Power supply mcb's ?

[hutchison]

hello all, im piecing together a cabinet for an nc controlled spindle, i need some help with choosing the right MCB's for internal transformers.

The supply voltage I will be using is 110VAC.
this input voltage to the cabinet supplies three main items amongst others;
1. 13.3A 110AC~130DC servo drive
2. 110VAC~5VDC @ 4.4A switching supply to power various interface cards.
3. 110VAC~12VDC @ 2A switching supply to power a spindle speed conversion card i.e; step and dir to 0-10 analogue volts.

what im trying to figure out is what is the best type of MCB i.e; type b? ,c , d, k? for each item that is going to draw power from the mains input to the cabinet.

#1 servo drive,
This has a 13.3A draw on the mains @ 110VAC, it produces up to 130VDc @ 12A (max 24A momentary!) output direct to a brushed servo. The output side has thermal and overcurrent trips to the motor and variuos other bits so no worries on the output side.

power supply #2 has an inrush current of 20A, the manufacturer recommends a 2A fuse for input protection. I will be fusing the individual boards that draw power from this 5VDC Psu using standard cartridge fuse rated just above each boards draw.

power supply #3 has an inrush current of 15A from cold, the manufacturer recommends a H3.15A fuse for input protection. Again I will be fusing the output here , i'm only drawing 1 output @ 200ma.

So.. i've been advised by a friend of mine to use mcb's on the input to these things, now i need to know which is the best type for each of the devices 1,2,3.

I was going to use a type 'K', 16A mcb for the servo input, this any good?

I was going to use type 'D' mcb's for the transformers, not sure on what ratings i would use.. any help / explanantions greatly appreciated

adam.

[gar]

080330-0739 EST USA

hutchison:

Yesterday I had not the slightest clue what you were talking about.

Today it is only slightly clearer.

Apparently an MCB is either a fuse or circuit breaker.

Any fuse or circuit breaker that is thermally based, meaning it functions based on the temperature change of some element in the device, has a trip time characteristic where the trip time is inversely related to current.

Some circuit breakers are magnetic only, meaning the trip current is relatively independent of time. Others are magnetic with some sort of dash pot, a mechanical shock absorber, to provide some time relationship. And there are breakers like Sq-D QO series that are basically thermal, but because of the current path thru the breaker have a magnetic effect for very high currents.

Devices like transformers and motors can tolerate substantial current over their nominal rating for a short time because of their thermal mass. Semiconductor devices like diodes are very much less tolerant of excess current.

For transformer and motor applications you want slow blow type devices. For power supplies with a capacitor input filter, most are of this type, there is a large inrush current and you also need a slow blow device. However, the rectifiers must be selected to tolerate this inrush. In all these case you may be able to use a NTC thermistor inline with the circuit protector to reduce peak inrush current. However, this is a power waste under steady state conditions. NTC refers to negative temperature coefficient.

Fuses and circuit breakers are not usually very precise devices and their operating point is a function of ambient temperature for thermal devices. You should probably size these to about 1.3 times nominal expected full load current. However, you might have to go higher based on the inrush current.

I have not specifically answered your questions but this should give you some background.

.

[Al_The_Man]

Adam, MCB's are a bit more of an expensive way to go over fuses, Most of the wiring components I use are Telemecanique, marketed here by Group Schneider.
If I use an MCB I use the C60 series, I believe these are both magnetic and thermal, and they come in B,C or D modes, which is the instantaneous trip point, I have not heard of K type.
The 3 magnetic modes are:
B curve = 3.2 ~ 4.8 times Amp setting
C curve = 7 ~ 10 times
D curve = 10 ~14 times
As far as I know the thermal trip occurs at 120% of rated value, and the instantaneous trip occurs at the BCD value.
Being made in France, they should be available over there.
If I need a fast blow protection, I normally go with a suitable fuse.
Al.

[hutchison]

Gar, yesterday I had far less clue what I was talking about than you, which is why i asked, today I also, am only slightly clearer, thank-you for the help!, its just what the doc ordered.

