RoadTest: Portable Multi-Channel Recorder
Author: shabaz
Creation date:
Evaluation Type: Test Equipment
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?: There are a lot of data loggers and chart recorders, but the DAS240 is unusual with the very wide range of voltage settings, providing the ability to get clean measurements even at very low voltage output sensors. Also, ease-of-use is miles ahead of older systems such as an old chart recorder I own, but also far easier to use than modern logging bench meters due to the easy touch display (although it could be improved - some icons were confusing initially).
What were the biggest problems encountered?: Some minor bugs (not service-affecting) reported to the manufacturer. It will probably need a protective case or bag before you can really deploy it in the field. Other than that, no issue using the product.
Detailed Review:
Recently I have been working lots with several data logging products, so I was interested when I heard about the latest BK Precision data recorder! It is a compact battery-powered unit (298 x 176 x 66 mm), about the size of a small laptop, but thicker. The front has a large touchscreen, and the interface connections are on the long side at the top.
I have purchased and used several BK Precision products over the past few years, and I’ve also reviewed a couple of BK Precision products too. I think they’re generally great value, flexible and easy to use, so consequently I get a lot of value out of the products.
The DAS240-BAT (the name means Data Acquisition System with internal battery!) from a birds-eye view is a portable device with a display, a 16-bit analog-to-digital converter, and a plug-in unit to allow up to 20 analog signals to be attached (the plug-in units can be chained for a total of 200 channels).
The DAS240-BAT will scan the channels (at up to 1000 measurements per second overall) and plot the data to the display, and record to memory. A very rough approximation would be a slow oscilloscope, but with the high resolution of a multimeter (around 4.5 digits).
It is fundamentally different from an oscilloscope and a multimeter however. It is easier to see its functionality through some example use-cases.
The DAS240-BAT provides several functions that can replace other instruments. One example is the chart recorder; you could replace these older pen-plotting devices with the DAS240-BAT. It has applications for temperature and humidity monitoring in labs or factories over days or weeks for instance.
It is a portable battery-powered device that is useful for field work. In the HVAC world and hot water supply boilers in large premises, a flexible device is handy for testing temperatures in pipes for long periods, to reduce the risk of diseases such as Legionella, due to bad temperature conditions, or for checking if equipment is arcing occasionally and logging the detail to justify replacement.
The scientific world can make use of portable data loggers for long-term experiments. I was keen to try to build a magnetic observatory with the DAS240-BAT.
Another modern-day use for portable high-speed and long-term data loggers would be to capture sensor data for use as training data for artificial intelligence (AI).
That’s a lot of diverse use-cases. And finally, although it is a field instrument I was also keen to explore the DAS240-BAT as a curve tracer for electronic engineering purposes.
This review covers some of the scenarios described above.
For a 3-minute 'Fun with Data Recorders' video, click here:
The DAS240-BAT has a folded all-metal shell with a large screen. The entire unit is fully touch-screen operated apart from the power button which is on the top edge.
There are two models currently. The all-on-one DAS220-BATDAS220-BAT model has built-in connectors for the input channels. The DAS240-BATDAS240-BAT reviewed here has a connector for attaching a remote 20-channel interface box, and many of them can chained for a total of 200 analog channels. There are 12 built-in logic channels, and 4 frequency or PWM measuring channels too. There are four alarm outputs as well.
The unit comes with a lot of accessories to get going straight away. In particular, it is supplied with a 20-channel interface box and cable, as well as 20 plug-in screw terminal connectors. It is supplied with a carry-strap to make the unit more portable. There were many accoutrements supplied, such as a little screwdriver and a touch stylus (which during my use I didn’t need on the capacitive touch-screen) and even a screen cleaning cloth.
The display resolution is 1024 x 600 and with the large display the user interface didn’t feel cramped. Viewing angles are average; as good as any typical test instrument these days. I felt the display is fine for the use-case.
