uC based reflow oven controller

manton · Post by **manton** » Mon May 25, 2009 6:26 pm

So I'm trying to understand why you would use a cycle time of anything smaller than 8.33ms. Even if you can get the SSR to turn on at a random time (and many SSRs will), it will not turn off until it hits the zero crossing, which occurs every 8.33ms.

I would think the best way to do this would be to phase control the SSR, by locking to the line frequency so that you can turn the SSR on at a know time relative to the next zero crossing. This would give you the best duty cycle control possible, but you still can't do it faster than using a 8.33ms period (by any method).

Mike

corvette123 · Post by **corvette123** » Mon May 25, 2009 10:32 pm

depends on the SSR

..

also there are other methods than an SSR, that is the suggested method.

ummm did you look at the graph?

doesnt get much better than that!!!

tnx

manton · Post by **manton** » Mon May 25, 2009 11:31 pm

I agree that the result is excellent, but I don't see any significant change between 2.5 and 5 ms periods, and my reasoning would explain it. In general, this process is fairly slow, so I can't imagine that there would be much benefit, if any, to controlling faster than 120Hz, and it wouldn't surprise me if an even slower control rate would work just as well.

Can you point me to an off the shelf SSR that can turn off at points other than the zero crossing? I agree that you can design one from scratch using IGBTs, or FETs, but most individuals will not do this.

BTW, Breville has a really nice counter top oven that I am considering using as a reflow oven. It uses IR heaters, has an output of 1800W, and has a convection fan, though it is larger than most (and expensive). But, it looks to be really solid, and could likely be insulated easily, which would allow for faster ramp rates.

Mike

corvette123 · Post by **corvette123** » Tue May 26, 2009 10:27 am

when i zoom in on the graph (with the zoom tool in the software) you can see some small differences.)

the main difference is to allow for slower SSR's that have zero cross and cant handle the 2.5 switch.

yes ive had a few customers that couldnt get their SSR to work with the 2.5 mS period.

so thats why the slow period.

however that value will be configurable in software in the new version of reflow tools 5.0 that will be coming out soon!!!

it should be pretty cool.

a major update and its just an upgrade.

tnx

-drex

corvette123 · Post by **corvette123** » Tue May 26, 2009 10:36 am

k type thermos back in stock

corvette123 · Post by **corvette123** » Fri May 29, 2009 12:38 pm

just got a new shipment of parts today.... stop on by and get our new reflow 3.0 controller!!

corvette123 · Post by **corvette123** » Fri May 29, 2009 9:28 pm

hmmm the views is upto like 62000 now

is it gonna overflow hahhahaha

safety0ff · Post by **safety0ff** » Thu Jun 04, 2009 8:59 pm

corvette123 wrote:hmmm the views is upto like 62000 now

is it gonna overflow hahhahaha

66 666 th view, and it didn't blow up...

corvette123 · Post by **corvette123** » Thu Jun 04, 2009 10:44 pm

hahaha so thats more than an int then huh

i wonder how high it can gooo.....

corvette123 · Post by **corvette123** » Thu Jun 18, 2009 8:51 am

its leveling out ~67000

corvette123 · Post by **corvette123** » Mon Jun 22, 2009 7:06 am

we are making leaps and bounds on the new reflow tools 5.0 software!!

-new USB framework fixes some bugs and x64 support.
-updated graph api
-female voice will "talk" to you announcing reflow stages (optional)
-tons of extra options to configure.
-unified firmware for LCD and no LCD and all options. all configurable from the windows GUI.

its gonna be pretty cool.

u can checkout the featurelist from our forums under the reflow section

corvette123 · Post by **corvette123** » Thu Jun 25, 2009 5:26 pm

well heres the latest feature list of reflow tools 5.0

we have moved everything into 1 firmware and have added so many features it aint even funny!!!!!

===========================================

current list of future Milestones
__________
Crystal fontz cheap serial LCD firmware with pushbutton support instead of keypad
scrolling graphs

Milestones reached so far
______________
-HID bootloader (no driver needed!!) + windows application to upload firmware to board
-support for x64!!!
-new USB driver framework fixes sleep, and some other USB bugs.
-updated graph API's fixes slow line draws and some other vista specific issues
-added female voice to signal start of reflow stages
-temp and time LCD readouts
-better graphing / updated the graph api... should be a bit faster now.
-stores program settings in registry and last position of some pulldown and selection boxes
-option to load all profiles from controller on 1st controller connect (automaticalluy)
-option to turn off voices (your wife may talk to you too much so why should your computer also??? hahahaha)
-added firmware options box
-3 pid algorithms tested (full on off / PID with on off / PID with PWM on off at extremities
-pid algorithms selectable in windows gui, all in 1 firmware.
-cooldown options by time or temp selectable in windows
-graph title shows all parameters for that run
-more accurate a2d math.
-selectable PID rates from 2.5 mS up to 1 second!!!
-gains now go upto 255 and are better mapped than before, so youll see the differences in the PID algorithm in full pid mode.

the pid rates are really cool, going from 2.5 mS to 1 second!!! and yes they affect the derivative term as well as the other ones, so its one more thing to play with.

3 algorithms to choose from including:

1. full on / off

2. PID with full on / off

3. PID with PWM and full on / off at extremeties <10% or >90% duty cycles.

so its coming along, will probably add crystal fonts LCD support and users can modify firmware from there... serial.

corvette123 · Post by **corvette123** » Thu Jun 25, 2009 5:28 pm

yes you can drive the PID from another source such as an IR thermocouple from www.omega.com that goes from 0-5 volts. then have the regular thermocouple report its data also!!!

all selectable in software. no programming for you guys to do.

corvette123 · Post by **corvette123** » Thu Jun 25, 2009 5:32 pm

there WILL be a linux version of the software as well using LIBUSB.

i dont know how pretty the GUI will be as i suck at linux programming, but well seee on that one

at the very least, a functional one.

i will use QT4 for the interface probably.

corvette123 · Post by **corvette123** » Tue Jun 30, 2009 2:47 pm

techFX reflow tools 5.0 is out!! go get it!

windows version that is.

will see what we can do for linux.

tons of great features.

-HID bootloader (no driver needed!!) + windows application to upload firmware to board
-support for x64!!!
-new USB driver framework fixes sleep, and some other USB bugs.
-updated graph API's fixes slow line draws and some other vista specific issues
-added female voice to signal start of reflow stages
-temp and time LCD readouts
-better graphing / updated the graph api... should be a bit faster now.
-stores program settings in registry and last position of some pulldown and selection boxes
-option to load all profiles from controller on 1st controller connect (automaticalluy)
-option to turn off voices (your wife may talk to you too much so why should your computer also??? hahahaha)
-added firmware options box
-3 pid algorithms tested (full on off / PID with on off / PID with PWM on off at extremities
-pid algorithms selectable in windows gui, all in 1 firmware.
-cooldown options by time or temp selectable in windows
-graph title shows all parameters for that run
-more accurate a2d math.
-selectable PID rates from 2.5 mS up to 1 second!!!
-gains now go upto 255 and are better mapped than before, so youll see the differences in the PID algorithm in full pid mode.