Test/review of Battery tester ZTS Mini-MBT

[SIZE=+3]Battery tester ZTS Mini-MBT[/SIZE]



The ZTS tester is fairly advanced battery testers that uses some sort of pulse test. This is one of the smaller models. I will see how good it is, how it do its magic and how it is build.



This model includes a small bag for it.



The front has one connection for a couple of battery sizes.




The other connection is hidden in a slot on the device (A neat solution).



There is also a power button on the box, a press will turn the device on for some time, power off is always automatic.



When a battery is connected the lights run left and right a couple of times, before a single led lights up showing the estimated remaining energy.




[SIZE=+3]Measurements[/SIZE]

• Power consumption when on and idle is 17.5mA
• Power consumption when off is below 1uA
• Power consumption when testing is 31mA
• Power consumption when showing result is 78mA (The relay is using a lot of power).
• Power on time is about 2 minutes
• Works down to about 4 volt on the internal batteries.
• The power led will flash when internal batteryes are too low.

To test the tester I discharged a AA alkaline battery in 100mAh steps and put it on the tester for each step (I discharged with 100mA).
Each time I checked the state on the ZTS I did multiple tests, if the tester changed between two values I recorded a value between, i.e. a indication of both 60% and 80% was recorded as 70%.
The dots above shows the ZTS indications, the green line is the correct remaning capacity.
This curve was done over two days, this means the battery got a lot of rest time at some of the measurement points.



I did the same test with a NiMN, (2000mAh Eneloop discharged with 500mA).
As expected it is very difficult to measure charge state of NiMH batteries.



The pulse loading is one 2 second pulse at around 150mA for alkaline batteries.



With NiMH the current is 350mA and the pulse is slightly longer.



LiIon batteries are pulsed with 700mA
Because there are different LiIon chemistries with different voltages, this cannot be very precise.



And CR123 is pulsed with 750mA



[SIZE=+3]Tear down[/SIZE]



It was easy to open, I just had to remove a couple of screws, but removing the circuit board required desoldering the two connections to the battery box.



The 5 level leds (D1 to D5) are smd leds and marked with the color. There is also a power led (D6).
Q1 is used to drive the relay and Q3 is probably used together with the power switch.
Z1 may be the reference.
It can also be seen that the circuit board can be used for another MBT model (There are holes for some other battery terminals).



On this side is the microprocessor (U1: PIC16F1507), the load resistors and a relay to disconnect the loads.







[SIZE=+3]Conclusion[/SIZE]

The device is not a precision meter, but it can give an idea about the power remaning in a battery
Compared to many other battery testers it has the advantage of a well controlled load, both current and time.



[SIZE=+3]Notes[/SIZE]

I got the device from a reader.

Here is some voltage tables for estimating remaning energy in LiIon batteries

HKJ said:

[...] LiIon batteries are pulsed with 700mA Because there are different LiIon chemistries with different voltages, this cannot be very precise.

Click to expand...

Indeed, it will surely be way off in many cases. Why did you skip graphing this case?

It is not clear how to interpret the graphs. What is the meaning of the Y-coordinate (voltage) of the ZTS dots? Is it the voltage under load during its test, or before the test? Does each dot represent one test or the average of many? How much rest did the cells have after your discharge before being tested on the ZTS?

Finally, what percentage range does ZTS claim for each color? (the graphs are meaningless without knowing that)

Thanks for the test. I've been interested in seeing this!
With Alkalines its fairly accurate between 0-60%
NiMH isnt very accurate at all. When showing 80%, it can even be 40% actual charge left. Thats a big difference.

Gauss163 said:

Indeed, it will surely be way off in many cases. Why did you skip graphing this case?

Click to expand...

Two reasons:
1) Time, I cannot automate this, but need to check the batteries many times over 1 or 2 days.
2) I would need to check at least two types of LiIon to get useful charts

Gauss163 said:

It is not clear how to interpret the graphs. What is the meaning of the Y-coordinate (voltage) of the ZTS dots? Is it the voltage under load during its test, or before the test? Does each dot represent one test or the average of many? How much rest did the cells have after your discharge before being tested on the ZTS?

