Chauvin Arnoux CA6526 Insulation Tester Review

Introduction

The CA6526 is the first insulation tester to be added to my portfolio for my grand scheme of testing it against the Keysight  U1461AU1461A that has failed me and the Megger  MIT420/2MIT420/2 that I road tested for element14 last year

Keysight U1461A failure

Megger MIT420 road test

As a comparison, the CA6526 falls in-between the Keysight and the Megger units. It offers similar functionality as the Megger unit and is primarily an insulation tester and not an insulation multimeter like the Keysight. Whilst the Megger unit lacked the download capability I desired, the CA6526 comes with a built in blue tooth function and Microsoft Windows compatible software, so I have high hopes for this unit. Unfortunately, the inclusion of blue tooth connection has come at the cost of a variable test voltage function, that was included with the Megger unit.

The CA6526 does not have all of the functions that come with the Keysight unit. I have used software from Chauvin Arnoux for some of their other instruments and I have found it to be a bit cumbersome and basic. It does not offer an iPad module that the Keysight offers so already has some loss of usability in comparison.

In terms of price, the CA6526, sits between the Megger and the Keysight units. It is significantly more expensive than the Megger unit, that is exacerbated by the issue that the remote control probe for it must be purchased separately. Both the Keysight and the Megger come with a remote control probe included in the kit.

Below is a summary of the functions of the three units;

The technical capabilities of those functions can be seen below;

It is clear that the CA6526 is aimed more towards competing with the Megger unit rather then the Keysight unit, which for the purposes of these charts has had the multimeter functions ignored.

Package Contents

The CA6526 arrives in a cardboard box that contains the meter itself, within a soft carry case. A pair of leads, a red crocodile clip and a black probe, set of six AA cells to power the meter. A quick start guide, safety instruction sheet, calibration certificate and two mini CD ROMS. One containing the user manual in a different languages and the other, the software for connecting the CA6526 to a Windows based computer.

I am not a big fan of the case supplied with the meter. The main problem is that the case has a cutout for accessing the meter terminals to allow the meter to be used whilst still in the case. However, this remains open with the case zipped up and would allow dirt into the meter terminals whilst not in use. More minor issues with the case are the restraint strap that sits over the function buttons, the  neck strap arrangement that allows the meter to be used hands free, but has to be unclipped every time to close the case and the accessory storage that whilst being a separated compartment has poor access that means all the accessories have to be tipped out to get to them.

Whilst the CA6526 does have the facility of a remote control probe, the probe has to be purchased as an extra, adding further cost to the instrument setup. The remote probe has a different connection arrangement to the other two units, so the remote probes are not interchangeable between them. The remote probe for the CA6526 is slightly different in that it has its own PP3 battery to power a work light within the probe. The probe will still control the meter without this battery installed. It is also unusual, in that the remote probe plugs into the negative terminal of the meter and not the positive, as with the other two meters.

This arrangement is also followed with the supply of a crocodile clip for the positive lead and a probe for the negative lead, encouraging the user to clip the positive to the earth of the circuit under test and the negative to the circuit live. This test methodology is old school, and used primarily for step tests which the CA6526 is not capable of carrying out. I also find it hard to believe that only one crocodile clip is provided for a meter that is designed to carry out a 10 minute timed test. Do Chauvin Arnoux really think that someone will hold a probe onto a circuit under test for 10 minutes? It may not cost much to obtain a set of crocodile clips, but given the cost of this unit and the functions it offers, a second crocodile clip really should be included.

Teardown

A brief look inside the unit was carried out to ascertain the quality of the build.

The outside case has a slight rubberised feel to it, the meter does not come in a detachable holster. The battery pack is housed behind a separate compartment with a cover that contains the tilt stand and some magnets that allow the meter to be calmed against a metal panel and used hands free. I really like that option. The battery cover is held in place with a plastic quarter turn screw. The battery arrangement is slightly unusual, having three cells down each side in a tube arrangement. This gives access to a small clip on cover that reveals an pin header, presumably for connecting to during the calibration procedure.

The outer case is held together with four torx head screws. The case can then be pulled in half to reveal an inner black case that houses the battery connections. The battery connections have a plug and socket arrangement onto one of the PCBs that allows the two case halves to be separated completely.

The meter terminals sit within their own plastic housing that is then connected to the PCB via internal wiring preventing any damage to the PCB from plugging in and unplugging the leads. The output circuit looks to have MOVs for over voltage protection, there does not appear to be any internal fusing, which is not unusual for an insulation tester.

A small modification looks to have been carried out, with a diode soldered across the legs of a transistor. Other than that, the board layout looked to be clear. Function selection is varied outbid a rotary switch with spring contacts connecting onto the switch pads of the PCB tracks. The function and main test button are also PCB track style, with the carbon contact housed within a rubber moulding.

