Building 3rd Fluke 5700A "Hulk-III" precision calibrator

Contents



This is our “experimental” calibrator rebuild project, using leftovers and spare boards from first two Fluke 5720A projects. Most of the boards have major failures/missing parts, so this is most interesting educational unit.

Remember, buying broken or even listed “tested good” used 5700A below market price and fixing it likely will end up as bad expensive exercise, when all hidden costs are considered and accounted for. Paid cost for such a unit WITH proper repair, long-term verification and demanding calibrations will easily exceed cost of buying a fully tested and calibrated unit from reputable source/distributor. So even if you buy complete calibrator for $5K, you can be easily be rewarded with just pile of unstable/non-fixable parts that would need another $XXK to get even just working again, leave alone meeting specifications.

So how would a dedicated person make the dream to have high-performance calibrator at home to come true, without selling both arms, legs and eyes to nearest Fluke Calibration branch?

No escaping the dream today – go build our own calibrator from scrap parts!


Image 1: Random bits and pieces, lego-version of the 57xx series calibrator hardware

Disclaimer

Redistribution and use of this article, any parts of it or any images or files referenced in it, in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

  • Redistributions of article must retain the above copyright notice, this list of conditions, link to this page (https://xdevs.com/fix/hulk3/) and the following disclaimer.
  • Redistributions of files in binary form must reproduce the above copyright notice, this list of conditions, link to this page (https://xdevs.com/fix/hulk3/), and the following disclaimer in the documentation and/or other materials provided with the distribution, for example Readme file.

All information posted here is hosted just for education purposes and provided AS IS. In no event shall the author, 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 willing to contribute or add your experience regarding instrument repairs or provide extra information, you can do so following these simple instructions

Manuals and documentation

Fluke does provide all main manuals, including service information for these older calibrators, right on the product page, without requirement of any registration or access requests, unlike Tektronix or Keysight web-sites. Thumbs up for Fluke, way to host the data on your product page, not hiding behind registration forms, “ask a consultant” or other big-dollar marketing obscurity.

Service notes

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

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

According to An evaluation of the 5700A Series II Artifact Calibration document, major changes in Fluke 5700A/5720A Series II are related to Artifact Calibration and high-frequency AC voltage. Artifact Calibration was improved on a number of ranges, to minimize biases. A new interpolation algorithm is also used for lower ACI ranges to provide better performance.

From hardware standpoint, Model 5720/5700 Series II calibrator has a modification to the in-guard microprocessor circuitry and firmware to minimize noise in the low millivolt DC ranges. Also, because of the tighter specifications of the 5720A, there are more carefully selected components, than in the standard 5700A. Lot of circuit tweaks and changes are on A11 REFDAC and A16 Power Amplifier inguard assembly modules. Many tweaks and minor changes on every board in 5720A level hardware address better reliability and stability of the calibrator. Digital processor board A20 firmware are also updated as a result to reflect added functionality.

Project concept and workflow

Over the course of the project author and other members of xDevs community have learned a great lot about inner workings and troubleshooting non-obvious soft-failure issues, such as out of specification output drift or excessive noise. We hope to provide great amount of details in this article about 5700A or 5720A, however some of this gained knowledge is restricted to internal use and will not be published. If you have specific question about 5700A/5720A/5730A you can try asking us online on this page.

Modularity and 5700A/5720A flexible diagnostics

Fluke designers kept everything very modular, relatively easy to troubleshoot and follow. Many of the local functions are contained within one PCBA/module, allowing great deal of per-module repair or diagnostics. Self-test procedure is also very elaborate and runs about 20 minutes to perform hundreds of test and point measurements to provide you comprehensive diagnostics. Most of the circuitry is tested and many faults can be detected just from the self-test messages. This is great help for the Hulk project here, otherwise without having “spare” working 5700A to make “reference” measurements it would be much harder to diagnost and correctly determine faulty conditions in many mixed cases.

Wise approach here is to test minimum amount of modules/boards first , to ensure voltage regulation/power supplies correct operation and all digital control/interfacing work. Experimentally we determined that to provide valid hardware configurations for functionality checks.

Config, Hardware setup Modules installed/used Provided functions
A, Inguard digital + power test A3+A4 + A19+A20 + A1+A2 Front panel + A21 Unit boots firmware, display VFD, hang on error “Guard crossing”
B, Inguard + outguard power test Same as A + A18 PCBA Unit hangs on “Guard crossing”, but now can check outguard power pre-regulator / filter
C, Outguard power test, cross-guard interface Same as B + A17 PCBA Unit boots, allow to navigate menus, but no any outputs/functions
D, DCV function test Same as C + A8 and A11 PCBA Unit boots, allow to output 220mV,2.2V,11/22V DC Voltage
E, DCV+Resistance function test Save as D + A7,A9,A10 PCBA Unit boots, adds Resistance and DC Current 220 µA – 220 mA
F, ACV function test Same as E + A12,A13 PCBA Unit boots, adds ACV and ACI functions, except 2.2A and 220V,1kV ranges, ACAL can be finished
G, HV/HC test Same as F + A14,A15,A16 PCBA Base configuration, adds 2.2A and 220V,1kV ranges. ACAL fully enabled
K, Wideband option K + Add optional A5+A6 PCBA Full configuration with 2.2mV – 3.5V 30MHz Wideband option
W, Boosted option Connect 5725A to expansion port Enables 11A DCI and ACI, extends V/F limits for 1kVAC range

Configurations in bold letters allow complete self-test and artifact calibration procedures to be completed without errors. Calibrator in this article will be rebuilt to F level. We have no plans to go further, as that would require finding rather expensive A14/A15 and WB modules. Since xDevs.com already have all these in main Fluke 5720A “Hulk-1” and “Hulk-2” calibrators, it was decided not to duplicate the functionality here.

