RoadTest: Portable Multi-Channel Recorder
Author: Gough Lui
Creation date:
Evaluation Type: Independent Products
Did you receive all parts the manufacturer stated would be included in the package?: True
What other parts do you consider comparable to this product?: Various portable chart-recorder and data logger units from Hioki, Graphtec, Grant, Fourtec, MadgeTech, TandD and their re-badged equivalents.
What were the biggest problems encountered?: Localisation problems with documentation, firmware and software, bugs in firmware and supplied software.
Detailed Review:
B&K Precision / Sefram DAS240-BAT Multi-Channel Recorder RoadTest
By Gough Lui – January 2020
It is with great thanks to B&K Precision/Sefram and element14 that I was chosen as one of three to review the B&K Precision/Sefram DAS240-BAT Multi-Channel Recorder. This piece of test equipment is a rather useful one for field work, as it has 20 channels (expandable up to 200) connected by pluggable terminal blocks using an I/O module that is attached by cable, with 12 channels of digital inputs, 4 channels of timing and 4 alarm outputs available on a separate DB-25 style connector. Best of all, it has the capability of running from battery and uses a 10” colour LCD capacitive touch-screen interface.
This review seeks to test the DAS240-BAT in all aspects of its operation, pushing it to its limits and verifying proper operation under a range of simulated field applications. It also covers the other contenders in the market, unboxing experience, user experience, remote control capabilities, instrument performance and a teardown of the device. The final section details the responses that I received from B&K Precision regarding a number of issues identified within the review.
More details can be found in the RoadTest-in-Depth blog articles linked at the end of each section, which I encourage you to explore. As usual, if you found the review informative, educational, interesting, entertaining or otherwise useful, I would appreciate if you would leave a comment, bookmark, like and/or share the article with others who might benefit.
Oscilloscopes, Data Loggers and Chart Recorders all perform a very similar function of being able to record and display input parameters over time (or parameter versus parameter), but do so at different scales. Oscilloscopes tend to be optimised for high-bandwidth rapid events that happen infrequently and are designed to capture a “buffer” of data at every trigger event (with short acquisition blind periods in-between), while data loggers and chart recorders are designed to capture data continuously to monitor more slowly-changing inputs. Oscilloscopes tend to have fewer channels and lower vertical resolution, favouring higher sample rates while data loggers and chart recorders tend to favour more channels being sequentially sampled and higher vertical resolutions with a lower sample rate. Data Loggers may be optimised for logging without a user interface, depending on a PC, while Chart Recorders formerly used paper as a recording medium leading to an inflexible and bulky solution. Modern multi-channel recorders blur the lines between Data Logger and Chart Recorder by offering both in the same box – such as the B&K Precision/Sefram DAS240-BAT. There are other devices which further blur the distinction, offering new combinations of channels, sample rate, memory depth and vertical resolution to meet various needs.
The B&K Precision/Sefram DAS240-BAT is almost the most expensive single unit in this round-up, but compared to the competition, it does stand out on a number of key parameters. Its integrated 20-base channels of analog input and 12-channels of digital input are amongst the best, with the expandability to 200-channels being unique amongst the competition. The input range is also the widest of all units, offering the best sensitivity for small voltages. The accuracy compares well with the competition, with the 1ms minimal sampling time being comparable to the best from many of the peers (although some are capable of as low as 0.25ms). But where the DAS240-BAT excels is in the offered channel modes, calculations, display size and resolution, storage capacity and battery life. This may justify its higher price tag compared to the competition, although if 10-channels is enough for you and expandability is not a requirement, the DAS220-BAT is relatively competitively priced compared to its peers while offering very similar advantages otherwise. Compared to the other options from Hioki, Grant, Graphtec, MadgeTech, Fourtec, TandD, the market seems to have many older solutions which have more limited monochrome LCD/small colour LCD user interfaces, small memory capacity and limited battery powered options. The only other interesting contender was the Hioki LR8410/8510/8511 combination which offered all-wireless Bluetooth-connected channel interfaces, but at a higher cost with other trade-offs.
