Introduction
It seems like such a long while since I have had an insulation tester on my work bench, I need to make amends and there seems like no other better way than to introduce the MTR105 from Megger. Previous Insulation Testers Reviewed;
Keysight Insulation Multimeter Failure
Chauvin Arnoux CA6526 Insulation Tester Review
Flir IM75 Insulation Tester Review
Extech MG302 Insulation Tester Review
Di-LOG DL9307 Insulation Tester Review
RS Pro RS-9985 Insulation Tester Review
RS Pro IIT1500 Insulation Tester Review
Uni-T UT505A Insulation Tester Review
Sonel MIC-30 Insulation Tester Review
GMC MetraHit Coil Motor Testing Multimeter Review
Brymen BM877 Insulation Multimeter Review
Hioki IR5048 Insulation Tester Review
I have had the MTR105 Rotating Machine Tester around for almost a year now, so have got to know it quite well. I have to say it is not your average insulation tester, whilst it is from Megger, it is actually part of the Baker Instruments family and is directly aimed at motor testing, so testing on generators may take it out of its comfort zone. The closest instrument to the MTR105 I have reviewed so far is the MetraHit Coil from GMC and that did not go too well.
In comparison to the other instruments offered in the Baker family, the MTR105 is quite a baby and for me is much closer to an advanced insulation tester rather than the rest of the line up seen above. Based on it's size and visual style, it very closely follows the Megger MIT420/2, that I have previously reviewed.
Unboxing and Overview
The MTR105 comes in a soft carry case with a separate compartment for the meter and a larger compartment for the accessories that opens up like a book to reveal two inner pouches and velcro straps to keep the accessories together.
There is a good set of accessories supplied with the MTR105 as seen below. Three silicone test leads for 3 phase voltage measurements with corresponding probes and grippers. A set of kelvin leads for use with the 4-wire resistance measurement function, a T-type thermocouple and a velcro hanging strap. Not shown, but also arrive with the MTR105, are the quick start guide, calibration certificate and a USB stick for data transfer.
The MTR105 does have provision for a remote probe, but the probe is not included with the accessories. It can of course be purchased at extra cost, as it is the same probe as used by the MIT series of insulation testers, but it does have to be the Megger remote probe, as the jack arrangement is unique to Megger.
As time has gone on, I have purchased two of the optional extras available, the mains charger kit and duplex probe leads for 4-wire resistance measurements.
The battery charger did come with a set of six NiMH cells for use with the meter. It will plug into the jacks of the MTR105 to allow the cells to be charged without taking them out of the meter. This of course, makes the meter unusable for the charging period. The duplex leads are very nice and are the same as the kelvin leads but just have a duplex probe at each end instead of the clips. The spring-loaded pins in the duplex probes can be replaced, which is another nice touch.
The downside to these accessories, is that I struggle to fit them into the case supplied with the MTR105. Either one will fit, but both of them stretches the case a lot. I also feel that the elasticated pouches are more of a hinderance than an aid. One is useful to store the grippers and probes, but trying to get coiled leads in and out of them is awkward, a velcro strap is all that is needed for these. This I feel is a bit of an oversight on such an expensive instrument. The duplex probes are not really essential, I just like them for measuring contact resistances where I cannot gain access with the kelvin clips, for a motor the kelvin clips are fine. As you will see later on though, rechargeable batteries are pretty much an essential item for this meter.
The MTR105 is styled similar to the MIT420/2 I have. Like the MIT420/2, the output jacks are located at the top of the unit and the control of the instrument is a mixture of a rotary selection switch and push buttons just below the screen. The MTR105 is the same height and width as the MIT420/2.
When laid down on its back, the MTR105 can be seen to be slightly fatter than the MIT420/2, but it is not noticeable in use.
One of the main differences to all the other insulation testers, and certainly the MIT420/2 and U1461A, can be seen when the MTR105 is turned on to reveal a full colour display.
