Lighting circuit design 1

[14041972]

I have been feeling lonely over the last few days until I staggered in to this site.
Boy I feel great to meet brilliant professionals like you. I will have to read all the stuff you have dealt with over the years because every sentence sounds valuable. Where have you been for so long? I know this is not a religious forum but God bless you all and may you prosper in all your endeavours.
MY QUESTION
Designing an up-grade for a factory perimeter lighting circuit.
I am supposed to start working on this project by next week but I am stuck.
I work in a manufacturing plant that occupies a big compound lit up by 54 pole lamps each with 400 watts sodium vapour lights.
All of them are switched on at the same time by use of 1-day light switch, 1 relay and 1 contactor.
I feel I can do better than this. I am planning to modify this circuit to work better, more efficiently, easy to fault find and friendly to non-trained users.
Currently: -
- The start up current is very high because all the lights start up at the same time.
- A semi-skilled guy must goes round the perimeter at night to determine whether there are any faulty or fused lights as there is no indication or monitoring system.
- There is no mimic panel to show whether the lights are working.
- I suspect that the power factor for this circuit must be poor (due to the characteristics of sodium vapour lamps) but I am still to confirm this.
For this reason I am planning an upgrade with the following objectives: -
· To divide the 54 lights in to 4 zones (zone 1 to 4) each with 13 lights (2 excess)
· Ensure that the lights in the 4 different zones come on at 4 different stages by use of timers and a day light switch. This will ensure that there is less current drawn from the supply during start up. Remember sodium vapour lamps take about double the normal current during start up but latter it comes down after the light is fully on.
· To have a way of monitoring faulty/fused light in the four different zones. I will do this by measuring the circuit current in each of the 4 zones when all the lights are healthy i.e. (Ah1, Ah2, Ah3, Ah4) – this will be my datum. Then I will use four C.T.s, which will measure four load currents when the lights are operating normally (Al1, Al2, Al3, Al4). The difference between the two (Ah1-Al1= Af1) will be faulty/fused lights current in zone 1. Repeat the same for the other zones. Then I will use a current to voltage converter to interplate how many lights are faulty. The results will be displayed on four different digital display units.
· Design and Install power factor correction capacitors that will bring the power factor very close to unity - if not unity.
· To have a mimic panel that will indicate when various zones are ON or OFF. I will do these using contactors in the 4 zones.
· To perform this control using a PLC.

My question is:-
Is my ambition practical?
Do my thinking show error?
Is there better ways of doing what I am trying to do?
What else can I add to this circuit because I feel it is good but I can still do better?
Your support is greatly valued.
Peter

 

[itsmoked]

Welcome 14.

If your remote indication is necessary then your scheme sounds fine to me.

A few points. I would probably just use three more eyes and skip the relays as delay relays can sometimes be confusing to later people whereas eyes would not be. All you need to do is point the eyes differently or shade each very slightly differently to have the units not start simultaneously.

Is there a PF penalty involved? Does anyone really care about it?

A PLC seems over the top to me....

 

[jraef]

Sounds like a decent approach, although I would not go so far with the monitoring. You can find curent monitoring set-point relay devices that have an over/under function, so you can just set them up to look for a bandwidth of acceptable current, and anything over or under that trips a relay contact that you use in your mimic panel. I also like itsmoked's idea of separate photoeyes because you can make them stagger start via sensitivity settings and eliminate the need for a time clock. One caution would be to have at least one zone come on a little earlier than necessary so that it is not too dark by the time the 4th zone is up to full output.

As to the pf issue, if you have older ballasts with poor pf, I would consider retrofiting them with new ones that have a higher pf instead of spending the money on caps. Some new ballasts are available with up to .9pf and offer higher efficiency so you save both ways.

Eng-Tips: Help for your job, not for your homework Read faq731-376

 

[itsmoked]

jraef; I like that upgrade ballasts idea! I think you can even extend bulb life a little with modern electronic ballasts.

 

[hottip]

14,

What you want is not practical. I would say if you want a practical approach : you can try to reduce the power consumption by adding a power economizer (eventualy by doing this you can reduce the power consumption with 20-30%)

I've been working in plants with a lot (I mean a lot!!!) of lighting poles and lighting fixtures. The only approach accepted by the managment was the economical one (reducing the power consumption). What you are saying is nice to have but for 54 poles is something to ambitious.

 

[HCBFlash]

Hi 14041972, nice to have you here! We all learn from each other you know.

...one thing I learned is that there are disadvantages in letting others figure out the birthdate of your favorite rockstar...
;-D


This is actually a motors forum,,, but PLCs, inrush currents, dynamic PF etc, for lighting sure doesn't have it's own forum.....

Uh, your system is pretty ambitious sounding. If you need to monitor performance, schedule maintenance, and obtain lots of data about such a system, via the internet or some other LAN/WAN, then it sounds very good. (actually pretty cool!)

Massive inrush? Street lighting uses individual fixture-mounted (twist-lock) photocells for that precise reason. But then I'm sure you knew that and ruled it out due to the problems that go with that.

Deal with PF as [highlight]much as possible[/highlight] at the ballasts. 54 fixtures isn't really very many, and if you do a full ballast / lamp replacement / fixture and lens clean one zone at a time you could handle that part as ordinary maintenance costs.
.
.

remember: An opinion is only as good as the one who gives it!

