xDevs.com | Wavetek/Datron 4950 restoration and review

Intro
Disclaimer
Manual references
Initial inspection and basic disassembly and service
Firmware
Custom FRAM adapters and EEPROM programmer adapter
Other possible issues
Tweaks and tricks
Calibration
Performance verification
- DCV Noise evaluation at shorted inputs
Restoration summary

Intro

We all know Keithley/Tektronix, HP/Agilent/Keysight, LeCroy and Fluke. However few decades back bit less known T&M manufacturer Datron was also famous for their high-performance DMM and calibrators lineup. Datron was acquired by Wavetek in 1987 for $15.8M US Dollars. Wavetek was still manufacturing original Datron DMMs and calibrators, and eventually sold it’s T&M division in year 2000 to Fluke for $40M. Fluke was happy to obsolete all and everything, replacing the void in the market with their own products. As of today there is no service provided for Datron legacy lineup, but lot of knowledge and technology from Datron was still reused in Fluke’s high-performance products, such as 8508A 8½ DMM and obsoleted now unique 7001 DC standard.

However Wavetek also made some niche specialized metrology gear, such as Datron/Wavetek 4920 AC Voltage Measurement Standard and Wavetek 4950 Measurement Transfer Standard. In this article the latter unit will be taken apart and serviced.

What is Measurement Transfer Standard exactly? It’s a special version meter, that is tweaked to deliver best accuracy on very limited range around full-scale of the measurement range. So as result this unit does not have full measurement ranges as a DMM, ranges of the 4950 MTS are limited to narrow bands both in amplitude and in frequency (for ACV and ACI) to meet only calibration points of the calibrator/DUT. With sacrifice of wide dynamic range and use additional circuitry and adding self-calibration parameters, it become possible to improve transfer accuracy for short term transfers, with it’s main application for metrology labs. Internally Datron 4950 resemble a love story between Model 1281 8½-digit DMM unit and Model 4920 AVMS for AC measurements.

Typical use for The Wavetek 4950 MTS is to allow the automated calibration of Wavetek 4800 series and other Wavetek calibrators on-site. The 4950 MTS is initially calibrated to traceable standards, then a pre-transport calibration check is done. The 4950 will then be transported to the calibrator, and the calibrator will be calibrated against the 4950 MTS (transferring the traceable standards to the calibrator). A post-transport calibration check is done on the 4950 MTS after it is received back at the issuing laboratory to verify the unit is still in specification. If it is in spec a calibration certification can be issued with good confidence.

Transfer Function	Transfer points	Accuracy, 30 day, within ±1 °C TCAL
DC Voltage	±100 mV	±3 ppm
DC Voltage	± 1, 10 and 19 V	±1.5 ppm
DC Voltage	± 100 and 190 V	±2 ppm
AC Voltage	10 Hz – 30 kHz, 1 V, 10 V, 19 V, 100 V	±10 ppm
AC Voltage	55 Hz – 30 kHz, 1000 V	±15 ppm
DC Current	±100 µA, ±1 mA, ±10 mA, ±100 mA	±7 ppm
DC Current	±1 A	±15 ppm
DC Current	±10 A	±20 ppm
Resistance	1-2 Ω, 10-19 Ω, 30-19000 Ω, 30-190 kΩ, 300 kΩ – 100 MΩ	±20-15, 5, 3, 5, 8-180 ppm
AC Current	100 µA, 1 mA to 1A, 10 A @ 10 Hz to 10 kHz	±40 – 200, ±200 – 1000 ppm

Table 2: Brief specifications for Model 4950

There are different options for 4950, such as:

Option ID	Description
Option 80	115V, 60Hz Line Operation
Option 81	115V, 50Hz Line Operation
Option 90	Rack Mounting Kit
Option 95	Rack Slide Kit
Model 4953	10 A shunt

Table 3: Wavetek/Datron 4950 options and versions

Unit in this article came with full package, including Datron 4953 10A shunt and original Wavetek shipping box.

Box also have temperature and humidity meter. Nice touch!

Shunt photos and peek inside:

Disclaimer

Redistribution and use of this article 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:

Redistribution of article must retain the above copyright notice, this list of conditions, link to this page (/fix/w4950/) and the following disclaimer.
Redistribution of files in binary form must reproduce the above copyright notice, this list of conditions, link to this page (/fix/w4950/), 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.

Manual references related to Datron 1281/1271, 4920 and 4950

Wavetek Model 4950, Instrument User’s Handbook, Issue 4, December 1998

Wavetek Model 4950 Specifications

We are not lucky enough to have schematics or service manual for Model 4950 MTS. If you have this information, please upload to our repository by following these simple instructions

Datron 1281 DMM documentation.

