of the story you may know, got infected with volt-nut virus and off we go down to the endless rabbit hole. Now there is a chance that I have more 8.5-digit DMMs than some of calibration labs might have :). And I'm still always information-starved.

But amount of old/broken/fixed accumulated equipment does not matter, if one can't get it used and calibrated. So to resolve this issue two things were started, one of which own design of DC voltage reference and DAC/ADC modules for experimenting and learning purposes. Second is calibration standards and equipment projects were kicked off to be able source stable DC, AC voltages and currents and resistance check points. Most obvious result of this are my LTZ1000(A)-based voltage reference modules, some of which already got into EEVBlog members hands. I sent tens of blank PCBs to community too for their own tinkering, hopefully to see more weird and funny stuff posted. And recently in 2016 I was able to perform few cross-checks and DIY calibrate both K2002's to act as transfer standards for volt and ohm into my home-lab. I'm not very interested in RF stuff (mostly because I don't have deep understanding of it, I recon), but that might change in 2017, if I get that 8753A VNA in functional condition.

Now before we dig deep into the teardown, I'd like to express my appreciation to daily lurkers of EEVBlog Metrology and T&M forums, especially Dr.Frank, ManateeMafia, lowimpedance, Kleinsten, plesa, quarks, VintageNut, HighVoltage, chuckb, lymex, Andreas, Edwin, doktor pyta, Alex Nikitin, and many more. Wish only you guys post more often all that exciting and fun stuff :)

Intro

Fluke calibrators are often in the dreams of many volt-nuts. This class of instruments is mainly found only in calibration labs and scientific installations and less often in analog design houses. Old but still mighty units like 5440A/B sometimes manage into volt-nut hands, despite their 4U full rack-mount size weight and price over $1000 USD :) But newer higher-spec 5500A/5700A series units are still even on used-marked in broken condition go for over $10K+ price tags. Even then their specs are barely reaching needs of average passionate volt-nut with DMM like 3458A to justify the cost. Of course having stable source of arbitrary voltages and currents is a great help in the lab, but that goal can be achieved by set of cheaper instruments. Over last decades Fluke multi-function calibrators become recognized and desired industry standards for many calibration and metrology labs all over the world, just like 8½-digit HP 3458A from DMM world. Started with Fluke 3xx series standards, today these instruments evolved into lab-in-the-box tools like Fluke 5730A, some of the very best commercial calibrators money can buy.

I was given a chance to have access and play with mighty Fluke 5700A, including DIY-level service, so prepare yourself :). Fluke 5700A is already discontinued model today, but that does not make it any worse. MFC such as covered here Fluke 5700A is a high precision instrument that can be used as stable source to calibrate electrical measurement instruments and sources. 5700A available functions and provided ranges allow to test, calibrate and adjust practically any handheld and bench multimeters with a reading resolution all the way up to 8½ digits. In corner cases with additional guard-banding Fluke 5700A can be used to test and calibrate metrology DMMs such as HP 3458A. This MFC in most of user cases allow to replace whole range of expensive and bulky calibration standards, and can be itself calibrated by only three reference standards - 1 ?, 10 K? and 10 VDC, similar to 3458A DMM. We will also dig a bit into design of the box, perform required service, repair damaged parts if any, and do some tests for it's performance. This also means we will not be doing expensive boards swap, unless absolutely unavoidable and necessary. If you think 3458A overhaul repair is expensive, you don't want to know pricing for 5700A...  :o

All information posted here is hosted for education purposes only and provided AS IS. In no event shall the author, EEVBlog, xDevs.com site, or any other 3rd party, including Fluke be liable for any special, direct, indirect, or consequential damages or any damages whatsoever resulting from loss of use, data or profits, whether in an action of contract, negligence or other tortuous action, arising out of or in connection with the use or performance of information published here. If you don't agree, better close this page right now.

As usual, all photos are clickable with high-resolution version. Please note, that if you desire to use photo for your own reference/article/site, it's OK to do so only if you include link to this thread or originator of this content. It's NOT okay to steal photos for your eBay listings, sell threads or your pages without a reference. This simple rule is to ensure that our time not wasted, and community can benefit from the performed documenting work in equal and fair manner.

Another note - if you get impression that buying used MFC (or even broken), like this Fluke beast for half of usual market price (which is often north of 20K$) and fixing it would be easy way to get high-end lab standard or nice profitable quick resale, consider mountain of hidden costs and caveats ahead. Accredited calibration for top of the line MFC is not a trivial task and cost many thousands of USD, can usually be performed only by professional labs with 90-day calibration loops and characterized standards and even then you need to have multiple verified stable standards like Fluke 732B/742A and 3458A's with maintained calibration plans to ensure long-term calibrator operation and specs. So think clearly, when you entering this very conservative market, as it is just endless black hole for funds. But volt-nuts should already know that by now, having that JJA in garage  :bullshit:.

Calibrator purpose, typical use case

What scenario would justify paying high dollar for the instrument like calibrator? Often these are used for calibration and adjustment of other measuring and generating instruments, such as DMMs, power regulators, amplifiers of different kind, ADC/DAC testing, reference module evaluation and so on. Calibration commonly defined as determination of ratio relative from known signal to unknown signal. In case of measuring device - known signal is applied to the DUT meter, and DUT meter's reading is compared to expected signal value. In case of source - unknown output signal is measured and compared to known signal. To provide good confidence and meaningful results, reference known signal should be at least 2 times more stable than DUT unknown signal, better more.

Let's take a brief example with 5.5 digit DMM and calibrator as reference to perform a calibration and decide if adjustment is needed. To keep things simple test applied only on single function and single range of DMM, 10V, as idea is same for rest.

1. Verified and in-spec (8 ppm) calibrator is used to generate 10.0000 VDC reference signal.
2. Reference signal is connected to DUT DMM, which has specification 30 ppm for 10VDC range.
3. Expected value from calibrator is from 9.99992 to 10.00008 VDC.
4. DUT DMM measurement result is 10.0013 VDC
5. Calculated error is 10.0013 / 10.00000 is 130 ppm (0.013%) +/- 38ppm.
6. Even in best case (10.0010 (-30ppm) / 10.00008 (+8ppm) ) DUT DMM is out by +92 ppm
7. Adjustment/repair of DUT DMM is required to meet the spec.

Of course, this test and comparison can be done without calibrator, with only known tested 10V source. But even basic DMMs usually have many ranges, such as 100mV, 1V, 10V, 100V, 1000V, have multiple functions like DC voltage, AC voltage, DC current, AC current, resistance, capacitance and so on, with their own ranges. To have set of required sources for all these references signals to check each function and range will require huge investment and lab space if we talking about testing even mid-range 6.5 DMMs, leave alone high-performance 7.5+ digit devices. Having everything combined in single compact 4U unit makes multi-function calibrators perfect tool with added ease of use.