Despite the unit by default coming with many extras, a few things are not supplied that would be worth purchasing. One is a Bluetooth WiFi adapterBluetooth WiFi adapter; it’s expensive ($100) but could come in very handy if Ethernet is not available and remote connections are still needed. Any generic USB WiFi adaptor will not work due to limited drivers installed. However, the Plugable.com USB 2.0 Wi-Fi Adapter model USB-WIFINT does work, and is under $15. The cost of the officially supported one is what it is – other manufacturer products cost just as much – I spent a similar amount on a plug-in Bluetooth interface for a multimeter from a different manufacturer. It is worth buying the $15 adaptor because using the instrument remotely with your mobile phone (see later in the review) is extremely convenient.
Another thing which needs to be addressed is the physical ruggedness. It is an instrument intended for field use, and it feels very naked without a carry case or bag. The display is large, and so there’s a risk of damage without a case. The carry casecarry case costs $100, which again is in line with other manufacturer accessories, but for a budget option I used a cloth bag with strap. I may insert a sheet of acrylic as a protector for the screen when in the bag. Anyway, the end result was a field recorder kit set-up that was personalized to suit my needs, and I can use the extra pockets to carry cables and connectors.
A supplied DB-25 connector can be used to interface to the logic channels and alarm connections. I sourced a 15-way cable to attach to that. I printed off a label for the rear of the instrument, to remind me in future of my wiring color-code.
There’s a tilt stand on the rear of the DAS240-BAT but it is quite basic with just one tilt angle. The input connectors and the power button on the other hand are rugged. It is great that they’re using reliable DB-25 and Mini D Ribbon (MDR) connectors, because these are easy to source if additional cables need to be made.
To summarize, I think the construction is fine, but more ruggedness on the display side, such as a supplied hard shell screen protector or carry case would have been welcome. Another justification for a case or a bag is that storage is needed for the ancillary items like the interface block and cables, so if you’re thinking about purchasing the device, you’ll definitely want to budget for a case or bag.
There are several ways of working with the instrument. The touchscreen can be used, or alternatively USB peripherals can be attached. I tested using a Logitech wireless keyboard and mouse.
Yet another way is to make use of the network connectivity (either the built-in Ethernet interface or optional WiFi USB adaptor) and control the instrument from a PC or mobile phone.
Working with the DAS240-BAT user interface is fast. The touchscreen is responsive, and it takes just a minute to be ready to view and capture data (longer if you’re setting up things like formulas or many channels). The main screens that are navigated in a normal workflow are shown below. From the home screen (top left) you can get into a grid-like channel configuration view. Tapping there brings up a large numeric keypad for fast entry (and it’s got practical things like an exponent button). Equations if required are chosen from a decent selection. Channels can be switched on/off and then two main types of charts can be selected from the home screen – plotting against time, or plotting against an input (XY mode). The file system is a little weird – the icons are confusing and I hope that changes in a later software release.
There are many additional (let’s say secondary) workflow screens too, as shown below. There’s a fair amount of chart customization, and the multiple meter display was great. It also shows min/max for each channel.
VNC (short for Virtual Network Computing) is a popular cross-platform method of viewing and controlling the screen of computers, so it was nice to see it supported on the DAS240-BAT. It means that the device can be controlled from any computer platform as if the user was sitting in front of the device. I tried it with RealVNC’s VNC Viewer app on Windows, and it was useful for quickly grabbing screenshots (although there is a dedicated screen capture capability on the instrument itself). However what I really liked was that for awkward situations, VNC can be installed and operated from a mobile phone. Outdoor testing or surveying scenarios could be made easier by leaving the DAS240-BAT in the bag or case, and just viewing and controlling from a mobile – much easier when on the move : ) For that to be convenient, it would require the USB WIFi adapter and optionally create a wireless hotspot - most Android phones support this functionality.
The DAS240-BAT has analog inputs for DC voltage, resistance and temperature, as well as digital inputs that can be used for pulse counting or frequency or pulse width modulation (PWM) measurement. It can sample (broadly speaking - there are some caveats) at up to 1000 times per second, which is fine for a lot of use-cases, but it would have been nice to see a higher rate.
