Comparative review of portable microwave devices Arinst vs Anritsu

[Kitdze 39]

 

An independent review was conducted to test two of the compact radio frequency measuring instruments developed by the company, Kroks.  The devices consist of a spectrum analyzer with a built-in signal generator and a vector network analyzer (reflectometer). Both devices cover a wide frequency range from 35 (23 for reflectometer) MHz to 6.2 GHz.
My objective was to determine the utility value of these meters, and to better understand the manufacturer’s comment that the device is intended for amateur radio use, since it is not a professional measuring tool. The word “amateur” hints that the devices may be closer to a display meter as opposed to a comprehensive measurement device.

It should be noted that these tests were conducted as an amateur user who would lack a thorough understanding of metrological studies of measuring instruments, or the basis of the standards of the state registry and related subjects. Radio amateurs are more interested in looking at comparative measurements of devices that are often used in practice (i.e., antennas, filters, attenuators), as opposed to theoretical “abstractions” commonly explored in metrology (i.e., mismatched loads, inhomogeneous transmission lines, or short-circuit lines).

To avoid the influence of interference in the comparative measurement of antennas, an anechoic chamber or open space is required. In view of the absence of the first, measurements were taken outdoors. All antennas with directional radiation patterns “looking” into the sky were mounted on a tripod to avoid displacement in space when changing instruments.

 

 

 

Low quality, Chinese-made adapters were not used due to the frequent lack of repeatability of contact during reconnection, as well as the shedding of the subpar antioxidant coating used instead of the usual gilding ...
To obtain equal comparative conditions, the instruments were calibrated with the same set of OSL calibrators before each measurement, in the same frequency band and current temperature range. OSL is “Open”, “Short”, “Load”, which is a standard set of calibration measures: “open measure”, “short circuit measure” and “matched load 50.0 Ohm”, which are usually calibrated vector network analyzers. For the SMA format, the Anritsu 22S50 calibration kit was used and normalized in the frequency range from DC to 26.5 GHz, a link to the datasheet (49 pages):

https://www.testmart.com/webdata/mfr...COMPONENTS.pdf

For calibration of the N type format, respectively Anritsu OSLN50-1, normalized from DC to 6 GHz.

The measured resistance at the coordinated load of the calibrators was 50 ± 0.02 Ohm. The measurements were carried out with proven, precision laboratory-grade multimeters from HP and Fluke.

 

To ensure optimal accuracy and equal conditions in comparative tests, a similar bandwidth of the IF filter was installed on the devices--the narrower the band, the higher the measurement precision and the signal-to-noise ratio. The largest number of scan points (closest to 1000) was also selected.

 

A link to the illustrated, factory set of instructions, will help to better understand the functions of the reflectometer:

http://arinst.net/files/Manual_Vector_reflectometer_Arinst_VR_23-6200_ENG.pdf

 

Before each measurement, all mating surfaces in coaxial connectors (SMA, RP-SMA, N type) were carefully checked, since at frequencies above 2-3 GHz, the cleanliness and condition of the antioxidant surface of these contacts begins to have a rather noticeable effect on the measurement results and stability their repeatability.

 It is very important to keep the outer surface of the central pin clean in the coaxial connector, and the inner surface of the collet mating with it in the mating half. All the same is true for “braided” contact. Such control and necessary cleaning is usually possible under a microscope, or under a high magnification lens.

It is also important to prevent the presence of crumbled metal chips on the surface of the insulators in the mating coaxial connectors, because they begin to introduce a stray capacitance, significantly interfering with the performance and signal transmission.

An example of a typical metallic clogging of SMA connectors that are not visible to the naked eye:

 

According to the factory requirements of manufacturers of microwave coaxial connectors with a threaded type of connection, when connecting, it is NOT possible to allow the central contact of the collet entering it to be turned. To do this, it is necessary to hold the axial base of the screw-on half of the connector, allowing rotation of only the nut itself, and not the entire screw-on structure. This significantly reduces scratching and other mechanical wear of the mating surfaces, providing better contact and extending the number of switching cycles.

