Can I automatically switch a fridge between 2 sockets?

[jabuzzard]

APC and others do automatic transfer switches that are design for handling single powered IT equipment in racks.

http://www.apc.com/products/resource/include/techspec_index.cfm?base_sku=AP7721

Never actually used one because it is generally cheaper to buy the kit with redundant power supplies in the first place which also protects from accidental disconnection of the power cords. Should be able to handle the load of a fridge.

 

[ivixor]

How about this contraption:

Get something like a mains powered motor with spring. A zone valve from a heating system would do.

Arrange two push-to-make switches of the kind you have to keep held down such that the valve presses one switch when powered, and then presses the other switch when unpowered.

Connect the fridge to both switches, and then you should have a nice delayed changeover to the other circuit when power is lost on the one which is powering the valve.

Of course.. one must take care never to press both switches at the same time...!

 

[bernardgreen]

That looks like a good idea.

I would not use the switches to switch power directly but instead to control a pair of electrically interlocked contactors. That would take care of a switch sticking in the ON position and the then the other switch being operated.

 

[JohnW2]

How about this contraption: Get something like a mains powered motor with spring. A zone valve from a heating system would do. Arrange two push-to-make switches of the kind you have to keep held down such that the valve presses one switch when powered, and then presses the other switch when unpowered. Connect the fridge to both switches, and then you should have a nice delayed changeover to the other circuit when power is lost on the one which is powering the valve. Of course.. one must take care never to press both switches at the same time...!

Yes, that's effectively just a slow-acting relay, but without the implicit 'interlock' one has with changeover contacts. In addition to the risk of pressing both switches, which you mention, the main 'danger' of that system would probably arise if one of the switches stuck in the 'on' position. That problem could probably only be overcome with interlocking which somehow sensed the mechanical position of the switch contacts - since, if power happened to be absent from the switch with the stuck contacts, no electrical interlock would know that the contacts had stuck.

Kind Regards, John

 

[JohnW2]

I would not use the switches to switch power directly but instead to control a pair of electrically interlocked contactors. That would take care of a switch sticking in the ON position and the then the other switch being operated.

As I wrote yesterday, when I tackled this issue the main problem I had was in relation tyo 'power up' situation. How would you arrange the electrical interlocking of the contactors such that if, at power up (as a result of one switch being stuck 'ON') power was coming from both switches, the desired thing would happen?

Kind Regards, John

 

[stillp]

John,

A mechanical interlock would ensure that if both coils were energised simultaneously there would be no change of state.

 

[JohnW2]

John, A mechanical interlock would ensure that if both coils were energised simultaneously there would be no change of state.

Yes, I realise that, but Bernard wrote specifically of an electrical interlock, and I'm very interested in knowing if/how that, alone, can be used to achieve the required functionality (so that the approach can be applied to standard relays, as well as mechanically-linkable contactors). I've struggled in the past with this question, and not come up with any totally satisfactory solution which does not involve time delays.

Kind Regards, John

 

[stillp]

Well, you did ask

How would you arrange the electrical interlocking of the contactors such that if, at power up (as a result of one switch being stuck 'ON') power was coming from both switches, the desired thing would happen?

and my answer was intended to show that I wouldn't rely solely on electrical interlocking.

 

[JohnW2]

Well, you did ask

How would you arrange the electrical interlocking of the contactors such that if, at power up (as a result of one switch being stuck 'ON') power was coming from both switches, the desired thing would happen?

and my answer was intended to show that I wouldn't rely solely on electrical interlocking.

Fair enough - so I presume that, like me, you cannot think of a way of achieving the desired functionality with electrical interlocking alone (without time delays). I therefore hope that Bernard will respond to my question, in the hope that he may have thought of an approach which has so far evaded both you and I!

Kind Regards, John

 

[JohnW2]

John, A mechanical interlock would ensure that if both coils were energised simultaneously there would be no change of state.

