1
\$\begingroup\$

I want to design a cheap but safe, low-wattage, always-on heater to keep a small space above freezing. (A small underground tunnel that the water pipe for an automatic horse waterer goes through.)

I've got NiChrome wire for the element, and I need a cheap way to step down 120V AC to about 50V AC @ 2.3A. (or 25V @ 4.6A or 17V * 6.9A , etc.) I will use a half-bridge rectifier for low-power and a full bridge rectifier for high-power if the pipe freezes and needs to be thawed. I figure a simple wrap-wires-around-a-bolt transformer should work for the job. Efficiency doesn't matter much -- it is supposed to generate heat.

My guess-work assumption is that I need to make the primary coil with enough turns so that its reactance will prevent the primary from burning up. (It is, after all, a wire connecting hot to neutral) Is it solely based on the inductance/reactance of the primary? Reactance + resistance? Reactance * resistance?

What factor must I design to ensure that: 1) The transformer won't burn itself up, 2) The transformer won't be absurdly inefficient or dangerous, and 3) this is a reasonable use of a cheap, DIY transformer. (IOW, if keeping this safe isn't realistic, I can either find a pre-built transformer for this or figure out a different way to provide power to the heating wire. This sounds fun to design, but I'm not too proud to be told not to do it.)

NOTE: For safety, assume that the heating coil powered by the secondary will eventually burn out and become an open circuit. And assume that this will happen but not be noticed for >2 weeks.

\$\endgroup\$
9
  • 2
    \$\begingroup\$ Why not simply size your nichrome wire such that it produces the desired amount of Watts from 120VAC vs. messing with a transformer? \$\endgroup\$
    – jwh20
    Jan 6, 2021 at 17:16
  • \$\begingroup\$ You should be able to buy a cheap 24vac transformer, they are really common in CCTV systems. \$\endgroup\$
    – Ron Beyer
    Jan 6, 2021 at 17:19
  • 6
    \$\begingroup\$ Don't build your own mains power transformer. Just don't do it. Given that you feel you can use a bolt as the magnetic core, even more reason not to do it. Don't do it. \$\endgroup\$
    – Andy aka
    Jan 6, 2021 at 17:27
  • 1
    \$\begingroup\$ Given the "wrap-some-wires-around-a-bolt" type of approach, simply buy a transformer that is rated to be safely used in the temperature and moisture levels you are going to use it outdoors, even if it is in an underground tunnel. It will also contain overtemperature cut-off and is rated to handle some abuse. You don't want it to become dangerous. \$\endgroup\$
    – Justme
    Jan 6, 2021 at 17:29
  • 2
    \$\begingroup\$ A bolt as the core shows that you know enough about electromagnetism to be dangerous, but not enough to be safe or effective. Buy a transformer. Or buy mains operated pipe-heating tape. \$\endgroup\$
    – Neil_UK
    Jan 6, 2021 at 19:47

4 Answers 4

1
\$\begingroup\$

The reactance will not be increased significantly by increasing the number of turns on the primary

A common way of making a transformer short-circuit proof is to reduce the coupling between primary and secondary to increase the leakage inductance. This may involve the way the laminations are arranged or even put the primary on a separate limb of the transformer with a shunt path between the primary and secondary.

Increasing the number of turns on the primary will increase the magnetizing inductance and reduce the no-load current.

\$\endgroup\$
1
\$\begingroup\$

My guess-work assumption is that I need to make the primary coil with enough turns so that its reactance will prevent the primary from burning up.

No. The leakage inductance will be what you get, it won't prevent an overcurrent.

What factor must I design to ensure that: 1) The transformer won't burn itself up

You add a circuit protection breaker designed for transformer protection.

  1. The transformer won't be absurdly inefficient or dangerous

If you buy a manufactured one it won't be inefficient. When you order it, look for safety transformer. It has more insulation between primary and secondary, so if one gets melted it won't make a contact with the other. Voltages below 34VAC are safe for direct touch.

  1. this is a reasonable use of a cheap, DIY transformer

The transformers are cheap, but it is a task for a specialist. I can't never build one transformer bettter and for the same price as it does a specialist, with all tools for potting, vacuum extracting air bubbles, certified insulation material, epoxy resins, ....