The colour of the display also changes as different functions are selected to match the colour marking on the selector switch. In the picture above, you can see the red display, that matches the insulation testing functions at the top of the rotary selection switch. At first, I though this was a bit gimmicky, but I have grown use to it as I have used the meter.
The leads, probes and grippers are all good quality, as would be expected of Megger. The grippers I find to be only just large enough, they max out on an M12 stud, and will not go around the nut. It also leaves them a bit small for clipping to the larger motor frames when trying to find an earth connection. Being 4mm standard safety connection, they can easily be swapped. By contrast, the Kelvin clips for the 4-wire measurement function have a much larger capacity and more friendly for larger motor testing.
| {gallery} Probe access tests |
|---|
Grippers on stud terminals |
Maximum capacity of gripper on M12 stud |
Kelvin clip capacity for 4-wire measurements |
Probes do not connect in 2.5mm Weidmuller terminals |
Probes connect in a 2.5mm SAK Terminal |
The probes are long reach when compared to many other probes supplied with other meters. They have a solid insulation conforming to GS38 requirements, there is no cap that can be removed to reduce the diameter of the probe. As a consequence, they will not connect into the smaller 2.5 mm terminals from Weidmuller. They will connect in to 4 mm terminals and above from the Weidmuller range. For the SAK range, the probes will work in all of the terminals I have available from 2.5 mm up to 35 mm.
Looking at the functionality of the MTR105, it compares well against the MIT420/2 but does lack some of the functionality of the U1461A in terms of DMM functions. Whilst the MTR105 does not have a 100 V insulation test voltage setting, it does have a variable voltage, so a 100 V test can be accommodated. The other element it lacks is the ability to read current, however, with a good voltage function, it is possible to utilise current clamps with the instrument.
The above table does not really do justice to the MTR105 and all of its capabilities. I therefore constructed a new table to break out the functions some more.
Using this method, it can be seen that the MTR105 has 5 unique functions when compared to all of the other insulation testers I have looked at. That is pretty impressive to be built into one meter. It does however, come with a hefty price tag in comparison to all of the other instruments. Based on recommended pricing, I can actually buy the both the Megger MIT420/2 and the Keysight U1461A together for the price of the MTR105.
The MTR105 has a good set of ranges for the functions. Insulation testing wise it matches the Megger MIT range and is a lot better than the majority of insulation testers out there. The addition of a 4-wire test capability gives a much lower resistance measurement range than any of the other instruments and should also offer improved accuracy. The capacitance range has been extended to a much wider range than the MIT series and is much more useful now. The addition of the temperature function provides an interesting twist in that it can be integrated into the insulation testing function, one of the unique features of the MTR105 that will be demonstrated later.
Performance Tests
Having a significant amount of measurement functions, there is a lot more to go through on the MTR105 in this section.
The insulation resistance accuracy tests show some deviation when higher resistance values are measured. In part, this has to do with the resolution of the instrument, that lacks a decimal point in comparison to the U1461A. This only appears to be part of the answer though and the unit appears to to lack some basic accuracy at the very high insulation resistance values towards the upper end of the instrument's capability.
What I find slightly peculiar, is that his is not shared by the MIT420/2. Overall this has an accuracy that matches the U1461A, so it seems strange that the MTR105 has this problem. If the test range is reduced to the limited insulation values seen on a lot of instruments, then the MTR105 fairs much better and is more in line with expectations.
When compared across all the instruments, it can be seen that the MTR105 is lagging quite behind in the accuracy table. However, at 2.25% in my opinion, this is still adequate for insulation testing on motors and windings.
The open circuit output voltage of the MTR105 is very well regulated. It is not quite as good as the Keysight unit, but is comparable with the MIT420/2 with all three units being significantly better than anything else I have tested. As previously stated, the 100V test is obtained from the variable voltage function of the MTR105.
As expected voltage regulation, tested at 500V, showed a very stable output.
The regulation seems to be slightly better than the older Megger unit, with it not dropping as much when under a greater load.