Datron Model 1281/1271 Datasheet and specification

Datron 1281 Users manual, Issue 6, Feb 1999

Alternative User’s manual

Datron 1281 Calibration and service manual, Volume 1 and 2, Issue 1, July 1989

Datron 1281 Calibration and service manual with schematics, Volume 2, Issue 2, January 1989

Datron 1281 Calibration and service manual with schematics, Volume 2

There are also useful manuals for similar Datron DMM, Model 1271. Credits for storing these manuals goes to KO4BB. Here’s mirror of them:

Datron 1271 Schematics

Datron 1271 Service manual, Calibration and servicing

Datron 1271 User’s manual

DJVU-format files can be opened with freeware WinDjView from Djvu.org.

Initial inspection and basic disassembly and service

The unit was purchased used, in operational condition. I was asked to go over the unit before it was put into service in our calibration lab. Physically the unit was found to be in very good condition. There was no damage to the exterior, all of the hardware was present, and all of the buttons mechanically operated OK. The covers were then removed for a visual internal inspection. The covers of 5 ceramic DIP ICS were separated from the body of the IC. According to the schematics these are resistor arrays. I inspected them under the microscope and no obvious damage was apparent to the bonding wires and the resistors measured good with a DVM. I located the missing covers in the bottom of the unit and reinstalled them using epoxy. Further inspection of the interior of the unit looked good, this unit does not have any vents to allow dirt or dust accumulation. All of the hardware was present, there was no obvious damage (IE burn marks or excess heat), and there was no leaking or bulging of the electrolytic capacitors. I powered the unit up for an initial functional test. There is a 6 hour warm-up period for the 4950. A battery backed ICM7170 real time clock keeps the time, and if the unit has been turned off for a short period of time a new warm up period will be calculated, so a full 6 hours will not be required every time the unit is powered off. After warm up I successfully ran the confidence test and took a couple of quick measurements to test the unit.

I usually replace all electrolytic capacitors, opto-isolators, single wipe sockets, batteries and fans during a refurbish. I also visually inspect the boards, and remove and reapply thermal compound. This unit only required new capacitors, opto-isolators, and back-up battery. Several of the 1uf capacitors had low capacitance and high ESR. Visual inspection revealed bad solder connections on the ??? board.

After assembly the unit would not pass its confidence test. Troubleshooting revealed a solder bridge on one of the capacitors. I re-soldered the connection and the unit passed all of its confidence tests.

Image 1: Wavetek 4950 front face

Display is custom, model DC405E2 from ITRON/Japan, same used in Datron 1271/1281 and 4920/4920M. Display size is 1×40 char, 5×7 DOT matrix, 5mm height. Itron does not have any stock for the display and do not have public documentation for it. But thankfully, we got and able to share full datasheet for DC405E2.

Image 3: Testing VFD display output and test connections

Image 4: DMM rear face

The reference used in the analog to digital conversion is derived from a specially conditioned Linear LTZ1000-based buried zener reference module. It contains the reference device and its associated support circuits, such as Linear LT1013 opamp and Vishay Precision Group BMF resistor network. All critical parts are packaged in hermetic metal cans to avoid humidity effects. The module is tested to be stable within ±4 ppm per year, produces noise less than 0.1 ppm_peak-peak, and has a temperature coefficient better than 0.15 ppm/°C. This temperature coefficient is held over a wide temperature span of 0°C to +70°C, and the reference exhibits negligible temperature shock hysteresis.

Datron 1281 (and lesser brother 1271) using interesting arrangement for primary power supply with two big toroidal transformers. First transformer converts mains to low voltage supply for earth-referenced digital domain, and powers low-voltage PWM DC-DC around second transformer, with switching frequency in sync with mains.

Patent data

Secondary outputs of this low voltage transformer are rectified and regulated with usual LDOs to provide clean analog supply voltages. What is the benefit in such double conversion? This design approach targeted to reduce capacitive coupling and keep power consumption by linear power supplies constant, so temperature inside meter’s chassis does not change in different conditions.

We already know, that temperature changes are our main enemy in high precision voltage and current measurements, so Datron had to make extra effort to reduce instability of temperature contributors, instead of adding fans or heat sinks around critical components. Similar idea technology was also used in Fluke’s 7000 DC reference standard and recent Keithley’s 7½-digit DMM7510.

Datron 1281 and 1271 are the using unusual arrangmenet for main voltage reference. Model 1281 using two identical Linear LTZ1000 based modules, labeled Datron 400763-1, with custom hermetically sealed Vishay metal foil resistor network, while Model 1271 using LTZ1000 module and LM399 module to save cost. In out units refereces were enclosed in black epoxy, and would be troublesome to remove from PCB for teardown.