That's why calibrators, like Fluke 57xx series are often used as working standards in calibration labs, often just two steps behind national primary standards. In case of DC voltage first top level is primary standard, such as Josephson-based Voltage Standard. Second level is bank of solid-state references, such as Fluke 734A/732B or "golden" stable metrology-level DMM like HP 3458A. Each underlying level is calibrated and compared to higher level, to determine error value, called measurement uncertainty. Then calibrator is tested to these higher standards and used as working standard for lower level comparison and instruments. Eventually every DMM , even those handheld have relation to the primary standard thru all middle levels, with added up uncertainties.

Another case of calibrator use is various precision and sensitive lab experiments in analog design. This is because high-end calibrators can provide much more stable signal, acting as reference source for voltage/current than most generators/supplies can provide. Common example of this need is ADC/DAC testing, opamps, voltage references, passive components characterization and so on. Today's even 5$ ADC can have 24-bit of resolution, which dictate difficult tasks for validation and testing such ICs to make sure they meet design specs.

Inspection and overview

Alright, enough theory, let's crack it open! We are dealing today with mighty Fluke 5700A Series I Calibrator today. Front panel plastic is yellow due to UV and age, but that's reversible and not a big deal breaker. No deep scratches or damages visible, that's good news.  :-+



It also has fairly recent calibration, dated August 2012. This is perhaps a good thing, if unit worked thru most of it's life and still was fine four years ago.



Both sides of calibrator are flooded with vent holes to get excess generated heat away. Given presence of active cooling in the unit, one should be ready to see plenty of dust inside. Label on bottom cover screw was removed before, so somebody had a peek inside, hmmm.  ::)



Rear side reveals installed rear terminals, which is an option and not always available. All connectors here are in good shape. Air filter is clean, so perhaps it was cleaned not so long time ago. This box was covered by Fluke's service plan warranty once, from 30 October 1992 till 29 December 1997, so it's likely that unit was manufactured and sold in 1992 but was not kept on continuous Fluke service plan since then. Still that some good 24 years till today, which is good and bad news. Good news - usually voltage references, resistors and parts are very well aged to get nice and stable performance. Bad news is that all electrolytic capacitors are likely at their end of life, and other parts perhaps got old enough to fail/reduce performance. That's if nothing drifts away/faulty like famous U180's in HP 3458A's.



Main HI/LO on front saw some wear and tear, but rear ones are just like new, if not accounting yellowed plastic color. These banana posts are pure copper, to provide low-EMF connection to ensure good stable connections. Only downside is that pure copper oxidize easily, so connectors must be often cleaned to ensure best contact. Interesting note, these connectors look very much like Lowthermal Model 2758 binding posts. I have bought some of these before, they are very nice, but cost about $23 USD, each.



Keyboard bit dirty, some of paint labels on rubber pads are gonsky, but all keys present and no major damage :). To clarify which series of calibrator we have, serial number can be used. Any serial number below 6565601 is first-generation Series I, while serial number from 6565601 are updated fancy Series II.

Manuals references

Fluke does provide all main manuals, including service information right on their product page, without requirement of any pesky registrations or access requests, unlike Tektronix or Keysight. Thumbs up for Fluke, way to host the data on your product page. Few of key documents mirrored below, just in case they get moved.

Brochure: The 5700A/5720A MFC: Taking Accuracy to New Level

Fluke 5700A/5720A Series II Operators Manual, May 1996, Rev.2

Fluke 5700A/5720A Series II Service Manual, Nov 2007 w/o schematics

The 5700A/5720A Series II High Performance Multifunction Calibrators - Extended Specifications, Rev. 04

And of course, the most important document for repairs/maintenance - 570-page service manual. Credits for hosting it freely going to KO4BB portal. We are lucky to also have schematics, list of parts, assembly drawings of the boards included in it!  ^-^

Fluke 5700A/5720A Series II MFC Service Manual with schematics, Rev.1,3/02, June 1996

5700A Series I and 5700A/5720A Series II difference

According to a great An evaluation of the 5700A Series II Artifact Calibration document, major change in Fluke 5700A/5720A Series II are related to Artifact Calibration. Artifact Calibration was improved on a number of ranges, to minimize biases. Also interpolation algorithms were changed for AC voltage, particularly at higher frequencies; corrections for the level dependent AC/DC differences at low frequencies (low frequency AC-sensor correction). A new interpolation algorithm is also used for lower ACI ranges. Digital processor board A20 and it's firmware was changed as a result. Model 5720/5700 Series II calibrator has a modification to the in-guard microprocessor circuitry to minimize noise in the low millivolt DC ranges. Also, because of the tighter specifications of the 5720A, there may be more selected components in the calibrator circuitry, than in the 5700A Series I and some other minor bug fixes and improvements.
« Last Edit: November 11, 2016, 04:40:41 am by TiN »
YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4206
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
Re: #2000 Post : Teardown and study of Fluke 5700A Calibrator
« Reply #1 on: November 11, 2016, 04:08:11 am »
Diagnostics

Initial checks didn't reveal big problems first, so tried to power the unit on to see if it reports any issues or failures. For initial checks HPAK 3458A, calibrated fairly recent in June 2015 was used as a reference meter. Checked input power switch settings and unit was powered on.



Initial results are very promising, even with only 30 minute power on time (which is not enough for gear like this!) readings are pretty close, considering 4 year old calibration on 5700A.

Output 10.000000 VDC, Measured 10.000005 VDC, Error +0.5 ppm
Output 10.000080 k?, Measured 10.000069 k? 4W, Error -1.1 ppm

Too good to be true? Calibration self-check reported largest change since last calibration is on 220VAC output, highest 1.0000MHz frequency. It's 105% of the specification. This is internal check, as standard configuration 5700A without help of 5725A amplifier cannot provide 1MHz 220VAC.



Simple test program in Python was written to talk with MFC to get some diagnostic data, as well as calibration constants check. Raspberry Pi with linux-gpib package installed and National Instruments GPIB-USB-HS was used. Same setup as in my detailed guide about setting Pi and linux-gpib.

