Go Down

Topic: reflow (toaster) oven now earning its keep (Read 5357 times) previous topic - next topic

mmcp42

I've been working on a reflow oven
(and hijacking other people's threads to boot)
so now decided to have my very own bragging thread

started with a 1 kW 9litre toaster oven
4 heating elements; 2 top 2 bottom

I added a K type thermocouple
(actually took a couple of goes as the ifrst one melted!)
I took it apart, cut off the smouldering ends, joined the new ends together
used heat resistant (fibre glass?) insulation this time

disconnected all the oven control gear and wired it via a triac
triac controlled by 'duino of course

software attempts to match the profile for reflow ovens

anyway - time for some pictures

here's the triac driver board
it's inside the oven to keep mains voltage well away from small children and idiots
(er and me :) )

2nd picture shows the much reduced internal wiring

3rd picture shows the first board "wot I made"
it's a micro 'duino

once Nick Gammon helped me with the bootloader - it worked!

4th picture is part of a range of camera kit we're selling
all the SMD parts sorted in < 10 minutes!


there are only 10 types of people
them that understands binary
and them that doesn't

mmcp42

finally last picture shows the sort of profile I'm aiming for, and achieving
there are only 10 types of people
them that understands binary
and them that doesn't

Nick Gammon


(actually took a couple of goes as the ifrst one melted!)


Poetic justice, eh? <grin>

Congrats. Triacs are something I haven't fooled with yet, sounds like fun.
Please post technical questions on the forum, not by personal message. Thanks!

More info:
http://www.gammon.com.au/electronics

mmcp42



(actually took a couple of goes as the ifrst one melted!)


Poetic justice, eh? <grin>

Congrats. Triacs are something I haven't fooled with yet, sounds like fun.


indeed!

triacs seem to be dead easy
I'm using a zero-crossing opto-isolator to drive it

all seems to work just fine :)
there are only 10 types of people
them that understands binary
and them that doesn't

mmcp42

here's the final circuit
(went through several iterations to get there - as you do!)

and here's the triac heater control

I've allowed for a fan to help the cooling down phase

if anyone is interested I'll post the sketch as well :)
there are only 10 types of people
them that understands binary
and them that doesn't

Nick Gammon


if anyone is interested I'll post the sketch as well :)


I'm interested. Does the opto isolator do everything (switching on the zero crossing) or does the sketch have to do that?
Please post technical questions on the forum, not by personal message. Thanks!

More info:
http://www.gammon.com.au/electronics

mmcp42

seems the zero-crossing bit does all that
you just send a logic high to the input side and magic does the rest :)

sketch here...

if there's any bit that needs explanation ...
there are only 10 types of people
them that understands binary
and them that doesn't

Nick Gammon

Please post technical questions on the forum, not by personal message. Thanks!

More info:
http://www.gammon.com.au/electronics

Coding Badly


RE: profile.jpg

Are you using PID to control the temperature?

Have you considered adding a servo connected to the oven door to speed up the cool-down cycle?  How about a fan?

CrossRoads

Nice work mmcp42 - I'll get mine assembled one of these months :)
Designing & building electrical circuits for over 25 years.  Screw Shield for Mega/Due/Uno,  Bobuino with ATMega1284P, & other '328P & '1284P creations & offerings at  my website.

mmcp42



RE: profile.jpg

Are you using PID to control the temperature?

Have you considered adding a servo connected to the oven door to speed up the cool-down cycle?  How about a fan?


not using a PID, yet
need to get the temperature ramping fast enough
at the moment there's nothing to control!

and yes, I have seriously considered a servo to open the door
also a fan
watch this space...

@CrossRoads
cheers - and thanks for the initial inspiration :)
there are only 10 types of people
them that understands binary
and them that doesn't

CrossRoads

Thanks, but I am just following in the footsteps of many others here.
Designing & building electrical circuits for over 25 years.  Screw Shield for Mega/Due/Uno,  Bobuino with ATMega1284P, & other '328P & '1284P creations & offerings at  my website.

mmcp42

that's what I like about this forum
so many generous people leaving footprints for lesser mortals to follow :)
there are only 10 types of people
them that understands binary
and them that doesn't

mmcp42

#13
Feb 13, 2013, 09:41 pm Last Edit: Feb 13, 2013, 09:57 pm by mmcp42 Reason: 1
shiny new temperature probe arrived today :)
doesn't work :(

took it to pieces (what else! :) )

found the problem

the two wires are insulated with what looks like glass insulation
then slipped inside a stainless steel "knitted" (thank you Mrs Noah) outer jacket

trouble is at the business end the trimming was awful
with strands of wire touching the probe
and shorting it out intermittently

so - trimmed it back - much easier said than done stainless steel is tough!
wrapped the end of the metal in heat shrink - just to hold it while I re-assembled

temperature readings now rock steady :)

but :(

as the probe is inside a metal sleeve, there does seem to be some inertia in the measurement now
the graph below shows the problem
- (apparent) delay in getting started
- overshoot at the intermediate temperature points

a naked thermocouple is much more responsive
(see the original curve posted earlier)

I may well add PID processing - which should stop the overshoot
next decision after that's all working

oh and for amusement here's a picture of the first probe
not completely heat proof methinks!

added link to probe
there are only 10 types of people
them that understands binary
and them that doesn't

Coding Badly

as the probe is inside a metal sleeve, there does seem to be some inertia in the measurement now


"Dead-time" in PID jargon.

Quote
I may well add PID processing - which should stop the overshoot


Not necessarily.  There is a conflict between getting the response you want, the dead-time, and overshoot.  If possible, you should first strive to reduce the dead-time as much as possible.  That will make working with PID much much easier.

Go Up