Prevent Simultaneous Compressor Start-up ?

[Minggat]

Understood. It is the starting load the is the issue. Could use just one relay on one unit. That would always make that the lead unit in a lead/lag array. Should cover the issue, simple and cheap.

[JayH]

Originally Posted by btrayfors

If I understand the OP's question correctly, it's not just a matter of cutting power to one compressor while the other starts.

With two A/Cs or two friges, you want BOTH of them running at the same time. If your inverter or other ac power source can handle both loads while they're running, but has problems if -- by chance -- both compressors attempt to start at the same time, then you can't just solve the problem by using relays which will cut power to one compressor while the other is starting. The reason is that the other one -- the one you cut power to -- could well have been RUNNING (not starting), and if you cut power to the compressor you can't just turn the power back on without waiting a "decent interval" or you could REALLY provoke a high amperage draw, as the compressor's head pressure would be high. Some installations have a "wait time" after power is cut, but others don't.

I don't know what the solution is, but I'm pretty sure it's not just a couple of relays. At least that's how it seems to me this late in the evening.

Thanks Bill, this is an accurate summary as I see the problem. That's why I was thinking four relays, with the second pair to disable the first if the compressors were already running. The relays should be fast enough to reduce the chances of simultaneous random starts to practically zero. It probably doesn't help for non-random starts, ie. when the inverter is turned on.

[JayH]

Originally Posted by goboatingnow

get a standard AC motor protection unit. set the current to the max the invertor is happy with, use the contacts to disable the other compressor while the fault ( ie too high current exisits). All ac wiring.

This is an intriguing suggestion, but I don't think I'm smart enough to understand it. I did some research along these lines and ended up looking at variable frequency drivers and soft starters. This looks like a very expensive solution with each unit costing about as much as the fridge it would control.

Can you recommend something specific so I can see what a "standard AC motor protection unit" is? Each fridge draws about 170VA when running.

[JayH]

Originally Posted by Minggat

Understood. It is the starting load the is the issue. Could use just one relay on one unit. That would always make that the lead unit in a lead/lag array. Should cover the issue, simple and cheap.

I appreciate all suggestions, but I don't understand how this one solves the problem. Are you assuming that both units are on the same thermostat? The units have different thermostats and different duty cycles (one is a freezer), so I don't think a lead/lag array will work.

[um saudade]

There are timer relays (not time delay) that will energize (or de-energize) a circuit for a given period of time. Can be called cycle timers and come in all kinds of configurations.

Use a cycle timer of 10 seconds or so activated by the line voltage to the compressor of each unit. Now use the normally closed contacts on the relay to break either the line power or compressor power to the other unit.

So when compressor A kicks on, it activates the timer relay for 10 seconds. During that time the line circuit to compressor B is held open by the timer relay. At relay timeout the contacts close and compressor B could start if the thermostat contacts were closed. Works both ways as in if B starts and A just happens to want to start a second or two later.

If you get paranoid with statistical probabilities you can have one set for 5 seconds and the other set for 10 and if they should start at exactly the same time, one would always time out and start first, keeping the second compressor disabled for another 5 seconds..

Um Saudade

[JayH]

Originally Posted by um saudade

Use a cycle timer of 10 seconds or so activated by the line voltage to the compressor of each unit. Now use the normally closed contacts on the relay to break either the line power or compressor power to the other unit.

So when compressor A kicks on, it activates the timer relay for 10 seconds. During that time the line circuit to compressor B is held open by the timer relay. At relay timeout the contacts close and compressor B could start if the thermostat contacts were closed. Works both ways as in if B starts and A just happens to want to start a second or two later.

Doesn't this mean that compressor B will be turned off each time compressor A starts? Could another pair of normally closed relays prevent that if compressor B is already running?

[um saudade]

You caught me! Yes that is one set of conditions not covered by my logic. Another set of contacts in series with the coil, held open by the running of the compressor would prevent the relay from cycling. This is getting a little complicated though as it would most likely entail yet another pair of relays.

Should be some easier logic. Could do it very simply with digital logic, not so easy with coils as you can’t have common tie points without the voltage activating all circuits tied to the point, as opposed to a simple NAND gate or NOR gate.

Have you seached for this circuit on line? There could be one available.

Um Saudade

[David M]

Are these compressors cooling the same compartment or different compartments? Or has this question already been asked? If its the same refrigerated compartment then setting the thermostat at two different temps would make some sense. If not then buying a second inverter would make sense since such a switching device might not exist at all.

[um saudade]

Depending on the duty cycles of the two compressors, if they don’t run for long periods of time to cycle, then a simple relay with the coil activated by one of the compressor lines or thermostat circuits would do the trick.