The short circuit current is very consistent and was at the bottom of the range of results in comparison to the other instruments. Given that the voltage regulation was better then the MIT420/2, I was expecting the short circuit current to be higher for the MTR105, but this was not the case.
Next, on to the 1mA load tests.
Again, very tight control was seen, with only a 10uA spread across the voltage range, better than the two main comparison units, but some of the other units I have tested have a comparable spread for the load current.
The discharge resistance of the MTR105 was quite a bit higher than a lot of the other instruments. This isn't a specific problem when testing out in the field of cables and motors. I did have an issue with it when testing my motor winding simulator, due to the capacitive nature of it. The MTR105 will switch to voltage mode if it detects that the circuit under test remains charged after the test stops. The reading can be recovered by disconnecting the meter, but obviously that still leaves a charge in the circuit.
The battery pack consists of six AA cells. A battery charge symbol is permanently displayed in the upper right hand corner of the screen. During normal use the battery symbol depletes at a consistent rate, but when connected to my power supply, I struggled to get a stable response.
The instrument cut out at 6.20V, giving a 31% usage of the battery capacity, which seems to be fairly typical for insulation testers that I have tested.
However, the speed at which the MTR105 eats into the battery is quite eye opening. The table below shows the power drawn from the battery pack for the various functions.
The meter looks to be drawing in the region of 1.2 Watts more than any of its competitors for each of the functions.
This is undoubtedly due to the colour TFT screen that has been fitted. Over the 10 months I have been testing the MTR105, I have actually worked my way through 5 sets of AA cells. I did eventually upgrade to rechargeable NiMH cells and purchased the main charger unit that plugs into the top of the MTR105 and allows the cells to be charged within the instrument. I would strongly recommend that anyone who wants to purchase this instrument goes down this route.
Winding Simulator Tests
My standard set of tests were carried out on the motor winding simulator to compare the results against the other instruments. However, the MTR105 offers a couple of twists to these tests.
Winding resistance measurements were found to be reasonably accurate in comparison to the other instruments, certainly good enough for a phase balance test. The MTR105 offers a two decimal place reading against the 3 decimal places offered by the U1461A.
However, the MTR105 offers a 4-wire resistance mode test that I have not seen before on any other insulation tester. I therefore decided to test this against a bench multimeter that offers 4-wire resistance measurement for a direct comparison. The test current for 4-wire is 200 mA on the MTR105.
Interestingly, the MTR105 did not appear to be as accurate, in terms of absolute values, when using the 4-wire resistance measurement function. The test current is quite a bit higher on the MTR105 than it is on the 8846A and I have had this cause issues on certain tests in the past. The table is also a little misleading as I have calculated the spread across all the different tests. If I look at the spread across each individual winding configuration, then the MTR105 is the more consistent meter. For further investigation, I tested the 4-wire measurement of the MTR105 against some low value resistors I have available.
This test produces a much more accurate result expected of 4-wire measurement. It also shows the excellent resolution that none of the other insulation testers have for resistance measurement. Displaying 0.53 mOhms on the MTR105 for a measured value of 0.5283 mOhms on the 8846A, is pretty impressive. Whether or not that kind of resolution / accuracy is required for motor testing may be debated by some.
The MTR105 offers both DAR and PI test ratios as do the U1461A and the MIT420/2.
Comparison of this test function against the other meters shows that there is not much to choose between them. I decided to test the extended insulation testing range of the MTR105 against some other PI simulators I have built up, that I have also developed a theoretical model for, one of which I have described in a previous blog;
5000V Polarisation Index Simulator
The first set of results tests the range of the instrument to 13 GigaOhms. At a lower test voltage, the MTR105 seems to drop off below the U1461A and the theoretical curve. Apart from one reading towards the end, the test at 500 V showed the MTR105 following the desired curve better.
The second test goes up to 40 GigaOhms. Here it can be seen that the MTR105 deviates away from both the expected and the U1461A measured curves at both test voltages.
