Online assistance for electrical trade people in New Zealand and Australia Login  |  Register  |   Forgot Password
Assistance for electrical trade people
 

 

 

 


Click here to send Forum Admin a pdf document for publication on this Topic

Documents must be less than 200k in pdf format

Posted By Topic: Transformers Humm

TEX
Aug 09 2012 14:37

On connection of a 240VAC to 18VAC the transformer starts to humm. No load on the output side of the transformer.

Eventually after 5 minutes the transformer has \"burnt\" out. Smells and is warm.

Continuity between N + G

Circuit is 1.5mm TPS from a lighting circuit.

When circuit closed

P -> N 234VAC
P -> G 232VAC
N -> G 0V

When circuit open ( but other lights still

P -> N 2.5V
P -> G 2.5V
N -> G 0V


   

me
Aug 09 2012 18:08

poor quality transformer, wots your application - security alarm?

   

Dave.I.
Aug 09 2012 18:16

Yep. If you\'ve got no load on secondary, and correct voltage on primary (did you have the primary fused?) Then it sounds like a faulty transformer.
   

TEX
Aug 09 2012 18:43

Yes security application.

Been through two BRAND NEW transformers. Never had an issue like this before. Tried it with the security panel plugged in and same deal. Hummm from onset and then after about 5 mins it is dead.

Yes the phase is fused inside the panel. Running off a lighting circuit with a 10amp breaker and three double fluros. The circuit was existing.

Thinking there might be a backfeed or something through the neutral via one of the fluros.

Got replacement tranny coming tomorrow and gonna run it off a plug in flex to determine if circuit fault.


   

me
Aug 09 2012 19:53

where are you from tex, if your in my part of the woods i have heaps of trannies i could give ya. even some tried and proved ones.


   

Colin
Aug 09 2012 21:11

Hi Guys.
Perhaps you could have a tank circuit due to the inductive/capacitance of the fluro\'s (under/over corrected power factor and the inductance of the transformer all being on the same circuit?
   

Apprenti
Aug 09 2012 21:20

Ferroresonance??
   

TEX
Aug 09 2012 21:30

Colin can you explain more????

I have spoken to a few people about this and they all suspect something to do with the fluros, but what exactly no one can explain.
   

me
Aug 09 2012 21:37

Tex, alarm system are commonly installed on lighting circuits, and therefore obviously fluros are encountered all the time, i have had maybe 4 trannies fail over the 30 years i\'ve been in the security industry, and all were due to either prolonged low mains voltage (network issues) or lightening. never from your situation. if fluros caused this issue, it would be a wider issue. get ahold of atlas gentech and get a tranny from them.
   

TEX
Aug 09 2012 22:45

Thanks \"ME\"

This is the first I have seen such an issue and thought i would ask if anyone else had a similar issue. Grabbed some replacement trannys from CSP this arvo so will do further testing tomorrow.

At first I thought it was a tranny fault since the panel had been sitting around for awhile, but then when the second one blew in the same manner I knew there was something else to it.

The building was recently rewired to a new meter and consumer number (Building split from single meter to 4 separate meters) which leads me to believe there might be some issues either in the earthing to related.

   

Silly
Aug 10 2012 08:16

question but are you 100% sure you have the primary and secondry the right way? Wouldn\'t be the first time its happened
   

E14143
Aug 10 2012 12:29

Transformers of that size just dont burn out.some will if faulty.I have heard of some trannies that require at least 25% FLC connected at Idle to stop over heating(strange but true).Talk of Fluos being down-stream seens odd also.Transformers do not like DC content,being transformers they will replicate input to output.For an Alarm Guy to carry an oscilloscope should be as easy as emptying the ash tray of coins and buying one.Check incoming Wave before you wave on another Tranny.run Tranny at home before client second guesses your ability.(Smile)I was reading posts about EX boxes.I will help where my ability can.Cheers.Brian H
   

Frostybird
Aug 10 2012 15:11

Would be interesting to see what primary current was with seconday unloaded. Also current with and without flourescents on to see if it changes. Also look at waveform with lights on and off.
   

me
Aug 10 2012 18:21

been thinking on this, was the first transformer that burnt out, new or had it been installed for a while at this location?. I\'m assuming your connecting the primary and secondary around the correct way ?
   

TEX
Aug 10 2012 22:07

First transformer newly installed.
Primary and secondary correctly installed. Secondary wires are like twice the size as primary and labelled so no way could have got that wrong.

Have not managed to get back to site today to check. Will be measuring current on secondary and primary with the load when I get a chance. If a load on the secondary is shorted then it in turn could burn out the primary, although in saying that, the humming was also occuring when the second transformer was plugged in with NO load.

   

me
Aug 11 2012 09:08

measure the output voltage
   

Rob
Aug 11 2012 09:16

Thinking aloud.

With no load on the secondary there is no current in the secondary winding and therefore it will produce no flux. The absent secondary flux would otherwise be in opposition to the primary flux (Lenz\'s law) and tends to desaturate the core. With open secondary terminals and no flux the core will quickly saturate. The potential (induced EMF) in the secondary winding will be much higher (albeit with no current flowing), perhaps higher than the insulation resistance rating of the secondary winding.