On the analog side, the primary measured quantity is voltage. The resistance ranges are limited to two ranges of up to 1 kohm and 10 kohm. The DAS240-BAT is primarily intended for connecting to sensors that output a voltage. A very wide range of thermocouples are supported too, amongst other temperature sensors. Current can be measured using an external shunt.
The chart below shows the voltage measurement accuracy, and it's very good. The horizontal axis indicates the voltage to be measured, up to 100V, and the vertical axis indicates the maximum error in volts, assuming that the best range is used for the measurement. The orange portion covers the DAS240-BAT performance, and the other lines are a comparison. It can be seen that above 100mV the DAS240-BAT performance is very similar to a decent mid-range hand-held multimeter, such as the Fluke 175. Compared to another data logger, the Hioki MR8901 card, the DAS240-BAT has better accuracy, although the MR8901 has a faster sample rate (but fewer channels).
It’s clear that the DAS240-BAT is quite unusual in that it has a lot of ranges even for a data logger. At the low voltage measurement end, measuring less than a millivolt, the accuracy compares well with the most advanced hand-held multimeters such as the U1282A. It won’t compare to a high-end bench multimeter, but then the DAS240-BAT is a field instrument, not a lab instrument. It has a 16-bit converter. The unusual low-voltage ranges makes it a excellent for trialling sensors. For a portable battery-powered field instrument, the DAS240-BAT has good performance. Bench multimeters may come with higher resolution, but then they are not battery-powered field instruments.
The diagram below shows an approximation of how the instrument may work (this is a guess!). The switching is done inside the expansion modules using solid-state devices (no relays) and as can be seen from the accuracy chart above, it works really well:
The DAS240-BAT has some advantages when it comes to sensor measurement compared to other measurement instruments, not least the large amount of supported channels. Often many channels can be required, to handle the sensors, monitoring of reference voltages and ambient conditions. Some modern sensors provide data that varies with many parameters, so it is important to capture the environmental conditions at the same time. The measurement ranges are highly configurable, to suit many different types of sensors.
For this particular test, I wanted to be able to capture data from two sensors, along with associated reference voltages so I needed at least four channels. The sensors would be oriented the same, but separated by a distance. This, combined with the DAS240-BAT would form a gradiometer system that would record the difference across that distance. Such gradiometers can be used to search for underground features by looking for slight anomalies in the earth's magnetic field.
The sensors were chosen to be Texas Instruments DRV425 fluxgate magnetometersDRV425 fluxgate magnetometers, which are sensitive enough to detect the earth’s magnetic field. By positioning them apart, if any mineral or object in the earth affected the magnetic field, then it could be detected from the difference in measurements.
The sensors output a voltage between 0 and 5V, and usually the level in the absence of any field is mid-way at a reference voltage of about 2.5V. There is a 100 ohm resistor on the sensor board that I swapped out to a 510 ohm resistor to increase the gain so that detecting the earth’s magnetic field in my region could cause a difference of 1.25V. Next, the sensors were oriented and aligned in position using a thick acrylic tube. The sensors were about 1 foot apart from each other. The alignment was done using wood dowel as a non-ferrous screwdriver to rotate one of the sensors to set it into a similar plane as the fixed sensor.
The sensors were connected up to the DAS240-BAT, and the input channels were set for +-0.2 V since I wasn’t expecting a higher magnetic field than this. It was possible to add a ‘function’ channel, which was set to be the difference between the two input channels, and then plot that function output with a range of +-10 mV. It worked really well. It was responsive, and by moving the acrylic tube around outdoors I could detect an iron pipe (a washing line support) from a couple of feet away. I know that far higher sensitivity is possible, but it requires a more thorough setup and accurate alignment. In any case, I was impressed with the capability. I was ready to go surveying – I just needed someone to carry a shovel!
