0
\$\begingroup\$

I have a very expensive 500va 1:1 toroid transformer with to isolate workbench from ground. But I'm afraid to use it for fear the winding may burn up when something shorted. I'd like to know what is the behavior when the secondary shorts. Would it heat up faster than the primary winding able to draw current and tripping the primary breaker? What value must be the primary breaker, and can you use secondary breaker too (and what value)?

\$\endgroup\$
  • 1
    \$\begingroup\$ Can you make a thermal cutout switch with epoxied thermistor set to 60'C ? Or would you prefer a resettable PTC for a conservative holding current on both primary and secondary? The heat generated should be equal if using same AWG and high mutual coupling. The PTC ought to heat up faster than the ferrite \$\endgroup\$ – Sunnyskyguy EE75 Jan 17 at 1:38
  • \$\begingroup\$ I just want it to trip if any of my circuit shorts without the windings opening in the secondary of the isolation transformer. \$\endgroup\$ – Jtl Jan 17 at 1:40
4
\$\begingroup\$

Overcurrent does not damage a transformer, it's over temperature that's the problem.

When a transformer output is short-circuited, there's a large increase in current in both secondary and primary windings. If this is allowed to continue, it will eventually overheat it, and destroy the insulation. However, it will take many, many seconds to reach a dangerous temperature, plenty of time for a fuse or breaker to respond, so it can be completely protected by using a suitable fuse or breaker.

It's more of a problem to protect against 50% overcurrent, as a fuse that's going to blow at 50% above nominal will run very hot, and may nuisance blow on switch-on transients. A time delay fuse is essential to resist the switch-on transients.

A breaker can be set more accurately than a fuse, and is a better bet if you also want protection from small amounts of overload.

Small overloads will take 10s of minutes to overheat the transformer, so the best protection against small overloads is the use of a thermal breaker attached to the transformer.

\$\endgroup\$
  • 1
    \$\begingroup\$ you don't say whether you're on 120 or 240v. However, even on the lower current 240v, with a 95% efficiency, your S/C current ought to be at least 20x your rated current, so >40A at 240v, will easily trip a 16A breaker. However, that only protects against a S/C. What happens with a 5A or a 10A load? Neither will trip the breaker, both will overheat the transformer. The 5A load should take long enough for a temperature switch on the transformer to operate, but 10A might be too fast, You ought to have a fuse or breaker of lower value, 5A perhaps, time delay to avoid switch-on transients. \$\endgroup\$ – Neil_UK Jan 19 at 6:47
  • 1
    \$\begingroup\$ You'll have to look at the efficiency rating of the autotransformers for each tap point, and work out the worst case. They are tricky as their efficiency and s/c varies by which taps you take the output from. Part of their advantage is that they 'reuse' part of the winding for both primary and secondary, which complicates the specification. Or do you mean variable autotransformer? When that's set to a very low voltage, you can easily destroy the brush and output winding without drawing much primary current, an output fuse is essential for those. \$\endgroup\$ – Neil_UK Jan 23 at 9:46
  • 1
    \$\begingroup\$ As transformers get smaller, they get less efficient, it's just the way that the geometry scales. However, even at 100VA, I'd hope it would be efficinet enough to blow a 1A primary fuse on a secondary short circuit. Try it with a 5A primary fuse. If it will take that out, then a 1A should be OK. Again it's whether the speed of heating gives time to protect the transformer temp rise with a thermal switch or not, if you want to protect against small overcurrents. Do you have something against secondary-side fuses? They are cheaper than an accident. \$\endgroup\$ – Neil_UK Jan 23 at 10:16
  • 1
    \$\begingroup\$ EGC goes to the chassis of the powered device, regardless of whether the supply comes direct, through an isolation transformer, or through an autotransformer. Secondary side fuse is more necessary for stepdown autotransformers, whether fixed or variable, as the low voltage side has a current rating rather than a power rating. For a 2:1 fixed stepdown, the argument is more moot. \$\endgroup\$ – Neil_UK Jan 23 at 15:50
  • 1
    \$\begingroup\$ depends on secondary load. If big caps after a rectfier, then still might need slow blow, if filament lamp then definitly slow blow. I thought we'd agreed that 16A breaker was excessive and you were going to use 5A? It depends whether you want to protect from low overcurrent through overtemperature, or just from short circuits, and allow the transformer to melt down with a small degree of overcurrent. Note if current is only 2x rated, ultimate temperature rise will be 4x rated. \$\endgroup\$ – Neil_UK Jan 23 at 21:29
2
\$\begingroup\$

Primary needs a slo-blow due to excitation current with worst case remanence and no-load excitation currents 1/2 cycle.

Secondary depends on surge current of loads like diode bridge caps so it also needs to be a slow blow for rated current.

This is to prevent false fuse failures.

Alternative protection could be an ICL slow start for rated current with fast blow fuses.

a 500 VA torroid will have a Zo= <5%pu so short circuit current is 20x rated current.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.