One thing to note that is although the resistance values differ, the actual DAR and PI ratios are much closer. When observing the test, the U1461A has a better resolution than the MTR105 for resistance value readings. Above 20 GigaOhms, the MTR105 seems to step up in 5 GigaOhms steps where as the U1461A still has one decimal place before finally dropping down to 1 GigaOhm resolution. It would appear though that the MTR105 does not use the displayed reading to calculate the actual ratio, as the measured values do not calculate out correctly.
Megger do not seem to place the same necessity on actual insulation resistance values during DAR and PI testing as I do. They use the two tests predominantly to obtain the ratios. Whilst the ratios are the key to the tests, I have always been taught to record insulation resistance values as well, so this performance is a little disappointing from such a high end instrument as the MTR105. What I also find seriously disappointing is the the MTR105 only records the values at the ratio time stamps and does not record intermittent values to obtain a curve. It was also noted that the MTR105 does not record the DAR value when carrying out a PI test, this has to be recorded manually or the tests carried out separately to obtain both ratios within the instrument. Whilst not an issue per se, care needs to be taken when conducting multiple insulation resistance tests on a winding, as a charge can build up and lead to false readings.
Besides the 4-wire resistance capability, the MTR105 offers enhanced features for insulation testing. The first is the guard terminal, that I have seen on a few other testers, but it is quite rare to find it on 1 kV insulation testers. Guard terminal use tends to be used when testing above 1 kV.
The picture above shows one method of using the guard terminal to remove the surface leakage over a bushing and allow the instrument to measure the true reading of the circuit to earth. The blue lead is the guard terminal and is reading the leakage current over the bushing via the copper braid collar wrapped around it.
Another unique test the MTR105 offers is an automated 3 phase insulation test that can be used on a motor with an open winding configuration. It is interesting, but I was a little disappointed with this test. When I test an open winding, I have always shorted the two windings not on test and then shorted them to earth.This way, the total leakage from the winding under test to earth and into the other two windings is measured. The MTR105 does not do this though. It merely carries out a timed test between two phases whilst leaving the third phase open. The other thing I did note is that this test always ran for 1 minute and could not be changed in the settings. This is just a 1 minute timed test, no DAR ratio is displayed, nor is the 30 second value recorded to allow for manual calculation.
The pictures show the 3 phase test being carried out on a 53 MOhm star configured resistor bank. During the test, the meter displays a count down timer and not the actual insulation resistance value. Just below this the display has three blocks identifying which two phases are under test. The blocks fill out as the meter progresses through the test sequence. In the grey bar at the bottom, the measured test voltage and leakage current are displayed.
At the end of the test, a combined results display is presented showing the resistance values, test voltage and leakage current for each of the three tests.
The final enhancement to the insulation testing function is a temperature compensated test. I have only seen this feature offered on very high end motor analysers, such as a Baker Surge Tester, such a feature is not present on any of the higher voltage insulation testers I have used either. Insulation systems on motors generally have a negative temperature coefficient. So when an insulation resistance is measured with the motor offline it will generally be cooler, and a higher insulation value will be obtained than what it would be when the motor is in operation.
To use the feature, two parameters in the settings menu need to be set and a temperature reading needs to be taken and entered either manually or via the memory function. The first setting is the class, which is set as either A or B. Class A is for older asphaltic insulation based systems and Class B is for the newer mica based systems. The second setting is to set the desire reference temperature the MTR105 will compensate the reading to. The IEEE 43 standard specifies a correction to a temperature of 40 Degrees Centigrade, but the MTR105 allows for a much wider selection if desired.
The slightly issue I have with this set up is that the information on this is not contained within the manual for the instrument. The information on the settings format was obtained from Megger who also sent a further document to me on using the MTR105, but I could not find this document anywhere on their website to download freely. I have been given permission from Megger to distribute the document, so have attached it to the blog. However, to me, the manual needs to contain relatively basics on how the meter operates such as these settings.
