On 16/03/2010 15:59, NT wrote:
On Mar 16, 10:50 am, Dave wrote:
On 10/03/2010 21:15, andrew wrote:

At work we have a 100A 3ph supply with several buildings being on separate
phases and only the workshop with 3ph. If the flats are the same and given
that there may not be enough peak capacity to supply the immersion and all
the flats at once; is there a device that would switch off the immersion if
the total load on the 3ph supply exceeded 80A and then only switch it back
in when it fell below 50A in any one phase?

Not really, no. By convention, the solution to your problem is to go to
the DNO and ask (pay) for a bigger supply to meet your demand.

Implementing automatic load-shedding on this scale would involve a
control panel manufacturer building you a bespoke solution, which would
consist of a set of current transformers and an energy analyser to
monitor current. The analyser would need to have presetable alarm
contacts which would operate a relay when the preset current level was
reached in any phase. The relay (along with some front panel override
controls, etc.) would operate a contactor which cut out the supply to
the heaters.

I would expect a commercial price in the thousands to implement this.

If you had a building management system (BMS) which was monitoring
demand, you could leverage this system to load-shed.


Or you could get some small scale shop to make one up for several
hundred. The technology's simple. It can be a lot simpler than the
above.

There's nothing complicated about the above.

3 relays with their coils rewound

OK, where are you going to get relays with their coils rewound?

so they switch on at around 50A
current (relay winding across a shunt, you dont want all 50A thru the
relay). You have one of these in each phase feed.
The switch contacts run in series such that you only get power out if
all 3 phases are loaded at under say 50A
And this power line operates a 3 phase relay, or 3 single ones, to
switch the heater feed on and off.

In what way would "3 single ones" be acceptable?


You can adjust switching sensitivity by altering the shunt.
Need to do calcs, this scheme may need tweaking.


NT

An interesting solution. However, in broad terms, your solution is no
simpler than mine. It has comparable design, construction, installation
and commissioning time to my solution and may well (by your own
admission) require more tinker time. It would likely save money on
parts, I agree. However, if you farmed it out to a contractor, the
labour charge would far outweigh the parts costs.

Note that whatever solution you employ, you would need to install a
separate consumer unit for the control and the heater, the problem being
that you need to monitor demand excluding the heater load, otherwise the
system will flap.