MFC spit next stuff:

Code: [Select]
Fluke 5700A - detected, Version: JA*
This is Series I unit
-i- Reading initial calibration data from 5700A
Unit last calibrated : 1308 days ago
Unit running time : 111438 hr
CAL CONST 6.5V reference voltage : +6.891366504559953E+00
CAL CONST 13V reference voltage : +1.379482200889601E+01
CAL CONST 10KOHM standard resistance : +1.000044087249608E+04
CAL CONST, Zero calibration temperature : +2.245000076293945E+01
CAL CONST, All calibration temp : +2.245000076293945E+01
Calibrator UNDER REMOTE CONTROL
Calibrator STABLE (settled within spec)
Calibrator is cooking report over SERIAL
User string PUD : #2055700A
Fatal errors history: "
 3/17/93  8:18:20  Fault   230: 5725 RESET (power-up or watchdog timer)
 3/30/93  7:07:49  Fault  4014: Sequencer Timed Out Waiting for 5725
10/05/94 18:53:17  Fault   218: 5725 +400V Supply Too Large
10/14/94 20:32:02  Fault   218: 5725 +400V Supply Too Large
11/07/94 12:27:47  Fault   218: 5725 +400V Supply Too Large
11/07/94 12:28:06  Fault   218: 5725 +400V Supply Too Large
 1/06/95  7:56:16  Fault   218: 5725 +400V Supply Too Large
 1/06/95  7:56:39  Fault   218: 5725 +400V Supply Too Large
 1/06/95 12:34:42  Fault   218: 5725 +400V Supply Too Large
 1/25/95  7:59:37  Fault   230: 5725 RESET (power-up or watchdog timer)
 1/25/95  8:00:18  Fault   230: 5725 RESET (power-up or watchdog timer)
 2/03/95  7:02:33  Fault   230: 5725 RESET (power-up or watchdog timer)
 4/07/95 12:53:33  Fault   225: 5725 Fan Not Working
 4/07/95 12:53:34  Fault   225: 5725 Fan Not Working
 4/07/95 12:53:35  Fault   225: 5725 Fan Not Working
 4/07/95 12:53:36  Fault   225: 5725 Fan Not Working
"
HP 3458A detected
 TEMP? = 37.5 C
Fluke 5700A readback output = 10.000000000 V
10V test result: 1.000001492E+01 VDC [deviation 1.4920 ppm]
Program completed!

p. Running time 111438 hours translates to approximately 12.7 years. Given calibrator's age - now we can see that half of the time it was running on, as a standard. No fatal errors were recorded, expect ones related to external Fluke 5725A amplifier in period from 1993 to 1995. It was a heavy worker, this box :)

Full serial log output, Fluke 5700A 28 March 2016, initial

Performed also automatic self-tests routines and the more thorough self-diagnostics which takes about 1 hour to complete. No test failures were reported, yey!. Cal check was also performed without errors. Tested DCV, ACV, DCI, ACI, resistance, and wideband outputs using same calibrated HPAK 3458A-002 and uncalibrated Fluke 8920A for AC. All readings were close to expected values from self-check report, even though the last calibration was some four years ago.

Repair/maintenance workflow

The exciting part! Of course, creat care taken on analog section PCBA's handling to avoid contamination and reduce leakage risks. This means using gloves and face mask (:D) to reduce chances of sensitive analog PCBAs contamination.

Location of all modules and boards is readily provided from service manual.



Time to remove the top cover. All internal modules and assemblies are further shielded and protected by solid metal frame parts with vent holes and openings for board levers.



Now we can see presence of A5 and A6 optional boards for wideband, so this calibrator does indeed have 5700A-03 option. Nice to have it, which allow ACV/ACI functions to be tested up to 30MHz. This is required for calibration of instruments such as Keithley 2001/2002.



Look from the top reveal all boards presence and lot of warnings regarding dangerous high voltages presence and charged capacitor cautions. Let's do same here in this article.


-1100 and +1100VDC voltages present inside of this calibrator when operation are dangerous and even lethal. Pay attention of what you are doing, and keep all connections stable and robust, not using hand probing unless absolutely required and safe. Use tools rated for CAT III operation while servicing high-power circuits in this instrument.

First we start with outguard section, boards A19 and A20. All modules will be covered in detail below, using dedicated sections.

Exposed now base board is all covered with dust and dirt, and urges for intensive dry air cleaning and careful inspection.



Middle shielded can contain mains primary transformer A22, custom made for Fluke. This power unit together with A19 PCB powers whole box. Pair of 80x80 mm DC fans mounted on the side to keep airflow around. There are dual HP (today Avago) Versa-link transmitters to inguard domain. This is very same optical isolation technique used in HPAK 3458A.





Extensive cleaning will be performed on chassis parts. DC FANs are ok, but not top tier brand, like was expected. Might change them in later.



Time to look into inguard area, with all boards and modules removed.





Check that relay love with own shield cover and dedicated cutouts in the frame. Wideband option SMB cable to N-type connector going thru filter choke to supress EMI.









Optoisolated interface to digital brains on A22 board.



A2 Front panel PCA

Interesting section in service manual covers on page 289 regarding non-uniform pixels brightness on VFD, which usually happen after static image displayed for long time. According to manual, this effect can be reversed and cleared up by lighting up the whole display and leaving it on overnight. To enable all segments, next procedure suggested: Enter _Setup Menus_, select _Self Test & Diags_, _5700 Self Diags_ > _Front Panel Tests_ > _Display/Control_ > _All On_.

Left VFD display with segment digital section is very bright, like new, which is great news, as new front panel PCA cost few K$! Graphical dot-matrix VFD on the right is dimmer, and have some visible ghosting:



I'll run recommended above procedure to see if it improves.





Front panel is solid metal frame behind plastic cover and keyboard with opening just for displays.



All terminals are beryllum copper, as expected for instrument capable of sourcing precision DC voltages at microvolt levels.



Front panel board is further shielded by solid metal cover.



A3 ANALOG Motherboard PCA

This board just have interconnects for all modular PCBs and some switching COTO relays and optical isolation interface for inguard side.



A4 DIGITAL Motherboard PCA



Interconnects for digital earth-referenced side of calibrator. It also feature connectors to transformer and mains input voltage selector switches.

A5 Wideband output PCA (Only on 5700A-03)









A6 Wideband oscillator PCA (Only on 5700A-03)







A7 Current/High resolution oscillator PCA







A8 Switch matrix PCA



This board have too many relays.
« Last Edit: November 11, 2016, 12:01:44 pm by TiN »
YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 
The following users thanked this post: klaus11, jonovid, villas, electron69

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4206
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
Re: #2000 Post : Teardown and study of Fluke 5700A Calibrator
« Reply #2 on: November 11, 2016, 04:09:25 am »
Here are the most interesting boards...

A9 OHMS,CAL PCA

Calibrator resistance references are located on this (for 1? 1.9? short resistance) and A10 PCBA (10? - 100M?). All resistors except highest 100 M? are routed with Kelvin 4-wire connections to cancel influence of routing/connection parasitic resistances. There is also reference 10K?/90K? 10:1 divider, lower part of which is used as internal calibration standard. During Artifact Calibration, this is the resistor which will get compared to external 10 K? standard.

Used Fluke proprietary resistors are designed and manufactured to have very low TCR, since the calibrator is specified at +/-5°C around the calibration temperature, usually +22°C or +23°C. By design temperature coefficient is targeted near zero at +22°C. The temperature coefficient of the thin film networks increases slowly to 1 ppm/°C at about 8°C from the zero TC at 22°C.