I found that the temperature compensation test only works with the 'spot mode' of the insulation testing. It will not work with 'timed', 'DAR' or PI test modes. Whilst not specifically a problem, it does increase the amount of testing that I have to do. I will always test a motor winding for 1 minute, as per the IEC standard. So to get a 1 minute temperature compensated test, I have to manually lock the MTR105 and time the duration myself. If I then wanted to get a DAR value recorded, I would have to switch to that function and repeat the test. Having the temperature compensation function enabled for timed and / or DAR tests would allow me to complete the insulation testing with just the one test and record all the data I want.
To activate the temperature compensated test, the temperature of the winding has to be entered into the instrument. The specific procedure for this is not in the manual, but I found two ways to do this. Switching the meter to the temperature function, the winding temperature can be measured with the probe provided. When a reading is captured, it can be saved to the relevant memory slot and then when turning the function back to the insulation test, the temperature is transferred over and displayed. The second way is to manually enter the temperature on the temperature function, available when the probe is not plugged into the instrument. Using this method, the temperature reading does not have to be saved to a memory slot to transfer it to the insulation testing function. Without a temperature measurement, the MTR105 will not allow the test to be started.
When the temperature compensated test is completed, the display shows the temperature compensated value, the motor class and the temperature the reading is compensated to on the main screen. Just below this, the non-compensated reading and actual temperature of the winding is displayed in the grey bar, which I find to be a nice feature.
The MTR105 also offers capacitance and inductance measurements, so I tested these out against some decade boxes as inductance and capacitance measurements do not work so well on the winding simulator.
The LCR mode offers an automatic mode, but can also be switched to manual inductance or capacitance modes. There are only two test frequencies available on the instrument, 120 Hz and 1 kHz. Whilst this appears to be quite limiting against a standard LCR meter, given that the MTR105 is specifically aimed at motor testing, I find the settings to be perfectly acceptable.
Capacitance wise, I found the the MTR105 lacked accuracy below 10 nF and above 100 uF, especially in comparison to the Keysight U1461A, although their meter also suffers down in the low nF range.
In inductance mode the MTR105 is on its own, no other insulation tester I have in my arsenal has this capability. Accuracy seems to drop off below 1 mH, but above that, up to its upper limit of 10 H, the accuracy was perfectly acceptable.
Measurement of the winding inductance is an alternative to the coil winding test function of the MetraHit Coil, I have previously tested. An inductance test isn't as effective as this type of surge test is at finding inter-turn coil faults. However, whilst I have found the coil test function of the MetraHit to be unstable, the inductance test function of the MTR105 worked on all the motors I tested it on, and in all the different winding configurations.
Motor Tests
I will demonstrate two motor tests here. The first is the usual 90kW motor I have available for testing in the motor stores.
There weren't too many issues with the tests using the MTR105. Insulation readings were obtained easily. The grippers were large enough to go around the motor terminal studs, but they would not have fitted around the nuts. Luckily enough an earth stud is present within the terminal box, so I did not have to find somewhere on the motor case to connect to, otherwise, again, on this size of motor, the grippers would not have been large enough.
On the 90 kW motor, the three phase insulation resistance test produced a good set of results and worked very well. For a motor with open winding configuration it is a little bit of a time saver.
Insulation tests to earth were conducted in the usual manner, without using the guard terminal.
I measured the motor winding resistance with the 4 wire methods, as it is quite a large motor and hence has a low DC resistance. I also took the liberty of measuring the inductance to see how well this works on an actual motor.
The clear advantage of the 4-wire technology for the resistance measurement can clearly be seen. All the other insulation testers I have used on this motor cannot discriminate between the different wiring configurations where as the MTR105 clearly shows the difference. Working to a maximum 5% deviation between the phase balance, much more certainty can be gained when using the MTR105 than any of the other instruments. The down side to this, is that the test has to be conducted at the motor terminals. Generally, when working in maintenance, all my tests are carried out from the motor control panel, so the cable is included in the measurements. This adds in extra resistance that makes the 4-wire test incompatible.
