Hotwire foam saw wiring

An _ordinary_ step-down transformer becomes quite inefficient at below-normal primary voltage. Core losses soar, and the transformer tends to overheat, shortening its life -- sometimes to "immediate death".

It would be simple and pretty inexpensive to build a "dimmer" for use on the low-side, so you don't vary the transformer input.

LLoyd

That, and the step-down transformer windings don't like seeing the chopped waveform from the switching transistor in the dimmer, and the transistor isn't thrilled about the big inductive load either.

High-watt wirewound rheostat, ceramic body. 25W and 50W readily available. You can also cut the power output of the secondary in half rather crudely with a few diodes to form a half-wave rectifier

Or get a Variac (variable transformer) and connect the secondary directly to the hot wire. Might want to plug into a GFCI receptacle for personnel protection, just in case someone touches the hot wire and a ground at the same time.

-->--

The one I built used a toy train transformer, we had some adjustable- tapped high-powered wire-wound resistors around, one of those in series with the hot wire did the job. The hot wire came from some other wire-wounds we had kicking around.

Stan

Core losses increase with lower primary voltage? How can this be, with constant frequency and lower flux density?

Ordinary light dimmers use triacs, which may not produce symmetrical AC output pulses. This can result in there being a DC component which can saturate a transformer primary.

This is pretty much a matter of the particular dimmmer and transformer in question. If it works in your case and nothing overheats, there's no reason not to do it.

Other possible approaches: use a dimmer circuit on the primary that uses an alternistor rather than a triac.

use a variac to control primary voltage. They're expensive new, but do show up as surplus.

if your hotwire current is less than 40 amps you could use a triac dimmer on the secondary side. 40-amp triacs are cheap and readily available.

For higher current, rectify the secondary and use PWM with suitable MOSFET's. 100+ amp MOSFET's are cheap and readily available.

You beat me to that question Don...Lloyd might just be on his way to discovering the solution to our world's worsening energy shortage.

A buddy from high school daze (a non techie) was bugging me about building his own heated wire foam cutter for some artsy projucts of his.

I suggested he get one of these Weller "hot knives" which cost about $20, which he did and sounds pleased with.

That just might be another way to skin the OP's cat (Pun unintentional, sorry.)

Jeff

As the core flux density diminishes below an optimum level, coupling to the secondary diminishes _rapidly_, causing much higher core loss per secondary VA than when operating at optimum. Specific silicon steels for transformers have different specific optimum operating flux levels and different saturation levels -- as do thick lams vs. thin lams. And of course, DC currents in either pri. or sec. coils mess add to the formula.

Since he's effectively trying to control secondary current - which won't vary tremendously from "too hot" to "not hot enough" for foam cutting, he's wasting more watts in the core per VA in the cutting wire.

That might or might not overheat the xfrmr... just depends on how conservatively designed it was. Of course, if he's trying to cut the secondary output WAY down, it won't hurt a thing.

LLoyd

I should have added that imposing too _much_ primary excitation is worse. When the core begins to saturate, core losses increase asymtotically to the vertical.

LLoyd

SORRY TO CHANGE THE THREAD,---BUT!!Some foams emit POISONOUS GAS WHEN HEATED !!! think it's the Urethanes--but not sure--if you see any compount like isocyanates,or similar--they emit CYANIDE

You sure sound like you know what you're talking about Lloyd, but I always thought the optimum flux level was determined by designing the transformer so that the sum of copper losses and the core losses at the intended operating power level was minimized.

Can you point to a cite which backs up what you say about specific "optimum operating" flux levels (not "stauration levele") for different silicon steels?

Jeff

I have no experience with multi-KVA transformers, so if you do then I'll defer in that area. In sub-KVA xfmrs operating below saturation, coupling coefficients are a fn of geometry with excitation level being a second-order factor at most.

I'll be honest and say I don't even have the literature any more, and don't have the perfect memory of what I once knew about transformer design by rote.

I was the senior lab technician in the prototyping and product testing facility for Florida Transitron (1971). We made everything from subminiature audio and 400Hz power transformers, to ferro-resonant rigs, to toroids, to multi-KVA cans with 3/4" square primary wire. My lab designed and built the test rigs that would push the envelope for every specification on the prints. I remember one of toughest ones was an EMF spec. for an IBM small can transformer. We had to have special mu-metal cans drawn, and small-geometry cermet headers fired to get within the spec.

But I can't cite the specifics for the steels anymore. I can't even remember the alloy names and numbers.

So I'll decline to argue with anyone who's memory is fresher or more recent than mine.

LLoyd

Ah, the essence of diplomacy. We can all feel like we are right!

Peace,

Jeff

So do cigarettes!

For that matter, there are not too many things that do NOT emit poisonous gasses when heated enough.

Cheers Trevor Jones

The DC component will not do the transformer any good. The best idea is the Variac (see below) Bill K7NOM

Cheap and dirty.... put some light bulb sockets in series parallel and in series with the wire. Screw in bulbs as necessary with different wattages to get the heat desired.

John

"Optimum flux level" depends on the design goals for the xfmr so there's no single answer for that. Economy and efficiency are not always the only or even most important parameters.

Coupling coefficients depend upon geometry and upon the relative permeabilities of the linking path (thru ferromagnetic material) and leakage paths thru air. The higher the permeability of the material, the tighter the coupling. Most ferromagnetic materials display monotonically decreasing permeability with increasing flux density, but there may be exceptions. A material with very high residual induction (square loop materials) may also exhibit greater transfer efficiency with higher induction, but such a material would not be used in a sinusoidal power transformer due to hysteresis losses. They have been used in inverters and Royer oscillators because, while xfmr losses are higher, switching losses in the semiconductors are significantly reduced.

1. if it works and meets your needs, proceed.
2. older dimmer circuits were easily destroyed by the back voltage from an inductive load, which is why they say "lamp only" - the ones used to run small motors though should be immune.
3. you really want to regulate current in the wire, or better yet temperatuer. If you measure the resistance of nichrome, it's temperature sensitive, and by remembering a variant of ohms law, Power = V^2/R, you will see that holding a constant voltage on the wire would give you constant temperature. This is best done by regulating output voltage. High current DC power supplys are cheap because no one uses the old ones any more - a vairable voltage 0 to 5 V power supply with 30 or 40 amps of current is likely to cost you well under .

This thread has got very technical but not really answered the original question.

If I've understood it properly Greg has successfully used a conventional power transformer fed from a domestic light dimmer to control the power input to his hotwire saw. He has seen some comment that this is not a good idea and wants to know why.

I think the simple answer is that this method is a simple, convenient,and power efficient method of power control. Providing the transformer accepts the light dimmer output waveform without overheating there is nothing wrong with it.

The downside is that light dimmers use triacs and are not always designed to cope with a transformer load. The two bad possibilities are a residual DC component in the output waveform or a the triac failing to recover properly at all dimmer settings resulting in jerky control or remaining permanently "on".

Residual DC increases the peak flux density with some increase in iron loss. However the DC component is usually pretty small and, unless it is a significant fraction of the normal magnetising current is unlikely to be a problem. One type that may be difficult is toroidal transformers - these mostly have a very low magnetising current and are unduly sensitive to DC component.

The inductive tranformer load gives less time for the triac to recover at each half cycle of the input power and with older, slower triacs it may remain "on" when it should be off resulting in jerky or failure to control control. Most modern dimmers will behave OK - it's just a matter of luck - suck it and see!

Jim