Understanding and diagnosing erratic BD50 compressor

[LifesBetterWhenYoureBeating]

I have a 6ft³, well insulated box, kept cold by a 16"x22" evaporator plate connected to a bd50 compressor. I'm in the tropics. It was installed a few months ago and the compressor always cycled for 4-5 min ON and 4-5 OFF, drawing around 4A. The thermostat is at the least cold setting as otherwise the gas pipe next to the compressor gets covered by a layer of ice. Lately I noticed the compressor sometimes stays on for more than 1h (both during the day and several hours after sunset), drawing up to 8A. The contents inside the box are pretty much the same as before, and I've seen this happen when the plate had a thick layer (1/2") of ice on the bottom half as well as a couple days after defrosting (when the plate had only a thin layer of ice)

Does anybody have any idea what could be the cause of this erratic behavior?

I'm also wondering how to decide when to defrost as so far I didn't notice much difference in the system's efficiency (temp/consumption) between running with a 1/2" layer of ice on the bottom half of the plate and a freshly defrosted plate.

Any other suggestions/ideas would be greatly appreciated as this is the first time I'm dealing with a fridge like this so am not familiar with what is normal/expected behavior

Thanks in advance!

[LifesBetterWhenYoureBeating]

Looks like it was a faulty thermostat

[Bowdrie]

It may seem counter intuitive, but ice is a decent insulator.
The evaporator absorbs heat from the air in the box, and with a layer of ice the heat must first be absorbed by the ice and transferred to the evaporator and the humidity continues to make the ice thicker.
Kind of like an igloo in reverse, the heat inside the igloo has a hard time escaping to the outside unless and until it can melt the ice.
The same process keeps the evaporator, (the outside,) from quickly absorbing the heat in the box, but the evaporator won't allow the ice to melt.
Without knowing the specifics of the mounting of the sensor for the thermostat it's hard to give a hard and fast answer.
Sometimes an "ice clogged" sensor will not allow the compressor to turn off until the "cold", as it were, can migrate thru the ice, (simplifying it, as it's heat that does the migrating).

[RaymondR]

Cycling 4-5 minutes is not a good sign.

Icing all the way back to the compressor is also not a good sign and indicates liquid refrigerant is evaporating in the line and may even be fonding it's way back to the compressor.

[LifesBetterWhenYoureBeating]

Originally Posted by RaymondR

Cycling 4-5 minutes is not a good sign.

Icing all the way back to the compressor is also not a good sign and indicates liquid refrigerant is evaporating in the line and may even be fonding it's way back to the compressor.

What would be the ideal cycle time? What could be causing the short cycle times?

Now with the new thermostat the fridge is plenty cool, drawing ~50A/day and the pipe next to the compressor doesn't have a layer of ice. Given that, should I worry about the short cycle times?

[RaymondR]

Could be just the fan running intermittently.

[stormalong]

Originally Posted by RaymondR

Cycling 4-5 minutes is not a good sign.

Icing all the way back to the compressor is also not a good sign and indicates liquid refrigerant is evaporating in the line and may even be fonding it's way back to the compressor.

Have to disagree. 50% cycle time is fine for an evaporator plate system. They are made to constantly cycle. My BD50 based AB cold machine has been doing that non stop since 2010.

[LifesBetterWhenYoureBeating]

Originally Posted by RaymondR

Could be just the fan running intermittently.

Hmmmm. I don't think so as the way I'm watching the compressor cycles is via the amp draw on my victron battery monitor