The MDL-150 battery capacity tester and DC electronic load
A great inexpensive unit that performs as expected and is really good value for money. User interface could use some improvement but otherwise I highly recommend it for hobbyists for occasional use. Oh yeah, throw away the dangerous power supply that comes with it and use a standard 12V 1A PSU with a 5.5×2.1mm barrel jack instead. Those FCC / CE / UKCA certifications are fake. Video review towards bottom of the page.
Introduction
I decided to build my own LiFePo4 battery pack from salvaged cells basically taking a 6P8S configuration 25.6V pack and turning it into a whatever capacity I could get at 12 volts. That meant there were 48 cells to test so I started looking around for a battery capacity tester which could automatically turn off the load when the low voltage point of the battery cell was reached. I knew there were for certain one or more dud cells in the pack as the 8 banks of cells were at different voltages and the battery didn’t give the expected capacity. Manually rebalancing the cells did work to some extent but I noticed one cell getting hot.
Hence the purchase of the MDL-150 which is unbranded and can be found on AliExpress, Amazon and eBay for an approximate price of £35 ($47USD) but it comes in a desktop metal case unlike the similarly priced testers that are just a bare circuit board and PC style CPU cooler. These typically have a single MOSFET and I was suspect they cannot handle the advertised rating and one particular common model blows up if you connect the load and forget to set the current to minimum beforehand. Also with them being a bare circuit board there’s a chance that it could get damaged or short out on something on the workbench.
Teardown
The first thing I did was take the MDL-150 apart and found that is based on the bare PCB model made by the same manufacturer – the DTL-150 pictured here.
However the MDL-150 has two MOSFETS on a rather small heatsink but the case is cleverly designed so that airflow is directed through it via ducting on the underside of the lid. Not much can be determined from the chips used onboard as they had their part numbers lasered off but there would be the usual op-amps and associated controller circuitry which appears to be based on an ARM microcontroller. The control circuitry is on a PCB behind the front panel with a ribbon cable connecting it to the main PCB sitting in the bottom of the case. With having two MOSFETS I felt this can have a better chance of supporting the advertised current and power ratings.
Overview
On the front there is a 2.4″ colour screen a “menu” button spelt incorrectly as “MEUN” (oh dear, what cheap junk have I bought here?) and a R/S button which starts and stops the load. There’s a rotary encoder for selecting menu options. The user interface does need some improvement – to get into the menu you press and hold the menu button for 5 seconds then the button selects the menu option. The rotary encoder selects the parameter currently selected. Pressing the encoder does nothing. You can exit the menu by either pressing and holding or pressing menu till exit is highlighted and then turn the knob right or left – it does not matter.
During normal operation pressing menu selects the various options highlighted by the option flashing e.g. selecting constant current, constant power etc. The rotary encoder knob then changes the currently highlighted option. Pushing the knob moves the cursor when adjusting the required setting. I suspect the awkward operation is because it is based on the DTL-150 which has buttons instead of a rotary encoder.
Finally there are two LED’s and the input connector for the load which supports 4 wire measurement for greater accuracy and to account for the voltage drop along the lest leads. The LED’s are labelled rather confusingly CHG and DSG giving the impression this unit can charge batteries. It can’t – the green CHG LED indicates that the test has finished and the load has reached the set time limit, current limit or set cutoff voltage. The red DSG LED indicates that the load is connected and the test is running.
The 6 pin input connector is standard with the outer 4 being used for the load with two pins for + & – providing for connecting heavier loads and the inner two the low current voltage measurement. I would have preferred banana jacks for the load input as this makes it much more useful and could be used with standard test leads. If you do use 2 wire mode the positive and negative sense pins need to be bridged to either of the load pins otherwise it won’t draw any current from the load. I would recommend 4 wire operation for greater accuracy. Speaking of which…
Calibration
The voltage and current measurements should be calibrated before use. This only has to be done once. To do this you must input exactly 30 volts then go to the menu and select voltage reference 30V. It may show not exactly 30 volts but to calibrate it move the encoder till “OK” flashes up on the display. The displayed voltage should now be accurate. The same goes for current however setting it is more tricky. You need to input a 3A current such as from a bench PSU with a current limit set and then set it to constant current and turn on the load with the R/S button. Then you need to repeat the above steps this time selecting current reference 3A. In my case the current reading was already accurate so I only had to calibrate the voltage.
Testing
So to test this unit I discharged each of the six parallel connected cells to 2.8V measuring the total amp-hours accumulated. The tester will remember the capacity so you can test higher capacity batteries over several sessions if you didn’t want to leave it running overnight. For example you could discharge 50% of the battery one day and the rest the day after. You clear the accumulated readings by going into the menu and selecting clear accumulative. The next test was to individually test the cells part of the 6 sets in parallel that gave a lower than expected amp hour rating so I could determine any faulty cells. 3 were found in total.
The tester worked flawlessly and didn’t emit the magic smoke and performed exactly as expected. The fan is incredibly loud and only comes on at full speed when the power drawn goes over 30 watts or the internal temperature reaches 35 degrees centigrade. There is no variable speed depending on temperature and it doesn’t turn off once started. A minor niggle. To sum up, for hobbyist use that needs to test the occasional battery or build their own DIY battery or even use it for testing power supplies it is great value for money. It has features that are only found in more expensive products and I would definitely recommend it over the bare PCB battery testers.
Very pleased with it.
Video teardown and review