As for your application, there are much simpler solutions. For example a semiconductor heating element. It has two wires, between them there is a PTC heating material, so self regulating. It can be shielded, so you can use a direct mains AC with ground fault interrupter GFCI and you're done. This is how industrial water pipes are maintained. An environment temperature thermostat can turn the heating on/off.

https://youtu.be/XzPMk2DXhME

\$\endgroup\$
1
\$\begingroup\$

I figure a simple wrap-wires-around-a-bolt transformer should work for the job.

Nope. Google "Eddy currents". A big solid piece of iron isn't suitable for this. You'll just heat up the bolt really, really hot.

I've got NiChrome wire for the element, and I need a cheap way to step down 120V AC to about 50V AC @ 2.3A. (or 25V @ 4.6A or 17V * 6.9A , etc.)

You can buy heating ribbon that's designed to wrap around pipes and gently heat them. There's no need to roll your own here.

If you do feel compelled to do so, use a long enough length of thin enough wire so that it's just right when you put 120V across it. If you want more heat, then use two sections, one for warm, one for hot.

But that ribbon is better, because it already works, already has a plug on the end, etc.

I will use a half-bridge rectifier for low-power and a full bridge rectifier for high-power if the pipe freezes and needs to be thawed.

Attaching a half-bridge rectifier to a power circuit that comes straight off the mains is a bad idea. Your power company assumes that your house is an AC load, that doesn't shove DC currents back at the pole transformer, A half-bridge rectifier does exactly that.

If you're bound and determined to use your nichrome wire idea, then connect it to a lamp dimmer that's rated for the full load the wire will present. Then just turn the dimmer down for "warm".

\$\endgroup\$
3
  • \$\begingroup\$ +1 for the ready-made heat tape. Should be safe, with no danger of shock if someone touches it. \$\endgroup\$ Jan 6, 2021 at 21:39
  • \$\begingroup\$ The ready-made heat tape is notoriously unreliable. Every time I've tried to use it, it burns out within a few months. My crazy heater ideas come from the fact that I want something more reliable than heat tape. \$\endgroup\$
    – SvdSinner
    Jan 8, 2021 at 16:34
  • \$\begingroup\$ If you aren't already, try ready-made heat tape from a commercial supplier. I'd look to Granger, MSC, or McMaster in the US, or a commercial plumbing supply house (as opposed to a home store). Plant maintenance folks aren't going to put up with that, and will be willing to spend the boss's money on something that costs twice as much (or ten times!) but lasts forever. \$\endgroup\$
    – TimWescott
    Jan 8, 2021 at 17:28
1
\$\begingroup\$

You can remove the HV secondary of a microwave oven transformer and wind a secondary to use it for this purpose if you want. Roughly you'll need about 50 turns on the secondary (cut and try) for 50VAC (about 1V/turn so lower voltages may be easier to wind). The downside is that you'll need thicker wire for the higher current if you go lower in voltage.

You can find various tutorials on the net, I'm sure some are better than others. Be exceedingly careful with anything connected to the mains, and involve someone local who is knowledgeable and is willing to help you before you wire it up to the power.

The advantages of the MOT is that it already has a properly designed primary and core that is more than adequate for your output power.

As Neil says, the disadvantage is that it's made very cheaply and will run very hot and is assuming it will be fan-cooled. That can be mitigated by rewinding a primary with perhaps 30%-50% more turns on it (or, more easily, a smaller number of, properly phased, set of turns added in phase with the primary), but that's more work. So it looks very good compared to a bolt with wire on it, but very bad compared to a proper transformer designed at conservative flux levels using grain-oriented silicon steel.

\$\endgroup\$
4
  • \$\begingroup\$ The advantages of the MOT is that it already has a properly designed primary and core that is more than adequate for your output power. The disadvantage of a MOT is that it is designed to be as cheap as possible, drawing huge magnetising current and running hot with fan cooling. it is not the thing to run continuously, or into a 100 watt load. It's strictly for rewinding for spot welders, kW load for short periods. \$\endgroup\$
    – Neil_UK
    Jan 6, 2021 at 19:44
  • \$\begingroup\$ @Neil_UK True enough, ideally you'd re-wind the primary with more turns. \$\endgroup\$ Jan 6, 2021 at 21:06
  • 1
    \$\begingroup\$ I tend to recommend adding a dozen or twenty extra turns to the existing primary, no need to totally rewind it. \$\endgroup\$
    – Neil_UK
    Jan 6, 2021 at 22:02
  • \$\begingroup\$ @Neil_UK Good point .. of course the added turns have to be phased properly and kept properly isolated from the secondary winding. \$\endgroup\$ Jan 6, 2021 at 22:04

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.