Click to expand...

The voltage is used for the red line, the ZTS dots are related to the green percent scale, like the green line.
As I wrote each dot is based on a number of tests, the resting period is unspecified (minutes to 10 hours for the alkaline).

Gauss163 said:

Finally, what percentage range does ZTS claim for each color? (the graphs are meaningless without knowing that)

Click to expand...

I did include this image:

HKJ said:

The voltage is used for the red line, the ZTS dots are related to the green percent scale, like the green line.

Click to expand...

So the y-coordinate of the colored dot indicates the percentage reported by the ZTS (average over some number of tests)?

HKJ said:

I did include this image:

Click to expand...

Your graphs show additional colors (e.g. red) There is not enough info to determine the ranges that ZTS claims for each color. In order to judge the accuracy of the meter the reader needs to know the claimed ranges for each color, e.g. green = 80-100%, orange = 80-60%, red = ... or something similar. Keep in mind that the reader may have neither the device or its instructions,

Gauss163 said:

So the y-coordinate of the colored dot indicates the percentage reported by the ZTS (average over some number of tests)?

Yes.


Your graphs show additional colors (e.g. red) There is not enough info to determine the ranges that ZTS claims for each color. In order to judge the accuracy of the meter the reader needs to know the claimed ranges for each color, e.g. green = 80-100%, orange = 80-60%, red = ... or something similar.

Click to expand...

You can just ignore the color, it is only included to make the chart look pretty and match the display on the MBT, but do not affect the percent.

HKJ said:

You can just ignore the color, it is only included to make the chart look pretty and match the display on the MBT, but do not affect the percent.

Click to expand...

My point is that the reader needs to know further info such as the resolution of the LED meter. Can it only report values that are multiples of 20%, or do they use some scheme with colors and/or blinking LEDs to report multiples of 10%, or some similar interpolation scheme? We cannot accurately calculate the error of the device without knowing these details (normally one graphs ther error since it gives at-a-glance some intuition on the accuracy and, furthermore, eliminates the need to know the answers to said questions),

You still have not explained the denotation of the y-coordinate of the colored dots.

Gauss163 said:

My point is that the reader needs to know further info such as the resolution of the LED meter. Can it only report values that are multiples of 20%, or do they use some scheme with colors and/or blinking LEDs to report multiples of 10%, or some similar interpolation scheme? One cannot assess the error of the device without knowing these details.

Click to expand...

I believe that I explained it here:

When a battery is connected the lights run left and right a couple of times, before a single led lights up showing the estimated remaining energy.

HKJ said:

I believe that I explained it here:

When a battery is connected the lights run left and right a couple of times, before a single led lights up showing the estimated remaining energy.

Click to expand...

Is that supposed to imply that the meter can only report 5 values: 20%, 40%, 60%, 80%, 100%? When you take the average of (how many?) readings, do you round to only those values? If you took 2 readings of 20% and 40% then which value do you round to (or do you take more readings?)

Gauss163 said:

Is that supposed to imply that the meter can only report 4 values: 20%, 40%, 60%, 80%, 100%?

Click to expand...

I would say 5 values.

Gauss163 said:

When you take the average of (how many?) readings, do you round to only those values?

Click to expand...

I did 5 test at each stop, if the meter showed two different leds I put the dot between the 5 fixed value. I did not do any average, a 40% 40% 40% 40% 20% or a 40% 20% 20% 20% 20% would both be marked with a 30% dot.

HKJ said:

I would say 5 values.

Click to expand...

Agreed (modulo typo).

HKJ said:

I did 5 test at each stop, if the meter showed two different leds I put the dot between the 5 fixed value. I did not do any average, a 40% 40% 40% 40% 20% or a 40% 20% 20% 20% 20% would both be marked with a 30% dot.

Click to expand...

Ok, now I think I understand better how to interpret the graphs. Thanks for elaborating (and, of course, for doing the review).