For more information, please see B&K DAS240-BAT RoadTest-in-Depth: Ch1 – Market Survey & Feature Introduction.
The unboxing experience for the B&K Precision / Sefram DAS240-BAT is a relatively straightforward and enjoyable one. Packed carefully inside a white cardboard box with a full-colour label on one side, the entire package is quite comprehensive including accessories to make it convenient for the new user. The unit did, however, seem to impact on the rear of the cardboard box, but still made it safely to me in Australia because the individual components are further protected by plastic bags, egg-crate foam, cardboard and bubble wrap compartments.
Included inside the box is the unit itself, a 20-channel module, set of 20 pluggable Phoenix Contact SM STB 2,5-5,08 terminal blocks, flat-headed screwdriver, channel expansion brace (with screws and washers), DB-25 connector (and backshell assembly), rubber-domed capacitive-touch stylus, port covers, shoulder strap, earthed power supply unit from Cincon, US-style IEC power lead, analog channel interface lead from 3M, CD-ROM manual, calibration report and information leaflet.
The unit itself feels solid and rigidly built with a metal chassis that doesn’t flex at all. It has a gorgeous 10” touch panel covering the front of the unit with a kick-stand at the rear that props the unit up to an angle that makes it convenient for use on a flat surface. Build quality is, however, somewhat rough with some of the edges between panels having slight gaps and rough edges that were verging on being sharp. Unusually, the unit weighed in at 1914g which is heavier than the datasheet stated 1.5kg, which is a significant discrepancy. The supplied calibration report also had some localisation issues and expired merely days after the unit was received. Hopefully retail units are better in this respect.
For more information, please see B&K DAS240-BAT RoadTest-in-Depth: Ch2 – Unboxing.
The DAS240-BAT has a solid, chunky-feel that inspires confidence with its aluminium frame and large rubber bumpers. This is, however, tempered somewhat by the large 10” LCD touchscreen that feels otherwise unprotected, small gaps between the LCD and frame, slightly sharp case edges from poor finishing, plastic strap anchor points, plastic rivet construction and abundance of ventilation holes that may lead to dirt ingress over time. An optional case is available to protect the unit while in transit.
Its design is rather user friendly, placing all the important connections at the top edge of the unit, making it easier to manoeuvre. The LCD is bright and large, although the TN technology used limits viewing angles and the glossy finish reflects glare and fingerprint smudges more easily. User feedback from an onboard speaker provides positive assurance of input, with large touch targets on screen and a supplied capacitive-compatible stylus. The single-position kickstand allows for more comfortable usage when resting on a desk, but is not adjustable for the optimal angle and is liable to collapse when too much pressure is applied from the front.
The channel modules provide the DAS240-BAT with a large level of expandability and flexibility, allowing the unit to scale from 20 to 200 analog channels, connected by a 1m cable that has captive features. The channel input blocks themselves use detachable rising-clamp terminal blocks which can be individually labelled and pre-wired, making channel swapping a breeze. The modules themselves are somewhat pricey, making the prospect of leaving them on site less appealing. Likewise, with the separation of the channel input and the recorder unit, one handed relocation becomes a little tricky without applying undue tension to the connecting cable. Digital inputs and outputs do not share the same convenience, instead relying on a traditional DB-25 connector shell which the user must wire to their own needs. There is an optional break-out cable, but it is not exactly affordably priced.