The photo below shows the output in orange, when a screwdriver is moved about a foot away from the end of the plastic tube.The purple and green traces reflect the near sensor and the far sensor respectively. The orange output is the difference between the two, shown at a higher resolution – it hit the limit of the display when the screwdriver was sensed.
The DAS240-BAT worked well for this purpose, however initially there were a few bugs. Basically, some of the channel settings seemed to alter or become corrupted after shutdown/startup, or if I did things like leaving function channels enabled while disabling the actual channels. However, nothing was catastrophic, there were no crashes, and I was able to set them up correctly again. I only saw this toward the beginning of the review period, so perhaps there is rechargeable battery back-up inside the instrument that needed initial charging. It will be reported to the manufacturer - it may be a software issue that can be resolved in an upcoming firmware release; the DAS240-BAT is a new instrument. In any case, for the past month I have not seen a recurrence of that.
Unlike desktop lab instruments that usually support a wordy protocol called SCPI, the DAS 240-BAT supports an industrial standard communication method called Modbus TCP. It is a protocol that is very lightweight and easy to use. Any client can query the DAS240-BAT which acts in server mode, by establishing a TCP/IP connection and sending a register read request. The DAS240-BAT responds with a few bytes for each channel that was requested. The received response could then be stored or converted into a usable value for display, logging or actioning in some other manner. Parameter names are not sent with Modbus TCP; it is up to the communicating devices to interpret values and manipulate as necessary to turn the data into a human-readable form.
To try it out, I wrote a short Modbus TCP Python library that can be used by typing import das240 into a Python program or Python interactive shell. Then, simply type a single line to retrieve a measurement from the instrument:
value = das240.read(“xx.xx.xx.xx”, chan)
where xx.xx.xx.xx is the IP address of the target device, and chan is a number between 1 and 200 for analog channels, and between 1001 and 1012 for the twelve digital channels.
The das240 python library also supports writing measurements to a file. To do that, use the das240.log command with the same parameters as above, but with a third parameter which is the filename.
Using this software, it is possible to integrate real-time measurements from the instrument into existing logging or management tools. Free examples include kst-plot which can read files and update charts in real-time as new measurements arrive.
The library can also be used to collect data and plot charts on demand using tools such as Jupyter Notebook. Jupyter is a web server that provides an interactive web page called a Jupyter Notebook, that can be edited from within the browser. Code (such as Python) snippets can run from the browser too. This means that long-running experiments can be browsed from a web page, and math formulas applied and modified.
For the example below, the DAS240-BAT was connected to a temperature sensor, and the Jupyter Notebook shown here was used to plot the measurements at any time throughout the day. This was really convenient to view from a mobile phone (albeit not a very mobile-optimised display), and I could manipulate the data, zoom in, and download it and e-mail it from the mobile too.
In summary it was quick and easy to work with the DAS240 and Modbus TCP in order to access measurements programmatically then use that data in a variety of ways.
The DAS240-BAT supports plots of multiple input channels versus a reference channel. This provides a nice way to connect the device-under-test (DUT) input and output or outputs to the instrument, and see how the outputs vary with the input. To try this out, I used a small motorized power tool and connected it to a current clamp (it outputs a voltage proportional to the current through the motor). The x-axis on the instrument would be configured for monitoring the current clamp output.
For the y-axis, I used a rotary encoder. It was attached to the shaft of the power tool. The DAS240-BAT was configured to display the encoder pulse speed in Hz (by using a formula the instrument can display the shaft speed in RPM too if desired).
It was possible to power up the motor, and see the speed and current at no-load, and also see the values change as the load on the motor was increased by applying pressure to the shaft. I didn’t test until the motor stalled for too long, because I don’t want to damage the tool.
This type of test is useful for confirming that the motor is drawing the expected amount of current at no-load and under a defined load or at stall. This information is also useful to be able to estimate the motor speed based on current consumption.
An excellent feature is the ability to upload custom graphic backgrounds for the XY chart. I created one to test my motor. It has a green outlined box where I expect to see the motor parameters under the no load condition. Attached to this review is a blank 500x500 bitmap with a 10x10 grid of faint dots, which can be adapted for any XY charts with the instrument.
