This page describes the mechanical and electrical type aspects of building the roaster. The electronics and programming are on a separate page.
The oven came from ebay, it's actually a Home-Tek HT459, apparently with a
retail price of 59 pounds. Mine is a graded return, apparently - I think
that means someone returned it, possibly faulty, and it has been checked before
being sold cheap.
In truth, the fixings looked like it had been taken apart a
couple of times, and slightly bodged back together where the self-tappers had
torn out too much material. Actually, it would have been fine for someone
that just wanted to use it as the manufacturer intended.
This is particularly suited to my purpose, because it has a griddle on the
top ...
... which is actually just a removable plate sitting in a hole in the top of
the oven. In use, it is simply heated by the oven elements just below it.
So I take out the lid and griddle and modify them slightly (with nibbling
tool and dremel) to take the elements
from a convection roaster (also from ebay).
The oven itself has no insulation - the bottom and back are just a thin sheet of steel. There's an air gap between the oven inside wall and outer case to each side, but apparently only to contain the electrics and door mechanism.
Therefore, I decided it would want some more insulation, so incorporated some 6mm cement-bonded board, which is a more-or-less fireproof board which is workable with hand tools (hardened saw, tungsten tipped tooling for routing). My plan is to provide a 25mm air-gap all round the oven chamber, with an outer enclosure of this board. The board is also damp-proof and rot-proof, which is handy, because the roaster will in due course live in an unheated barely weatherproof outbuilding.
The convection roaster top is assembled around the modified griddle and lid,
with addition of a cement board.
The roaster top has been rewired to take out the thermostat and timer. The holes will need plugging in due course. The fan and elements are each now powered individually, though with a common neutral. That is, the wire into the unit now has an earth, live to elements, live to fan and a common neutral. These will be controlled from elsewhere - there's no switching or thermostat in the roaster.
The neons are still connected - the 'power' is now a 'fan on' and the
'heat' is still 'element on'.
The oven electrics are all in one end.
On the front panel (top to bottom on the left of this picture) is:
The rotisserie motor is obvious in the middle, on a long stand-off,
presumably to try and keep it somewhat cool.
A quick circuit diagram - laid out with components arranged roughly as
they are in the oven.
The switch is arranged to sink current from the other components - meaning that even when not selected, all the elements and the rotisserie motor are live (which surprised me - I'd have expected them to be isolated when not in operation). They are isolated when the timer is off, because the timer is on the live.
The selector switch can be set to operate either upper or lower or both
sets of elements, or the upper elements with the rotisserie motor.
I've gutted all the controls and rewired so I have a wire to each element and
a common off the other end. Due to re-using the high temperature insulated
wires originally in place, my common neutral wire is red.
I've also fitted an internal cement board (with
25mm air gap from the oven inner wall). The cement board is fixed in with
high temperature silicone sealant (from, can you guess? ... ebay).
The drum is from expanded metal mesh, with some aluminium sections and all held together by pop rivets. It's assembled onto the rotisserie fork prongs (after they've had some trimming and bending) so it gets driven round.
All the metal-working is by hacksaw, nibbling tool and file, with bending generally done by clamping in a vice and using an adjustable spanner closed down on the plate to get a grip. The sheet used for the end deflectors is only 0.8mm and bends reasonably by hand. The only power tool used was a drill for the holes.
I've put some inclined baffles / deflectors at the end of the drum. I don't
want to have to deal with trap doors and so on, so the idea is that the drum
is mostly open at one end (for pouring beans in and out), but with deflectors
to encourage the beans to stay in.
This does mean the drum must only rotate one way, and as an oven, the rotisserie motor runs in either direction randomly. I've dealt with that by building a crude ratchet mechanism onto the motor shaft - if the motor starts up backwards (ie, not in the desired direction), the mechanism jams and the motor reverses. This sort of motor does that (reverses if it jams) - I'm just making use of that characteristic.
I want it to handle 250g batches. Like this, it can run with 300g of green
and only shed a few beans. I think I can improve on that with some tweaking
of the bending.
The oven re-assembled. Roaster on the top, a big hole cut in the end of the
case - this is where I will mount my controls. Other than that associated
with the roaster on top and the internal board,
no added insulation yet. That's probably the next building operation.
It was at this point that I did the uncontrolled roast reported on the main
roaster page and described on the
roast notes page.
This is how it's set up for the first controlled tests and roasts - controller
still on blocks of wood, but wired to the mains and connections made to the
mechanicals via the connector strip between them.
The laptop (a 9" screen netbook running openSuSE) has got both the USB programmer (so I can reprogram the controller) and the controller output connected to it.
This is the setup that produced
the first proper roast.
and this is as far as it has got ... tune in again for another thrilling
installment some time.
back to the main oven roaster page
back to the top roaster page
back to Ian's contents page
back to www.astounding.org.uk top contents
page
To comment on anything (please do) email ian@astounding.org.uk