Thanks, I have been reading about mcb's and power supplies and what ratings are advised so i shall try and reply as best i can. When i mean MCB's i mean a breaker with which i can reset. I do not want to use fuses as although they can be replaced easily they cannot be replaced as easy as throwing a switch and thus no more parts are needed if I select correctly. The input side will generally not change(or will it? see below*), but I may add things onto the outputs or change them at a later date so I can easily swap a fuse rating then. Also the outputs from the small power supplies are small draws, up to 500ma, so an MCB is not readily available at those low ratings. That is the logic that made me choose the mcb's, is there anything wrong with that, please let me know.

telemec' are readily available over here al, although i am finding ABB breakers are cheaper about $14 / £7 for each of the ratings (i think) i want. I just choose a type K breaker al, as thats about the best explanantion in radio spares i could relate my application to, What i was trying to figure out was if 16A was o.k to put on the servo drive 110VAC supply side:

http://uk.rs-online.com/web/search/s...duct&R=5140147

if i can tackle the two power supplies first.. surely if the manufacturer recommends 2A for the 5vdc supply, and a high blow (H3.15A) for the 12vdc supply, i should just use them ratings in C type mcb's right?

*one other thing i read, if i can draw your attention to the first paragraph, please, will this affect my set up?:

"Sizing of Fuses & Circuit Breakers;
So that the rating of the fuse or breaker can be determined, the nominal input current of the power supply needs to be established. If the application has more than one power supply or other mains powered equipment these will need to be taken into account.

To determine the input current, we need first to determine the input power and, in AC systems,

remember to take into account the power factor.
Input Power = Output Power / Efficiency
Input Current = (Input Power / Input Voltage) / Power Factor

Choose fuse or CB rating at least 1.5 x Input Current - Time Lag

It is advisable to use a time-lag fuse or breaker to avoid nuisance tripping on start up. The 1.5 x input current rating is to overcome the ageing effects of fuses. Fuses are often labelled FF, F, T, TT (ranging from super fast to long time lag). For power supplies it is recommended that time lag T or TT types are used to prevent nuisance tripping.

Circuit breakers are often rated A-K (very fast to long time delay). For power supplies, C or above would be recommended.
"

[gar]

080331-1823 EST USA

hutchinson:

You wnat to use a circuit breaker instead of a fuse for your AC input protection to have a resettable device. This makes sense if you expect many overloads, also might be true if if you want to use this as a manual on-off switch.

I have used a P&B slow trip magnetic breaker on a power supply that we use in an industrial environment. When designing this supply a goal was to have the nominal rating of the breaker somewhat above the maximum steady state rating of the power supply transformer. I believe the transformer was about a 160 volt-amp rating, and the primary was 120 V. 1.33 A or so. To prevent tripout from turn on inrush we would have needed something like a 15 A breaker. Totally unrealistic as this would provide no useful protection. We used a 2 A breaker and an NTC thermister to limit peak inrush. With this there are no false trips from turn on. However, if you turn it off and immeadiately back on it will trip. This is because the thermister is still hot and a low resistance. Waiting maybe 15 to 30 seconds is enough to allow successful turn on.

I suggest this might be a useful direction for you.

A more expensive and interesting approach is based on an invention of mine done for Mechanical Products about 45 years ago. Mechanical Products at that time was a manufacturer of hot wire type thermal breakers. Probably most were for aircaft applications. The design for this electronic breaker was to tolerate large inrush currents yet have a very precise trip limit. We could take an inrush 100 times the nominal setting of the breaker and yet trip 1 % over that nominal rating.

A normal incandescent lamp may have a peak inrush about 10 times its steady state current. A transformer may range from less than this to something larger depending upon the core material, air gap, and the residual flux state and input voltage and polarity at turn on.

A suitable breaker for transfomer protection may not help much for your diodes. Also you need to decide if you want protection for minor steady state overloads relative to the transformer rating.

Besides transformer inrush current you have the capacitor load.

The low cost approach will assume you are protecting against short circuit problems, rather than minor overloads.