The photo below shows the no-load characteristics of the motor, and the values of current and speed as a load was applied by pressing against the shaft.
Although the device natively supports thermocouples and platinum resistance probes, I wanted to use a thermistor. My use-case was to measure ambient temperature. I constructed an air temperature sensor using a so-called linear thermistor which actually consists of two bonded-together thermistors. Arranged in a circuit along with some precision resistors, the very non-linear normal thermistor characteristic becomes linearized (this isn’t strictly necessary because in fact it can be straightforward to use the non-linear characteristic directly, and convert to a temperature using a formula. However, each method has benefits and disadvantages. I went with the linear thermistor method for now). For more information, it is described here: Building a Dual Thermistor Air Temperature Sensor
The output from the circuit can be directly converted into degrees Celsius, using a y=mx+c formula. The formula for the specific circuit shown above is degC = (70.366 * V) – 34.152. It was possible to type in that formula when configuring the channel on the DAS240-BAT. Although I didn’t need very frequent measurement, I decided to sample the output at 100 times a second! I set the sample time to 10 msec, and applied a 5 Hz filter.
I logged to internal memory. I liked that when configuring the device, it clearly indicated that I would be able to record for 57 days at the chosen rate.
After about five hours I stopped the logging and made a copy of the data to a USB memory stick and uploaded it to my PC. The supplied Sefram Viewer software allows data files to be viewed and exported into several formats including comma separated values (CSV). The software was straightforward to use, and I liked it - it’s perfect for creating fast reports with annotated charts. You can also view recordings directly on the DAS240-BAT, but frankly I didn't like the implementation - it would be very difficult to go through very large record files using the touchscreen scroll and zoom tools - it's much better to either use the PC-based Viewer software, or alternatively just export to CSV and use the tools of your choice.
Since I’m more comfortable using CSV files, I exported it and then imported into Matlab. I was impressed how detailed the recorded measurements were. From here I could begin to understand how the ambient temperature had large, slow variations as the heating switched on and off, and smaller fluctuations whenever there was human activity in the room.
Zoomed in, it was nice to see how noise-free the measurements were. The vertical scale is one hundredth of a degree Celsius per division.
With this level of performance, the DAS240-BAT makes an excellent instrument for temperature logging.
The DAS240-BAT supports plots of multiple input channels versus a reference channel. This allows stimulus to a device-under-test (DUT) to change over time, and provided that stimulus is connected to the DAS240-BAT as the reference channel, the instrument will be able to plot multiple parameters that could be related to the stimulus in some way. I was more interested in automating such tests, so I decided to combine a couple of portable instruments; the DAS240-BAT, and a Hioki process calibrator (it is used to generate reference voltages or currents). The process calibrator is quite old, and relies on RS232 which is a pain to configure and operate in comparison to Ethernet of course. Anyway, I decided to check something which I’d wondered about for ages; just how closely matched are diode forward voltages matched? This has uses, because diode matching or binning is important for high-end mixer circuits, and LED matching allows for efficient parallel configurations for lighting. I connected up five diodes to the inputs, (but, due to a loop count mistake in my coding I only captured data for four of them). I placed my DAS240-BAT Python library, and a RS232 library onto my server, and then connected to it from a web browser. From the Jupyter Notebook within the browser, I typed the code to interact with the libraries and plot the charts.
It is really simple to use – just three lines of Python code (highlighted below) will connect to the DAS240-BAT, capture data from any desired channel and store it to a file. I used a loop to repeatedly dump channel data to a file, as the diode curve stimulus was adjusted (using the process calibrator). All the remainder of the code in the Jupyter Notebook screenshot below is related to controlling the process calibrator via RS232 (using a USB-RS232 adapter).
I used a Python charting library called Plotly to analyse the data. The chart below shows the characteristics for the BAT54 Schottky diodesBAT54 Schottky diodes that were tested. The results are extremely clean, no filtering was used.