When in use, the user interface responds well to user input, switching between displays relatively briskly. Basic operations can be achieved easily with careful attention to the user interface, with audible feedback given for some key operations such as starting or stopping recordings. However, I found the software on the unit to be its biggest Achilles heel. The software suffers from some strange design choices which take some getting used to – this includes the overuse of ambiguous iconography to depict choices, reversed OK/cancel button order, localisation issues with the text (e.g. Valid instead of Enable), confusing naming (e.g. Logical, Logical Function, Function Channel) and placing options in slightly odd locations (e.g. recorder lock in Additional Settings, record options in Trigger). The software felt like it was not well refined, with certain useful features not available – e.g. interval triggers and with some issues with complex features – e.g. combining channel math with a math channel results in channel configuration corruption and unexpected results, record length limits were rounded, switching between screens resulted in incorrect plot jumps, cursors would read incorrectly if zoomed in then out again, and more. Even simple features like disabling the feedback beeper would not persist after a reboot and would be reset if a self-calibration was performed. The relationship between the hardware capabilities (e.g. hardware voltage ranges) and the software capabilities (i.e. arbitrary defined range, minimum, maximum linked with plotting) was not made clear. Viewing of recordings was also limited to some degree, lacking any ability to perform vertical zoom or searching for a given digital or analog input condition, which could make looking for a particular event rather tedious. Zooming was limited to using buttons and cursors, with no gesture-based multi-touch support. It was also possible to cause the software on the recorder to crash into a hard lock situation or to fail to recognise its internal storage. In this regard, it seems that the DAS240-BAT’s hardware is capable, but the software still needs quite a few improvements before I would consider it completely reliable and user-friendly (especially for English-speaking users).
The supplied documentation consists of a CD-ROM and a calibration certificate. The CD-ROM contains user manuals in many languages, but does not appear to contain the software relevant to the DAS240-BAT. The disc itself has a number of odd quirks, such as not working on high-lettered CD-ROM drives, relying on Internet Explorer and containing shortcut .lnk files. In all, I would recommend avoiding the disc altogether, in favour of the online resources. The calibration certificate received was also within days of being one year old. When asking B&K Precision for clarification, they mentioned that the certificate was provided as proof that the recorder was functional and compliant as shipped but could not guarantee the certificate had any remaining validity when unpacked by the user. They recommend calibration at yearly intervals or less, subject to customer requirements. While this was understood, it is also slightly disappointing as some other manufacturers do offer assurances that provided the unit is stored within published specifications, that the certificate would be valid for a year from when the unit is put into use.
Online resources were rather scattered with different versions of software available from B&K Precision and Sefram’s websites. Downloads were split between product pages and software update pages. The published datasheet is both a brochure detailing key features and a table of specifications. It was a bit light on some key details, such as the logic level input threshold, RPM accuracy, pulse count mechanism (edge/level) and minimum pulse width for pulse counting/frequency/RPM measurement. There are also some inconsistencies, including the 4A power supply specification which does not match the main unit or the supplied adapter and the claimed weight of 1.5kg is well below the measured 1.9kg. Finally, the list of options is also incomplete. The manual is fairly similar in this regard, as it is relatively brief and light on details. While it is relatively easy to read as a result, there are a number of poor expressions and minor typographical errors scattered throughout. I think that this is also an area for improvement.
For more information, please see B&K DAS240-BAT RoadTest-in-Depth: Ch3 – User Experience.
The recorder features an inbuilt 10/100Mbit/s Ethernet interface with the option for a USB 802.11b/g/n Wi-Fi interface. Configuration of the interface is available via DHCP or static IPv4 addressing. Over this connection, the DAS240-BAT provides access to files via HTTP (web browser) at up to 4.8MB/s over Ethernet and 2.6MB/s over Wi-Fi and via FTP at up to 8.5MB/s over Ethernet and 2.9MB/s over Wi-Fi. Access to the screen and user interface is provided via VNC server, which is protected by a password up to 8-characters (visible in the configuration menu), allowing for remote front-panel operation. A Modbus Server allows for access to channel readings directly via a de-facto standards protocol. An inbuilt NTP client allows for time synchronisation over the network as well. This arrangement generally works well to allow the device to be used and accessed remotely.