The 90M? resistor that makes up most of the 100M? standard designed to have TC <10 ppm/°C. According to Fluke documents and test results, 1? and 1.9? resistors have a temperature coefficient of a few ppm per °C at reference temperature +22°C.









A10 OHMS,MAIN PCA





A11 DAC PCA

The key board in the unit, where the reference magic lives.



Heart of main DC voltage reference performance and stability are two SZA263/LTFLU-1 refamps, located on heated substrate with TaN resistor networks in shielded can on this board. Similar reference arrangement idea was used in old Fluke 5442B, but with famous teal-colored wire-wound resistors on regular FR4 PCB.











A12 Oscillator control PCA



A13 Oscillator output, PCA







A14 High voltage control PCA





A15 High voltage/High current PCA







A16 Power Amplifier PCA











A17 Regulator/Guard crossing PCA





And no design is perfect, even 30000$ calibrator from Fluke. ;) Lonely TO-220 regulator or transistor just hanging in there on jump wires on bottom side of the board.



A18 Filter/PA PCA



A19 Digital power supply PCA



This 5700A-1604 board, A19 with part number 761056 have Revision F and taking AC outputs from A22 mains transformer secondary windings and generate various DC voltages.

There are nice accessible test points on top edge, which we will probe during service. According to date code on parts, such as Q1 RCA410 pass transistor, this board is manufactured in 1992. This makes it 24 year old, which means all electrolytic capacitors are up for replacement.

High voltage +75 and +35VDC rails regulated by discrete transistor regulators, main pass elements of which are easily spotted by large heatsinks on right top corner of PCBA. Reference voltage for these supplies are generated by set of zener diodes (VR6 and VR7 for +75V and VR14+VR15 for +35V). +75V rail additionally controlled by PS_SD and RESETL logic inputs, likely to prevent dangerous +75V be supplied before main processor initialize and get everything ready for main operation.

Lower voltage, +12 and -12VDC generated by regular 7812 and 7912 integrated chips. These also provide power for two 12V DC fans, providing forced airflow in the unit. +5V supply for digital outguard logic is generated off U3 Sanken SI-3052 LDO, which is 2.0A +5V regulator in TO-3P package, very similar to 7805.

All power rails are input-protected by PCB-mounted fuses, F1-F5 located right next to card's connector. Every power input and output are routed via that single dual-row 64-pin connector P41 at bottom edge. This board have array of electrolytic capacitors, all of which we will replace later.

There is some gunk coming off silicon spacers located on bottom board. Due to exposure to high temperatures silicon got decayed and oozed it's chemicals down following the gravity path. Nasty stuff, so this will all be cleaned.



Here is also A19 PCBA schematics from service manual.



A20 Processor PCA

Service manual lists all 5700 series calibrators as ones with same hardware except processor board A20.
This specific A20 5700A-3006 board is P/N 761072, with native 5700A Revision F marked by ink, and Rev.G by PCB copper etch.



CPU board is based on well-known and popular Motorola MC68HC000P8 processor, tied to 28C256 EEPROM, four firmware 27010 EPROMs and 128K of SRAM (four TOSHIBA TC55257BPL-10). There are option sockets for additional 2 x 27010 EPROM and extra 2 x 32K RAM. Firmware stickers say it's Revision J. We will reed these D27C010 ROMs later on programmer. There is bunch of 74-series logic around and few 22V10 PAL and GALs. Main clock is provided to CPU from 7.3728MHz XO.

3V Lithium battery BT1 backup only realtime clock, based on Intersil ISL7170, which is good news after all the calibration data lost shenanigans on HP 3458A, Tektronix 7000 scopes and such. Calibration data is likely live happily in socketed U13, Xicor X28C256P.

Tactile switch SW1 located near top right corner is for CPU RESET controller, which is generated old known TL7705 IC. Same one we saw before on Keithley 2001 and 2002. Non-reset state is shown by lonely red LED CR1. If it's lit - we good to go.

Interconnection with backplane done via pair of 64-pin dual row connectors. Lonely LM324N general purpose opamp is used to monitor fans health and drive fan alert signal in case of fan failure. Board also have load of test points, which we can check if necessary.

Also interesting to note UART controller 68C681 U31 which have serial interface at RCVB (TP10) and TXB (TP9) points. Likely this is main serial interface between outguard and inguard domains of MFC.

Firmware

Main firmware stored in four 27C010-200V10 ROMs, U15-U18 positions on A20 processor board. Calibration ROM is stored in non-volatile Xicor X28C256P-25. Calibration ROM does not depend on battery, it's simple SRAM, so no calibration data corruption even for old units, unless ROM itself become bad (which is rare, but still happens).

So far I have only one firmware version, for Model 5700A Series I, Rev.J, Build 11/15/93 10:52:39.
* 5700A Rev.J binary (combined)
* U15 ROM CHK:00922940
* U16 ROM CHK:0081BC9D
* U17 ROM CHK:00B940B6
* U18 ROM CHK:009E9081

Combined binary images were merged using simple python tool:

Code: [Select]
# xDevs.com Firmware combine tool
# https://xdevs.com/fix/f5700a

import os
with open('5700a_s1_u15_761262_revj_00922940.bin','rb') as a:
    with open('5700a_s1_u16_761361_revj_0081bc9d.bin','rb') as b:
        with open('5700a_s1_u17_761379_revj_00b940b6.bin','rb') as c:
            with open('5700a_s1_u18_775247_revj_009e9081.bin','rb') as d:
                with open('5700_J_FULL.bin','wb') as x:
                    for cnt in range(0, 262144):
                        x.write ("%s%s" % (b.read(1),a.read(1) ) )
                    for cnt in range(0, 262144):
                        x.write ("%s%s" % (d.read(1),c.read(1) ) )

Quick glance on result file revealed some hidden strings. Easter eggs? :)


                       
Easy to read and backup calibration ROM as well, using external programmer, such as TL866. It's wise to backup calibration ROM before attempting own re-calibration or sending unit to service, just in case.

Inguard microprocessor ROM (27C256) located on A17 PCBA, position U64.
Dump of 5700A Rev.A In-guard ROM P/N 881698 here: Binary CHKSUM: 002E740A

As usual, many other instruments firmware dumps are available on my xDevs.com documentation site. If you have another Fluke 5700A firmware (or any other instrument firmware), please feel free to upload.

Repairs and diagnostics

A18 High voltage/High current PA

Few 1W 220K resistors were measured low (~90 K?) and were replaced with 3W 220K, as expected by Fluke BOM. These resistors set current and affect control for pass transistors on high-voltage supply regulator for PA.

Below is schematic fragment with marked components.



