Go Down

Topic: Reflow oven (aka Toaster oven) modification (Read 10601 times) previous topic - next topic

pwillard

Quote
No need for PID and all that complexity.


You may have concluded this... but in my opinion,  the PID would serve you well and could eliminate the need for a fan altogether if you do it right.

mmcp42

here's what I am achieving with simplistic control

algorithm is
if temp < target then
turn on heater
else
turn off heater

the ramp up slopes are less than I would like
the cool down is even slower
I open the door a smidge as soon as reflow is done
the sudden drop at the end is door fully open
there are only 10 types of people
them that understands binary
and them that doesn't

mmcp42

next step is to add a fan in the floor
a) to aid cooldown
b) might improve ramp-up as well!

I notice that th heaters are on full time during ramp up
is my oven under powered?
or is the profile "close enough for government work"?
there are only 10 types of people
them that understands binary
and them that doesn't

rvasque

Quote
next step is to add a fan in the floor
a) to aid cooldown
b) might improve ramp-up as well!


My understanding of a convection fan (during heating) is it circulates the hot air more evenly (so there are no hot spots and shadows).
I've read to achieve a fast ramp up (200C in 2 minutes), insulating the glass door, or you need to have more powerful heating element.

(my oven has 4 heaters, and can achieve the 200C in 3 minutes from a totally cold oven, for the first batch)

Quote
the cool down is even slower

Yup... this is my problem. 
Even opening the door isn't doing it for me....
I have a big oven, 16cu.ft., big enough for a 12" pizza), so more heat trapped inside too.

For others using a small bread toaster oven, opening the door is sufficient for them. I think I really need an exhaust fan, or something.

mmcp42

#19
Feb 11, 2013, 06:37 pm Last Edit: Feb 11, 2013, 06:48 pm by mmcp42 Reason: 1
mine is 9 litre capacity with 1 kW of heating (4 elements 2 above, 2 below)
glass door is double skinned
but the oven body is just sheet metal
I've been running with its "triusers" off while I get things working
maybe putting the outer cover on will help :)
there are only 10 types of people
them that understands binary
and them that doesn't

CrossRoads

I use a 4 element oven, 2 top & 2 bottom, 1500W, observations with k-type thermocouple to digitlal multimeter are that it heats up fast enough.
skyjumper found the same a while ago - 3 elements not quite enough.
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.

rvasque


I use a 4 element oven, 2 top & 2 bottom, 1500W, observations with k-type thermocouple to digitlal multimeter are that it heats up fast enough.
skyjumper found the same a while ago - 3 elements not quite enough.


+1. I agree.

Using 1500 watts here too.

Gerhard34

4yi, here is the temperature curve of my reflow oven (850 W with 2 elements, k-type thermocouple; PID )

MarkT

My cool-down curve is achieved by:

1  switch off

2  open the door

3  carefully slide the wire mesh shelf out as far as I can (don't want to disturb
components while solder still molten)

4  waft a piece of card gently to produce a draught over the board for a while.

Takes about 1 minute for the board to be cool enough to handle (just).

If using a fan it would need to blow otherwise it will melt itself, and will need to be
quite small (to prevent components being blown about.  A short steel tube to keep it
mounted remotely from the heat might be wise.
[ I will NOT respond to personal messages, I WILL delete them, use the forum please ]

cmiyc

Simply opening the door is probably causing a faster cooldown than 6°C/second, which is a excellent way to damage components.  Especially ceramic caps.  And the nice thing about cracked ceramics, they rarely show signs of failure right away.

You really need to allow the board to come down from the peak soldering temperature before exposing it to room temperature air.
Capacitor Expert By Day, Enginerd by night.  ||  Personal Blog: www.baldengineer.com  || Electronics Tutorials for Beginners:  www.addohms.com

MarkT


Simply opening the door is probably causing a faster cooldown than 6°C/second, which is a excellent way to damage components. 



I doubt it, the oven stays hot for a long time, its walls are radiating heat efficiently and
until the board is outside to some extent you won't get much cooling. The board itself holds a
lot of heat and limits the rate of cooling of the components.  Even fanning a board for
a minute leaves it hot enough to burn fingers.
[ I will NOT respond to personal messages, I WILL delete them, use the forum please ]

Constantin

A couple of observations...

I happen to use a heavily-modified toaster oven with a rocket scream PID controller. As best as I can tell, the unit ramps up as desired, i.e. it follows the reflow pattern recommended for my Kester 'green' solder paste (i.e. no Lead). The only part that is likely 'non-compliant' so far is the cooldown phase, i.e. I have to figure out how far to open the door so that the temperature drops as intended, hopefully without cracking a ceramic cap. If you'd like to see some pictures, see this thread.

As for the construction process, I disassembled the oven and removed all controls. A ebay-sourced SCR (2x20A) is perfect for this formerly-'Black and Decker' $30 toaster oven since the four quartz-tubes are OEM-limited by a fat diode. The SCR achieves the same thing, but with control and I used the second channel for the convection fan. On the front face, I mounted a leftover plastic enclosure that I salvaged from a Neuton mower whose motor had gone bad. It holds the Arduino 2009 and the Rocketscream shield perfectly. I stuffed the (formerly empty) voids between the inner and outer oven cavity with 2000*F rated insulation.

Inside, I further reduced the mass 'visible' to the quartz lamps by stuffing the same insulation between the lamps, filling the rear of the cavity where no PCB would ever go, etc. The unit still gets hot during operation, but nowhere as hot as it would get without the benefit of insulation. I even insulated the door window, leaving just a little peep-hole to limit heat loss through the glass as well. I presume this is why the oven has no issues ramping up as needed - there is 'spare' capacity and the quartz elements are not on 100% of the time. At full power, it draws around 1340 Watts, IIRC.

If anyone in the Boston area is interested in looking at the thing or would like some leftover insulation, I'd be happy to share.

Constantin

... and one more thing... does it necessarily make sense to go through all this effort when you can buy a certified reflow oven on Amazon/eBay for about $285, delivered? Consider that the materials alone (oven, insulation, heat sink, reflow shield, K-thermocouple, etc.) probably came out to about $125.

Sometimes the journey is just as much fun as the destination.  :)

Go Up