The unit is supported by DASLaB and SeframViewer. The DASLaB package allows for remote re-configuration of the device, file management and plotting of values. The SeframViewer application allows for reading the .rec files generated by the recorder, plotting the data, performing basic analysis and annotations and exporting the data for analysis in third party applications. Unfortunately, I ran into problems with both software packages including problems with IP-address configuration issues, recorder set-up corruption and incomplete displays in DASLaB; and some semi-transparent windows and .csv export issues with SeframViewer. Both software packages lack polish and suffer from numerous localisation errors when used in the English language. The downloads on both B&K Precision and Sefram’s sites differ in versions, so it can be hard to ensure you have the latest version of the software. The downloadable FlexPro software listed on their site is a third-party data analysis package without any supplied license and testing of this software was not undertaken. Their GitHub provides sample code for Modbus connection using National Instruments LabVIEW and Python, although it seems the Python example using umodbus was not actually customised for the DAS at all. The LabVIEW code was not tested, however, the specifications were sufficient to design your own Modbus client noting a minor error in address calculations in the documentation.
Unfortunately, it seems that the network connectivity is not as robust or as secure as it should be. Aside from the password-less access via HTTP to the stored data, the FTP anonymous access allows for file modification, deletion and upload as well. The recorder’s Modbus server does not appear to report over/under-range conditions properly, while the NTP client does not appear to work with hostnames when a static IP is configured as no DNS servers can be entered. Even more interesting, the unit serves an unauthenticated Telnet-style interface for configuration pulls and pushes, participates in SMBv1 sharing with an unknown username and password, requests and configures its own IPv6 address (which may expose it to the internet directly in some cases). Add to this the fact that I was able to crash the software running on the DAS240-BAT and crash the networking stack, taking the recorder offline entirely, this does not suggest that the unit would be safe to connect to networks containing untrusted peers.
For more information, please see B&K DAS240-BAT RoadTest-in-Depth: Ch4 – Remote Control Interfaces.
Verification of the DAS240-BAT’s performance was done with the help of a Rohde & Schwarz RTM3004 Oscilloscope, NGM202 Power Supply and Arduino Uno board, with copious use of Python-based automation techniques.
The power requirements of the DAS240-BAT were determined to range from a peak of 1.248A when charging when powered off, to a peak current of 1.746A when charging and powered on. It seems likely that true peak requirement is closer to 2A or 2.5A which is more than the 1A printed on the rear of the case, but less than the 4.5A power adapter it is supplied with. As a result, it seems likely that the supplied power supply would have a long reliable service life due to this. Battery recharging time from empty appears unaffected by unit operation, coming in at around 7 hours 20 minutes regardless with a total delivered energy of 112Wh.
Analog channel accuracy appears to be well within the specifications, although at the lower ranges, it was difficult to draw firm conclusions due to the large error contribution from the NGM202 compared to the DAS240-BAT’s specification. Unfortunately, as I do not possess any better test equipment, further tests of the lower ranges of 1V and below is difficult. Above this, it seems clear that the unit is able to meet and beat the claimed accuracy levels with the resulting error remaining within the boundaries even when discounting the NGM202’s contribution. The unit also appeared to be quite stable from cold, not exhibiting any clear warm-up trend at the standard room temperature.
Digital timing channels were tested with 5V-level signals, with the counter found to be operating correctly provided the pulse widths were sufficiently wide, having the occasional erroneous value in case of high frequency/short duty cycle pulses. The frequency measurement range appears possible to set values outside of the datasheet specifications up to 200kHz however, readings were suspect especially below 10Hz and above 4kHz depending on the duty cycle of the signal with inverse-duty showing similar characteristics (i.e. 1% vs 99%, 5% vs 95%, etc). One outlier was also recorded with regards to frequency. Duty cycle measurements, however, were found to be unreliable and in significantly higher error (up to 6%) compared to datasheet specification, being particularly poor for signals below 50% duty. This seemed to be frequency independent, suggesting a potential software issue with the way the DAS240-BAT calculates duty cycles and has been reported to B&K Precision for future improvement.
