www.TK560.com Forum Index www.TK560.com
Vacuum Forming, Movie Prop, Sci-fi and GIjOE Forum
Log in to check your private messagesLog in to check your private messages

Log inLog in  RegisterRegister

Profile  Search  Memberlist  FAQ  Usergroups
Conventional (IR) Oven Design

Post new topic   Reply to topic    www.TK560.com Forum Index -> Vacuum Forming & General Stuff
View previous topic :: View next topic  
Author Message

Joined: 04 Sep 2006
Posts: 705
Location: Austin, Texas

PostPosted: Sat Oct 07, 2006 6:06 am    Post subject: Conventional (IR) Oven Design Reply with quote

I explained some basic oven design issues in the "Hobby-Vac -- Construction Plans" thread:


I made a small error.

Thinking about it, I think it may matter some whether the side walls of a reflective box oven are shiny, rather than just reflective. (However, aluminum flashing seems to be shiny enough to do a pretty good job.)

The basic idea is that you should think of Infrared "heat" as light, and think about reflecting light off of shiny (mirror-like) surfaces or matte surfaces (like a flat white wall). After all, infrared "heat" isn't really heat at all---it's light but at a slightly lower frequency than we can see. When objects absorb it, rather than reflect it, they turn it into real heat. (Mechanical motion of molecules bouncing against each other.)

As long as you use surfaces that are very reflective, you can think of the problem of distributing infrared "heat" as a lighting problem.

A matte white wall actually reflects a bit more light than a typical mirror, but since it scatters the light, you don't get an image out of it, just spread-out illumination.

The side walls in a simple box oven are not being evenly illuminated from all directions. They're being illuminated mainly from the back (coil) side of the oven, and the light is traveling more or less toward the front side of an oven.

With mirror-like (shiny a.k.a. "specular") reflection, light hitting the surface is reflected at the same angle to the surface, flipped over. So, for example, if light is traveling 30 degrees to the right of forward, when it's reflected it will go 30 degrees to the left of forward. In this picture, the asterisk is the emitter (heating coil) and p is a point on the plastic, and the vertical line is a shiny wall:
*   |
 \  |
  \ |
  / |
 /  |

From point p's perspective the light that is emitted 30 degrees to the right of forward will appear to be coming from 30 degrees to the right, from a mirror image of the light, through the wall:


*   |   * <- apparent position of element, reflected off wall
 \  |  /
  \ | /
  / |
 /  |

(The heating element is radiating in all directions, so there's also the obvious direct path from * to p, and a bunch of other paths that don't end at p---i.e., they directly or indirectly hit other points on the plastic.)

This principle works just as well with infrared as with visible light, so the mirror images will provide heat when the infrared is absorbed. You can heat things indirectly just like you use indirect lighting.

This principle applies at all angles, so if you have a row of several heating elements and a mirror-like side wall, you'll get mirror images of them, too. If the wall is perpendicular to the row of elements, the row will appear to continue past the wall. If you have two opposed mirror-like walls, they'll go on indefinitely due to reflections of reflections.

So if we actually have this (with four emitters and two shiny walls):
              |               |
              | *   *   *   * |
              |               |
              |               |
              | ------------- |

it will "appear" to the sheet of plastic at the bottom as though there is a continuous row of emitters coming from way to the left and extending way to the right:


*   *   *   *   *   *   *   *   *   *   *   *


(I've only shown the emitters that the plastic can "see" by direct paths or a single reflection here; due to reflections of reflections, the row should go on forever in both directions.)

This is why a shiny-sided rectilinear box with evenly-spaced elements can provide very, very even illumination.

If we have a shiny back wall (opposite the heating elements from the plastic) like this:

               ---------------   <--- shiny back wall
                *   *   *   *    <--- heating elements
                -------------    <--- plastic

the plastic will also see the elements reflected in that, as though they were in the same direction but further away. (The same distance past the wall that the actual elements are from the wall.)

                *   *   *   *    <--- apparent position of
                                      reflected elements                             
                *   *   *   *    <--- heating elements
                -------------    <--- plastic

And those reflections will also be visible through the shiny side walls, so it will "look" to the plastic like there are two rows of elements, one further than the other, that go on indefinitely in both directions:

*   *   *   *   *   *   *   *   *   *   *   *

*   *   *   *   *   *   *   *   *   *   *   *


Now I'll get back to the issue of shiny (smooth) walls vs. matte (reflective-but-not-shiny) walls.

If the back wall is matte, that's all to the good. Rather than seeing distinct images of the heating elements reflected in the back wall, the plastic will see an evenly-illuminated rectangle. This rectangle will be repeated by the side walls, so there will appear to be a huge uniform wall going on indefinitely in all directions. That provides even more even lighting than a row of discrete points or bars going on forever. But as long as the plastic is significantly from the elements as the spacing between them (or further), it doesn't really matter.

(Since the reflections of the back wall appear even further than the direct images of the heating elements, it won't do a whole lot of good to smear them---if the plastic is close enough to the heating elements that the elements are too far apart to illuminate evenly, you'll have more of a problem with the direct illumination than the reflections, and a matte back wall won't fix that.)

The side walls are a bit different. They're being illuminated from the back (coil) side of the oven, both directly and indirectly through reflection off the back wall. You don't want them to scatter that IR light in all directions, including back toward the elements and the back wall. None of these surfaces are perfectly reflective, so if you bounce the light around too many times before it hits the plastic, you will lose some efficiency. You want most of the light to hit the plastic directly or after one or two bounces.

So it's good for the side walls to be shiny---but how shiny?

We're not trying to make images here, so we don't have to reflect things exactly. If the light paths are randomly perturbed a little bit, it doesn't matter. A little bit of light might not hit the "right" point on the plastic because it bounced off at a little different angle than it came in, but on average that's okay because some other bit of light will be randomly directed there, when it should have hit somewhere else.

It averages out, mostly, as long as the light coming in is mostly reflected at roughly the right angle.

What that means is that you want something shiny enough that you can see very blurred and smeared "images" in it. If you hold it at an angle to a light source, you should be able to see that light source as a bright blob.

Aluminum flashing is that shiny---it's got a satiny finish that is more shiny than matte, so it should do fine. (And in my little Sunbeam grill over-and-under, it seems to do great.)
Back to top
View user's profile Send private message Visit poster's website
Display posts from previous:   
Post new topic   Reply to topic    www.TK560.com Forum Index -> Vacuum Forming & General Stuff All times are GMT - 6 Hours
Page 1 of 1

Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum

phpBB "skin" by DewChugr

Powered by phpBB © 2001, 2005 phpBB Group