This scenario would point to either poor design or poor selection. Perhaps the transformer was designed to be within an appliance where the primary is switched rather than the secondary.

A suitably rated resistor placed across the secondary in parallel with the switched load will allow some current to flow at all times thus keeping the secondary voltage within design limits.

Now where\'s my pocket protector......
   

ElroyL
Aug 11 2012 22:50

I\'m going to try to elaborate a bit on what I think Colin and Rob are talking about.

CIVIL.

Capacitive Current leads the Voltage.

Inductive Current lags the voltage.




Explanation:

In an AC circuit with a resistive load, and an inductor and a capacitor connected in parallel with the supply, the capacitive reactance and inductive reactance (1) from both devices respectively, alter the characteristics of current flow in that device, creating a current flow that is out of sync with the voltage - this is called \'reactive\'(2) current, which is generally unwanted (1). The two devices will pass the \'reactive\' current backwards and forwards between themselves.

Power factor is when you try to balance the inductive and capacitive reactance, by selecting or adding whatever device is needed to bring the reactive current flowing back to your lines company to as little as possible.


From my understanding, the closer this is to Unity power factor (1), the faster the transfer of the reactive current, which as I (kind-of) understand, can create dangerously high voltages across the supply, which is why we only power factor correct to 0.95, not 1.0.


E.g. like holding two tanks of water, with a hose connecting them, and lifting one up and down, which with gravity causes water to flow from one to another - the faster you do this, with \'momentum\' if you like, builds up, you get the timing right, and the pressure on the water can become quite high, it starts to get quite violent in the transfer of water, like two sea waves becoming a larger one, the same occurs with voltages across the inductor capacitor circuit increasing to dangerous levels.


(1) and if not close to being balanced, passes this reactive current back to the lines company, requiring larger cable sizes, etc, which they don\'t like

(2) this current does no real \'work\', so is unwanted

(resistive load plus reactive load equals higher cable capacity)

Sorry guys, appreciate feedback on this, got a bit lost as you can see.
http://en.wikipedia.org/wiki/LC_circuit
   

Rob
Aug 12 2012 09:30

...perhaps ditch the tranny and use a switchmode power supply instead.
   

me
Aug 12 2012 16:37

rob, most alarm systems have switch mode power supplies inbuilt on their motherboard, which should assist greatly with the tranny, i would suspect that either the trannies are of very poor quality or the tranny has been connected around the wrong way, as these trannies are normally designed to last for a very long period of time, not run hot, and run 24 hours a day. if there was something wrong at this site, then everything electronic or wire wound would be burning out. i would suspect that it would be unusual to connect one of these trannies around the wrong way as the primary normally has brown/blue connection leads and the secondary has two yellow leads for direct connection into the alarm PCB. if you connected one of these trannies around the wrong way i could expect the humm & short burnout time that has been discussed. These trannies also dont normally humm at all.
   

Colin
Aug 12 2012 21:51

Hi Guys,
Thanks for the technical explanation Elroyl,Couldn\'t have put it better myself,while on the subject of fluro\'s and power factor I have in my garage at home approx 10 fluro\'s (mix of single and doubles)that while they were switched on and then the 1050 Hz ripple pulses came through the 6 amp lighting MCB would trip.It could not be reset until the ripple pulses had finished.Loading on MCB is 4 amps,swapped it with another 6 amp MCB from house swb-problem gone.
cheers Colin.
   

AlecK
Aug 13 2012 09:13

I fancy Rob\'s theory over the resonance one. But experimentaion will tell which, if either is correct.

Most fluos these days have their (400 V)PF correction in series, so trying the tranny on a different circuit would remove that influence.
Unlike (230 V) parallel-connected fluo caps, which used to cause havoc by absorbing ripple signals. Some lines companies had / have rules about connecting parallel - corrected fluos for this reason.

Incidentally, most security systems are non-compliant, because no segregation between ELV and LV. Yes they are equipment, and probably no SDoC required; but licensed persons are (or should be) connecting the LV wires. and signing a CoC to the effect that what they\'ve connected is electrically safe and compliant with ESRs & Wiring Rules.

Have seen one instance of resonance, we\'d been upgrading old tresonant-start fluos to switch-start so we could use the skinny, lower-wattage tubes while retaining the \"old\" look of the fittings - ran out of whatever size PF caps we\'d been using, so used a different on on the last fitting. Must have struck PF = 1(or close to it) because as soon as it turned on the cable insulation in that fitting turned to smoke due to overcurrent, but the mcb never tripped.
   

TEX
Aug 14 2012 07:36

Transformer replaced. Connected to a flex and to a standard power point socket and no humm. Fitted back into the cabinet and back into the circuit and the things starts humming again. Quickly switch off all the lights in the circuit one by one until the hum is gone. Isolate to one fluro fitting. Dropped the fitting out of circuit and the unit is working as it should. Will need to do a bit more testing to the fitting to see where the problem is. Thanks to all that replied.