Jump to content

MAX31855 based reflow oven controller

Recommended Posts

  • Replies 61
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Popular Posts

First, Much thanks to Rickta59 for helping me understand various aspects of the MSP430 in hardware, and software. Then to Spirilis who wrote the SPI initialization, and usage code in an attempt to hel

The one with the LCD on it is the one I'm testing with, the other one is the fully populated board. They're both identical I think, except the one I'm testing with have no TRIACs, and the one I'm test

BTA26-800BWRG triac = $3.20 each                                     this will easily handle the heater current MOC3063S-TA1 zero crossing triac driver  - 47 cents each   3.3 volt turn on makes a smo

Very good, we'll see what Richard wants to do first.


For everyone who has been patiently waiting. The boards should be here in about ~a week. So when they come in I'll be taking in shipping information, at which point we'll need cost of the boards, plus shipping. Before we ship.


As for code, I still have not finished anything on the list except for some of the LCD initialization and usage code. For those of you who watch the forums closely you may have noticed i ran into some trouble with the Nokia 5110, and operating at 16Mhz CPU/DCO. With the LCD working and functional I can not get my g2553 to run at 16Mhz, but it will do 8mhz, with a basic clock control register divisor of ZERO ( DIVS_0 ). At this speed, the max31855 is not fast enough to keep up. So, for now, the LCD speed app will be 1Mhz. This should not be a problem, but . . . UART version *only* is fine and good up to 16Mhz, as the divisor does not effect USCI_A->UART.


Anyway, I will try to get two version workings, then who ever can try to play with or get both UART, and LCD operating at 16MHz if they like. Just be warned, if you're not happy with C++, a lot of the peripheral device code will be C++ template classes. This is because a lot of the code that is known at compile time, can be compiled at compile time, instead of run time. Which makes the code smaller, faster, and more efficient. Anyway . . there you have it. Ill try to document the code well after I get it working good.

Link to post
Share on other sites

Heads up for everyone on the list for a board. We still have not received the boards just yet. But I will message here, and PM everyone when they arrive.


Also, for everyone interested  we're considering doing a mini group buy on parts needed to build this thing. If anyone is interested please let us know !


Thanks in advance :)

Link to post
Share on other sites
  • 2 weeks later...

Hello everyone we're still here !


Sorry for lack of communication the last week or two, but we just got our parts in today, That, and we're still missing a couple of parts that wont show up for a few more days yet.


Basically I've had the code running on the device for three days now. Using a mirrored Launchpad plugged into the board. But do keep in mind, the board is not populated 100%, and there is still some stuff I am unable to "smoke test" up to this day. Again, still waiting on parts. We want to make sure this controller is going to work without any hitches before sending off boards.


The good news is that the firmware has been running 3 days straight, no hitches, but without temp comparison  and without button code on it yet. No buzzer code yet either, or fan "control" but these things should be fairly easy to implement.


Mainly at this point its runs, and puts the relevant information out on the Nokia 5110 LCD. So when it comes time to do the same for UART, well for me, the UART code is easier to write. Perhaps not much but there you have it.( for those who would prefer to use Serial output instead of using a Nokia LCD ). Possibly they both can be used at the same time, but in this case, the way my code stands, the MCU would be limited to 1Mhz, Which really is not a big deal.


Anyhow code working, and "prototype" functional. Will keep everyone informed as to how it progresses further.




Actually the LCD code will be limited to 1Mhz period. As there are two/three SPI devices, and the 5110 does not seem to play nice with other devices on the SPI "buss". Possibly I may find a work around for this, but honestly it is not a top priority as DCO==1Mhz has not proven to be a problem yet.

Link to post
Share on other sites

OK sorry for no updates. Due to a few errors and one big oops i will scrap this batch of boards. This is the first board i have made with Allegro all my past boards were made with orcad layout. There was a learning curve and even though it is easier there was an error in footprint pins vs schematic pins in the voltage regulator ( 2 and 3 swapped ) I swapped 2 pins ( MOSI AND MISO ) because of me not paying attention to the pictures on https://github.com/energia/Energia/wiki/Hardware. I used the pinout for the MSP430G2452 ( WTF? why would TI swap the 2 pins anyway ) I also noticed that my power connector needed to face south not west so i can put the case back on the oven. Those errors were not that large and i have a proto running fine but the real deal breaker for me was putting the display upside down ( AHHHHH ) So here is a copy of the revised schematic for your review. I was asked by Yrkoon to add a 2 pin header for access to the serial port pins on the msp430. I added a 2 pin header so people that don't use a wall wart power supply with a barrel connector to just solder power wires to. I used mostly through hole components because i have so many of them it just makes sense. For those who want the new board i will toss in all the through hole resistors and caps for free. if anybody wants the connectors http://www.ebay.com/itm/10-x-2-Pin-5mm-PCB-Universal-Screw-Terminal-Block-Connector-300V-16A-GS007S-/350726604695?pt=LH_DefaultDomain_0&hash=item51a8ef4797 i can toss them in at my cost or you can order them on ebay

I ordered the Nokia displays on eBay as well DO NOT GET THE BLUE BACK LIGHT ONES they have an incorrect pin out. Please let me know about anything you want added to the design for you mad experimenters out there before i send off the boards for manufacture again. Thank goodness they are cheap @ 5 bux each plus shipping. I will order 10 boards unless i get more people wanting more i will wait a few days for all of you to chime in then i will send out the order. Please email me with your quantity required @ wulfman@wulfman.com     Good things are worth waiting for.


Link to post
Share on other sites

That's sad. Hope the next one works out.



I will order 10 boards unless i get more people wanting more i will wait a few days for all of you to chime in then i will send out the order. Please email me with your quantity required @ wulfman@wulfman.com     Good things are worth waiting for.


Could you please start a GB thread here for this with pricing details instead of custom orders. It will help me monitor it and also keep it separate from the main dev thread here. 

Link to post
Share on other sites

Yeah, I have a "full" working one in front of me with everything including Piezo buzzer on it. Even upside down LCD ! We dont have the actual "long" K-Type thermocouple sensor thats going into the oven with this(yet), but we do have a short K-Type sensor attached to it now. It works very nicely.


Hmm gives me an idea, I can try to take a picture of this running with output on the LCD.

Link to post
Share on other sites

The one with the LCD on it is the one I'm testing with, the other one is the fully populated board. They're both identical I think, except the one I'm testing with have no TRIACs, and the one I'm testing only has 1 max31855 on it.


As a matter of interest. max31855 are ~$8 each, so you may want to consider only using one. At least initially. We're going to test with two thermocouples, but in the end it is my gut feeling that it will only add cost, confusion related to the data output, and complexity. Plus there is only really room for one thermocouple output on the screen as you may be able to see, unless you want every_single_line on the LCD full up, and less readable.


Also at some point, I want to give PWM control of the TRIACs a good test someday, but in reality it is not a priority. I have seen a video on youtube where one such creator of a different reflow controller did really well in thermal control using PWM. His output over shot the peak target temperature only by 1-2C, which is really exceptional. Anyway, his code used PWM to do this ( PID controller ).








Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • Create New...