PDA

View Full Version : Kirchoff Law resistances



Peter_1
21-12-2003, 12:58 PM
Posted this in another thread:

Same story for heating resistances in evaporators (each year, we find installations from our competitors with this fault)

Situation: a lot of heaters in an evaporator + flexible drain heater and the starpoint isn’t connected to the common (neutral we say) on a 3 x 400V net. Suddenly, one heater fails and then the show begins. Always good for someone…. we in this case.

Did you encounter also this fault where colleages didn't connected the starpoint (on a 3 x 400 V +N net of course)?

Are there other countries where 3 x 220 V is still used like in Belgium?

DaBit
23-12-2003, 11:02 AM
During normal operation of a 3-phase heater in 'star' mode, the neutral is not carrying any significant current. The only situation where current flows into the neutral is when the three heating elements are not balanced. Thus, in the case of 3 good heating elements it is not necessary to connect the neutral.

If one of the elements fail (open connection) and the neutral is not connected, there will be 400V across two heating elements in series, reducing power dissipated in each of the remaining heater elements somewhat (200V per heater instead of 240V, thus remaining capacity is about 70%. Total heater capacity thus drops from 100% to 46%).

If one of the elements shorts, the situation is worse; the power dissipated in the two remaining elements will rise, leading to failure.

Peter_1
23-12-2003, 03:04 PM
And if you have 5 heaters or 11 or 13 like we have in several evaporators?
What will happen then you think.
An evaporator with 3 heaters (total) will be a small evaporator.

DaBit
29-12-2003, 10:02 AM
Well, the elements are always in series or parallel so it can be reduced to the three-element problem.

Peter_1
29-12-2003, 10:19 AM
Originally posted by DaBit
Well, the elements are always in series or parallel so it can be reduced to the three-element problem.

Not correct in my opinion.

If you have 13 resistances in an evaporator (4/4/5) and one of the 4, there is a burn out, you become 2/4/5.

This is what we call un unbalanced net where the greatest voltage will be over the smallest resistance. You will have more then the 240 V allowed voltage over the remaning resistances and I can assure, it don't take long to burn also the others.

Agree?


Peter

DaBit
29-12-2003, 11:00 AM
Agree, but basically this is the same failure scenario as I described in the 3-heater example. It is only less severe since there is still resistance left.

But really, I don't see why heaters should use the common/return line. It only increases cabling and connector cost. I would prefer the delta interconnect scheme.

Peter_1
29-12-2003, 05:31 PM
Originally posted by DaBit
Agree, but basically this is the same failure scenario as I described in the 3-heater example. It is only less severe since there is still resistance left.

But really, I don't see why heaters should use the common/return line. It only increases cabling and connector cost. I would prefer the delta interconnect scheme.

I think I have to clarify a little better.
Together with my son, I've done a setup in Multisim, a realy great and marvelous software package he uses very often.

Setup: a 3 phase source/380VAC and connected 12 resistances of 100 ohm. (4/4/4)
In each line an ampmeter and over the resistances 3 VOLT meters.
In one line, we placed 4 switches (the upper 4) so that we can switch off each resistance separately.

You can see the corresponding currents and voltages going up and down on the different displays.

Another second setup with an unbalanced wiring of the resistances (4/3/2), very common in the field!

We toke a printscreen each time after opening a switch and pasted it in Word. Because the file is too big to post it here, I will post it elsewhere so that you can see whats happening regarding voltages and currents.

To be honest, I realy don't know anymore how - it's too long ago- you can calculate this manualy via Kirchoff and Thévenin. I wonder how this was/is done.

Second question concerning the suggested delta connection:
If we have a 380 V net as here in Belgium, you only can connect a multiple resistance evaporator in star, delta is impossible. Because as far as I know, you only can find resistances in 220/240 V versions.

So my conclusion: the neutral must always be connected as soon as ther are more then 3 resistances or unequal resistances.

Peter.

You can find the pictures at http://users.skynet.be/sky71808/3.htm (Hope it works, it's almost 1MB big)

Posted them in March aso on www.tld-nv.com/peter/3.htm

DaBit
29-12-2003, 08:10 PM
Well, without inductances and capacitances, it is simple. At a joint, the sum of currents is 0 (I1 + I2 + I3 .. +In = 0, Kirchhoff). The voltages are the same (U1 = U2 = U3 .. = Un).

Then we have Ohm's law, which describes the relationship between voltage, current and resistance (U=I.R, I=U/R, R=U/I)

Next thing to do is to set up the set of equations, and solving them. When interested I can scan a few pages from a textbook and mail them.

Peter_1
31-12-2003, 12:50 PM
Someone an idea what I can do to post the drwing with the Multisim printscrees definitive elsewhere?
I will need to remove this perhaps from my limited web space.

Dabit,

I also have schoolbooks which describe different cases, but always with DC sources but none with AC sources from which the amplitudes are additional shifted.

The problem becomes more complex because the voltages over the resistances is not always equal.

I searched my old schoolbooks and the only thing we did was exercises with different DC sources.

If you can provide info with AC sources, it's always welcome.

But the problem was very quickly solved with Multisim.
Perhaps the disadvantages of computers and calculators, you think less than before.

Peter

DaBit
01-01-2004, 03:37 PM
Originally posted by Peter_1
[B]Someone an idea what I can do to post the drwing with the Multisim printscrees definitive elsewhere?
I will need to remove this perhaps from my limited web space.

I have plenty of webspace at my disposal, so you can mail them if you want.


I searched my old schoolbooks and the only thing we did was exercises with different DC sources. If you can provide info with AC sources, it's always welcome.


My schoolbooks also describe the AC cases. I am currently still busy moving to my new apartment, but as soon as I have the scanner up and running, I can scan & mail some pages.

In case you have access to a university library, look for books on 'netwerktheorie' (dutch). Mine are from Ir G. van der Zee.



Perhaps the disadvantages of computers and calculators, you think less than before.

Not entirely true; since you have to bother less with the details, you can move the thought process to a higher abstraction level. Which is also valueable.