The digital logic channels were tested by iterating all possible 12-bit values using the Arduino, showing the signal could be resolved even at the maximum 1kHz sampling frequency. Testing of the alarm output was also undertaken, with some alarm events based on this signal input being delayed or missed entirely as it seems the DAS240-BAT could not keep up with the input. Digital signal level thresholds were measured to be about 3.3V, with reliable zeroes below 3.18V and reliable ones above 3.41V making digital inputs suitable for use with 5V TTL and high-level signals, although incompatible with 3.3V TTL.
Closer testing of the alarm outputs was able to show that alarm outputs could only be generated at a rate about 14Hz, with a propagation delay variable from 4 to 10ms which would be fast enough for many applications except perhaps the most demanding research and development needs.
For more information, please see B&K DAS240-BAT RoadTest-in-Depth: Ch5 – Instrument Performance.
A lot was learned by applying the B&K Precision/Sefram DAS240-BAT in simulated field conditions. A rather nice discovery was that the battery life was able to exceed the manufacturer’s claims – my testing indicated a worst-case scenario of about 12 hours 51 minutes and a best case of 24 hours and 50 minutes, far exceeding the 10 hours and 15 hours claimed by the datasheet. This provides a good buffer against battery capacity degradation over time, allowing for a longer working life.
The unit was applied to various monitoring tasks. The first was individual cell monitoring of a six-cell Ni-MH pack during charging and discharging. The isolated channels allow for a math-free way of monitoring the condition of each individual cell, allowing two bad cells to be identified where the pack’s overall voltage curves would not have given a clear indication of problems. The DAS240-BAT’s CAT I 100V safety rating does limit its applicability to larger battery banks, however.
A simulated field test was also completed outside in the hot summer of Sydney, Australia on a day when the temperature was expected to reach 45 degrees Celsius. Testing of thermocouple and voltage inputs with scaling factors applied was successful, albeit with unusual results recorded when the temperature of the channel connection module exceeded 40 degrees Celsius, likely indicative of a limitation in the compensation capabilities of the reference junction. Software flaws in both the recorder firmware math channel handling of a channel with change-units applied, and in re-scaling of a channel with change-units applied were also experienced, suggesting the more complex functionality is as yet unpolished.
Testing of the RPM functionality of the frequency measurement mode offered by the K-channels (Logical Function Channels) was successful, allowing for the measurement and determination of the relationship between input voltage of a brushless DC fan and its RPM. Unfortunately, limitations in the recorder’s software meant that X-Y plots between dissimilar channel types (A-channels and K-channels) is not possible, meaning that this experiment instead had to rely on analysing recorded data on a PC instead. Likewise, it was also discovered that the Modbus TCP return value was not the measured RPM but instead a percentage value of the maximum RPM configured in the channel menu, which is not the expected behaviour.
Testing of fast-changing signals using a high-voltage differential probe and mains electricity was able to demonstrate the 1kHz sample rate reconstructing the sine wave input, however, limitations in the sample-by-sample interpolation-free display made it difficult to analyse the results. Likewise, no channel functions were available to compute RMS amplitude, frequency or process AC signals in any other way. It seems more suited for analysing slow-changing near-DC type continuous signals.
Testing of small amplitude signals was successful, by hooking up a channel input to a two-terminal polar chest strap and using the unit similar to an ECG machine to measure the waveform from the heart to show the heartbeat. Likewise, testing of X-Y mode was also demonstrated successfully for measurements of resistance where the DAS240-BAT was responsible for injecting the bias current.
For more information, please see B&K DAS240-BAT RoadTest-in-Depth: Ch6 – Simulated Field Tests.
