Reflow oven controller colaobration

[FriedMule]

I have got an idea on an external universal controller that can be used without having to "taking apaaart" any oven.
I was thinking if wee all together could put our heads together and construct a great controller we all can use? :-)

My idea is that if we could build a electronical "veriac" with a heat sensor.
It should function by a program that follow the four steps in the heating profile:
1) Voltage enough to the heating elements almost glows (Ramp to soak)
2) Turn down the voltage (Soak)
3) Voltage enough to the heating elements almost glows (reflow)
4 Turn off the oven and alarm for opening the door (cooling)


All that can be done outside via the AC cable. The heat censor is maybe an other problem but could it be made wireless?

[rstofer]

It just isn't a big deal to add an SSR to the internal wiring when disconnecting all the existing controls.  You need to add insulation between the walls if you expect to get to temperature any time soon.

This is a pretty decent controller.  I have the previous version mounted in a metal box with the display connected remotely.

http://www.rocketscream.com/blog/product/tiny-reflow-controller/

[FriedMule]

It just isn't a big deal to add an SSR to the internal wiring when disconnecting all the existing controls.  You need to add insulation between the walls if you expect to get to temperature any time soon.

This is a pretty decent controller.  I have the previous version mounted in a metal box with the display connected remotely.

http://www.rocketscream.com/blog/product/tiny-reflow-controller/

You are right but you loose all warranty by modifying your oven, you do also risk the problem from the IR heaters that can darken your PCB if the heat elements is allowed to glow fully.

[sokoloff]

Losing the warranty on a ~$50 (or even $150) toaster oven would almost literally be the least of my worries in a project like this.

[rstofer]

You are right but you loose all warranty by modifying your oven, you do also risk the problem from the IR heaters that can darken your PCB if the heat elements is allowed to glow fully.

I bought the Black & Decker InfraWave Toaster Oven for the very specific purpose of building a reflow oven.  I never even thought about the warranty!  Not a consideration!

I have toasted bare PCBs and PCBs with soldermask with no damage whatsoever.  The lamps are a long way away from the board.

I usually buy MiniBoard PCBs from ExpressPCB (much more expensive than Chinese boards) and I can cook them on a GE hot plate I bought from Target.  I just put the board on the hot plate, turn the heat all the way up and wait for the solder to flow.  Then I use a pair of tweezers to remove the board before turning the hot plate off.

The only requirement for the hotplate is that it be flat and without voided areas such that the entire board touches the plate.

I use either method with equal success.

[Peabody]

I'm currently working on a version of this for my local open source hardware group.  But at this point I'm working with the $10 Mainstays hot plate from Walmart, topped with a circular saw blade.  See attached pics.

The idea I had was to use an MCU-controlled SSR to cycle the AC power on and off.  The cycle would repeat every 5 seconds, and average power would be determined by the duty cycle (i.e 2 seconds ON, 3 seconds OFF).  The duty cycle will be adjustable in 100 steps (50ms increments).  That's my equivalent of a variac or rheostat.

I'm going to use a TM-902C thermocouple to follow the temperature, and try to develop a sequence of duty cycle steps that produce the right temperature profile.  Then that exact sequence could be repeated without the need for any temperature feedback.

The "controller" will have a rotary encoder and two-digit display through which the duty cycle can be changed.  I'm currently writing software (in assembler) for a TI MSP430 processor (because that's what I have on hand).  But the SSR won't arrive for a week.  So it will be a while before I have an idea of whether this will work.

This wouldn't require modifying the hot plate, and I think the controller could also be used with an unmodified toaster oven.  The cycle period doesn't have to be 5 seconds.  It could be anything.

I'll be happy to share any of this.  I just don't know at this point whether there is anything useful to share.  And someone would probably need to translate the code to Arduino.  All my code is in the interrupt service routine.  The processor sleeps between interrupts.  I don't know how well that works in Arduino land.

Apparently a lot of people perfer hot plates over toaster ovens.  Apparently they get more consistent results, and fewer burned boards or parts.  But I don't see how that could be the case without some control over temperature.  Anyway, we will see.

If you want temperature feedback, the Rocketscream Tiny controller looks interesting, at a good price.

[FriedMule]

Have found out that it is already made: https://uk.beta-layout.com/estore/order_product_details.html?wg=1&p=613
Even Dave has made a video about that:

[perieanuo]

I have got an idea on an external universal controller that can be used without having to "taking apaaart" any oven.
I was thinking if wee all together could put our heads together and construct a great controller we all can use? :-)

My idea is that if we could build a electronical "veriac" with a heat sensor.
It should function by a program that follow the four steps in the heating profile:
1) Voltage enough to the heating elements almost glows (Ramp to soak)
2) Turn down the voltage (Soak)
3) Voltage enough to the heating elements almost glows (reflow)
4 Turn off the oven and alarm for opening the door (cooling)


All that can be done outside via the AC cable. The heat censor is maybe an other problem but could it be made wireless?

Sorry, it has be done already in an arduino project for reflowing oven and author 'tune' pid parameters.
But I never verified if the values obtained are ok...I built the board, it works with a heating plate (don't laugh it was my one and only patient for testing...)
On the other hand, if you pull out a better one and maybe on some modern and fast hw with min 12 bits dac, I say go for it.
Regards,pierre


Envoyé de mon iPad en utilisant Tapatalk

[hamster_nz]

I just put a SSR in an extension cord for mine... with lots of insulation tape to make it approach vaguely safe. You can see it on the right of this photo. No modification was used on the oven - just switch it on.

The element is cycled off and on based on a thermocouple reading, so the surface never gets too hot.

I suspect you don't want to attempt PWM on the heater element, as I assume SSRs are not designed to act as dimmers. 

I always think that full PID is an overkill for this. The system isn't that complex - heater on, temperature goes up, turn it off, and after couple of seconds it starts to cool. The thermal lag isn't that long.

Have the setpoint trace the profile, and then use a simple model of "if the short term trend continues for the next few seconds, will I be above or below my setpoint?", and use that to switch the element.  You just need to pause the set point at the peak temperature until that temp is reached.