For the inductance test, I found I could get a reading on all three motor winding configurations. The closest I have to this is the coil test on the MetraHit Coil. This could not give a reading star configuration where as the MTR105 was successful. In fact the MTR105 was successful in capturing inductance readings from all of the motors I tested, proving to be much more reliable than the MetraHit Coil.
In practice the MTR105 offers uni and bi-directional testing of the winding resistance, recording the reading in both directions and then showing an average on the main reading when using bi-directional testing. The MTR105 can be set up to test manually using the test button or automatically that detects the lead connections and carries out the test the moment a circuit is detected. This latter method is very useful when utilising the duple probe test leads.
The second motor test is in video format and is on a small motor I have set up to be fed from an inverter. The video is just over 15 minutes long and goes through the complete suite of tests the MTR105 has to offer, so is much more realistic.
Unfortunately, at the end of the video the 3 phase voltage measurement does not appear to work on the inverter. This has been tracked down to poor filtration of the inverter output. Assembling a crude sinusoidal filter on the output of the inverter and the MTR105 started to pick up voltage readings. I then tried to create my own filter for use just on the meter. I struggled with getting hold of a good combination of capacitor and inductors with the right voltage measurement that would fit into a little box suitable to plug into the top of the MTR105, so this did not work quite as well as I wanted.
| {gallery} Inverter Output Voltage Measurement |
|---|
Unfiltered output of inverter captured on an oscilloscope |
Filter added to output of inverter |
Erratic voltage measurement of inverter output |
Bespoke filter made to plug in to MTR105 |
Inside bespoke filter |
Inverter output voltage measurement with filter |
Whilst the filter made to run the motor produced better results, the plugin filter did work to a certain extent but was more erratic. With the inverter at full 50 Hz output, the reading was quite stable and the phase rotation was also displayed. At lower frequencies, some erratic behaviour was observed. If a more industrial inverter was being tested that had a built in sinusoidal filter on the output, then I think the MTR105 would work just fine. These kind of filters are quite rare in my experience, usually the inverters have a line filter installed to stop harmonics getting back into the supply, and leave the motor to fend for itself. These line filters will not help out the MTR105.
However, the MTR105 does have a true RMS function on the single phase voltage measurement, so this option can always be used to get a voltage value, but obviously phase rotation of the inverter output could not be obtained. The three phase function does work fine on a standard mains supply, typical for a direct on line motor starter.
The MTR105 does not have a a direct current measurement function, but using a clamp adapter a current reading can be obtained. This works well for single phase mode and seems to be accurate enough, as seen below. I also tested in three phase mode, but the MTR105 does not appear to have mV resolution in three phase mode and so a reading could not be obtained. This was a little disappointing as current balance is something that I would measure for a motor more than voltage balance.
It would be great if Megger could resolve these couple of issues with the voltage function. Even better if they could add in a clamp mode for the meter that automatically converts the voltage from the current clamp to a current reading. Doing this for the three phase function would make the instrument even more unique.
Software Functionality
The MTR105 has onboard memory that test results can be saved to as they are completed. This memory function works differently to any of the other meters I have reviewed. Before being able to save results, a 'memory slot' must be created on the meter. Tests can then be saved to the pre-selected memory slot with the push of a single button. The memory slot is named via a keypad screen that is worked through using cursor and selection keys, and I found this a bit laborious to carry out. Once entered though, the slot stays allocated until it is deleted.
The software is a bit of a dark secret with the MTR105. There is no mention of the software within the manual supplied with the instrument, it just explains how to save the data to the provided USB stick. In the brochure there is a single line that details being able to download the instrument to the PowerDB Software. I had to upgrade my version of the software to achieve this functionality. Once loaded, the MTR105 appears in the menu and results can be extracted from the USB stick into a single test sheet. The test results are exported as text files for each individual test into a folder that carries the same name as the memory slot, making it easy to keep track of the results.