Unfortunately, few amateurs are aware of this procedure or the damages caused by scratching the already thin layer of the contact’s working surfaces. This is evidenced by the large number of YouTube videos from the so-called "testers" of new microwave equipment.

 

In this test review, all the numerous connections of the coaxial connectors were made strictly in compliance with the above operational requirements.
In comparative tests, several different antennas were measured to check the reflectometer readings throughout different frequency ranges.

 

                     Comparison of the 7-element Uda-Yagi antenna of the 433 MHz band (LPD)

Antennas of this type always have a rather pronounced back lobe, as well as several side lobes.

To maintain the purity of tests, all environmental conditions of immobility were especially observed--up to locking the cat in the house—while photographing the different display modes. This will assure there is no movement that ends up in the coverage area of the black lobe, thereby introducing indignation into the graph.
The figures contain photos of four modes from three devices.

The top image is from the VR 23-6200, the center from the Anritsu S361E, and the bottom from the GenCom 747A.

VSWR Charts:

Charts of the return loss:

Volpert-Smith Impedance Charts:

Phase graphs:

 

As reflected in the images, the test graphs are very similar, and the measurement values have a dispersion within 0.1% of the error.

 

                                           1.2 GHz coaxial dipole comparison

VSWR:

Return Loss:

Volpert-Smith Chart:

Phase:

Likewise, and according to all three devices, the measured resonance frequency of this antenna were within 0.07%.

 

                                              3-6 GHz horn antenna comparison

An extension cable with N-type connectors were used which slightly introduced a non-uniformity in the measurements.

The task was simply to compare the devices--not the cable or antenna. 

If there was a problem in the path, then the devices should show it accordingly.

Calibration of the measuring plane taking into account the adapter and feeder:

 

VSWR in the band from 3 to 6 GHz:

Return Loss:

Volpert-Smith Chart:

Phase graphs:

 

                                                            Comparison of the circular polarization antenna of the 5.8 GHz band

VSWR:

Return Loss:

Volpert-Smith Chart:

Phase:

 

                                 Comparative measurement of VSWR of the Chinese LPF filter 1.4 GHz
Filter appearance:

 

VSWR Charts:

 

 

                                             Comparative feeder length measurement (DTF - Distance to fault)

I decided to measure a new coaxial cable, with N type connectors:

Manually measured 3 meters 5 centimeters with a mechanical meter.
The devices showed:

Here, as they say, comments are superfluous.

 

                                        Comparison of accuracy of the built-in tracking generator
On this gif picture, 10 photos of the readings of the frequency meter CH3-54 are collected. The upper halves of the pictures -  the readings of the test VR 23-6200.

The lower halves - signals from an Anritsu reflectometer. Five frequencies were chosen for the test: 23, 50, 100, 150 and 200 MHz.

If Anritsu served the frequency with zeros in the lower digits, then compact VR served with a slight excess, numerically increasing with increasing frequency:

 

Although according to the technical characteristics of the manufacturer, this cannot be a “minus”, since it does not go beyond the declared two categories after the decimal sign.

                              Pictures collected in a gif about the internal "decoration" of the device:

 

Benefits:

The advantages of the VR 23-6200 are its low cost, portable compactness with full autonomy which does not require an external display from a computer or smartphone, and its fairly wide frequency range displayed in the marking. This is not a scalar meter, but a fully-fledged vector meter. As demonstrated from the results of comparative measurements, VR is somewhat equivalent to the larger, reputable and expensive devices. In situations that require climbing onto the roof (or mast) to clarify the condition of the feeders and antennas, its compact size makes it preferable over a large and heavier apparatus. FPV racing equipment (radio-controlled flying multicopter and airplanes, with on-board video broadcasting to glasses or displays) using the 5.8GHz frequency bands could benefit largely from this meter for selecting the optimal antenna (from a collection) or reconfiguring (straightening and/or adjusting) an existing antenna damaged following a crash. The “pocket-sized,” meter with its low end weight (which can easily hang on a thin feeder), makes the device convenient for many field applications.

 

Small cons are also noticeable:

1) The greatest operational drawback of the VR 23-6200 is the inability to quickly find the minimum or maximum markers on the chart, not to mention the search for the “delta”, or the auto-search for subsequent (or previous) minima / maxima.