Next, I tried four red LEDs (Vishay TLUR6401), and charted the characteristic up to 1 mA. Again, the results are extremely clean for the range tested. A tiny bit of filtering, or better wiring, could eliminate the very slight ripple seen on the purple trace below.
Beyond 1 mA, again the slight differences between the devices became more visible:
To conclude, I was impressed how easy it would be to capture data from lots of semiconductor devices near-simultaneously with the instrument.
The DAS240-BAT contains additional functionality (such as PWM measurement, and digital logic inputs) that this review has not covered in any detail, although it was used for some testing, for example for the motor speed tests for a brief verification that it worked. There’s a decent amount of capability with the instrument, but really it excels at one main thing, which is DC voltage measurement with charting capability. It performs that task extremely well, and it can be put to use with temperature sensors or any other commonly encountered sensor, especially because of the immense amount of voltage ranges, and the equation capability to convert the captured data into usable units. The charting capability is reasonably flexible (actually it’s pretty good, but I’m picky there – I like spending an hour just to tweak charts on a PC anyway : ) Thoughtfully by the developers, it is possible to modify colours for all 20 channels, and line thickness can be changed too.
It’s a competent instrument at what it does, and I hope the few non-critical bugs are ironed out in a later software release. The external build quality seems adequate, but really the omission of a front panel clip-on cover or similar, isn’t good – a case or bag will be essential. This is not an instrument I could throw into a vehicle without doing that.
During my time using it over the past two months, I found the ultra-low voltage ranges very useful – the data acquisition is very clean provided care is taken with wiring. Since I was working with very low-level signal differences for the fluxgate magnetic sensors, I used screened ‘starquad’ audio wire for example, and twisted wires throughout all the tests. The instrument is fast to operate – it is running within fifteen seconds from power-up, and the touchscreen controls were always responsive. For simple data acquisition it is possible to be capturing data within a minute. The user interface is straightforward, but some of the graphic icons are confusing. Some of the language could be changed slightly too – it’s a multi-lingual product. However, it’s not complicated to use, and I didn’t need to dig into the user manual much, apart from consulting connection tables for assembling the digital logic port wiring.
It would be nice to see more features, such as the ability to enter arbitrary formulas (although it is already pretty good) and to store them off. Copying settings from one channel to another would save some time too, if such a feature existed. Some usability stuff needs changing too – for instance I didn’t like how constants for equations were rounded off in the display (but the correct value is stored and used). However, right now using the user interface is fairly fast, and perhaps it benefits very few to over-complicate it (I found myself using the instrument as a handy voltmeter at times, even if I thought I didn’t need the charting capability, rather than reach for the real multimeter!). For the ‘power users’ the ability to capture data to a PC or mobile phone allows for on-the-fly complex charting and calculations if desired anyway.
I have had the instrument powered up non-stop for a few days running off the DC power brick, and the instrument gets very faintly warm, barely noticeable. In terms of practical time with battery power, after an initial full charge, I ran it only off batteries for the first week or two, and with several hours of use every few days, the battery eventually dropped to 30% after those two weeks.
Is it worth buying? As mentioned, for voltage based data recording and charting, it’s extremely competent – way better than I expected from a field instrument, and it stands up very well against typical chart recorders, and modern multimeters. As a result, If you’re having to work with sensors, such as evaluating them in the field, then I think it is a highly attractive product. For long-term measurements for electrical issues, I think the instrument is usable, but lacks some features (such as AC current and power measurement) – but it’s not intended for that use-case anyway.
Thanks for reading!
Top Comments
Hi Gene,
Thanks!
For anyone wondering why the review has no hyperlinks at all, there's some issue.. I'll raise it on the Feedback&Support, and hopefully update with links in the coming days.…
Great review Shabaz, the example use cases really demonstrated the utility of the instrument.
For someone like me who, when this roadtest was offered, didn't really understand what it could be used for, your report has explained those basics and so much more. Great graphics, clear photos and explanations…