Ehile testing optocouplers with second Datron 1281, it was noticed that error codes were different on second meter if optocouplers were left out of circuit. Possibly one of the daughterboards was failing? A quick check of all cables to the main DC board showed one that was not connected and stuffed under the main board. Not good, as this shows someone was diagnosing an issue that required the board to be possibly removed from the frame.

After reconnection cable, this 9005 error code went away and first DCV readings were acquired. It looked good with the input shorted, however the OHMs readings were open. Another search for connector issues showed the input connector was hastily plugged in and had missed a pin on the DC board, pushing it aside. A quick reconnection and all the ranges appeared to respond as expected with the short on the input. A screw was also missing that held the connector in place.

A quick self-test gave what probably were the original errors, DCV ERROR 2182 DC 100mV range zero magnitude and NVRAM error. The errors are broken down based on the type of test. The Fast test failure code is explained in section 2.7.3 for True Zero Checks. It tests the mean value with the inputs shorted on the 100mV range.

Conditions are for test 2182 are : -250µV < Mean 100mV Zero < +250µV.

The NVRAM error is most likely due to a bad old battery, similar to what we already know from repairing old instruments like HP 3458A, scope CSA7404 and R&S SMT03. Good rule of thumb is to replace all batteries if instrument is older than 10 years.

Without further diagnosing of the error, the inguard and outguard supplies needed to be tested as the parts have datecodes from 1988 and the lithium backup battery is dated from 1987. Its voltage was measured at 3.7V which is pretty good for a nearly 30yr old battery.

The schematics for both the inguard and outguard supplies were scanned. Test points for the DC board were easily found as Datron did not use silkscreening on their boards and used unmasked copper for the TP numbers on the DC board. The Digital board, however, has no markings and does not match the component layout drawing in the service manual.

Neither boards have any REV markings on the top, but might be located on the bottom side.

Digital board

This PCBA have main meter’s brain, good old 16 bit Motorola HD68000P8 processor, manufactured by Hitachi. It’s indeed very popular chip, present in many instruments of those years, such as HP 3458A, Keithley 200x, etc. It is NMOS based 5V ASIC, specified to run at 8.0 MHz clock and able to address 16 MBytes of memory space.

There are multiple different versions of this PCBA for Datron 1281. Another kind of board is shown below:

%(imgref)Image X: Digital board, courtesy of www.amplifier.cd

Lot of 7400 series logic can be found around processor, having glue function for various I/O interfaces and control signals. All DIP-packaged ICs are replaceable, with using nice collet-type sockets. Re-seating all chips would be a good idea, considering age of the instrument and unknown history.

DCV board assembly

All but two capacitors were replaced on the DC board (12 of 14). A couple of gotchas were found during the change out. The three 1000 µF caps had one lead bent offset to the bottom. It appears the mounting holes were too close together and did not match capacitor pin spacing. Another gotcha is that it is important to note which lead is offset as there are some components that will prevent it from going in properly. Two of the capacitors actually would touch each other and this is solved with the offset leads.

Another gotcha is that most of the four 100 µF capacitors leads are not soldered straight to the board. At least one positive lead is soldered to the pcb in through a pad but it is not usually shaped for the polarity (square-vs-round). The other leads are soldered straight to components with insulating sleeves. Two of these capacitors were relatively easy to remove by clipping them at the mating lead. The old lead was left on the part and the new lead was soldered next to it. There was concern that additional heat could have caused issues with the aging parts. The other two capacitors were surrounded by other parts preventing them from being changed easily. It was decided to continue troubleshooting at this moment and replacing them would have been done later.

The 470 µF cap was replaced with a higher rated voltage. The new part is as tall as the three 1000 µF caps. The lid has polymide tape directly above the three parts so the same precaution was added above the 100 µF cap.

The remaining capacitors had no additional gotchas.

An IET DE-5000 was used to measure the 1000 µF capacitors for value and ESR. The old parts were at least 20% below rated capacitance and the ESR was approx 1Ω @ 120Hz. In comparison, the new parts were all within 10% and the ESR was < 0.1 Ω.

Compressed air was used to clean the dust and debris from the DC PCB. The board was reinstalled in meter and the FAST SELF-TEST was performed several times over the period of several hours. All tests completed now without errors.