Fluke 5700 “C” Hulk-III build

Let’s go ahead with worklog on repairs and testing, module by module, part by part.

A2 Display front panel PCBA

This board was purchased broken. It is still useful, as calibrator must have controller response from FP to complete boot sequence and operate. Without A2 board calibrator will never finish booting and will not respond to any commands over GPIB or RS232 bus either.

Other than shattered custom VFD glass there seem to be no other damages. Good for us, left VFD designated to display current calibrator status and output signal amplitude/frequency is perfectly fine.

For now we can remove shards and pieces of glass to operate board as is.

Original dot matrix right VFD was manufactured by Noritake Itron, Model DM256×64. Sadly, this is custom design and not available for replacement that easy.

Maybe in future we can buy another A2 donor board but with shattered left display? :)

A4 outguard motherboard

Board has internal layers damage from burn/arc flash.

Some smashed connectors too:

Also high-voltage transformer output was shorted to +44VAC rail. So this required excavation down to layer 5, milling all carbonized epoxy FR4 material and replacing broken connections with copper straps.

To allow spacing and insulation between “layers” additional Kapton tape was applied between connections.

Similar process done for mains switches area.

A9 OHM calibration PCBA

This is Fluke 5720A version OHM CAL board.

Here we can see missing Z1 network under the cage. It will be interesting to retrofit replacement…
This network is used for differential amplifier and has next elements:

Network resistor Z1 element Nominal value Pins
Rx 1 kΩ 1-2
Rx 75 kΩ 2-3
Rx 1 kΩ 4-5
Rx 75 kΩ 5-6

Second issue is destroyed reference 10k/90k network.

Network resistor Z5 element Nominal value Pins
Rx 10 kΩ 1-2
Kelvin junction 0 Ω 2-3
Rx 90 kΩ 3-4

A19 Outguard PSU PCBA

No repairs were necessary on this circuit board. For preventive maintenance all CC resistors and capacitors were replaced anyway.

Original rubber pads desintegrated into gray goo. Typical for old 5700A boards. It was easy fix by cleaning with some IPA and replacing with brand new silicon pads. Purpose of these pads to prevent A19 touching adjacent A20 CPU board.

A18 inguard power supply board

This board acquired as broken with missing parts and FR4 damage. Condition as received:

Burnt PCB around 220kΩ high power carbon composition is dead giveaway for bad board. This is very common issue for 30-year old 5700A and 5720A units. If your unit does not have burnt A18 PCB – then it’s either never been powered on for lengthy duration or have newer revision PCB where Fluke populated good metal film or PWW resistors.

P902 DIN connector had 3 pins bent, so that had to be rectified before board would be installed in the motherboard backplane. Always check DIN connectors for bent/damaged pins to avoid further mating connector damages.

Before any troubleshooting steps begin we must replace all consumable parts, such as electrolytic capacitors (ALL!) and carbon composite power resistors that tend to fail badly with age. All replacements parts purchased from Digikey online, using BOM below.

Part Description Digikey P/N Qty RefDes on A18
cap C16,C15,C14,C3,C202,C13,C2,C20,C21,C6,C9,C19,C18,C7,C8,C5,C4,C12,C11,C22
cap C201,C203,C204,C206,C202,C205
cap C207
res cc R5,R215,R207,R4,R208,R239,R210,R207,R238,R225,R229,R205,R216,R240,R224,R236,R235,R237,R209,R254,R255,R234,R2,R3,R1
fuse F3,R204,R203,R202,R201
diode VR211, VR215, VR219, VR210, VR231

Here is for comparison how modern 5720A latest revision A18 board should look like:

After sniping all bad parts out we got quite a pile already.

After replacing all CC resistors and electrolytic capacitors board was installed in calibrator and powered on. All low voltage DC rails checked out okay. Not so lucky with high-voltage section supply. Without smoke or sparks input 200mA fuses F202 and F203 were blown open. So this part will need more troubleshooting.

A17 Inguard regulator power supply and inguard crossing

This board received as broken/unknown condition with missing metal shield/airflow duct cover.

TBD

A11 reference/DAC board

First outguard power on test

For this test 110VAC supplied from Chroma 61604 AC source to calibrator. Calibrator is populated with A1,A2,A3,A4,A19,A20,A21 boards only. None of the inguard analog PCBAs installed.

Yay, it’s first power on for Hulk-3 5700A project! During this test calibrator consumed ~24W (without fans).

Summary on spent resources

This project took about $6k USD just in parts, so it does not make financial sense to go down this road. It will be a while till this investment pays for itself, but there are many benefits with 24/7 available multiple 7½ precision calibrators for analog design and various experiments.

And everything started with just innocent buy of 5700A A11 PCBA for $200 to tinker with. Who knew that idea to make stand-alone LTFLU reference get converted into this build? “Volt-nutting” at it’s finest, as even not every calibration lab actually own fully-featured 5720A.

Fortune favors the bold. Perhaps hobbyist with that level of commitment could already buy used and fully working Fluke 5700A, which could need only minor repairs (hopefully). But that was not the goal of this xDevs.com project. It is more interesting to understand (and share via such article) what goes into design of high-performance instrument like Fluke 5700/5720A series multi-function calibrators, with hands-on approach during complex repair tasks.

Discussion about this article and related stuff is very welcome in comment section or at our own IRC-chat server: irc.xdevs.com (standard port 6667, channel: #xDevs.com). Web-interface for access mirrored on this page.

Author: Illya Tsemenko
Created: April 8, 2021, 5:07 a.m.
Modified: Jan. 2, 2023, 10:27 p.m.

References