EBC-A20 power supply and fan replacement

[Mercruiser]

I have a new EBC-A20 Battery Tester Electronic Load (as seen on Off Grid Garage and others). I'm really happy with the functionality of the device. I could not home-build a functionally equivalent device for the $89 delivered from China price on Aliexpress.

The expensive part of making a home-build equivalent is getting a remote voltage sense power supply, which compensates for voltage drop in the battery leads. I can only find this feature on really expensive bench power supplies.

Fan Replacement

After using the A20 for a couple of weeks, the fan's sleeve bearing started seizing up. I bought a replacement from Mouser. The A20 uses DC Tubeaxial Fan, 60x60x20mm, 12VDC. Mouser has lots of those for sale. Instead of the original sleeve bearing, I paid an extra $3 or so to get a ball bearing fan. I like quiet, so I sorted the results by Noise and chose the quietest fan that was reasonably priced at $12: NMB Technologies PN 06020SA-12J-AA-00

https://www.mouser.com/ProductDetail/472-06020SA12JAA00

This fan works well and is a lot quieter than the original. At 12.7 CFM (0.36 m3/min) and 2800 RPM, this fan probably pushes less air than the original. But it seems to be plenty good for me. The air outlet temperature rise when charging or discharging at full load is minimal.

The hardest part of the project was re-assembling the case, due to the long header pins. See Andy's video at 9:30 (min:sec):

Next post: Power Supply Replacement.

 

[Mercruiser]

Power Supply Replacement

While testing my BMS working with an external current shunt, I noticed some strange errors in the current reading on the BMS. After trying a few experiments, I began to suspect that electrical noise was the issue. So I fired up my OWON VDS1022I USB oscilloscope and connected it to the power strip that supplies everything on my workbench, including the A20.

Note that I connected the scope leads to the output of a 120:12 transformer, hence the Vrms = 12V

Here is what I saw. Wow! This is a lot of noise on the AC supply.



To isolate the source of the noise, I started unplugging things on the power strip until the noise went away. Once I unplugged the EBC-A20, the noise disappeared. Here is what we have WITHOUT the A20 power supply connected:



Here is the noisy EBC-A20 power supply:


Next post: Replacement Power Supply

 

[Mercruiser]

Replacement Power Supply

The EBC-A20 uses a very common power supply brick that was popular with notebook computers. Specifically a 19V, 90W supply with a 2.5 mm Barrel Plug (2.5 mm id x 5.5 mm od).

There are lots of notebook power supplies on Amazon and others. Most of them are generic devices marketed with brands I've never heard of. Instead of spinning the roulette wheel on one of these, I wanted to order a PS from a company I have used many times: Mean Well. Being an electrical engineer, I like to buy products with actual Data Sheets with actual specifications. That why I like buying from Mouser, Digikey, Jameco, etc is that they tend to sell products with actual specifications. This is in contrast to most of the stuff on Amazon and Ali that just have a few bullet points for a "specification".

I ordered the GST90A19-P1M from Mouser for $31 plus $10 for 2nd day air shipping.

https://www.mouser.com/ProductDetail/709-GST90A19-P1M

The new PS works great and I have no more noise on the AC:

 

[gotbeans]

ball bearing is a good middle ground, definitely better than sleeve and quieter
You want a fluid dynamic bearing next time though. It'll last several-dozens of times longer than ball bearing and also it is quieter.

 

[Mercruiser]

View attachment 181355

ball bearing is a good middle ground, definitely better than sleeve and quieter
You want a fluid dynamic bearing next time though. It'll last several-dozens of times longer than ball bearing and also it is quieter.

I'll keep that in mind next time I buy a fan. I'm considering replacing the loud fan in the EBC-A40L. The fan is working correctly, but they installed a very high velocity fan. It is way overkill for my electronics lab in the basement of my house. The ambient temperature is never over 70F. The A40L fan would be good for full sun in Phoenix on a summer day. ?

 

[gotbeans]

Noctua makes fans that are near dead silent but they're a premium. I've had one running on my server about 20 years 24/7 though so the FDB (fluid dynamic bearings) really work well. I only buy as my other fans die slowly over time I guess I'll have all noctua some day.
Some other brands these days do it as well, they were just the first in the pc world

I usually have music going so ZKE fan doesn't bother me but yea it's louder than even my server fans are lol..

 

[Reigate]

I have a new EBC-A20 Battery Tester Electronic Load (as seen on Off Grid Garage and others). I'm really happy with the functionality of the device. I could not home-build a functionally equivalent device for the $89 delivered from China price on Aliexpress.

The expensive part of making a home-build equivalent is getting a remote voltage sense power supply, which compensates for voltage drop in the battery leads. I can only find this feature on really expensive bench power supplies.

Fan Replacement

After using the A20 for a couple of weeks, the fan's sleeve bearing started seizing up. I bought a replacement from Mouser. The A20 uses DC Tubeaxial Fan, 60x60x20mm, 12VDC. Mouser has lots of those for sale. Instead of the original sleeve bearing, I paid an extra $3 or so to get a ball bearing fan. I like quiet, so I sorted the results by Noise and chose the quietest fan that was reasonably priced at $12: NMB Technologies PN 06020SA-12J-AA-00

https://www.mouser.com/ProductDetail/472-06020SA12JAA00

This fan works well and is a lot quieter than the original. At 12.7 CFM (0.36 m3/min) and 2800 RPM, this fan probably pushes less air than the original. But it seems to be plenty good for me. The air outlet temperature rise when charging or discharging at full load is minimal.

The hardest part of the project was re-assembling the case, due to the long header pins. See Andy's video at 9:30 (min:sec):

View attachment 179606

Next post: Power Supply Replacement.

Thank you! Mine was delivered with a very noisy fan and have changed it to a similar one as yours and it is now quiet.

I calibrated the tester at 2.50 and 3.65 volts using a variable power supply and an accurate volt meter and I am happy that the tester displayed voltage is within 2mv throughout the target range. The trouble I am having is when I start testing a cell with a load, the tester displayed voltage is no longer accurate when compared to the volt meter. The leads for the tester are equipped with two wires each, one for load and one for voltage sensing so in theory, there should be no voltage error under load? I thought that the manufacturer could have miswired the plugs which would explain the voltage error under load however I took both the positive and negative plugs apart and the sense wires are connected the the correct pins which then run to the voltage sense mini connector on the motherboard?

I am wondering if I just have a faulty tester or am I just missing something?