Here’s BOM list of parts used for repairs:

Digikey Part Number	Vendor	Part Number	Specification	Customer Description	Quantity
493-1647-ND	Nichicon	UHE1J471MHD6	470 µF/63V	10000H/105 °C 7.5mm/sp	2
493-1622-ND	Nichicon	UHE1H102MHD6	1000 µF/50V	10000H/105 °C 7.5mm/sp	4
493-6802-ND	Nichicon	UHW1E222MHD6	2200 µF/25V	10000H/105 °C 7.5mm/sp	2
732-9242-1-ND	Wurth Electronics	860040778011	270 µF/63V	10000H/105 °C 5mm/sp	1
493-6817-ND	Nichicon	UHW1H221MPD	220 µF/50V	10000H/105 °C 5mm/sp	1
732-9325-1-ND	Wurth Electronics	860240578010	470 µF/35V	10000H/105 °C 5mm/sp	1
732-9281-1-ND	Wurth Electronics	860240474005	100 µF/25V	8000H/105 °C 3.5mm/sp	4
493-14361-ND	Nichicon	UBT1J330MPD	33 µF/63V	2000H/125 °C 3.5mm/sp	2
732-8966-1-ND	Wurth Electronics	860020772005	1 µF/63V	2000H/105 °C 2mm/sp	4

Now time for option boards, located on bottom side of instrument chassis

Option 10 – True RMS AC Converter

No issues were noted with the AC assy. It was inspected, photographed, and dust/debris were removed before it was reinstalled.

Option 20 – 2-Wire & 4-Wire Resistance Converter

After removal of Option 20 PCBA remnants of spider with it’s home was detected. That likely caused some excessive currents flow thru webs and poor bug, causing measurement issues.

No issues were noted with the Ohms assy. It was inspected, photographed, and dust/debris were removed before it was reinstalled.

Option 30 – Current Converter

Current shunt resistor network consist of separate set of resistors:

RefDes	Value	Specification	Manufacturer
R111	800 Ω	0.1%
R112	90 Ω	0.1%
R113	9 Ω	0.1%
R114	1.0 Ω	0.1%	Mann Components
R115	0.1 Ω	0.1%	Mann Components

Pair of teal colored precision 4-wire shunts are manufactured by Mann Components Ltd (Ayton Road, Wymondham). Mann Components was an English resistor manufacturer with sales of about $2M. It was bought by Vishay in early 1983.

During reassembly of the bottom half of the meter, it was discovered that the Current assembly had some difficulty with the three connectors at the front of the pcb. Looking at the wires an interesting issue was found. It appears that the two outside connectors were placed in the opposite location. The wires are a little short and there is some noticeable stress. Looking for other photos on the internet, a ’94 era 1281 had the connectors located in the right position.

Other repairs

The lithium battery was replaced with a TL-5101P. The power was left on and a battery powered soldering iron was used.

The power input filter was also replaced as a precaution. The old part is a Waycom WF 120-3/05 and it was replaced with a Schaffner FN9222-3-06. The ground cable is exposed so additional heat shrink was added to the P and N terminals.

The power supplies were remeasured. All supplies are close to as-found except the -35V supply. It was now measuring -35.6VDC with 0.13Vp-p noise. The +15V supply measured 15.1 with 0.22Vp-p.

Quick check after assembly, with short on input terminals:

Input DCV +10V test:

Repair workflow, unit #2

This second unit actually happens to be the the first Datron 1281 acquired. It was chosen to repair last repair due to the number of errors.
The NVRAM in this meter is the same as the first 1281 but the battery is a few years newer.

Repair procedure is very similar to previous box. This unit comes with Options 10, 20, 30 and 70, so it have full set of boards and modules to play with.

This unit does not have mount hardware for 19” rack (Option 80) like previous, but comes with foot to angle it on the bench for easier operation. It is still older version of Model 1281, still carry Datron Instruments logo, not Wavetek-Datron like on newer meters.

Model 1281 is rare DMM providing separate active resistance guard terminal, unlike HP 3458A or Keithley 2002. Fluke 8508A have combined Guard terminal, switchable between passive or active mode.

Rear panel is same as first unit, with all the usual connectors and without any fans or vent holes. Keylock to enable calibration mode on this unit is damaged by previous owner, likely by attempts of using wrong key. This is rather unusual today to see such hard keylock to protect calibration, but back then it was easy go-to solution.

Of course it’s just single pole switch and any dedicated calibration pirate cannot be stopped by simple lock, as photo above able to indicate well. ROM/RAM chips have same style paper-written labels with version/serial number.

The fast self-test returned the following errors including test limits from the service manual:

Error code	Limit min	Value description	Limit max
DCV 2182	-250uV	Mean 100mV Zero	+250uV
Ohms 2752	+0.01FR – 4%	Mean Magnitude	+0.01FR + 4%
DCI 2572	-50ppm of FS	Mean Magnitude	+50ppm of FS

The 2752 and 2572 test both use the Ohms 10mA current source. The current source was tested using Keithley Model 2002 as control meter. The 2002 showed a low noise source but the value was nowhere near specifications. Most likely the calibration constants are too far off.

It is hoped that the capacitor replacement will cure the DCV 2182 Error as it did on the first meter. The power supplies will need to be checked first, as usual with old equipment repairs.