The LMag and SWR modes also lack a marker management capability that is often useful. The marker must be activated in the corresponding menu and manually moved to the minimum of the curve in order to calculate the frequency and magnitude of the SWR at that point. Perhaps the manufacturer will add this function in a subsequent firmware update.
1 a)  The meter appears not know how to reassign the desired display mode for markers when switching between measurement modes.

For example, I switched from VSWR mode to LMag (Return Loss), and the markers still show the value of VSWR, while logically they should display the magnitude of the reflection modulus in dB, that is, what the currently selected graph shows.
The same is true in all other modes. In order to read the values corresponding to the selected chart in the marker table, it is necessary to manually reassign the display mode for each of the 4 markers. It seems to be a trifle, but I would like to see some “automatism”.
1 b) In the most popular VSWR measurement mode, the amplitude scale cannot be switched to a more detailed one, less than 2.0 (for example, 1.5, or 1.3).

2) There is a small feature in inconsistent calibration as if always “open”, or in “parallel” calibration.

That is, there is no sequential ability to record a calibrator measure readout, as is customary on other VNA devices. In the calibration mode the device successively prompts itself which one the (next) calibration measures should be installed and reads it out for accounting.

ARINST, at the same time, grants the right to choose all three clicks of the measure record which imposes an increased requirement of attention from the operator when carrying out the next calibration stage. The need to press a button that does not correspond to the end of the currently connected calibrator may be confusing for some users and introduces the possibility of making an error.

Perhaps in subsequent firmware upgrades, the creators of such an open "parallelism" of choice, "changes" the same in the "sequence", to exclude a possible error from the operator. After all, it is no accident that large instruments used precisely a clear sequence of actions for calibration measures to exclude such an error from confusion.

 

3) Very narrow calibration temperature range. If after calibration, Anritsu is provided with a range (for example) from +18°С to +48°С, then on Arinst it is only ±3°С from the calibration temperature, which may be small during field work (outdoors), in the sun, or in the shadows.
For example: when calibrated after lunch and working with measurements until the evening, the sun has gone, the temperature has dropped and the readings are now inaccurate.

A stop message should pop up stating the need to “recalibrate due to going beyond the temperature range of the previous calibration”. Instead, erroneous measurements begin with a biased zero, which significantly affects the measurement result.
For comparison, here's how the Anritsu reflectometer reports this:

4) The room is normal, but for an open area a very dim display.
On a sunny day, nothing is readable on the street, even if you shade the screen with your palm.
Display brightness adjustment is not provided at all.
5) I have the desire to solder the hardware buttons to others, since some do not respond immediately after being pressed.
6) The touchscreen in some places is unresponsive, while in other places too sensitive.

 

 

                                    Conclusions on the VR 23-6200 Reflectometer

 

Overlooking the few minuses when compared with other budget, portable and freely available solutions on the market (i.e., RF Explorer, N1201SA, KC901V, RigExpert, SURECOM SW-102, NanoVNA…) the Arinst VR 23-6200 appears to be the valued choice—due to the expensive pricing schemes of other devices, their non-universal and limited frequency bands, and a toy-like display screen.  Despite the modesty low price, the VR 23-6200 vector reflectometer turned out to be a surprisingly decent and portable device. If the manufacturers had modified the minuses in it and slightly expanded the lower frequency edge for short-wave radio enthusiasts, this device would occupy the global sector podium and offering affordable coverage from 2 MHz for SW (160 meters) up to 5.8 GHz for FPV (5 centimeters) --preferably without gaps in the entire strip as shown in the RF Explorer example:

We will most likely see cheaper solutions in the future to accommodate the wide range of frequencies.

At the time of this review (July-August 2019), I believe this Arinst reflectometer to be the best portable, non-expensive, and commercially available devices in the global marketplace.

[Kitdze 39]

- Part two

                      Spectrum Analyzer with Tracking Generator SSA-TG R2

The second device is no less interesting than the vector reflectometer.
It allows you to measure "through" the parameters of various microwave devices in the mode of 2-port measurements (type S21).