The B&K Precision / Sefram DAS240-BAT feels to be a sturdy and chunky unit held together by plastic rivets. Disassembly is simple and can be done without tools, voiding the warranty/calibration seal. Internally, the unit is built around a Toradex Colibri T20 System-on-Module, with an Analog Devices AD7694 16-bit 250kS/s PulSAR DAC and ADR02 Ultracompact, Precision 5.0V Voltage Reference as the core of the analog solution. This is supported by a number of other components including a number of Xilinx FPGA/CPLDs, solid-state optoisolated relays, isolators, real-time clock, LCD panel driver and power converters. The digital side appears to rely on a set of level shifting transistors and parallel-load shift-registers, with power supply via a resistor and PTC self-resetting fuse. Power is supplied from a Jauch 18650JP 3S3P pack rated at 10.8V / 8.85Ah / 95.58Wh constructed using Panasonic NCR18650A cells of 2950mAh (min) / 3070mAh (typ). The channel module makes plentiful use of IXYS OMA160 and OAA160 OptoMOS Relays for switching each channel and module. Storage is provided by a Transcend Premium 32GB microSDHC Class 10/U1 card.
The unit itself runs a Toradex-provided build of Windows Embedded Compact (CE) 7 from 2010 with a custom-developed application from the internal ROM image that houses the OS. The unit hosts VNC using Efonvnc Server WinCE Edition, with ModBus TCP served by the custom application and the remainder by the operating system’s included servers. I would not recommend attaching the device to an insecure or shared network as many of these servers do not need credentials for access (FTP, HTTP, ModBus) or are themselves insecure (SMBv1). The fact that I was able to crash their custom application to gain access to the OS suggests the security is less than optimal. The support for Wi-Fi connectivity appears to be limited to those supported by netrtwlanu.dll which include a range of Realtek-based adapters with specific product IDs, mostly from the 802.11n generation.
For more information, please see B&K DAS240-BAT RoadTest-in-Depth: Ch7 – Teardown.
The B&K Precision/Sefram DAS240-BAT Multi-Channel Data Recorder is aimed at users who want to have expandable universal digital/analog/timing data-logging capabilities in a portable and convenient battery-operated form factor that does not require interfacing to a PC but also does not sacrifice the graphical representation provided by a chart recorder. To this end, the DAS240-BAT is, on paper, a formidable competitor in this field with competitive specifications across the board, in return commanding a higher price.
The unit comes well-packed with a full complement of accessories, including a shoulder strap, screwdriver for terminal blocks. DB-25 connector/backshell for use with the digital I/O and capacitive touch stylus. It seems a lot of thought was put into ensuring the package is complete and “ready to go” out of the box.
When put to use, the unit was a bit of a mixed bag. The large LCD was bright and spacious, with a user interface containing large touch targets and reasonably quick user response times. On the downside, the TN technology display resulted in more limited viewing angles, with the glossy finish reflecting glare and the frame surrounding the screen having a slight gap. While the unit felt solid, the screen felt vulnerable and the finish on the casing was not smooth. The inclusion of a kick-stand definitely makes it easier to use on a table-top, but with a single position detent, it was difficult to prop up at the ideal angle and was liable to collapse if sufficient pressure was applied from the front. The terminal blocks were of good quality, rising clamp style which makes for easy connection and disconnection from the detachable channel modules, however, these expansion modules might not be as affordable as one may desire. Likewise, the digital DB-25 interface requires hand-terminating wires to make it useful, with the optional break-out cable also commanding a not-insignificant price.
But the biggest Achilles heel of the unit’s experience would lie in the software that runs on the recorder. In many places, there are issues with poor localisation and odd design choices including overuse of potentially ambiguous icons, confusing naming, reversed OK/cancel ordering and features in unexpected locations. There were also a few software bugs with incorrect cursor readings, jumping of plots, problems with setting accurate recording intervals, and more. This suggests that there is a lot of room for improvement with regards to the firmware.
Unfortunately, the documentation for the unit also seems a bit sparse, with localisation issues and even an occasional technical error. Their datasheet is light on some potentially important details, while also containing errors for key values such as the weight of the recorder. The supplied CD-ROM does not contain the software for the recorder, with some issues functioning on specific configurations, while online-downloads are scattered across three different webpages on two different sites with different versions depending on which site you visit.
