Oscillating relay for DC backup

[pwnell]

I am using this part from Amazon: https://a.co/d/04hzd7Qk

It uses this relay: https://semiconductors.es/datasheet/HK14FD-DC12V.html

I have 13V bench power supply with 3A current limit connected to IN1.  I also have a 12.8V LiFePO4 battery connected to IN2.  Lastly I have a DC Electronic Load connected to OUT, set to 20W CP.

When I turn off the bench power supply, the unit switches to the LiFePO4 battery just fine, about 17ms switch time.  That is 2ms longer than the max spec for the relay - I suppose it is due to the onboard capacitors.  When I turn the bench power supply back on it switches back to it and all is well.

However when I increase the DC load to 32W or higher (still under 3A), when switching from bench power supply to battery all is well as before, but when I turn the bench power supply back on the relay oscillates on/off very quickly - as in tens of times per second.  Why does this circuit oscillate? I have not connected any diodes / other components to this circuit.

The relay is rated for 30VDC @ 5A so I am well within the specs.

[xvr]

However when I increase the DC load to 32W or higher (still under 3A), when switching from bench power supply to battery all is well as before,

At this point, check the power output of the bench power supply. I'm guessing it drops below 12V and back. Perhaps the reason is the long and thin wire from the power supply to the relay board.

[pwnell]

I will certainly test that right now.  An interesting fact that I forgot to mention is that if I turn the load off, then switch to the bench power supply, then turn the load on, it works.  It is when the load is on and the relay switches from battery to the bench power supply while the load is kept on, that it starts oscillating.

[pwnell]

Just tested it - at 35W load the DC volt at the terminals of the relay is 12.8V so only .2V drop - not enough to cause the relay to be out of spec. Any other ideas?

[pwnell]

I am starting to wonder if the six caps on the Amazon sourced relay board are not to blame. Below is a probe when the battery is disconnected, bench power supply on IN1, DC load off.  I switch the bench power supply on and record the voltages.  Yellow is probe at OUT, cyan is probe at IN1.


It seems like the rise on IN1 (cyan) corresponds to the 6 onboard capacitors charging up.  When they exceed the voltage  threshold of the relay coil, the relay turns on.  This connects IN1 to OUT, and since OUT has no load the capacitors are not discharging through the load, however they  have discharged through the coil of the relay to store energy in the magnetic field.  However once that was done, the caps charge up again and stay charged, no drain on the input.

When I place a 35W load on OUT, and then turn on IN1, I get below:


Here the same initial charge cycle occurs for the onboard caps.  However once the relay closes and places IN1 on OUT, the caps are immediately drained.  That coupled with the drain from energizing the coil causes the input voltage to be too low to keep the relay coil energized - thus it opens.  Once it opens the caps charge up again, closing the relay etc.

Does that seem like a reasonable explanation?

** UPDATE: I desoldered the caps - this helped, I can now place 10W of load without it oscillating, but 15W+ is still too much. So there goes that theory. 
 

[xvr]

Cyan trace (from bench power supply) clearly indicates that power supply drops it output under load. Looks like it enters CC mode (if any) or just turned off under overload.
Check the current from power supply (by shunt resistor, for example) - will it really under 3A?
Connect battery to IN1 - would it oscilate?

[pwnell]

Sure, but why? The coil is rated at 12VDC @ 45mA or something.  Why would 2.5A load on the output cause it to drop voltage on the input?  Unless the characteristics of my Keysight EL34143A electronic load is highly capacitive, in which case the inrush of current might be an issue?

On battery it works.  I will check shunt resistor.

According to the bench power supply it never goes out of CV mode. It also does not trip on OCP of 3A.

[pwnell]

Figured it out.  The keysight EL34143A in CP mode causes the voltage to drop due to it adjusting the current to achieve CP, and even though it is really good at this it is not fast enough to not cause issues with the logic design of this Amazon sourced circuit.  That interferes with the whole purpose of the circuit.  Putting it in CI mode works like a charm.  The clue for me was when I saw this: (purple is a DC current probe connected to IN1).  No passive component can cause I and V waveforms like that.  So it has to be the DC load in CP mode:

[Solder_Junkie]

There are issues with relay switching times.

If a relay is used without a diode across the coil, you get a large back EMF Voltage generated when the relay is de-energized. It is common practice to add a diode across the coil to stop this back EMF, which can damage switches and semiconductors (transistors). The addition of a diode effectively shorts out the relay coil when power is removed, and in doing so increases the drop out time.

Generally when using relay switching for power backup, whatever is being protected needs to have sufficient capacitance to hold the supply Voltage while change over takes place. For example, I have a UPS by APC and it is specified with a maximum transfer time of 10mS, typically 6mS, yet I have seen reviewers complain that the short interruption in supply causes their equipment problems... I would guess that any equipment worth it's salt should have sufficient smoothing to cope with a 10 mS break in supply.