For example, you can test the performance and accurately measure the gain of boosters, amplifiers, or the amount of signal attenuation (loss) in attenuators, filters, coaxial cables (feeders), and other active and passive devices and modules, which cannot be done with a single-port reflectometer.

This is a full spectrum analyzer, in a very wide and continuous frequency range, which is far from common among inexpensive amateur equipment.

There is an additional built-in tracking generator of radio frequency signals (also in a wide range) and a help to the vector reflectometer and antenna meter.

This allows to check for carrier frequency deviations in the transmitters, spurious intermodulation, clippings and others.
The inclusion of a tracking generator and a spectrum analyzer when adding an external directional coupler (or bridge), makes it possible to measure the same VSWR antenna--though only in the scalar measurement mode--without taking into account the phase, as it would be on the vector one.
Link to factory instructions:
This device was mainly compared with the combined measuring complex GenCom 747A, with a limit on the upper frequency to 4 GHz. In addition, a new power meter of the precision class Anritsu MA24106A participated in the tests, with correction tables wired at the factory for the measured frequency and temperature, normalized to 6 GHz in frequency.

Own noise shelf of the spectrum analyzer, with a matched “dummy” at the input:

A minimum of -85.5 dB (with Att -20dB) was in the region of the LPD (426 MHz).
Further, with increasing frequency, the noise threshold also increases slightly, which is quite natural:
1500 MHz - 83.5 dB. 2400 MHz - 79.6 dB. At 5800 MHz - 66.5 dB.

                        Measurement of the gain of the active Wi-Fi booster, based on the XQ-02A module

 

A feature of this booster is an automatic switch that, when supplied, does not immediately keep the amplifier in the on state.

Experimentally sorting out the attenuators on a large device, we managed to locate the threshold for turning on the built-in automation.

It turns out that the booster switches to the active state and begins to amplify the transmitted signal only if it is more than minus 4 dBm (0.4 mW):

For this test, the small device simply did not have the output level of the built-in generator, which has an adjustment range documented in the technical specifications from minus 15 dBm to minus 25 dBm.  It is already minus 4 dBm which is much greater than minus 15 dBm.

Yes, it is possible to use an external amplifier but the task would be different.
With a large device, I measured the gain of the “switch on” booster which turned out to be 11 dB in accordance with the performance characteristics.
The small device then managed to find out the amount of attenuation on the “off” booster with the power supplied.

It turned out that a de-energized booster 12,000 times attenuated the transmitted signal to the antenna.

For this reason, once flying and forgetting to supply power to the external booster in a timely manner, the long-range hexacopter flying 60-70 meters stopped and switched to auto-return to the take-off point.

There was then a need to know the magnitude of the pass through attenuation of the turned off amplifier. It turned out to be about 41-42 dB.

                                                  1-3500 MHz noise generator

A simple amateur-made noise generator, Chinese production.
A linear comparison of readings in dB is somewhat inappropriate in view of the constant change in amplitude at different frequencies which is caused by the very nature of the noise.

Nevertheless, both devices managed to find very similar, comparative graphs of the frequency response:

Here the frequency range on the devices were set equally, from 35 to 4000 MHz.  

As for the amplitude, one can see quite similar values are obtained.

 

                               Feed-through frequency response (measurement S21), LPF 1.4 filter

 

This filter was discussed in the first half of the review. There it measured the VSWR. 

Here, the frequency response of the transmission is clearly visible to include what attenuation is missed, as well as where and how much is cut.
                                                                                                          

Here it is shown in more detail that both devices almost equally removed the frequency response of this filter:

At a cut-off frequency of 1400 MHz, Arinst showed an amplitude of minus 1.4 dB (blue marker Mkr 4), and GenCom minus 1.79 dB (marker M5).

 

                                                  Attenuator attenuation measurement

For comparative measurements, I chose the most accurate, branded attenuators—other than the Chinese version in view of their rather large scatter.
The frequency range remains equally from 35 to 4000 MHz. Calibration of the two port measurements were carefully performed with the obligatory control of the degree of surface cleanliness of all contacts on the mating coaxial connectors.

0 dB calibration result:

 

The center of the specified band (i.e., 2009.57) was used as the median sampling frequency.