Wire one set of contacts normally open to compressor A and one set normally closed to compressor B. Wire the highest duty cycle compressor to have the priority. Now when the High priority compressor is switched on it will run until it cycles and the other compressor will not energize.

The possible down side is when A thermostat closes the relay can and will turn off compressor B for the duration of the Compressor A run time. It should be no problem for compressor B as the other compressor will run for enough time that the head pressure will be down and there will be no stalled compressor start.

Simple to hook up and try, and easier on the inverter as the max load is always one compressor only.
　
Um Saudade

[Goofy Foot]

This has became way complicated.

Duty cycle needs to be defined?

There is some sort of switching relay that turns the "unit" on or off without unplugging it, thermostat is wired to it. Identify this switching voltage, line? 12 volt, 24 volt?

If both compressor go to one refer box a two stage thermostat is the proper, most efficient way.

Something like this:
http://cgproducts.johnsoncontrols.co...T-1927110.html

If you have a refrigerator and a freezer with independent compressors and set points you could use a smart relay, which is a relay with a limited PLC in it. You would wire your thermostats to the inputs, the outputs would be wired to the compressor start relays. Then write a small piece of logic that would lock the non-running compressor out while the demand compressor starts or do nothing if it is running. If duty cycle is problem you could have a timer that would turn the compressor off for a given amount of time. You could also set up a short cycle timer or use one of the spare inputs to shut down the compressors if the battery voltage drops, alarms, lost of options.

Something like this.
EZ Intelligent Relays

Or use a PLC like this if you want to have more fun. Pico Programmable Logic Controllers from Allen-Bradley

If you want to look at it from the AC plug-in side, you will need to put current switches on the AC leads to the motor input. These could go to the inputs of the PLC or the smart relay. You could then write the same lockout logic while starting timer thingy that would energize a lockout relay on the non-starting compressor until in-rush current has dropped.

You should be able to get a smart relay for $100-$150. Just match the voltage/ amperage/ gigowatts to your equipment.



Sorry my grammar sux.

GF

[daddle]

If your compressors start once per hour and the startup surge is a 2 second event then you will experience inverter failure about once per month and every time the inverter is turned on with both thermostats requesting cooling.

Is a preventer worth it? Probably.

[GordMay]

Originally Posted by Goofy Foot

... Or use a PLC like this if you want to have more fun. Pico Programmable Logic Controllers from Allen-Bradley ...

Wow - I haven’t heard mention of the old Westinghouse Numa-Logic “PICO” brick in years! Wouldn’t have thought they were still available under the A-B brand.

[JayH]

Originally Posted by Goofy Foot

This has became way complicated.

It has. I thought I was asking about a common problem that surely someone had solved before.

Originally Posted by Goofy Foot

If you have a refrigerator and a freezer with independent compressors and set points you could use a smart relay, which is a relay with a limited PLC in it. You would wire your thermostats to the inputs, the outputs would be wired to the compressor start relays. Then write a small piece of logic that would lock the non-running compressor out while the demand compressor starts or do nothing if it is running. If duty cycle is problem you could have a timer that would turn the compressor off for a given amount of time. You could also set up a short cycle timer or use one of the spare inputs to shut down the compressors if the battery voltage drops, alarms, lost of options.

Something like this.
EZ Intelligent Relays

Thanks for the link. Those relays look like they can handle way more complex situations than mine. I think this is the way to go rather than dealing with multiple relays and timers.

I haven't gotten much valuable (to me) feedback about soft starting / motor control / start capacitors, so I don't know if addressing the problem by reducing the startup load is a viable option.

[daddle]

Considered your problem for a few minutes...logic was my career until I went sailing a few years ago...unless there is some pre-packaged device you have a considerable amount of work to do to solve the problem properly. The input signals from the thermostats need to be differentiated into START and STOP signals and synchronized with a clock to avoid race conditions......etc...

An easier solution might be to only let one compressor run at a time. This will work well if neither runs for very long AND their total run time is less than full time. Use a simple relay in the thermostat loops to have one, the master, turn off the other. If they both run for 20 minutes an hour, or a few hours a day, or 5 minutes on and 15 minutes off, this will be simple and effective. I don't think the typical fridge really cares if there is a 10 minute delay in cooling while the master runs. Or if it's a holding plate system a few hours.

Maybe?

[Goofy Foot]

Originally Posted by JayH

I haven't gotten much valuable (to me) feedback about soft starting / motor control / start capacitors, so I don't know if addressing the problem by reducing the startup load is a viable option.

Depending on the inverter and motor combinations it might work but it is very hard to predict in-rush current/ voltage drops and time during random semi-simultaneous motor start ups and if the inverter could recover fast enough before it tripped.


GF