Unit was cleaned from dust and carefully reassembled. No further issues found. :)

Calibration

Fluke 732B10 VDC+/-1.5 ppmDCV,ACV,DCI,ACI
Fluke 742A-1    1 ?                +/-10 ppm    Resistance 1?, 1.9?
Fluke 742A-10k 10 k?            +/-4 ppm    ACI,DCI, 10? to 100 M?

Actual standards values used (right after calibration in August 2016):

Fluke 5700A CALStandard valueDeviation
  10.000000 VDC              9.9999323 VDC        +0.0 ppm
  10.000080 k?          10.0000505 k? 4W     -0.2 ppm
  1 ?                  0.99997505 ? 4W +23.1 ppm
  1.9 ?                  0.99997505 ? 4W -22.1 ppm

Now calibrator is adjusted to standards within 30 days official vendor's calibrations and ready to rock!
« Last Edit: November 11, 2016, 04:24:45 am by TiN »
YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 
The following users thanked this post: villas, electron69

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4206
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
Re: #2000 Post : Teardown and study of Fluke 5700A Calibrator
« Reply #3 on: November 11, 2016, 04:10:44 am »
Now this part is bit of deeper dive into key metrology important part of calibrator, it's DC reference and DAC subsystem. To do proper testing and study, separate Fluke 5700A A11 PCBA was bought, in borked condition. It's not worth the risk to damage calibrator by poking around working A11.

DC Reference source in most of Fluke calibrators is Ref-Amp.

Ref-Amp? What is it?

Ref-Amp consists of an NPN transistor in series with a zener diode. When biased properly, the combination has a extremely low temperature coefficient. The reference voltage that the can vary from +6.5 VDC to +7 VDC, depends on actual production bath. Also since both zener and transistor are located on same substrate and enclosed in hermetic package, they are tightly thermally coupled and protected from ambient humidity. This allows to improve stability over long time spans.

There are two well-known examples of this device family, Ref-Amps used by Fluke in high-end instrumentation equipment. These are Motorola SZA263 and Linear LTFLU-1H/LTFLU-1AH

The Motorola SZA263 is a "Ref-Amp" that was made by Motorola, and later discontinued after getting out of high-volume semiconductor business. This forced Fluke to find a partner to design a replacement for the obsolete SZA263, as Motorola would not sell the design files and masks of original design. Linear Technology was happy to help big customer like Fluke with the design, and that's how LTFLU-1AH Ref-Amp emerged. It's pin-2-pin and function compatible with the original SZA263, but the LTFLU-1AH use bit aluminum alloy for the interconnects, which is not the same one Motorola used. This rendered in different long-term stability, visible by positive drift over time on SZA263, but negative in LTFLU-1AH (same as LTZ1000/LTZ1000A and the LMx99 ICs).

If you have a bank of 4 x 732A's and a bank 4 x recent 732B's that you get calibrated once a year, you will clearly see these drift patterns. Care need to be taken though, as early batches of Fluke 732B were still using left-over SZA263's, but later all production 732B were updated to LTFLU-1AH.

Main differences summary between the SZA263/LTFLU-1AH IC and the market-available Linear LTZ1000 are:

* Different package. LTFLU-1AH is 4-pin hermetic can, the LTZ1000 in an 8-pin TO-99. They are far from drop-in compatible.
* SZA263/LTFLU-1AH has the transistor for temperature compensation in series with the zener, not parallel as LTZ1000 design.
* Due to different manufacturing process, Motorola SZA263 have positive long-term drift, while Linear LTFLU-1AH has negative long-term drift.
* Opposite to LTZ1000A, there is no on-die heater in SZA263/LTFLU-1AH
* SZA263/LTFLU-1AH require much more time (vs LTZ1000 design) and care for support resistor matching and tempco testing

Last item is due to different tempco compensation transistor arrangement in SZA263/LTFLU-1AH circuit, which needs more attention for temperature and current compensation than LTZ1000 circuits. LTZ1000 design is fine even with standard datasheet reference schematics, without any analog black magic or voodoo.

All this above of course less a problem than zero availability of LTFLU-1AH, as Fluke have exclusive rights for this design and chip, and Linear is not allowed to sell Ref-Amp on open market. So Motorola SZA263/Linear LTFLU-1AH are less friendly solution to implement a stable DC reference, as even if you get LTFLU-1AH chip, lot of time and money for resistor matching and temperature testing is required.

Die photos of LTFLU-1AH and comparison to other references

LTFLU-1AH



Courtesy of branadic (Dipl.-Ing. A. Bülau) from the EEVBlog

On the die photo we can see much more complex design than just diode and transistor.

LTZ1000ACH

Here's die photograph of Linear LTZ1000ACH chip:



LM399

This is National Semi version of LM399.



Courtesy of branadic (Dipl.-Ing. A. Bülau) from the EEVBlog

Instruments list using Ref-Amp

Known instruments to implement SZA263/LTFLU-1AH as primary DC reference

* Fluke 8840A DMM
* Fluke 8842A DMM
* Fluke 731B DC Voltage Standard
* Fluke 732A DC Voltage standard
* Fluke 732B DC Voltage standard
* Fluke 5440B DC calibrator
* Fluke 5500A Multi-function calibrator
* Fluke 5520A Multi-function calibrator
* Fluke 57LFC System Calibrator         
* Fluke 5700A Multi-function calibrator
* Fluke 5720A Multi-function calibrator
* Fluke 5730A Multi-function calibrator
* Fluke 5790A AC Measurement standard   

There is also great thread here on EEVBlog created by zlymex, revealing guts of various DC voltage standards.

Overview of A11 DAC PCBA from Fluke 5700A MFC





Fluke 5700A Service manual  has schematics and short description of DC reference and DAC module operation theory, so it's worth to refresh memory reading related sections to get familiar with DAC module design.



PCB with blue soldermask is 6-layer FR4, with most of the routing buried on inner layers. There are only few ground mesh fill polygons, around digital components, such as MCU, clock generator and ADC. Stripline routing might serve two purposes here - act as additional protection sensitive analog traces from environment and surface contamination. Book Basic Linear Design from Analog Devices provide some good ideas how PCB leakage and bad layout could introduce errors and issues in sensitive analog design.

DAC assembly provide adjustable stable DC voltage output, from 0 to +11VDC and features multiple sub-assembly units.

* DC Reference hybrid (HR5)
* DC Amplifier hybrid (HR6)
* DAC Filter SIP
* DAC mainboard

Fluke used PWM-controlled method to generate adjustable voltages. There are also few auxiliary support circuits on-board, such as sense-current cancellation block, linearity tune control, negative offset control. Block diagram below on Image 4 can help to understand overall function of circuitry on A11 DAC PCBA.