The number of scan points was also equal, 1000 + 1 each.

One will notice that the measurement results of the same instance of the attenuator at 40 dB are only slightly different.

Arinst SSA-TG R2 showed 42.4 dB, and GenCom 40.17 dB, ceteris paribus.

 

Attenuator 30 dB

Arinst = 31.9 dB   

GenCom = 30.08 dB

A roughly similar small variation in percentage was obtained when measuring other attenuators.

But to save reader time and space in the article, they were not included in this review, since they are similar to the measurements presented above.

                          Min and max track

Despite the portability and simplicity of the device, the manufacturers have nevertheless added useful options such as displaying cumulative minimums and maximums of changing tracks--which is in demand with various settings.
Three images are displayed in a gif photo using an LPF filter of the 5.8 GHz band as the example into the connection of which switching interference and disturbances were deliberately introduced:

 

The yellow track is the current curve of the extreme sweep.  

Red track - the maximums from past sweeps collected in memory.
The dark green track (after processing and compressing the pictures is gray) - respectively, the minima of the frequency response.

 

                             Antenna VSWR Measurement

As mentioned at the beginning of the review, this device has the ability to connect an external directional coupler (Direct coupler), or a measuring bridge offered separately (but only up to 2.7 GHz). The OSL calibration is programmatically provided to indicate to the device a reference point for VSWR.

A directional coupler with phase-stable measuring feeders were shown that were already disconnected from the instrument after the VSWR measurements. 

Here it is presented in an expanded position so one can ignore the discrepancy to the apparent connection. The directional coupler is connected to the left of the device, but inverted marking back.

Thereafter, the supply of the incident wave from the generator (the upper port) and the removal of the reflected input of the analyzer (the lower port) will be correct.
The combined two photos show an example of this type of connection and removal of VSWR in the previously measured above circular polarization antenna type "Clover", in the range of 5.8 GHz.

 

 

Since the possibility of measuring VSWR is not one of the main purposes of this device, it can nonetheless be viewed from the picture of the readings of the display, while there are still reasonable questions. A hard-set and unchangeable display scale of the VSWR graph includes a large value of as much as 6 units.

Although the graph shows approximately the correct display of the VSWR curve of this antenna, it is shown with a numerical value. 

The exact value on the marker (i.e., tenths and hundredths) are not displayed. Only integer values are shown (i.e., 1, 2, 3 ...).  As such, it remains an understatement of the measurement result which is acceptable to generally understand a suitable antenna or on damage—although it may be more difficult to use for fine-tuning and antenna.

 

                          Integrated generator accuracy measurement

Similar to an VR 23-6200, two decimal places are declared in the technical parameters.

It would be naive to expect from a budget-pocket device that there is a rubidium frequency standard on board. *smiley smile*   
The inquisitive reader will nevertheless be interested in the magnitude of the error in such a tiny generator.

  But, since the attorney precision frequency counter was available only up to 250 MHz, he limits himself to viewing only 4 frequencies below the range, just to understand the error trend, if any. It should be noted that at higher frequencies, photographs were also prepared from another device.

But to save space in the article, they are not included in this review since the numerical confirmation of the same percentage is a percentage of the error in the lower digits.

 

Four photos at four frequencies were collected in a gif picture to also save space: 50.00; 100.00; 150.00 and 200.00 MHz:

 

The trend and the magnitude of the available error are clearly visible: 

50.00 MHz has a shallow excess of the generator frequency, namely 954 Hz. 

100.00 MHz, respectively, a little more, +1.79 KHz.

150.00 MHz, even more +1.97 kHz

200.00 MHz, +3.78 KHz

Further up the frequency was measured by a GenCom analyzer, which turned out to be a good frequency counter.

For example, if the generator built into GenCom missed 800 hertz at frequency of 50.00 MHz, then not only the external frequency counter showed this, but the spectrum analyzer itself measured exactly the same:

Next, one of the photos of the display, with the measured frequency of the generator built into the SSA-TG R2, using the example of the midpoint of the Wi-Fi range of 2450 MHz:

To reduce the space in the article, I also did not upload the other similar photographs of the display, instead they briefly squeezed the measurement results over the ranges above 200 MHz: 

At  frequency of 433.00 MHz, the excess was +7.92 KHz.

