weird lighting transformer issue 1

[xJAyx]

I am working on a lighting project which is using 230VAC through some Lutron dimming modules to toroidal transformers (230:12 VAC).
I have seen some very weird things happenning with the primary fusing on the transformers. Basically if a short occurs on the secondary side of the transformer, it does not always blow the primary fuse. There is a much larger current supplied than the VA raing of the transformer should allow, but somehow the circuit holds at a certain current, which is often not enough to blow the fuse. This happens both when attached to the dimming module, and when the transformer is connected directly to the line voltage. The current limit seems to be somewhat affected by cable length, which I initially thought was a result of voltage drop in the cable, but the voltage drop calculations and measurements don't add up with the measurements. I can't understand what could be limiting the current. HELP!

 

[ScottyUK]

Ohm's Law?

The wiring on the LV side contributes some resistance, and the transformer itself contributes both resistance and inductive reactance. As you only have 12V available to drive current around the LV side it should be clear that a fairly small amount of resistance can have a significant effect on the current which flows.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[xJAyx]

I agree, but what I am seeing doesn't correspond. At no load, the transformer output is 12.6A. At 210W load, with 4.75m of 2.5mm cable, my output at the end of the circuit is 10.49V. Based on this, at 210W I have a total resistance of 0.121 ohms, which should also take into account the inductance of the coil. So with a dead short and .121 ohms resistance I should be seeing 99A. this would translate to 5A on the primary. I don't have the curve for the primary fuses, although I have been trying to get it (if anybody has data for an ELU M2A250VE please post a link or send it to me) but this is 250% of the fuse rating. When I tested I was getting ~60A with the added resistance of the terminal block I was using to short the secondary, this woul dgive ~3A on the primary (both calculated and measured) and I tested it for about 30 seconds (until the terminal block was too hot to continue) at 150% for 30 seconds any fuse should be blowing.

 

[ScottyUK]

Not often I look at clearing times for little fuses, but a quick look at Bussmann's website shows a 2A(T) fuse clearing in 30 seconds at approx 4A. I couldn't find much on ELU fuses either - is ELU a manufacturer or a type?

Have you considered that a real-world fault is unlikely to be as 'perfect' a short as you made for the test?


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[xJAyx]

I am still waiting to get some information on the fuses from the transformer manufacturer. ELU is a brand, and I can find distributors with google, but contact info for ELU or actual data for the fuses.

As far as a real world short, I have been thinking about that situation, and at the moment the only possible solution I can come up with that I am comfortable with is to fuse all the secondaries. The major problem is that there are about 1000 of these on the project and they aren't all that easily accessed.
FYI this is on a megayacht which was started in Taiwan and I am working on finishing it at a shipyard in Croatia.

 

[ScottyUK]

I agree about that being the best solution from an electrical perspective - when lighting transformers were still iron and copper instead of electronic types it was common to see them with one primary fuse and (say) four secondary fuses.

BTW, make a good job of my yacht!


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[zeusfaber]

Random thought: Could you be saturating the transformer core, giving you less current than you thought you were getting?

A.

 

[sreid]

You have the 31 feet of 10 ga. wire which has a resistance of 1 milliohm/foot so 31 milliohms of wiring resistance. That makes the transformer impedance about 90 milliohms which to me seem quite high for a 1/2 ohm load.

 

[xJAyx]

The calculated numbers are backed by measurements on the installed circuits. The load, at 12V is .7ohms, the resistance in the cable is .0164 ohms per meter (manufacturer spec Vdrop of 16.4mV/Am = 16.4mohm/m at 50Hz), so .0771 ohms of wiring resistance.

 

[rtronics]

Are you sure the transformers don't already have a thermal fuse built in with the windings?
If not, then get a 130 degree thermal fuse inserted into the primary winding. This will blow before it catches fire, and is the common method these days.
If the heavy secondary current on SC is a real problem then secondary circuit fuses will be necessary.
Regards, Ray.

 

[xJAyx]

The problem isn't inside the transformer, it's the high current in the secondary as you mentioned. Most of the circuits are 2.5mm^2, unless they were required to be higher for voltage drop.

 

[HCBFlash]

1000 fuses and blocks is certainly a tough situation, but I'm curious as to why secondary fusing using a feeder/branch scheme isn't seen to be viable to you.

While overcurrent device coordination would become involved (only a single level though) a relatively firm and predictable maximum secondary current value can be attained.

While some members positively bristle at cross-posting, you might get some more insights on a different forum. Re-posting this thread might work without "getting shot at".