Basic Calibration and Function Tests

A series of tests were conducted to look at the basic performance of the unit in terms of insulation testing.

Looking at the basic test results shows that the open circuit voltage of the CA6526 is much worse than both of the other two units. The specification for the the CA6526 is -0% to +20%, so is within the manufacturer's tolerances, but is just not as accurate as the two competitors. In terms of short circuit output current, this appears to be slightly lower for the CA6526 than the other two instruments, but is in accordance with the requirements of the IEC standard. The 1mA load capability is also IEC compatible for all three units.

The difference on the voltage regulation between the three units is quite clearly seen in the data table and the corresponding plot. With full load applied (33kOhm) the CA6526 has the lowest output voltage that quickly rises as the load resistance is increased.

The CA6526, does not have live battery indication on the screen. Battery level is displayed as the meter is switched on. This was tested by setting the battery voltage at different levels using a DC power supply and recording the percentage level as the unit was switched on. Low battery is achieved at 6.64V and the meter turns itself off automatically. Power consumption from the battery was measured for different functions. The most load is drawn during a low ohms continuity test that gives a 200mA test current, an insulation test with a 1mA test current draws about half the power of the continuity test. All other functions drew between 0.3W and 0.6W. Use of the backlight adds about 0.3W of load.

A bench test of the insulation resistance measurements was carried out and shows that all of the readings are well within the manufacturer's tolerances for all of the ranges Plotting the deviation from the insulation values produces some interesting results There is little correlation of the insulation values across the test voltage ranges of the CA6526 The  MIT420/2MIT420/2 produced much more consistent results Comparing the insulation values between the two instruments at the 500V test voltage shows the flatter response of the Megger unit in comparison to the CA6526 It must be noted that the Megger likely benefits from its lower resolution at the higher insulation values giving the appearance of a better performance

The instrument is easy to use and accessing the various functions relatively obvious. It does have a slight quirk in that at the end of a timed insulation resistance measurement, the instrument automatically goes into the 'hold' function. This allows the arrow key to be used to cycle through the voltage, current and resistance parameters for the reading. To take another measurement, the instrument has to be taken out of the 'hold' function, I don't find this as intuitive as the other units on offer.

I did note that only the PI or DAR value is saved to the local memory and not all of the other parameters. The Megger unit also allows the voltage and resistance values to be read back from a saved record to memory.

The display is clear and easy to read, a back light facility improves the screen contrast when viewing in darkened areas. However, it does not have the clarity offered by the Megger unit. The display on the megger also uses larger digits for the main reading which improves the clarity compared to the CA6526.

Software Function

The final element was to test the software. This utilises bluetooth functionality to communicate between the computer and the instrument. The software offers either a download function for test data saved on the instrument directly or live data collection where the instrument is controlled by the software and the data is passed directly to the computer. This latter method provides more test information, but has the disadvantage of requiring the computer to be at the test location. Any test function of the instrument that can be saved to the local memory can be downloaded by the software but only insulation tests can be performed when the instrument is connected to the computer and controlled by the software.

To connect the using the bluetooth communication, the instrument is switched over to the bluetooth / memory function and the bluetooth enabled using the pushbutton. The function knob is then turned back to the desired test function but will retain the bluetooth connection.

The instrument comes with the basic Megohmmeter software, that allows the instruments memory to be downloaded and tests to be conducted. Each set of tests is saved to its own unique file. Basic data can be exported to a CSV file for each test conducted. Chauvin Arnoux also sell their Dataview software that offers extra functionality and allows different readings to be compared to one another and full PI curve data to be exported to a CSV file. This offers useful functionality for trending test results over a longer period of time, rather than the spot tests that the Megohmmeter software is setup to record.

Both software packages save the tests attached to an object. The objects and tests can be renamed to something more suitable if required as demonstrated in the first screenshot below.

I did find that when the instrument was used to carry out a PI test on its own, only the PI data was saved. However, when the PI test was conducted using either the Megohmmeter or Dataview software, the DAR test ratio was also recorded, saving on the amount of tests that would need to be carried out or obtained by manually recording the test data. The Dataview software collects a total of 60 readings across a 10 minute PI test, which is more than enough to produce the PI curve plot.

Conclusions

The Chauvin Arnoux CA6526 is a very capable insulation tester with good build quality and functionality on offer. It does provide good record saving functions but to get the most out of it, the DataView software really needs to be purchased and this puts the complete kit in the price bracket of the Keysight instrument. For my needs, the instrument needs to have a variable test voltage function for the insulation testing. Chauvin Arnoux does offer a unit with a variable test voltage, but this comes in at twice the price of the unit tested.

I will put the CA6526 to one side for a while whilst I look at some other insulation testers, but it will definitely Uber a unit that I will take up to see how it performs testing the generator rotor.