Output DC voltage is generated by 5-pole discrete filter, which has two precision square waves of different amplitude as the input. First input channel of the filter is CH1, and it's amplitude is from 0V to reference voltage, which is around +13 VDC. This is coarse adjustment channel. Second CH2 channel is operated similar way but goes up only to attenuated reference voltage at +0.78 mVDC. This is fine tune adjustment channel. Filter is designed as LPF with bandwidth 30 Hz, and input square wave frequency is 190 Hz. So the output is clean and filtered DC voltage, derived from controlled PWM CH1 + CH2.

Filter output does not have capability to drive large currents, so output stage on separate ceramic hybrid takes filter output and provide driving capability for DAC output. Hence after all, circuit output voltage can be predicted and calculated by simple formula:

V[OUT] = Duty[CH1] * VREF[13V] + Duty[CH2] * VREF[0.00078V]

p. Use of this combined PWM scheme allows us to have efficient way to generate arbitrary voltage levels without use of very expensive resistor networks and expensive complex multi-bit DAC ICs. PWM duty cycle resolution of PWM generator used in Fluke calibrator is 0.0024%, which provides resolution of CH1 = 309 uV/bit and CH2 = 18.5nV/bit. PWM signal also electrically isolated by optocouplers.



Dark color covers around HR5 and HR6 hybrid assembly are not metal, but metallized plastic. Main purpose of these covers is to prevent stray airflow around reference circuity and DAC chopper amplifier. This is important due to parasitic thermocouple EMF-generation effect present with any thermal gradient within the board. Also since hybrids are actively heated, enclosure helps with thermal stability of inner thermostat area. Don't forget, Fluke 5700A dissipate plenty of power during operation and has two large fans to provide airflow around power amplifiers and high-power components and boards.



ADC section and PWM generator/digital controller are separately enclosed in metal cage shield on the top side. This serve dual purpose, to keep generated RFI/EMI enclosed and localized and to provide additional shielding from external fields. These shields are grounded  to module power ground plane.



Thermally-controlled HR5 hybrid network with DAC output buffer.

   

 



Thermally-controlled HR6 hybrid network with Ref-Amps to generate stable 6.xV and 13.xV reference voltages.

   

 



 

Main DC reference hybrid is built using sandwich of ceramic substrates. Main thin substrate has two Ref-Amp, hybrid resistor network, few opamps, temperature resistors and circuit tracks. Back side of HR5 has large area 27 &Omega; resistor acting as a heater. Then there is glued spacer to act as a heat-spreader and coupled to it hermetic hybrid resistor with clear quartz window for laser trim access.



All parts except Ref-Amps, which as Motorola SZA263 are SMT-mount. Two Ref-amps are used to provide higher +13 VDC (6.5 + 6.5) reference level to further reduce amount of noise and improve stability of the DAC. Excellent temperature coefficient of used Ref-Amps is achieved by using stable collector bias current of their transistors provided from stable thin-film resistor hybrid between Ref-Amps. This design allow to have very small output voltage impact from circuit component errors, so it's output stability directed almost entirely from Motorola SZA263 performance.



To prevent output drift with respect to ambient temperature whole assembly is heated to constant +62 &deg;C by external circuit block on main A11 PCBA. Temperature feedback element is thermistor RT1, located right near the Ref-Amp packages. As ceramic is a good thermal conductor, any change in temperature of hybrid will drive correction signal to Q2 on main board and have circuit adjust power to 27 &Omega; film resistor to adjust temperature back to set-point. There is also thermal runaway protection, implemented with second thermistor RT2, which activates Q9 to bypass base current of Q1 to avoid overheating. This protection kicks in once substrate temperature reaches +67 &deg;C, and normally should be never used.

Interesting to note that improved Fluke 5720A also has changes in this HR5 DC reference hybrid, as we can see thanks to zlymex.



Pair of Motorola SZA263 chips replaced by Linear LTFLU-1ACH, and LF351 opamps are replaced with Linear LT1006 and TL071C.



Hybrid laser-trimmed resistor network is smaller and have different configuration as well.

Very similar dual Ref-Amp assembly used also in Fluke's AC measurement standard, Model 5790A on A16 DAC board.



Bit simpler version with only one Ref-Amp is used in Fluke 5500A/5520A and system calibrator 57LFC. There is thermal control over ref-amp assembly in these calibrators either. So as a result DCV performance of these calibrators is "only" 11 ppm annual, with daily stability 2 ppm. Fluke 5700A has 8 ppm annual, and updated 5720A/5730A are half of that, 4 ppm.



Of course if better temperature stability is achieved with removal of all airflow, actual Ref-Amp can provide much better performance. This is proven by +/- 2.0 ppm/year and 0.3 ppm/day specifications and actual performance of double-oven Fluke 732B reference assembly.

Let's take a brief look on 732B reference assembly block diagram:







And hybrid resistor magic:



 

 

I plan to take better macro photos of all these resistors, by going real close with my 150mm macro lens and taking 12-20-photos with focus trim for stacking, so we can see every tiny feature and detail on those.

That will take few evenings of time, easy!

Power supply requirements of A11 PCBA

In this section we determine what requirement are to get A11 PCBA operational and working, without calibrator mainframe itself.

CAD dimensions of A11

This section will cover physical measurements and dimensions to design a suitable enclosure to fit A11 and other support electronic parts.

Backplane design and signaling

This section will cover electrical and physical design of interconnect backplane board.

Power supply design for A11

This section will cover design of power supply board to power up A11.

Bring-up for DAC/REF

This section will first tests.

Comparison with LTZ1000 reference

TBD...
« Last Edit: November 11, 2016, 04:32:37 am by TiN »
YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 
The following users thanked this post: villas, LaserEng, electron69

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4206
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
Re: #2000 Post : Teardown and study of Fluke 5700A Calibrator
« Reply #4 on: November 11, 2016, 04:12:32 am »
5700A Performance verification

Full performance test up to 5700A's 24-hour specification is many weeks challedge even for very well equipped lab, leave alone hobbyist home cave. I'd suggest once again to read well written and detailed Artifact calibration function study, which present test methods for this instrument.

Here are some quick tests, for resistance measurement results obtained from 24-hour calibrated 3458A DMM with pair of Fluke 5440 test leads. Settings of DMM are: NPLC 100, DELAY 1 and OCOMP ON for Rvalues <1M?, AZER ON.

