Parallel Supplies

[Knobhead]

I know it’s a rotten question.

DW would you run an unequal parallel feed?
I wouldn’t. But economics claws its way in to everything.




Until later in my working life accountants didn’t enter the equation. Tony wants, Tony gets.

I went in to engineering planning and procurement. Tony wants, Tony can get stuffed! It’s not coming out of this month’s budget!


BTW if anyone thinks of going in to procurement DON’T. You’re the whipping boy for both sides!

 

[Darkwood]

With the nature of my jobs Tony i very much doubt id be ever in a position of needing to consider wiring unequal parallel cabling/loading but if it did ever crop up i find absolutely no issue with it as long as loading on each core is monitored respective to its rating..... but if your in a position to be asking the questions regarding whether its allowed or not then saying no is often the simplest answer as Appendix-10 and even 11 will highlight to the OP with the calculations and considerations required to ensure a safe set-up.

 

[Engineer54]

With the nature of my jobs Tony i very much doubt id be ever in a position of needing to consider wiring unequal parallel cabling/loading but if it did ever crop up i find absolutely no issue with it as long as loading on each core is monitored respective to its rating..... but if your in a position to be asking the questions regarding whether its allowed or not then saying no is often the simplest answer as Appendix-10 and even 11 will highlight to the OP with the calculations and considerations required to ensure a safe set-up.

I can't think of any situation, where i would be happy to allow parallel cabling of differing sizes, unequal lengths, and/or of differing cable construction. God only knows what would happen in the future as additional loads etc, creep into the equation.... As far as i'm concerned it's an absolute No, No!!

 

[Darkwood]

I can't think of any situation, where i would be happy to allow parallel cabling of differing sizes, unequal lengths, and/or of differing cable construction. God only knows what would happen in the future as additional loads etc, creep into the equation.... As far as i'm concerned it's an absolute No, No!!

As a rule im in agreement it was just to explain it isn't in fact against regulation just comes with a lot of calculation and protective arrangements to be considered.

Any parallel cabling system is fundamentally at a disadvantage regarding overload protection of the separate cores as a number of issues can arise to create unequal loading but namely faults, i would never implement an unequal loaded system into any design without it been a last resort and all other options gone whether it arises from variable size cores or differing paths/lengths ....in a nutshell i probably never will but its to be noted it is an acceptable method when done correctly.

CT monitoring of each core with full linked 3ph tripping is one way to go, this give a higher degree of protection than the standard equal size/run parallel system would have so in reality it might not be the normally chosen method but is in fact better protected, they are other way to protect unequal loaded systems that IMHO fall short of my comfort zone but they still are acceptable within the BS7671 and BS7769 will give guidance in correct calculations.

Again i'm not promoting the system or trying to nudge the OP he might be able to use such a set-up just that the general response to say its a big NO NO! isn't true its just 99% of us would never entertain employing the use of such a method but it shouldn't be said that it should never be done.

 

[Engineer54]

As a rule im in agreement it was just to explain it isn't in fact against regulation just comes with a lot of calculation and protective arrangements to be considered.

Any parallel cabling system is fundamentally at a disadvantage regarding overload protection of the separate cores as a number of issues can arise to create unequal loading but namely faults, i would never implement an unequal loaded system into any design without it been a last resort and all other options gone whether it arises from variable size cores or differing paths/lengths ....in a nutshell i probably never will but its to be noted it is an acceptable method when done correctly.

CT monitoring of each core with full linked 3ph tripping is one way to go, this give a higher degree of protection than the standard equal size/run parallel system would have so in reality it might not be the normally chosen method but is in fact better protected, they are other way to protect unequal loaded systems that IMHO fall short of my comfort zone but they still are acceptable within the BS7671 and BS7769 will give guidance in correct calculations.
Again i'm not promoting the system or trying to nudge the OP he might be able to use such a set-up just that the general response to say its a big NO NO! isn't true its just 99% of us would never entertain employing the use of such a method but it shouldn't be said that it should never be done.

Come on now, CT protection monitoring of each core of a parallel circuit?? Now who in their right mind would ever employ such a method to get over two or more unequal parallel cable sizes!!!! Apart from anything else, it'll be a very expensive fix, and probably beyond the capabilities of many modern day electricians, outside of the industrial sector. There are unconventional ways around many things, but would anyone really go to those sort of lengths, ...i don't think so!!

The use of differing cable sizes in a parallel circuit is for all intent and purposes, is and has always been a complete No, No!!

 

[Darkwood]

Yes probably overkill E54 but it would do the job and on a large scale upgrade of existing could actually be the most cost effective method but as i said i was just throwing it out there that it isn't against reg's if calculated correctly - but agree beyond your average Electrician out there so yes following the rule of thumb always same csa, length and route.

 

[SPARTYKUS]

Interesting thread chaps.

 

[Engineer54]

Yes probably overkill E54 but it would do the job and on a large scale upgrade of existing could actually be the most cost effective method but as i said i was just throwing it out there that it isn't against reg's if calculated correctly - but agree beyond your average Electrician out there so yes following the rule of thumb always same csa, length and route.

Why would anyone, least of all a Design Engineer leave a parallel circuit in place with differing CSA on a large scale upgrade, that would be defeating the whole point of the upgrade.