The test sheet can be modified in some areas to make it more individual. Company logos can be changed and the location and details of the item under test altered to suit. I could not alter the name of an individual test function, which was a little disappointing for me. I would have liked to have given U, V and W phase names to the various tests to identify which winding was being tested, as it is, the client receiving the report would have to assume that the tests were in a U, V, W order.
What is nice is that different results from previous tests can be loaded into the test sheet. In this manner a trend of test results over a period of time can be displayed on the one test sheet, to show how a plant item has, or has not, deteriorated from test to test.
In the example of a test sheet from the PowerDB software above, the results of testing the 90 kW motor are displayed. I am not sure why, but the Ohm symbol has been replaced by the ampersand symbol when the test sheet was printed to PDF. I am not sure if this is a software error or an issue with my particular computer.
It is also worth noting that the results are actually saved as text files, therefore is is perfectly possible to pull all the data into Excel and produce your own test sheet, as shown below for one of the PI Simulators. Just a little more time consuming to do.
Build Quality
On the exterior, the MTR105 is typical of the build quality of an instrument from Megger. The case has a very solid feel to it, with the rubber coating moulded onto the plastic. The rotary and pushbutton switches have a positive feel to them. The unit does feel to be a little top heavy in the screen area. When on its tilt stand, pressing the pushbuttons below the screen makes the unit tilt backwards. It hasn't fallen over yet, but I imagine if you hit one of the buttons a little to hard, it would fall over.
The battery and fuses are contained within their own compartments and can be swapped out without accessing the electronics of the instrument and affecting a calibration certification. Both of the compartment lids have a rubberised seal around them to prevent dirt ingress and have captivated machine screws, with corresponding threaded brass inserts in the case, to hold the lids down.
The two halves of the case are held together by five self tapping screws, one of which is hidden behind the bottom battery cell in the battery compartment. The lower half of the case has a ledge with a lip seal all the way around, again preventing dirt ingress into the meter. This makes prising the two parts of the case apart a little difficult, indicating how well the seal works.
Once apart, a single screw can be undone to remove the complete multi-tier assembly of PCBs away from the instrument case. The battery pack is connected to the PCB via spring loaded contacts, so there is no wiring present.
The input jacks of the instrument feed into a separate board which also contains the input protection, two 500 mA HRC fuses, gas discharge tubes and varistors. The jacks are soldered directly onto the PCB, but the housing is a solid fit into the case and takes the stresses of inserting and removing the leads.
These two smaller PCBs and jack housing are connected to the two main PCBs via headers and some PCB stand-offs and can therefore be split away from them to reveal a set of miniature relays. These relays appear to be selecting the jacks for the various different tests when using the meter. They can be heard to operate whilst tests are undertaken.
The relays are Axicom and I have seen them in use in a few other instruments. They are not carrying out rapid switching and also not switching much current, so I presume that they will have a reasonable life span. Being on a separate board, it looks like it could be replaced relatively easily if there are issues.
Removing the smaller PCB stack also reveals one of the main MPUs for the MTR105, an ST Micros STM32F767. A micro SD card was also observed. The firmware can be upgraded via the setting menu and a USB stick without opening up the case, so I am not sure if this does contain the firmware or if it is just acting as memory for the test data to be saved to.
A second MPU is located on the larger PCB and is also an ST Micros Processor, this time an STM32L476.
On the other board is the unique high voltage transformer arrangement employed by Megger. I have seen this arrangement within their MIT series of instruments. It appears to have the actual windings created within the PCB with the core of the transformer clamped through the PCB. This is a much sturdier design than the more traditional transformer soldered onto the PCB.
This board also contains the selector switch on the opposite side, which is a totally enclosed design, unlike a lot of other insulation testers.
All the boards very found to be clean, with no evidence of solder residue. All the components were soldered direct onto the PCBs and there were no signs of post production modifications or repairs.
