Panel building, was Terminating 1mm T&E

[SUNRAY]

In the main I tried to build a panel so one terminal for one wire.

Nice idea... But I do have to ask: How?

I was asked to estimate for a job which came in around £50K then the contract appeared which went into great depth, including one wire per termination. I quickly turned it down.
.

 

[ericmark]

The din rail mounted terminals had link bars so you could have a bank of line and bank of neutral

it did not always work out, I started building this one then found some of the contactors were not of size expected and also one of the drives had an EMC problem and had to be moved out of the main panel. If you look at centre of block

you can see where the connector strip screwed in,

the number of cables were reduced in the panel shown as using ASii to and from PLC, seem to remember up to 32 blocks each with 4 in and 4 out so 256 in/outs with just two cables, yellow was control black was power all 24 volt DC. It has been nearly 20 years since I built panels and programmed PLC's mainly Mitsubishi and I was taught how to use the more complex PLC units but never used Allen-Bradley outside uni.

 

[SUNRAY]

I know all about link bars etc but this:

Was just one of 256 channels (2048 terminals or 10.2m of rail space) in a panel and I cannot see a sensible way of wiring that without doubling up on wires somewhere.
This

Goes some way towards it but it doubles the number of terminals used and that hasn't yes accounted for the 768 (or nearly 4 metres) of terminals for the 0v line at the relays.

No I think I was well rid of that one.

 

[mattylad]

Methinks he meant 1 wire per terminal block, which is achievable esp if you use links and stacked terminals should there be a lot of them.

 

[SUNRAY]

Methinks he meant 1 wire per terminal block, which is achievable esp if you use links and stacked terminals should there be a lot of them.

Possibly but what a silly expense and hard work.

EDIT: Also, but a seperate issue, some of my customers/consultants don't allow stacked terminals.

 

[bernardgreen]

Some cage clamp terminals do not make good connections if there are two wires of different diameter in the clamp

 

[SUNRAY]

Some cage clamp terminals do not make good connections if there are two wires of different diameter in the clamp

View attachment 185252

Oh yes that is a regular problem.

In control panels there is commonly a door interlocked isolator (to prevent the door being opened with power 'ON') but it is common practice to have a control circuit connected to the live side of the isolator. I've often had to fix the fault where a small cable like 2.5mm² is not clamped beside a much larger cable.

 

[Jackrae]

From a servicing point of view, one wire per terminal, with each wire being identified (at both ends) is far superior to a bunch of (usually unidentified) wires fighting for space under one screw or clamp. It all comes down to doing a full design job before starting the build and costing the job accordingly. If the customer wants you to meet a specification and are unwilling to pay for it then a cheap job is all they can expect.

 

[Simon35]

Some cage clamp terminals do not make good connections if there are two wires of different diameter in the clamp

View attachment 185252

Generally easily avoided with double bootlace ferrules or blade/fork lugs used back to back.

 

[SUNRAY]

Looking at the drawings for a random panel here is a precis of the enables and interlocks for the heating pumps, boilers and associated items. For convenience I've colour coded the nunber of terminations 'per wire'. Personally I don't see any sensible way of avoiding 2 wires in a double ferrule without adding silly splitters or sets of terminals all over the panel with all the additional wire runs filling up the trunking.

So how would others wire this without double ferrules?

EDIT to correct colour

 

[SUNRAY]

Generally easily avoided with double bootlace ferrules or blade/fork lugs used back to back.

I agreed except I've always been under the impression one should not mix wire sizes in a ferrule.
When dealing with vast difference in size like 2.5mm² and a 240mm² I'd probably look at a ring tag for the smaller wire.

 

[Simon35]

I agreed except I've always been under the impression one should not mix wire sizes in a ferrule.
When dealing with vast difference in size like 2.5mm² and a 240mm² I'd probably look at a ring tag for the smaller wire.

Hardly a typical example though. Generally I would expect to see back to back lugs for different wire sizes.

 

[SUNRAY]

Hardly a typical example though. Generally I would expect to see back to back lugs for different wire sizes.

It's very, very, common to power the BMS system this way and the wire sizes are absolutely not atypical

 

[Jackrae]

It's very, very, common to power the BMS system this way and the wire sizes are absolutely not atypical

But a decently designed system should have separate connection points for the power cable and the sense or control line. Torquing up a power lug with a lightweight ring terminal as part of the sandwich is a potential (no pun intended) weak spot.

 

[SUNRAY]

But a decently designed system should have separate connection points for the power cable and the sense or control line. Torquing up a power lug with a lightweight ring terminal as part of the sandwich is a potential (no pun intended) weak spot.

It's very easy when the terminations for the supply are bolted lugs rather than clamped. I have some agreement with the rest, however adding an additional joint in a large cable is no mean feat and one which many will do all they can to avoid, due to space and reliability. Specifing 2 feeds, like 1000A 3ph and 16A 1ph, into a panel rarely works either due to isolation issues and when it does the additional [external] isolator seems to be a target for later accidental operations.