I know what you're saying, but as i said earlier there are always way's to get over an existing problem, but 9 X out of 10 the additional costs and hassle just aren't worth not doing the job properly... lol!!
If the reg's condone such an installation, i haven't come across it. I'll have to have a look later today, if i have time.

 

[Darkwood]

Why would anyone, least of all a Design Engineer leave a parallel circuit in place with differing CSA on a large scale upgrade, that would be defeating the whole point of the upgrade.

I know what you're saying, but as i said earlier there are always way's to get over an existing problem, but 9 X out of 10 the additional costs and hassle just aren't worth not doing the job properly... lol!!
If the reg's condone such an installation, i haven't come across it. I'll have to have a look later today, if i have time.

I agree and see why your reaction to my original comment now, im jumping between forums and was on auto pilot the variable sized cores comment was relative to single cores in parallel across phases; if i remember rightly you are required to maintain equal csa on the same phase but can in certain set-ups have other phases of different csa (although not as a standard install of a 3ph supply with linked O/L protection across the phases). I didn't mean variable sizes between the same phase.... relative to 2 or more multicore cables in parallel then this doesn't apply to that senerio as yes equal csa required deffo....

It was merely an extreme example to back up my suggestion that parallel systems do function correctly and safely with unequal load sharing if designed carefully and protected properly.

This i got carried away here and went on a non-constructive tangent but ill see if i can reference what i mean another time as its time for the pit.

 

[Knobhead]

Can we stop the squabbling.

Equal parallel feeds = Good
Unequal parallel feeds = Not good, but with proviso’s acceptable.

I ran a 240mm² PVCSWA to bolster two 0.3 PILCSWA’s. Not ideal but it got our arse out of the fire. The PILCSWA terminations had been done with wax (only time I’ve ever seen this). They were so hot they were spitting wax out of the termination.

Needs must at times!

The switchgear catching fire is another story :smilielol5:

 

[Spoon]

Can we stop the squabbling.

Booo. The post count is only 50... we have lots more to go..

 

[Engineer54]

Can we stop the squabbling.


quabbling.
Equal parallel feeds = Good
Unequal parallel feeds = Not good, but with proviso’s acceptable.

I ran a 240mm² PVCSWA to bolster two 0.3 PILCSWA’s. Not ideal but it got our arse out of the fire. The PILCSWA terminations had been done with wax (only time I’ve ever seen this). They were so hot they were spitting wax out of the termination.

Needs must at times!

The switchgear catching fire is another story :smilielol5:

Are we squabbling?? I didn't think that we were...

Are you saying this set up was actually left in place as part of the factory installation??

 

[Darkwood]

This aint squabbling this is pub banter... no - one gets hurt or ridiculed unless they pinch my beer! :8:----:veryangry2:---:13:

 

[Knobhead]

Are we squabbling?? I didn't think that we were...

Are you saying this set up was actually left in place as part of the factory installation??

For many years. I said it wasn’t ideal but it got things cooled down a bit.

0.3”² = roughly 210mm². I had ½ a day to get the new cable in along with a new GEC 1200A ACB and two GEC 1600A bus section switches in to a SWS switchboard.

Must be nice to work with all new equipment that never gets over stressed. Over time things get up rated and you don’t get to know until things start cooking. One compressor on this MCC was uprated from 300HP to 450HP a couple of pumps 60HP to 120HP. We were getting a quart out of the original pint post.

As a said it must be great I do new installs only.

 

[Engineer54]

For many years. I said it wasn’t ideal but it got things cooled down a bit.

0.3”² = roughly 210mm². I had ½ a day to get the new cable in along with a new GEC 1200A ACB and two GEC 1600A bus section switches in to a SWS switchboard.

Must be nice to work with all new equipment that never gets over stressed. Over time things get up rated and you don’t get to know until things start cooking. One compressor on this MCC was uprated from 300HP to 450HP a couple of pumps 60HP to 120HP. We were getting a quart out of the original pint post.

As a said it must be great I do new installs only.

Believe me, new installs can just as easily go pear shaped too!!

 

[Knobhead]

I’ve got to admit I’ve enjoyed keeping things going and doing it on the cheap.

As for new stuff going wrong, a brand new Brush switch board was delivered. I refused point blank to have anything to do with it. ⅔ of it was scrapped before a single cable was connected, replaced by GEC gear. All the bus-bars replaced in the bits we did keep. It was a cock up on a monumental scale.
That was over 25 years ago, I still have a giggle about it.

 

[Chr!s]

The two differing impedance of the cables will sort out the current sharing, its protecting the the cables against fault where the problem lies.

 

[Darkwood]

The two differing impedance of the cables will sort out the current sharing, its protecting the the cables against fault where the problem lies.

This is the case for smaller cables but once you go large you have a lot more than measured impedance to consider as the reactive component of the impedance becomes a major player and positioning/arrangment and distance apart all become crucial to the effects of load sharing and need to be implemented in design to the best we are able.

 

[Chr!s]

This is the case for smaller cables but once you go large you have a lot more than measured impedance to consider as the reactive component of the impedance becomes a major player and positioning/arrangment and distance apart all become crucial to the effects of load sharing and need to be implemented in design to the best we are able.

Measured Impedance?

Impedance Triangle R,Z and X