convert AB thermostat to ISEC Smart Energy Control

[jviss]

Ah, got it, thanks. On the last point, minimum RPM is not necessarily optimum. So if not absolute minimum speed, what is the algorithm they employ to determine speed?

I could augment my dual-Inkbird solution to run at max compressor speed with a charging source present, as it's just a resistor in the thermostat circuit that determines that. But optimizing while on batts would require a microcontroller (or a true Rube Golberg assemblage of discrete components).

 

[jviss]

So, I purchased the Danfoss AEO controller. The only place I could find one was a Polish vendor on eBay, about $240 inclusive of shipping. I also will be installing the Inkbird temperature controller in place of the stock thermostat. It will interface via the A-B box's RJ-11 connector, via an RJ11 breakout panel I have and a standard phone patch cord. (This allows me to leave the stock thermostat in place and easily switch back if something goes wrong.) The only remaining piece is a circuit to control to a lower temp when a charging source is present, and I have a circuit and parts for it, but just not the will to do it yet; and, now I'm not convinced it's worth it. Finally, I will install a low voltage disconnect for the 'fridge circuit, cutting it off at 12.2V or so.

The AEO is cool. It servos towards a 30 minute "on"-time for the compressor. This is considered an efficient on-time goal, preventing short-cycling, or too-long on-times.

https://assets.danfoss.com/documents/DOC226586436896/DOC226586436896.pdf

 

[Mark Maulden]

Geez...what about just an on/off timer on input power...

 

[Hayden Watson]

It looks like it does a lot more than just on/off.

 

[jviss]

Geez...what about just an on/off timer on input power...

Good question. An on/off timer on the inout wouldn't accomplish the goal of having an energy-efficient cycle time. In addition, it would interfere with temperature regulation. For example, how would you know what on/off interval to select? And, at what compressor speed? And in varying conditions, the timer wouldn't know to come on early if the box temp goes too high.

This device learns the characteristics of the box and the heat load. The algorithm it implements is pretty clearly described in the linked data sheet.

If this helps, once the box temp is stabilized, the starting compressor speed for an on-cycle is analogous to the nozzle size on an oil burner. You want the burner to heat the boiler in a reasonable time, but you don't want it short cycling, either.

 

[Mark Maulden]

But, are you down in the menutia? Is there a problem with box temp varying a few degrees? My refer is on more than off. It is ~90 here though. The evaporator is frozen solid. I keep turning it down and maybe the t-stat is defective. Been thinking a timer just to keep the evaporator frozen and see how cold the beer is..

 

[jviss]

But, are you down in the menutia?

I don't know what that means.

But, are you down in the menutia? Is there a problem with box temp varying a few degrees? My refer is on more than off. It is ~90 here though. The evaporator is frozen solid. I keep turning it down and maybe the t-stat is defective. Been thinking a timer just to keep the evaporator frozen and see how cold the beer is..

That your evap is frozen solid and it's running on more than off means your t-stat is set too cold. It could be the t-stat, or it could also be the location of the thermocouple, or perhaps a lack of air circulation in the box.

Where is the thermocouple attached to the evap? Is there a fan in the box?

Mine is new to me, and the refrigerator a year old replacement. I can't figure out the stock, analog thermostat. It's hard to see the dial marks, there was no index until I made a mark with a Sharpie. (and it's easily knocked to a random position.) Now, even set at 2 it's very cold, and it short cycles - on for only 8 minutes at a time. I don't think the box fan is running, but I don't know anything about how these run. And, it seems to be using lots of power. Hence the quest for greater efficiency.

 

[Mark Maulden]

Menutia, my “definition”..if you have a vial and put a liquid in it, you will notice the level is higher around the circumference than in the middle. This is due to surface tension between the vessel and liquid, and that difference is Menutia. I think I spelled it right. In my world, which is the RF electronic design world, we say, “it’s in the noise”. In other words, is it worth worrying about..you could spend a lot money etc to optimize only to find out the box is not insulated well which I think is one of my issues. You could “optimize”, and it would be interesting to see the results. My thermocouple is clamped on the evap where it is “supposed” to be. There is no fan in the box. I had to have a “fill up” in BC and the tech said no leaks but you should install a fan. Also my evap is installed flat at the top with a removable shelf underneath. Thought about drilling some 1/2” holes in the shelf..I’m going to be very interested in your results..

 

[NYSail]

I have a AB ice box that feeds the refrig..... on mine I have two separate controllers made by Control Products Inc. seem to do a good job......

 

[jviss]

Menutia, my “definition”..if you have a vial and put a liquid in it, you will notice the level is higher around the circumference than in the middle.

Ah, meniscus.

 

[Hayden Watson]

And now there is a new contender in the thermostat wars. http://www.suremarineservice.com/Galley/Refrigeration-Accessories/SLFO-DCDC-122401.html

 

[jviss]

And now there is a new contender in the thermostat wars. http://www.suremarineservice.com/Galley/Refrigeration-Accessories/SLFO-DCDC-122401.html

Wow, I wish I had seen that before I bought the 101N0300 controller for more money!

 

[Mark Maulden]

So, Hayden, you have your order in right?

 

[Hayden Watson]

So, Hayden, you have your order in right?

Not yet. I still can't decide which one of the three to go with. They each have advantages. The ISEC is the cheapest and does a lot. The Fridge Optimizer does basically the same thing with digital readout and more control of temp, defrost and usage. The AEO takes a different direction and regulates based on box history, learning the characteristics of the system and forcing long run times and rest times.
Fortunately (or not) I don't need to make THAT decision just yet because I can't even get my fridge to do basic cooling right now. I have a tiny leak that I cannot find. It looses cooling after a couple of weeks and needs to be recharged. I have an electronic sniffer that cannot find a leak so I am thinking it must be intermittent or something.

 

[jviss]

I looked into the FO (Fridge Optimizer) in more detail. It's very well done, and a very good value. Unfortunately, the compressor speed is either min or max, i.e., 2000 or 3500 RPM. I think that if you combine it with the AEO you would get the best of both, since the setting of min compressor speed for the AEO controller puts it in AEO mode. But then, that would be nearly $500 for the combo. It's a shame FO didn't implement the AEO algorithm. But, since their firmware is upgradable, maybe they will some day.

More:

• besides running at max compressor speed when a charging source is present, they don't say if there's an alternate set point on the tstat for this, to avoid freezing things you don't want frozen;

Correction on the above: they do have a "reduced temp" setting you can set to choose the tstat setpoint when charging.

• the logging features are nice, but too short in duration: 24 hours for temp history, etc. Should be weeks, if not months!

 

[Hayden Watson]

As an update on this thread, I ended up installing an Inkbird T-stat like the one that jviss had discussed. I then added a switch and a 1500-ohm resistor in parallel the wire from the Inkbird to the "T" terminal. When the switch is closed, it adds 1500-ohm resistance to the C-T circuit which runs the compressor on the highest speed. When the switch is open, the resistance drops to nearly 0-ohms which make the compressor run at minimum speed. I usually leave the switch open but will close it and manually turn the temp down about 4º when motoring to cold-sink the icebox with the available power. I also replaced the loud factory fan with a 120mm PWM fan with PWM speed controller. This allows me to use a higher quality and capacity fan but run it a a lower and more efficient speed. In my area, I need to reduce the fan flow to avoid a condition called "supercooling" of the coil where too much heat is removed which make the gas not fully evaporate in the icebox which reduces the cooling ability. AB recommends putting a piece of cardboard over half of the condenser if the air temperature is under 60º for this reason.