Btw, it would be interesting to determine if any further accuracy can be obtained by averaging samples. Do you still have the 5 sample points, or did you discard them because you determined that they are useless for such purposes?

Gauss163 said:

Btw, it would be interesting to determine if any further accuracy can be obtained by averaging samples. Do you still have the 5 sample points, or did you discard them because you determined that they are useless for such purposes?

Click to expand...

I did not write down each sample, only the result for all 5 samples.

Mostly the latest samples will be at a lower value, the reason can be because the battery is at the edge of a lower value, but it can also be because the battery has rested and recovered some voltage, a few tests will eat some of the recovered voltage.

HKJ said:

I would say 5 values....

Click to expand...

Or perhaps 6 (including " 0 % " or no response) ... ?

Although I don't have this "Mini" version, I have the larger MBT-1 and (IIRC) I think there can be a battery state that triggers the "scan" but shows no capacity indicator.

Of course, with a completely empty cell, even the scan may not trigger, and thus this could also be considered another ( 7th? ) output state.

archimedes said:

Or perhaps 6 (including " 0 % " or no response) ... ?

Although I don't have this "Mini" version, I have the larger MBT-1 and (IIRC) I think there can be a battery state that triggers the "scan" but shows no capacity indicator.

Of course, with a completely empty cell, even the scan may not trigger, and thus this could also be considered another ( 7th? ) output state.

Click to expand...

You are correct, the 20% has two indications: steady red or flashing red. I assumed the flashing red was 0% (The last dot in both charts).

Thank you for that very thorough assessment of the ZTS Mini-MBT Battery Tester. I had been considering buying one, but I'm less inclined to do so after seeing the results of your tests.

I HAD been interested in the ZTS product because I wanted a more useful assessment of the actual status of batteries than can be obtained by using one of the inexpensive (usually $10.00 or less) battery testers that are commonly available, which I understand do nothing more than test voltage. ZTS seemed to be the only one claiming to do anything more.

Are there just NO reliable testers for determining a battery's actual condition?

If there are, are any reasonably priced for home use? I felt that $50 was about the outside limit of "reasonable price" for a home battery tester, so this ZTS unit seemed to be right at that outer limit. But I can't see paying that much for a tester that has as much variance as your tests showed.

Can anyone recommend a good battery tester for home use?

Thanks,
Paul

Rather more complex than most might need, but perhaps the West Mountain Radio CBA ?

Hello Paul,

Welcome to CPF.

I use my ZTS (full size and not the mini) all the time. The principle behind the ZTS is that they do discharge tests and record the voltage under load. This information is loaded into the tester and is used to compare the cell you are testing to what their data shows.

They build their data using quality name brand new cells. Your specific cell may have different IR which will give you different results. They regulate the pulse used so you get similar results regardless of the state of charge of the batteries that are powering the ZTS tester.

This tester is not precise. By that I mean that if you are using a cell with 2000 mAh capacity and the tester is showing 60% that does not mean that you have 1200 mAh left. However, if you test your cell and then find that you can use it in a device for something like 90 minutes then the next cell you put in will behave about the same. When you check your cell and it is testing at 60% you have about 90 minutes of life left. I guess I am indicating that it is repeatable.

This assumes that you don't change brands or lot number of the cells you are using. When you change brands, you are back to square one.

Unfortunately ZTS doesn't publish they reference voltage charts and don't tell exactly what cells they use to develop those references.

In general, if I am going for the weekend I prefer to have cells at least at 60%. If I am going for a week I want 100%.

Tom

pdalton said:

Are there just NO reliable testers for determining a battery's actual condition? If there are, are any reasonably priced for home use?

Click to expand...

If you seek a battery "fuel gauge" that is reasonably accurate and general (multi-chemistry) then you won't find one at the consumer level. It is very difficult to precisely estimate capacity for chemistries like NiMH that have large voltage hysteresis and (relatively) flat discharge curves. That's why the above ZTS and similar simple devices have huge errors.

Thanks for the review on this tester. I have the larger model-and it's been worth the price.