This impression continues on when we look at the remote-control interfaces and software provided. The unit has a number of interfaces including Ethernet and Wi-Fi through a supported USB dongle, over which HTTP, FTP, VNC and Modbus TCP access is available and NTP can be used for time synchronisation. The interfaces functioned well, however, it was discovered that the device’s connectivity did not seem as secure or as robust as it should be, with password-less access on HTTP and FTP, the latter capable of file modifications, over/under-range conditions not being reported over Modbus TCP, unauthenticated Telnet-port interface for configuration pull/push, SMBv1 file-sharing server with no known username and password, IPv6 autoconfiguration by default and the ability to crash and take the recorder’s whole networking stack offline. I cannot recommend the use of the device on unprotected/untrusted networks as a result.
The supplied DASLaB package works for basic reconfiguration, file-management and plotting tasks, although did suffer issues with IP address entry and more advanced configurations leading to corruption. The supplied SeframViewer allows for plotting and basic analysis of the .rec files made by the recorder and exporting the data to .csv, .txt and .xls files. In my experience, the software suffers again from numerous localisation issues, a broken .csv export due to delimiter clashes or crashes of the software. While basic tasks may be accomplished, the software lacks stability and polish.
With regards to instrument performance, it seems that the analog channels perform to specification with a stable reading that did not seem affected by warm-up. The digital timing channels, however, were a mixed bag with the software allowing ranges outside of the datasheet to be selected. Even within the datasheet’s parameters, problematic readings for frequency below 10Hz or at very high/low duty cycles were observed. Duty cycle measurements were also unreliable, especially below 50% duty cycle where an error of up to 6% was observed. The digital logic channels appear to have a threshold around the 3.18-3.41V mark, meaning that it would be compatible with 5V TTL signals up to 24V digital signals. Alarm outputs toggle at up to 14Hz with a 4-10ms propagation delay, while the instrument consumes about 1.746A when charging and powered on. The unit charges in about 7 hours and 20 minutes regardless of whether the recorder is active or not.
While in simulated field conditions, the observed battery life performance exceeded datasheet claims, with 12 hours 51 minutes recorded in the worst case and 24 hours and 50 minutes recorded in the best case. Advantages of having isolated channels were demonstrated, although the 100V rating could prove limiting when working on battery arrays. Simulated outdoor field testing showed thermocouple readings were possible, but only up to reference junction temperatures of about 45 degrees C where the unit suddenly lost the ability to read compensated temperature. Testing of the RPM mode was successful, although plotting of dissimilar channel types does not appear to be supported at this time and Modbus TCP reports a percentage rather than RPM. Finally, testing of fast-changing and small amplitude signals were successful, although the recorder is perhaps best suited for slower continuous style signals owing to the lack of sample interpolation on the display.
A teardown reveals a Toradex Colibri T20 System-on-Module (Windows Embedded Compact (CE) 7), Analog Devices AD7694 16-bit 250kS/s PulSAR DAC and ADR02 Ultracompact, Precision 5.0V Voltage supported by a number of Xilinx FPGA/CPLDs, IXYS OMA160 and OAA160 OptoMOS Relays, isolators, real-time clock, LCD panel driver and power converters. The digital side appears to rely on a set of level shifting transistors and parallel-load shift-registers, with power supply via a resistor and PTC self-resetting fuse. Power is supplied from a Jauch 18650JP 3S3P pack rated at 10.8V / 8.85Ah. Storage is provided by a Transcend Premium 32GB microSDHC Class 10/U1 card.