At a frequency of 1200.00 MHz, = +22.4 KHz.

At a frequency of 2450.00 MHz, = +42.8 KHz (in the previous photo)

At a frequency of 3999.50 MHz, = +71.6 KHz.                                           
But nevertheless, the two decimal places declared in the factory characteristics are clearly maintained across all ranges.              

 

                     
                              Signal amplitude measurement comparison                                               
In the following gif picture 6 photos were collected where the Arinst SSA-TG R2 analyzer measures its own generator at six frequencies randomly selected.

50 MHz -8,1 dBm; 200 MHz -9,0 dBm; 1000 MHz -9,6 dBm;
2500 MHz -9,1 dBm; 3999 MHz — 5,1 dBm; 5800 MHz -9,1 dBm

 

Although the declared maximum amplitude of the generator is not higher than minus 15 dBm, other values are actually visible.

To find out the reasons for such an indication of the amplitude, measurements were taken from an Arinst SSA-TG R2 generator, with an Anritsu MA24106A precision sensor with calibration zeroing at a matched load, before starting measurements. 

A frequency value was entered each time for measurement accuracy, taking into account the coefficients, according to the correction table included from the factory for frequency and temperature.

35 MHz -9.04 dBm; 200 MHz -9.12 dBm; 1000 MHz -9.06 dBm;

2500 MHz -8.96 dBm; 3999 MHz - 7.48 dBm; 5800 MHz -7.02 dBm

The amplitude of the generated signal by the SSA-TG R2 integrated generator shows the analyzer measurements are very worthy (for the amateur accuracy class).

The generator amplitude displayed at the bottom of the instrument’s display, as it turns out, is simply “drawn”. In reality it produces a higher level than it should in adjustable limits from -15 dBm to -25 dBm.

A level of doubt creped in as to whether the new Anritsu MA24106A sensor was malfunctioning until a comparison was made with another laboratory system analyzer from General Dynamics, model R2670B.

The discrepancy in amplitude was minimum and within 0.3 dBm.
The power meter on the GenCom 747A similarly showed the available excess level from the generator:
 

 

At the level of 0 dBm, the Arinst SSA-TG R2 analyzer slightly exceeded the amplitude indicators from different signal sources with 0 dBm.

At the same time, the Anritsu MA24106A sensor shows 0.01 dBm from the Anritsu ML4803A calibrator

It seems an inconvenience to adjust the attenuator attenuation on the touchscreen with your finger.

The list ribbon skips and/or returns to its extreme value. 

It seems to be more convenient and accurate to use the old-fashioned stylus for this task:

When viewing the harmonics of a low-frequency signal of 50 MHz and over practically the entire working band of the analyzer (up to 4 GHz), a certain “anomaly” was encountered at frequencies around 760 MHz:

With a wider band at the upper frequency (up to 6035 MHz), the Span turns out at exactly 6000 MHz add the anomaly is also noticeable:

At the same time, the same signal from the same built-in generator in SSA-TG R2, when applied to another device, does not have such an anomaly:

Since this anomaly was not noticed on another analyzer means that the problem is not in the generator, but in the spectrum analyzer.

 

The built-in attenuator for attenuating the amplitude of the generator, clearly attenuates in steps of 1 dB, all of its 10 steps.

Here at the bottom of the screen you can clearly see the stepped track on the timeline, showing the attenuator performance:

Leaving the generator output port and analyzer input port connected, I turned off the device.

Turning it on the next day, I found a signal with normal harmonics at an interesting frequency of 777.00 MHz:

In this case, the generator was left turned off which was confirmed in its the menu.

In theory, nothing should have appeared at the output of the generator, if on the eve it was turned off.

I had to turn it on at any frequency in the generator menu, and turn it off there.

Thereafter, the strange frequency disappears (and will no longer appear) until the next time the device is turned on.

Surely, in the subsequent firmware, the manufacturer will fix such a self-inclusion at the output of the generator while in the off position.

If the cable between the ports is missing, then it is unnoticeable that something is wrong--unless the shelf of noise is slightly higher.