Conclusions
The MTR105 Rotating Machine Tester from Megger has taken me a lot of time to work through, it is a very respectable attempt at a one stop instrument for motor testing. The instrument has 25 functions available for motor testing, five of which are unique to this instrument in comparison to all the other insulation testers I have reviewed. To give you an idea, below is a picture of the MTR105 against the host of instruments I would need to carry around with me to match its functionality.
Even with this set up, I still do not have a temperature compensated insulation test function and whilst the Sonel insulation tester has a guard terminal, it does not have a variable test voltage function that the MTR105 posseses. The huge issue though is the price tag that the MTR105 has pinned to it. The test setup I have pictured comes in at around £1,650 compared to the £2,040 for the MTR105.
Sonel MIC-30 - £708
Fluke 9062 - £225
Fluke 971 - £225
Uni-T UT233 - £180
RS Pro DT-5302 - £106
Voltcraft LCR-300 - £200
There are some further savings that could be made by selecting cheaper alternatives to the Fluke phase rotation and temperature probes. There are a couple of advantages that the multi-instrument setup has, the UT233 can also measure current, power and phase angle. The 971 also gives a humidity reading, which also affects insulation test results. The DT-5302 has a mA scale, useful for measuring 4-20mA transducer output signals.
This isn't just about the amount of instrumentation needed, there is work efficiency also at stake. Powering up different instruments and swapping leads around all add to an increase in test time that can be reduced using the MTR105. I do feel however, that there could be improvements in the MTR105 functionality to improve work efficiency even more.
Having to carry out multiple tests for DAR / PI ratios and temperature compensated tests could be reduced by making the temperature compensation function available for all of the insulation testing functions. Saving DAR data during a PI test could also be added. Not saving the plot data for a PI test I find really frustrating, especially as the MTR105 is such a high end instrument. The three phase insulation test is a nice addition, however, adding in the fourth terminal for an earth connection would reduce the lead swapping and test time even more. It would also have been great if Megger could have added in a three phase resistance and inductance measurement function to compliment the testing.
The three phase voltage function needs to be looked at. Inverters are more popular in industry now and I find it surprising that it did not function correctly. There are work arounds, that I no doubt could improve on over time, but I don't expect to have to do this on such a high-priced, top-end instrument. A current measurement function using clamps would be a great enhancement for the unit, especially to measure three phase current balance. It would also give Megger a possibility of a further sales option for a current clamp set.
Built quality is up to Megger's usual high standards, along with the built-in safety functions. Whilst the grippers are a bit small, the accessories in general are all good quality. The case could do with enlarging, whilst I can just about squeeze everything in, I have a feeling the case will either rip or the zip break at some point in the future. Getting accessories in and out of a packed case, can be frustrating. Personally, I find the unit much more easier to use than the multiple menu driven options of instruments like the U1461A. Whilst the memory function is awkward to set up, I really like this approach when it comes back to downloading the results, as the motor under test is immediately obvious. The software works well and is quick and simple to use. It would be nice to be able to have a bit more control over the test titles so that there were more representative of the actual test.
For me, I think the MTR105 would be a great enhancement for a motor manufacturer or a rewind shop. Being able to carry out a suite of tests and provide a test certificate would make for a very professional approach. They will also have the amount of testing requirements that justifies the high cost of the instrument. I don't think the MTR105 is quite there for use as a site maintenance tool, motor testing is just part of a site maintenance routine and predominately is done from the motor starter which negates some of the test capabilities of the MTR105. Sadly, as it does not record PI data, it isn't what I would take up to a generator rotor for insulation testing.
What intrigues me more, is the potential reaction from other instrument manufacturers and whether to not the likes of Chauvin Arnoux, Sonel, Fluke or Hioki come out with motor testing meters to match the MTR105 and the MetraHit Coil.
 | Excel MTR105 Test Comparisons.pdf |
 | GuideTo Motor Testing_en_V0L.pdf |
Top Comments