Yes, I realise that, but Bernard wrote specifically of an electrical interlock, and I'm very interested in knowing if/how that, alone, can be used to achieve the required functionality (so that the approach can be applied to standard relays, as well as mechanically-linkable contactors). I've struggled in the past with this question, and not come up with any totally satisfactory solution which does not involve time delays.

...and my answer was intended to show that I wouldn't rely solely on electrical interlocking.

... which leads to a supplementary question. Are there any readily-available (preferably '1-module', DIN-mountable) and not-too-expensive contactors available that would allow the required mechanical interlocking? It is the difficulty in finding such animals which has led me in the past to strive (not completely successfully) for fully-satisfactory solutions using only electrical interlocking - and a quick look around now still hasn't identified anything 'readily available and sensibly priced' that is really suitable.

Kind Regards, John

 

[bernardgreen]

Electrical interlocking ( normally closed contact on A in series with coil B and normally closed contact on B in series with coil A ) would work 99% of the time.

If both switches were on when both supplies became available then there woud be a race between the contactors to energise and lock out the other. That could become un-predictable, both contactors energising at the same time would be possible for a transient period.

Using the same interlock on a pair of relays acting as masters to two slave contactors almost certainly would prevent both contactors energising at the same time.

Contactor A coil via series normally open on master A, normally closed on master B and normally closed on contactor B

Contactor B coil via series normally open on master B, normally closed on master A and normally closed on contactor A

Even better would be to use relays that were slow to operate and fast to release for the masters.

 

[JohnW2]

Electrical interlocking ( normally closed contact on A in series with coil B and normally closed contact on B in series with coil A ) would work 99% of the time.

Indeed, that's the simple/basic way to do it - but it's the other "1%" (or whatever) which makes me slightly uncomfortable - mainly because it might worry others inspecting or inheriting the system; it's a level of risk that I personally would probably be happy to live with. [particularly given that, in my present application, both phases are fed to the relays via 3A fuses, one or both of which would hopefully operate before anything nastier happened in the case of the phases becoming 'joined']

If both switches were on when both supplies became available then there woud be a race between the contactors to energise and lock out the other. That could become un-predictable, both contactors energising at the same time would be possible for a transient period.

Exactly - that's the scenario I was asking about.

Using the same interlock on a pair of relays acting as masters to two slave contactors almost certainly would prevent both contactors energising at the same time.

That's what I ended up with - but it is still not foolproof. It's obviously impossible to quantify 'almost certainly', but it is clearly less that 100% - and, again, I am essentially concerned about how others would view it.

Even better would be to use relays that were slow to operate and fast to release for the masters.

Indeed - as I said, if one uses time-delayed relays, it becomes much easier (even without the master/slave approach). However, I wanted to try to avoid that, so that I could use bog-standard relays which I was fairly confident were always going to be fairly easily sourced - with particular view to the day when someone other than myself might be 'in charge' of the equipment.

Per my most recent post, do you have any ideas about 'readily available and not-too-expensive' mechanically-interlockable contactors which would be suitable?

Kind Regards, John

 

[stillp]

Per my most recent post, do you have any ideas about 'readily available and not-too-expensive' mechanically-interlockable contactors which would be suitable?

If I had this problem, I'd use something from Schneider's Tesys D range. Perhaps a little bigger than you'd like but could be fitted on a DIN rail in a suitable enclosure.

 

[JohnW2]

If I had this problem, I'd use something from Schneider's Tesys D range. Perhaps a little bigger than you'd like but could be fitted on a DIN rail in a suitable enclosure.

Thanks. I'll have a look.

Kind Regards, John

 

[JohnW2]

If I had this problem, I'd use something from Schneider's Tesys D range. Perhaps a little bigger than you'd like but could be fitted on a DIN rail in a suitable enclosure.

On the basis of my initial look, quite apart from my other requirements, I'm not at all sure these qualify as 'readily available'! Do you know an 'easy' retail source?

However, in the course of my looking, what I have come across is a plethora of 'on-delay' (and 'off-delay') relays - which, as I've said, might offer an acceptable 'electrical' solution, without the need for a mechanical interlock.

Kind Regards, John