And after forcibly turning the generator on and off, the noise shelf slightly becomes lower, but by an inconspicuous amount.

This is a minor operational minus, the solution of which takes an extra 3 seconds after turning on the device.

The interior of Arinst SSA-TG R2 is shown in three photos collected in gif:

Comparison of dimensions with the old Arinst SSA Pro spectrum analyzer to a smartphone display (top):

 

     Advantages: 

As in the case of the Arinst VR 23-6200 reflectometer, which was reviewed previously, the Arinst SSA-TG R2 analyze is the exact the same form factor and miniature dimensions--but it is a serious assistant for the radio amateur.  Unlike other SSA models, it does not require the use of an external display such as a computer or smartphone.

A very wide, solid and uninterrupted frequency band, from 35 to 6200 MHz.
I have not investigated the exact battery life, but the capacity of the built-in lithium battery should well accommodate a long battery life.
A rather insignificant error in measurements for an instrument of such a miniature class which is more than enough at the amateur level.
Physical hardware and firmware (and repairs if ever needed) are fully supported through the manufacturer. It is already widely available for purchase and ready to ship--as may not the case with other manufacturers.

     Cons were also seen:
Unaccounted for and not well documented is the spontaneous supply to the output of a signal generator with a frequency of 777.00 MHz.

I somewhat expect this to be eliminated in the next firmware release. 

When a user is informed of this feature, it is easily eliminated in 3 seconds by simply turning the built-in generator on and off.

The touchscreen requires a bit of getting use to since not all virtual buttons turn on immediately with a slider if you shift them.

If you immediately poke into the final positon without first moving the slider, then everything works as intended.

This is probably not a minus, but rather a “feature” of the drawn controls, specifically in the menu of the generator and the slider for controlling the attenuator.
When connected via Bluetooth, the analyzer, as it were, successfully connects to the smartphone, but the frequency response graph does not output, such as the outdated SSA Pro.

When connecting, all the requirements of the instructions were fully complied with, described in section 8 of the factory instructions.

It was thought that once the password is received, a confirmation of the connection is displayed on the smartphone screen, then this function is possible only for upgrading the firmware via the smartphone. But no. Clause 8.2.6 clearly states:

8.2.6. The device will connect to the tablet / smartphone, a signal spectrum graph and an information message about connection to the device Connected to ARINST_SSA will appear on the screen, as in Figure 28. (c)     

Manual:  http://arinst.net/files/Manual-Spectrum_Analyzer_Arinst_SSA-TG_R2-ENG.pdf

 Yes, confirmation appears, but the track doesn’t.

Repeatedly reconnected, each time the track did not appear.  And from the old SSA Pro, right instantly.

Another disadvantage of the notorious “versatility”, due to restrictions on the lower edge of the operating frequencies, is not suitable for short-wave radio amateurs.

For that, for RC FPV, completely satisfy the needs of amateurs and professional, even more than that.

    Findings:
In general, both devices left very positive impressions, since in essence they provide a complete measuring complex, at least even for the level of advanced radio enthusiasts. While pricing policy was not addressed, it is nonetheless noticeably lower than its closest analogues on the market with such a wide and continuous frequency band.
The aim of the review was simply to compare these testers with more advanced measuring equipment, and to provide readers with photo-documented displays and enough information to form their own opinions. 

This review is not intended for the purpose of marketing or advertising. 

It is simply a third-party assessment and publication of observation results.

Products webpage:  http://arinst.net/arinst_vr-23-6200.php
Sale eBay: https://www.ebay.com/usr/arinst
Sale Aliexpress: https://www.aliexpress.com/store/907723?spm=a2g0o.detail.1000061.1.2a38254cx8hk0J
Sale in Japan: http://shop-online.jp/ElectronicsDIY5/index.php?body=spec&product_id=1202524&category_id=149110&PHPSESSID=fba6b223f76a00c0c757da8f97009856
Sale in Russia: https://kroks.ru/shop/network-equipment/
Sale in China and Taiwan: http://www.ts-corp.com.tw/telecom/arinst-ssa-tg-r2

Thank you for looking.