Fluke 5700A OHM (actual) 3458A Measurement value Deviation
  0 ?   (0.000000)  0.00000288 ? 4W      +2.88 ppm       
  1 ?   (0.9998249) 0.9997906 ? 4W        -34.3 ppm       
  1.9 ? (1.8995690) 1.8994901 ? 4W        -41.52 ppm     
  10 ?  (9.999934)  9.99985546 ? 4W      -7.85 ppm       
  19 ?  (18.999101) 18.9989842 ? 4W      -6.14 ppm       
  100 ? (100.00187) 100.001919 ? 4W      +0.49 ppm       
  190 ? (189.99523) 189.995162 ? 4W      -0.356 ppm     
  1000 ?(0.9999919K) 0.99999252 k? 4W      +0.62 ppm       
  1900 ?(1.8999967K) 1.89999877 k? 4W      -1.08 ppm       
  10 k? (10.000080K) 10.0000968 k? 4W      +1.68 ppm       
  19 k? (18.999695K) 18.9997209 k? 4W      +1.36 ppm       
  100 k?(100.00133K) 100.001386 k? 4W      +0.56 ppm       
  190 k?(189.99286K) 189.993439 k? 4W      +3.04 ppm       
  1 M?  (1.0000017M) 1.00000047 M? 4W      -1.22 ppm       
  1.9 M?(1.8999562M) 1.89994836 M? 4W      -4.12 ppm       
  10 M? (9.999382M) 9.9990808 M? 4W      -30.12 ppm     
  19 M? (18.999034M) 18.999738 M? 4W      -37.10 ppm     
  100 M?(100.00848M) 100.02146 M? 2W      +129.79 ppm     

More to come...

DCV Noise evaluation

TBD... :(

DCV Linearity evaluation

TBD... :(

Tempco evaluation

TBD... :(

Connecting 5700A to 5220A current amplifier

TBD...

Currently we don't have native Fluke 5725A amplifier, but have much older 5220A amplifier. This also can be used to boost current range capability of calibrator system. As no interface cable was included, we have to make one ourselves or get one from ebay :). For this task Fluke Y5002 cable document will be used as reference, as well as schematics from Fluke 5700A Series I Service manual.

Summary

That was a pure precision enjoyment to play with 5700A. With this calibrator help both of my Keithley 2002's were also calibrated in August 2016, guarded with 3458A. Eventually more stability data and test results would emerge out of this, just need to get rid of some ongoing projects first. And as usual, here are reference links to more details on topic.

I didn't think that infamous HPAK 3458A restoration worklog article would be surpassed in amount of documenting work, but seems like this 5700A one has all chances to do so :)

* xDevs.com : Fluke 5700A MFC detailed article
* xDevs.com : Stand-alone use of F5700 A11 DAC
* xDevs.com : Time Electronics 9823 MFC article
* xDevs.com : HP 3458A DMM repair worklog
* xDevs.com : Keithley 2002 second unit repair worklog
* EEVBlog : The LTFLU (aka SZA263) reference zener diode circuit
YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 
The following users thanked this post: daqq, electron69

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4206
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
Re: #2000 Post : Teardown and study of Fluke 5700A Calibrator [56K warning!]
« Reply #5 on: November 11, 2016, 04:27:27 am »
Here's video from stream last week when I started design of own backplane for A9+A10+A11 boards.



I plan to have some updates on this once a week or better, I hope.
YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 
The following users thanked this post: electron69

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4206
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
Re: #2000 Post : Teardown and study of Fluke 5700A Calibrator [56K warning!]
« Reply #6 on: November 11, 2016, 04:28:31 am »
And few reserved posts for future goodies.
YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4206
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
Re: #2000 Post : Teardown and study of Fluke 5700A Calibrator [56K warning!]
« Reply #7 on: November 11, 2016, 04:30:31 am »
I had to publish stuff in multiple posts due to that 20000 chars limit set by Dave :D
Hope you enjoy this :)

Btw, if anyone want to repeat the feat, there is one on auction eBay now for only >6800$  :-X
« Last Edit: November 11, 2016, 05:01:53 am by TiN »
YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 

Online TheSteve

  • Supporter
  • ****
  • Posts: 3175
  • Country: ca
  • Living the Dream
Re: #2000 Post : Teardown and study of Fluke 5700A Calibrator [56K warning!]
« Reply #8 on: November 11, 2016, 05:07:39 am »
Amazing stuff TiN - I can't imagine the time it took to present everything in such a wonderful fashion.

Two questions though, when will you have a Josephson junction array and do you still have a day job? :)
VE7FM
 

Online Vgkid

  • Super Contributor
  • ***
  • Posts: 2600
  • Country: us
Re: #2000 Post : Teardown and study of Fluke 5700A Calibrator [56K warning!]
« Reply #9 on: November 11, 2016, 06:15:45 am »
Really nice job on the writeup, I'm not done reading yet.
viewing this on my phone doesn't give this thread justice.
If you own any North Hills Electronics gear, message me. L&N Fan
 

Offline jonovid

  • Frequent Contributor
  • **
  • Posts: 894
  • Country: au
    • JONOVID
Re: #2000 Post : Teardown and study of Fluke 5700A Calibrator [56K warning!]
« Reply #10 on: November 11, 2016, 06:21:21 am »
 :popcorn: :-+
Hobbyist with a basic knowledge of electronics
 

Offline klaus11

  • Supporter
  • ****
  • Posts: 156
  • Country: 00
Re: #2000 Post : Teardown and study of Fluke 5700A Calibrator [56K warning!]
« Reply #11 on: November 11, 2016, 08:00:38 am »
These Dynamic 28-pin Toshiba could fit into HPAK3458A

Toshiba[/url]
« Last Edit: November 11, 2016, 08:53:24 am by klaus11 »
HP3458A, HP3245a, Keithley 2000, Fluke 87V, Rigol DP832, TEK TDS5052B, HP33120A
 

Offline lukier

  • Supporter
  • ****
  • Posts: 627
  • Country: gb
    • Homepage
Re: #2000 Post : Teardown and study of Fluke 5700A Calibrator [56K warning!]
« Reply #12 on: November 11, 2016, 10:24:28 am »
Serious stuff.  :-+ Thanks for the details on the PWM DAC.

BTW Noob question maybe. I get the point of copper binding posts, low EMF etc. But looking at the cards it seems all the signals travel over the backplane, via DIN connectors, solder joints, etc. Is that not important here because the backplane and cards temperature is assumed constant and possible junctions are balanced?
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4206
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
Re: #2000 Post : Teardown and study of Fluke 5700A Calibrator [56K warning!]
« Reply #13 on: November 11, 2016, 10:28:25 am »
Two questions though, when will you have a Josephson junction array and do you still have a day job? :)

Surely, once I win a 500K$ lotto, first thing to buy is a JJA  :phew:.

