Suoer SON-20A 20Amp digital charger review
Up for review is a Suoer brand battery charger. Product link I bought this one locally as we have a need for charging car batteries when needed. I wanted an automatic one so that those not technically inclined could use it and not boil a battery to death when an incorrect setting is used. Ofcourse, before using it, I wanted to see how it fares as it is very cheap so I took it apart and did some tests... which did not look promising.
The product description on the website is as follows:
Omni - directional battery protection functions
1. The power will be turned off automatically when the battery gets fully charged
2. Three- phase charging mode makes it charge quickly
3. Reverse connect protection and short circuit protection
4. It can judge whether there is a battery, and will shut down automatically if there is no one
5. No spark appears when connecting, extremely safe
6. Hanging style design makes it easy to install
1. Positive and negative pole clamp connector
2. Suspension wear mouth
3. Cooling air outlet
4. Work indicating lamp
5. Electric plug
Now for the unboxing.
Product as received.
The charger in bubble wrap. Warranty card and manual underneath.
I was expecting it to be light as it is a switchmode supply but not this light.
Case is made of plastic, not extruded aluminum.
Input/Output cables and fan intake.
Three switches. One for power, another selects 12/24V operation, last one selects fast or auto charge mode.
As Dave of EEVblog says it, "Don't turn it on, take it apart!!!"
Removing the fan, we are greeted with a heatsink.
The metal thing glued in between the fins is a normally open thermal switch. This controls the fan to turn on when the heatsink gets hot. There is no other control for the fan.
Fan label. This is unusual that it is a 30V unit.
Good luck finding a replacement if you want to change it.
PCB top view.
Topology is a simple half bridge supply controlled by a TL494 / KA7500 derivative which is similar to cheapie PC supplies.
Switching supply control section.
A VERY tiny fuse. The current rupture rating is questionable in this one.
Note the absense of any inrush protection and EMI filtering at the input!!!
The power LED on the front is also directly connected to the mains through a 1/4W 470K limiting resistor! running >300V peak across a 1/4W resistor is asking for trouble.
This is the primary side snubber circuit.
A high power dissipation device in between large components then mounted flush on the PCB. Not a good design practice.
Rear heatsink with four devices.
Two MJ13009 as primary switches, and two MBRF20150CT dual diodes in parallel for output rectification.
A friend of mine that has worked with a company that makes offline supplies said that sharing a heatsink between main switchers and output rectifiers is NEVER done for safety reasons, even if the devices have insulators between them.
A view of the current shunt resistor.
It is both used by the current meter and current feedback circuit.
Output switching done with a RU75N08R mosfet for reverse polarity protection.
I guess they decided it is cheaper to solder on to the faston lugs than use quick connect crimp terminals.
Sub board holding the display.
There appears to be a SOIC package uC under the 7seg display. The soic8 part on the right is an LM358.
This looks dangerous.
Take a close look at the clearance and creepage distances between the input and output sides.
In most parts, the creepage is a measly 4mm or so but there is a part near the gate drive transformer (near the center of the picture) where the primary and secondary sides are about 1mm apart!!!
Now on to some tests. Tests are: Full load current, thermal images, device temperatures and output ripple.
Maximum power output when the charger is set to "auto" mode.
4.5A max, 58.8W max
Maximum power output when the charger is set to "fast" mode.
9.0A max, 118W max
Thermal image when the unit is running idle without any load connected.
You can see the primary snubber resistor is already uncomfortably hot.
Meter reading at 800mA output current draw.
It is only displaying 100mA!!
Ripple at 12V about 2A output current.
This shows a very bad loop instability at low currents. This explains the buzzing sound I could hear when I was initially testing it charging a car battery.
Meter reading at 5.0A current draw.
Meter accuracy is still questionable.
Ripple at 12V 5A output current.
The ripple is smaller but there are random blips in between the sawtooth wave. This was with the charger set to fast mode as I can only get to 4.5A max with the so-called auto mode.
Primary side snubber temperature at 5A output current.
Now the snubber resistor is really hot. This is a metal oxide part rated at about 2W but stuck in between two large components and then flush mounted to the PCB. In proper designs, it is recommended to mount these above the board for ventilation.
Main switching transistor temperature at 5A output current.
One transistor is getting hotter than the other. It shows that either there is a switch timing unbalance between phases(unlikely) or this one is not mounted firmly to the heatsink (more likely!).
At 9.0A output draw, the meter shows 9.6A.
No matter how I tried, I cannot go over this to reach the 20A nameplate rating. It seems Chinese rating is 2x actual.
Running my electronic load in constant voltage mode, I can simulate a deeply discharged battery. The charger simply runs at 9.0A max constant current into a drained battery.
Ripple at 12V 8A output current.
It is not visible but there is a small sine wave ripple that can be seen.
Ripple at 12V 9A output current.
Nearly the same as above but a little smaller amplitude.
Primary side snubber temperature at 9.0A output current.
Now the snubber resistor is really hot. I bet this will not last very long.
Main switching transistor temperature at 9.0A output current.
The switching transistor is also really hot now. If the case temp is at 125C, I bet the die temp is much hotter. The fan is running at this point and facing the heatsink so it SHOULD be cooler, not getting this hot.
Initial conclusion: The 24V tests are to be done next time. Due to the feedback loob instability, I decided to investigate it further before going to 24V testing. I don't want to let the magic smoke out yet as during 12V operation, it can only do half of its nameplate rating and the loop instability is causing me concern. Too much ripple can overheat the output capacitor and cause it to explode.
The charger does not exhibit the specified "Three- phase charging mode makes it charge quickly" thing. Instead, it only does CC/CV mode where the max current limit is set by the fast/auto switch and CV is factory set at 13.65V. There is also no auto shut-off when the battery is full. The display just reads zero when the current drops below 800mA.
Repeating the product description with my comments...
Omni - directional battery protection functions
1. The power will be turned off automatically when the battery gets fully charged - yeah right! no such feature in mine.
2. Three- phase charging mode makes it charge quickly - Also not available in my unit!!!! only does CC/CV
3. Reverse connect protection and short circuit protection - decided not to test this as I don't want to let the magic smoke out... yet.
4. It can judge whether there is a battery, and will shut down automatically if there is no one - The output mosfet just switches off without a battery connected but the switching supply is still running.
5. No spark appears when connecting, extremely safe - Uh, I can see sparking when I connect the clips to the battery!!!!
6. Hanging style design makes it easy to install - Hang it somewhere out of reach until it is redesigned to be safe for use.
1. Positive and negative pole clamp connector - uh? All chargers I've seen have this included.
2. Suspension wear mouth - what!?
3. Cooling air outlet - oooh cool.
4. Work indicating lamp - and it is also directly connected to mains!!!
5. Electric plug - see no.1
18 Jul 2017:
Here it is about 2.5years later. It stopped working so let's see what happened.
Board is really dusty. Fan has stopped turning and there is no output.
There appears to be about 21V on the fan supply but either the mechanical thermo switch or the fan itself has failed as the fan not running was said to be the cause of failure.
There is some water damage, I have no idea how they used this!
The hot resistor did not seem to have heat damage around the PCB.
There is no output voltage but we are getting 43V across a 35V capacitor. The reverse polarity mosfet does not seem to be working.
I have not gone much in depth on the damage and I'm not sure if I will bother because of the quality of the unit, maybe get a better one to replace it.
Page created and copyright R.Quan ©14 Dec 2014.