Let us consider your 13 A load. Assume 1 cycle, 20 MS, peak inrush of 150 A. Looking at one set of SQ-D QO curves we might need to consider 150/5 = 30 A as a suitable breaker. If we can knock that inrush down some, then a 15 A breaker may be realistic. Assume 170 V peak from a 120 V sine wave, then to limit to 34 A we need a 5 ohm resistance, or 2.5 ohms for 64 A.

Looking at some data see
http://www.ametherm.com/Inrush_Curre..._datasheet.htm
The SL32 5R020
Has a 5.0 ohm room temperature resistance,
is rated for 20 A,
is 0.034 ohms at 20 A,
is 0.082 ohms at 10 A.

At 10 A the voltage drop is 0.8 V and lost power is 8 W.
At 20 A the voltage drop is 0.7 V and lost power is 14 W.

Might be something to consider for experiment.

After a time delay you could short the thermistor with a mechanical relay contact.

See if any of this makes sense.

.

[hutchison]

i do understand a little but i want the solution to be a simple affair, no ntc's or extra relays if possible. i think i need to read up some more for now, or just stick with plain simple fuses!, 2 amp probably high blow etc, and a H3.15A for the small supplies.

As for the servo drive supply side, and those 'in rush limiters' , do they go in line with the live feed, simply?.

also, would a mains filter have any bearing , can i put a mains filter in-between transforrmer and servo drive

[Al_The_Man]

Originally Posted by hutchison also, would a mains filter have any bearing , can i put a mains filter in-between transforrmer and servo drive

I would be inclined to put the filter on the primary side of the transformer.
Transformers themselves offer some filtering, fitted to the primary side would offer any PWM noise being imposed back on the supply line.
Al.

[gar]

080402-0649 EST USA

hutchison:

A normal input noise filter will have essentially no effect on inrush current.

See http://www.electronics-project-design.com/fuse.html
for a general discussion on fuses.

The following is a nice discussion based on a quick glance:
http://www.your-book.co.uk/fuses.htm

Look at these curves:
http://cooperbussmann.com/pdf/4fd38e...063efd6287.pdf

.

[hutchison]

Originally Posted by Al_The_Man I would be inclined to put the filter on the primary side of the transformer. Transformers themselves offer some filtering, fitted to the primary side would offer any PWM noise being imposed back on the supply line. Al.

if i put a filter on the primary side will it not keep all the noise in the cabinet as you mentioned before? i think it would be better keeping it in between servo. it is almost un-avoidable for me without buying a huge bare bones transformer which is quite expensive. I am having to use an external transformer which feeds the whole of the cabinet through a 110v inlet port then into the 2 small power supplies and servo drive. I could not find a 2000w 240-110AC bare transformer for cheap enough, and my cabinet is not really big enough to house such a large number, so i will be feeding the cabinet with something like this: and hopefully filtering only just before the servo drive itself, so it does not interfere with the other power supplies, and the mains supply.

http://www.airlinktransformers.com/t...ransformer.asp

and hopefully filtering only just before the servo drive itself, so it does not interfere with the other power supplies, and the mains supply.

i'll take a snapshot which will make things a lot clearer, i'm probably not explaining best, thanks for the help so far

thanks for the links gar, i will look through those.

[Al_The_Man]

If your System is fed from one common transformer, which is the way I read your post, then If anything, I would put your Filter, if it is L/C type, directly on the input supply to the CNC Controller, whether it be PC based or other.
Al.

[hutchison]

here you go, i hope it explains my hopeless ramblings. by the sounds of your post i've placed the filter in the right spot, when it arrives. i'll have to move it all around a bit i have fuses to fit in yet for the small DC PSU's outputs, and plenty of cables to find a clean space for.

[Al_The_Man]

I see you have set up for a 110vac supply, Do you have a 240/120v isolation transformer? Being as you are posting from UK.
If so, are you setting up a local Grounded neutral at all?
I noticed you have used Din rail Grounding terminals, these are handy for some terminations, but I still prefer to set up a common ground bus/plate.
All my shields and common grounds go to this.
Do you have an external PC controlling the system, is it fed off of the 110vac?
Al.