In the end, I would have to say that the DAS240-BAT is a capable device in terms of hardware, which works for most basic tasks out-of-the-box, but still feels relatively unpolished with a number of software bugs, localisation frustrations and functional deficiencies. I have made note of these issues/feature requests and have dutifully reported them to B&K Precision for investigation and potential rectification, which may take some time to fully resolve. B&K Precision has mentioned that they do value the feedback and are committed to addressing the most pressing issues. It is hoped that in the future, the DAS240-BAT and its supporting software can mature to a more stable, robust, reliable and user-friendly platform, but for now I think buyers will have to take some of these issues into consideration when they’re considering a purchase and perhaps look for updates to see when and if anything has changed.
Thanks again to B&K Precision/Sefram and element14 for selecting me as a RoadTester for this product.
Throughout the course of the review, many noteworthy discoveries were made that suggested there were areas for improvement for the documentation, firmware on the device and software supporting the device. I used the e-mail contact form to report issues to B&K Precision as they were discovered, with the first contact on 12th January 2020 and periodically thereafter to report new findings.
I offered them an opportunity to make responses and improvements that would be incorporated into the review provided they reached me before the end of 20th February 2020 (a week prior to the RoadTest Review due date, to provide adequate time to check any fixes, document it within the review and publish). In my contacts with them, they did stress the importance of making any fixes carefully due to risks of introducing new bugs and regressions, thus I did not expect that the short timeframe of a RoadTest would be sufficient for the remediation of all identified potential issues.
Understandably, while I have maintained contact with the staff at B&K Precision, due to unforeseen circumstances outside of their control, they were not able to have a finalised response by the initial deadline. I have instead delayed publication of the review until they were able to supply some form of formal response to identified issues.
Their response with regards to issues identified were extensive. Instead, I have summarised their responses in the following table:
Issue | Response |
Math channel corruption | Corrected in last version |
Software vs Hardware range selection unclear | There is no hardware offset generated. So the range used is the upper closer to the measured value. |
Brightness control is very non-linear | It will be corrected |
Possible to set trigger time/date already passed without warning | It will be corrected |
The file length limit setting rounds values | Corrected in last version |
Lack of thermocouple burnout detection capability | It will be corrected |
Frequency measurement unstable from 1-10Hz | It will be corrected |
Duty cycle measurement is severely in error | It will be corrected |
Changing screens during recording results in traces jumping up or down | It will be corrected |
While recording, screen flickers "Calculation in Progress" | Problem corrected |
Cursor reading incorrect when zoomed in and out again | It will be corrected |
Screen off time actually controls screen dimming | The appropriate word will be used, it will be corrected |
Number '2' misaligned on the on-screen number keypad | It will be corrected |
Localisation issues in firmware | It will be adapted when products will be sent to BK |
Capitalisation of month | It will be fixed on a new firmware version |
Splash screen has the wrong slash on the URL (should be / not \) | It will be corrected |
Modbus TCP server does not report over/under-range | Corrected in last version |
Entering numeric values by host keyboard over VNC does not work | It will be corrected |
Lack of VNC remote cursor feedback | Software not developed by Sefram, no modification possible |
VNC password corrupted out of the box. | It will be corrected |
Password-less access to services, general network security and robustness | This point is taken in consideration for future investigation |
Network IPv6 autoconfiguration without any control | This point is taken in consideration for future investigation |
PCB construction quality - Kynar wire and inductor | Will be improved for the next production with new supplies |
In light of this, it seems B&K Precision and Sefram have been diligently working through the feedback provided and have begun to address many of the issues identified in the review. It seems likely that a new firmware and documentation will be released as a result in the near future. Besides the issues listed, there were a number of more minor feature requests and issues which are not currently being addressed at this time, which is also completely understandable.
The transparency of response from B&K Precision must be applauded, as well as their drive to improve their product based on feedback from their users. It shows that they do care about supporting their products beyond the initial point of sale. I look forward to the release of updated firmware and verifying the improvements for myself.
Top Comments
Hi Gough,
Great review! It was very interesting reading your take on the product - awesome exploration of the features.
A complete and thorough review. But it seems like the software was insecure to a large extent. The device is usable as a standalone. Can the device store and replay graphs later when brought back to the…