 

[Laplacian]

Slightly off topic, but check this out:

http://homepage.usask.ca/~amn076/lights.wmv

 

[itsmoked]

Laplacian;

Ah man!
That made my day..!
The power of PLCs, a-what?

HCBFlash; Good point there on per fixture PF control.

If.... as has never been answered, there is any reason what-so-ever for worrying about it at the OP's location.

 

[14041972]

Thank you for all the responses. I now know how to continue with this project.
To HCBFlash thank you for the birthday comment…you are full of humour. My apologies for asking a good question in the wrong forum. This is now happening to me the second time. The first time I walked in the city for a long time looking for a place I could buy a kilogram of pork. Afterwards I found a meaty place, got in and placed an order but he chap behind the counter got upset (when I named what I wanted) to a point of ejecting me out of the butchery. I had not noticed that it was a ‘HALAL’ butchery – Muslim suitable.
itsmoked –there is a penalty if the p.f is poor so we try to perfect it but of course within economic reasons. I checked the P.F.and noted that it is lagging at .65.
The reason why I was in favour of a plc is that there is an existing, under-utilised PLC in the same panel - monitoring alarms. So I thought it would be economical to use it since only half of the inputs and outputs are used. I had thought of using different daylight switches but one engineer discouraged me saying that it is wiser to have all control components in one panel so that in an event of a fault you will not have an electrician walking round the perimeter with a step ladder and black plastic bags to simulate darkness.
But I did not understand what HCBFlash meant by ‘individual fixture-mounted (twist-lock) photocells’. We have ordinary lamp posts- with a single pole circuit breaker at the bottom. There are no other controls in the post.
Nevertheless I must tell you that I spoke about the project in our morning meeting today and everyone looked amazed. My boss asked to tell him how I gathered all this information within a very shot time. I feel jealous to tell him the truth.

 

[itsmoked]

14; Thanks for the response. So PF is a penalty for you. Then you might do as jraef suggested and find new ballasts that improve the PF and do a payback calculation. You could also do one based on your 'adding all the caps at one place' scheme. Do check the efficiency improvement that new electronic ballasts would provide.

I would still be leery of using the PLC unless its whole point is other lighting or grounds functions. You do not want greatly different functions running in your PLC. It can cause real problems. Furthermore PLCs are not as serviceable as regular switch gear. If the I/O card chokes on the PLC it could take a week to get a replacement.... No lighting for a few days will erk someone I am sure.

 

[14041972]

itsmoked Did you understand what HCBFlash explained about twist lock. A I mentioned we have normal lamp posts different from the ones they use in the city streets. Unless there is something I need to learn.

 

[itsmoked]

Yeah he was just referring to typical individual street light sensors. In a city it is generally simpler to just have each street light individually controlled. No need to wire street lights together. No single point failure that stops a bunch of lights from operating. And the fact that since component variation, sensor position, dirt, and temperatures will automatically stagger all the street lights from coming on at once. A lot of those fixtures have a twist-lock sensor that makes them easy to repair and test.

 

[14041972]

Good, I thought as much. But in many cities I still see street lights comming on at the same time making me think that they are operated by a centralized timer or a daylight switch. But I get the point.

 

[EEJaime]

Often times when we do municipal street lighting the controls are cascaded. That is that no matter what the power source is for the particular sector of street lighting, the fixtures are energized by means of a control contactor which has it's coil energized by the energization of the previous block of lighting. For industrial facilities, (we recently completed the perimeter security lighting project for a major airport with approximately eight miles of dual fenced perimiter security zones). There are now low-cost, digitally programmable Lighting controllers which will work with any power source. You can zone these anyway you want, you can monitor the zones for load, you can pre-program start, off, weekends, holidays, etc.... Look at Lighting Control and Design, (LC & D) at

http://www.lightingcontrols.com

There are several systems, one of the smaller ones should be sufficient for your project.

 

[HCBFlash]

Sorry I hadn't checked back in lately.

(are you familiar with the term "cobra-head"?)

Many outdoor luminaires have a receptacle for a twist-lock device as part of the luminaire head itself. It's on the top of the unit, and a encased photoelectric switch is plugged-into it, or a "shorting cap" in it's place. Look at a luminaire mfr's website. Hubbell comes quickly to mind.

When I say "photoelectric switch" this is a line-powered relay, so there is a "Hot - neutral - switchleg" wiring arrangement, which makes for a 3-bladed twist lock device. Now you understand about the "shorting cap".

If your PLC is being used for facilities operations (as opposed to a particular machine or process) I don't see why it shouldn't be used. Just be wise about the wiring practices;
- use relays to interface with external power sources
- run control circuits wiring separately, and clearly identified.
- make certain that all panels are marked properly, identifying all power sources, etc.

.
.
.
Maybe you could use the nickname "Porky" after that other incident eh? I'm sure there's some good anecdotes behind the "flash" in my own handle....




.

remember: An opinion is only as good as the one who gives it!

 

[itsmoked]

Cobras are ...um...nice.

They keep the rat population in check!

 

[IRstuff]

itsmoked,

BTW, the video is elsewhere called Wizard of Winter. The controller is from Light-o-rama:

http://www.lightorama.com/

TTFN

 

[itsmoked]

Yeah IRstuff thanks!

I found that sight earlier and was quite surprised by the low prices for the controllers. Really, the software alone is worth using the controllers. Otherwise I would just go with OPTO22 boards and parallel ports etc..