And yes, I do have day job, full-time electronics design engineer making PC motherboards and graphic cards and some other stuff  :popcorn:

lukier

That's what external sense is for. And calibrator is still not quite nV-level instrument, so assuming there are no large ambient temperature variations, inner temperatures should stay rather constant. And thanks to ACAL and self-cal for zero most of errors can be significantly reduced right before the critical measurement/calibration to happen. I'm sure someone more experienced, like Dr.Frank can comment in better detail on this.
YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 

Offline lukier

  • Supporter
  • ****
  • Posts: 627
  • Country: gb
    • Homepage
Re: #2000 Post : Teardown and study of Fluke 5700A Calibrator [56K warning!]
« Reply #14 on: November 11, 2016, 10:38:07 am »
So I guess the auto-cal here is implemented by the PWM DAC described and some comparator in a form of successive-approximation ADC? No need for 3458A's U180 and its linearity? Must read the service manual over the weekend :)
 

Offline mimmus78

  • Supporter
  • ****
  • Posts: 676
  • Country: it
Re: #2000 Post : Teardown and study of Fluke 5700A Calibrator [56K warning!]
« Reply #15 on: November 11, 2016, 10:41:58 am »
Ohhh I reported the post to Dave and moderator. Pornographic content is not allowed here!
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4206
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
Re: #2000 Post : Teardown and study of Fluke 5700A Calibrator [56K warning!]
« Reply #16 on: November 11, 2016, 10:45:19 am »
So I guess the auto-cal here is implemented by the PWM DAC described and some comparator in a form of successive-approximation ADC?

Not SAR, it's integrating ADC with custom ASIC still, but it's much less resolution and slower than one in the 3458. This is OK because it is used as null-meter only, comparing stable DAC output (which is adjustable) to external measured signal.

mimmus

Maybe then Dave would add unicode support for forum engine, so we can all enjoy proper ohm symbols instead of question marks..
« Last Edit: November 11, 2016, 10:47:16 am by TiN »
YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 

Offline villas

  • Contributor
  • Posts: 6
  • Country: br
Re: #2000 Post : Teardown and study of Fluke 5700A Calibrator [56K warning!]
« Reply #17 on: November 11, 2016, 11:00:56 am »
Great post, TiN. As always. Thanks for sharing your time and knowledge.
 

Offline HighVoltage

  • Super Contributor
  • ***
  • Posts: 4524
  • Country: de
Re: #2000 Post : Teardown and study of Fluke 5700A Calibrator [56K warning!]
« Reply #18 on: November 11, 2016, 11:58:40 am »
Very impressive!
Do you ever sleep?

Based on your many posts I started to repair equipment as well, a while back. It is so much fun and so educational.
Thank you, just amazing!
There are 3 kinds of people in this world, those who can count and those who can not.
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4206
  • Country: us
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
Re: #2000 Post : Teardown and study of Fluke 5700A Calibrator [56K warning!]
« Reply #19 on: November 11, 2016, 12:04:26 pm »
Sleep is overrated. I was cooking this for a while, but today I couldn't hold it anymore.

This all because of your photo with stash of 34401s measuring 10V. I'm trying to catch up to make same one ever since...
YouTube | Metrology IRC Chat room | Live-cam | Share T&M documentation? Upload! No MB limit, firmwares, photos.
 

Offline Dr. Frank

  • Super Contributor
  • ***
  • Posts: 1907
  • Country: de
Re: #2000 Post : Teardown and study of Fluke 5700A Calibrator [56K warning!]
« Reply #20 on: November 11, 2016, 01:17:28 pm »
Really nice article as #2000!

Good thing, that somebody 8) 8) 8)  ;) salvaged this full manual for the 57xx to KO4BB a few years ago, before FLUKE stripped off the schematics...


Both ACAL instruments, 3458A and 57xxA have an emf offset calibration routine for gross errors, but I think it will anyhow be necessary to somehow compensate for additional offset fluctuations, maybe after RT changes.
The copper jacks help greatly with external connections in this sense, but other calibrators also work properly without.

The ACAL routine of the 57xx is not described in detail, like for the 544xA DCV calibrator.

As far as I understood the 57xx manual, this calibrator lacks the Hamon type 100/25/2/1:1 resistor divider of the 544x, that provides about < 0.1ppm gain factor calibration.
So the 57xxA should rely only on the linearity of its DAC, which is not as precise, I guess, as the 3458As ADC.
Therefore, for DCV, the 544xAs range autocalibration should still be more precise.

Frank
« Last Edit: November 11, 2016, 02:42:14 pm by Dr. Frank »
 
The following users thanked this post: saturation, TiN, lukier

Offline Echo88

  • Frequent Contributor
  • **
  • Posts: 619
  • Country: de
Re: #2000 Post : Teardown and study of Fluke 5700A Calibrator [56K warning!]
« Reply #21 on: November 11, 2016, 02:16:35 pm »
Wow! Impressive article TiN  :-+

Also those many ceramic hybrid-packages sure are eye-candy. Will take a lot of time to dig through all this material, especially the Service Manual.
 

Offline dr.diesel

  • Super Contributor
  • ***
  • Posts: 2197
  • Country: us
  • Cramming the magic smoke back in...
Re: #2000 Post : Teardown and study of Fluke 5700A Calibrator [56K warning!]
« Reply #22 on: November 11, 2016, 02:26:08 pm »
Wow thanks TiN! 

Offline Henrik_V

  • Regular Contributor
  • *
  • Posts: 111
  • Country: de
  • “ground” is a convenient fantasy
Re: #2000 Post : Teardown and study of Fluke 5700A Calibrator [56K warning!]
« Reply #23 on: November 11, 2016, 02:31:07 pm »
Thank you , thank you , thank you

Great 2k post  :-+

A really nice 'Easteregg' can be seen during Zero-Cal : Snails, cars, etc run over the display :D
I have a 5700A @ work to perform internal DC and mainly AC volt calibrations to some gear ... but I never took it apart, my boss always get nervous whenever I have a screwdriver in my hand  ;D . However he allowed me to take it to the MakerFair in Hannover each year to provide some units to everyone who pass by his DMM (>120 calibrations each time :-DMM)

And while we are at it: Does someone  know how the AC frequencies are generated?
More specific, reason why I ask: While performing AC calibration on DAQ cards we usually perform a sine fit and look at the residuals. At certain frequencies one can clearly see phase jumps in the residuals.  So the 5700 has some sort of accumulated phase error register, that results in added or skipped DAC values..  . In order to avoid that: Does  somebody know how one could pick output frequencies that avoid these jumps?   
Maybe something like : the 5700 has an internal clock frequency f_clk , use a M/N factor (both integer) add the internal x and/or divide by another y  and you get the 'clean' frequencies ?     ..  :)

Greetings from Germany
Henrik

The number you have dialed is imaginary, please turn your phone 90° and dial again!
 

Offline quarks

  • Frequent Contributor
  • **
  • Posts: 834
  • Country: de
Re: #2000 Post : Teardown and study of Fluke 5700A Calibrator [56K warning!]
« Reply #24 on: November 11, 2016, 02:47:07 pm »
Great post :-+

I once tried to buy one Fluke 5700, but did not get it.
Instead I got a Wavetek 4808.
« Last Edit: November 11, 2016, 02:48:59 pm by quarks »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf