T962A Controller with 5" Touch TFT (using *STM32*)

T962A Controller Drop-In (using *STM32H7*)

This is a drop-in replacement controller board for the T962A reflow oven, using the STM32H7

Similar projects worth following
STM32H7 CPU @400Mhz, 2MB on-chip flash

5.0" 800x480 Capactive-touch TFT display
(optional 7.0" 800x480 Capactivie-touch TFT)

64MB external SDRAM (LCD frame buffers, etc)

4 Max31855 on-board TC Sensors (K-type)

MicroSD socket for FatFS MicroSD (storing images, profiles, etc)

MAXIMUM profile lengths of 30 minutes

Hand draw full 30 minute profiles

Load and Save profiles to/from MicroSD card

built-in CPU temp for CPU fan control

PWM control of CPU fan (6Khz PWm)

120VAC Phase-cut control of COOLING FAN for true speed control
(via 8-bit cpu DAC and FL5160 AC Phase controller)

PWM control of heater/SSR (10Hz PWM)

USB & RS232 on-board interfaces

5v & 3.3v switching power supplies

Unused STM32 I/O pins brought out to debug pads

** board is fully grounded and can be screwed/grounded directly to oven case **

** AC is fused on board for short protection, etc **

** USB & RS232 Not yet coded up in software **


Just about complete with a first version of this project fully working, should have source code and a compiled binary up sometime this week... doing some code cleanup and gui improvements..

There of course is still much improvement to be done to the GUI, etc, but thankfully that can be an ever improving concept, as the ST 'ewWin' library allows you to design almost endless graphics ideas, so new ideas and things can always be added/changed!


T962A Controller w/touch TFT - Bill Of Materials (updated to version/latest 3.3C of the board)

sheet - 19.16 kB - 07/12/2019 at 14:39



Eagle version 7.7.0 schematic file (NOTE: You can also get these files in the github repo link on this page)

sch - 2.09 MB - 07/06/2019 at 00:41



Eagle version 7.7.0 board file (NOTE: You can also get these files in the github repo link on this page)

brd - 669.33 kB - 07/06/2019 at 00:41



Version 3.3C schematic in jpg format

JPEG Image - 4.39 MB - 07/03/2019 at 00:20



Steps for programming the controller flash

Adobe Portable Document Format - 2.51 MB - 06/18/2019 at 22:12


View all 8 files

  • 07/15/2019

    PhillyFlyers6 days ago 0 comments

    Wow, this is awesome:

    Actually, probably going to use this one, as it has WI-FI & BLUETOOTH..

    (and it's only $3.80, vs. $2.70 for the wi-fi only one)

    I wanted from early on to have a simple way to do Ethernet/Wi-fi on this thing, but didn't have enough pins on the STM32 to use their built-in lan interface, but with an all-in-one wifi module, I can add this on without having to do any TCP/IP crap, as it's all built into the module..

    I'll look at seeing how easy it is to add this on as a cable-up connection to the UART2 header already on the current PCB, as well as seeing if it's easy enough to add this footprint to the bottom side of the PCB and possibly do another PCB update..

    we shall see, but the reason I really want network of some kind, is at least to do:

    1) network monitoring of real-time reflow, etc

    2) Software upgrades over the network, so you don't need to open up the oven at all!

    And for $2.70!!!   that is insanely cheap..

  • 07/14/2019

    PhillyFlyers07/14/2019 at 18:22 0 comments

    Working on two documents currently, to help installation and a user's guide... so hopefully this week I will add these two docs (here as well as in the github repo)

    1) Installation Guide

    2) Users Guide

    I will have pictures of all the GUIs/Menus, and some explanation of each, so you can better understand how it all works, and how to navigate around, etc..

  • 07/06/2019

    PhillyFlyers07/06/2019 at 11:06 2 comments


    Version 1.0.0 is up, posted the source code and compiled binaries. 

    Got 5.0" and 7.0" both working, I will update the build docs soon to reflect the two resistor changes needed for the 7.0" backlight voltage..

    but for now it's all good, and I will keep testing and running version 1.0.0, and of course keep updating (and making new versions) to make the GUIs better, etc, but for now, this should be a good starting point.

    I will also make a 'user's guide' soon to show how to use menus, etc, and update the build instructions a bit more to show exactly the difference in mounting 5.0" vs 7.0" ,etc, plus more pictures...

  • 07/04/2019

    PhillyFlyers07/04/2019 at 13:43 0 comments


    Happy 4th!!

    Have a good working codebase that I'll upload as a version 1.0.0, probably tomorrow after I get a replacement 7.0" screen, as I sort-of mangled the prior one with all the tweaking and moving around trying to get board layouts right.
    (I've managed to destroy 2 or 3 screens total now.. lol)

    Took a bit longer than expected, but I needed to clean up so many bugs I encountered..

    Just as an FYI, since I haven't given too much details yet, (I plan on writing up a USER GUIDE soon), the way I created this you can have profiles that can be up to 30 MINUTES max in length!  pretty sweet..

    And they are 5 second interval points (whereas stock oven code was 10 seconds for each point), could've gone higher resolution, but in this type of environment, oven with heating elements, the heater reaction time is so slow, it's a total waste of time and bandwidth having finer point resolution, as it does nothing...

    so 5 seconds seemed way more than enough, even 10 secs is still prob. fine, can always go back to that if it turns out 5 secs is a waste as well.. anyhow...

    1) you can 'manually' draw profiles via the touch screen, 1 page at a time (3 total pages), since I can only fit about 12 minutes of profile on a screen at a time

    2) you can then 'edit' the entire profile, and change temps, etc, just like old oven software (1 page at a time)

    3) running the profiles, which can be up to 30 mins length, and can then view the run data 1 page at a time when it's over

    4) all sorts of settings and things can be changed in the settings menu, etc, and there is an 'About' menu that gives you build number, memory usage, etc..
    and much more... all settings saved in a 'config.ini' that is on the microSD.. as well as profiles are saved/loaded/deleted from the microSD..

    5) Also you can configure you own background 'wallpaper' (800x480) JPG image that is loaded from the SD card... since at the main menu there are buttons on the side for doing things, but majority of the screen is your background wallpaper, to make it look cool!

  • 06/29/2019

    PhillyFlyers06/29/2019 at 12:34 1 comment


    This board revision is perfect, RTC and battery is all working, will verify tomorrow once my ML2032s show up (the recharagable CR2032 variant) that the charging and everything holds.

    But board is good, LCD 5.0" & 7.0" both now work, looks awesome.  I am very very impressed with JLCPCB manufacturing, and it only cost me $40 for the board!...and you got 5 of them :)

    I'll update the schematics, eagle files, etc tomorrow... and hopefully in the next few days be posting the working code.

  • 06/21/2019

    PhillyFlyers06/21/2019 at 10:53 0 comments

    Finally got the 7.0" TFT working, wow does it look awesome, but it's big!  There was a discrepancy between the TFT datasheet and STMicro's datasheet, still not sure whom is right.. but whatever... it works..

    Since the 7.0" looks so good, and works the same, I decided to see how to best make the board easy to switch between one or the other, so I had to slide the connectors around a tiny bit... plus since I was making a change I added a socket for 'ML2032' (rechargable CR2032 battery), and changed out the RTC clock to a true low-power crystal like the STM32 eval boards use.

    This allows the board to have true battery backup for the Running Clock, so the thing can actually have a true time clock/calendar that doesn't lose it's settings :)

    Just a nice extra I wanted to make work.

    I'm cleaning up some more bugs and edge-cases that I identified, so will put up an initial software version prob. next week now, as I want to test it out some more, and possibly get that board revision back next week to test out..

  • 06/16/2019

    PhillyFlyers06/17/2019 at 00:54 0 comments

    I should be posting an initial version 1.0 of the code this week, as well as production hex/binaries for easy flashing.  The code is pretty good now, will have plenty to improve/work on as this moves forward, but it's at a nice spot now.

    I will work on updating the build instructions this week, to include pics/mods to the oven case, as well as installing the LCD itself.

    Also will put up a 'flashing instruction' doc, step by step on how to flash the board with the FTDI cable, using a production .bin/.hex file...

    Should have some goodies posted up this week..

    The 7.0" screen looks SO good on this oven, only bad part is the one I received may be bad, as I get ZERO picture on it, the lcd backlight works fine.  So I have to see if Newhaven can get me another, or if there is something else I'm missing..

    But wow does it make a much better look than the 5.0" as to how it fits...

  • 06/09/2019

    PhillyFlyers06/09/2019 at 15:13 0 comments

    Finished and published initial version of 2 'building instructions' for the procedure I recommend for assembling the board.  You certainly don't have to follow this approach, but it's how I've been building mine, and is the safest way to prevent you from blowing up the expensive processor.

    Also, I made a minor update to the PCB, now labelled version 3.1C (previous version was 3.0C).

    It's just a 1-wire addition for adding a second pin to receive the RTC clock signal.  I wanted to have a separate timer fed by the RTC clock so I could have a good 'seconds' counter and not have to keep resetting the real RTC clock.  AS I would like to at some point have a battery-backed-up RTC clock circuit so that the oven can retain a real time clock for however long the standard CR2022 or whatever battery can last!

    But currently there is no battery backup, so for now it will still lose real clock settings once you turn off power.

    ** Note: ** you do NOT have to do this mod if you already built this PCB, and don't feel like doing it.  I will provide a compile switch as well as two production '.hex' files:

    1) WITH the second RTC clock enabled (if you did the mod or built the latest 3.1 PCB)

    2) WITHOUT the second RTC clock

    I should have reflow running soon, I am working on this particular GUI's now and working the the graph display for it all..

  • 06/05/2019

    PhillyFlyers06/05/2019 at 13:26 0 comments

    Still making good progress on the GUIs, getting much closer now..

    I do have plans to write up some documents to post on here, to help with:

    1) Building instructions (some general instructions on best approaching to assembly the PCB)

    2) Programming the board through the UART with STM USB Flasher & FTDI cable

    3) How to add the 7.0" TFT instead of default 5.0" TFT

  • 05/29/2019

    PhillyFlyers05/29/2019 at 09:21 0 comments

    FYI, it is MUCH MUCH nicer when using the NON-BRACKETED LCD screen, you can just use super-sticky adhesive tape on the back of the screen, to stick it right to the oven front panel.

    I updated the pic in the project with how it looks, only a small bit of the lcd ribbon cable sticks out through the slot with it mounted in this fashion.

    Now the only bad part is my oven has 4 ugly drilled holes from my mounting experiments, LOL.. oh well, I'll fill them in with some gray caulk to try and make it not so ugly.

    i.e. definitely get this version of the TFT, ie without the mounting bracket:

    (it's also in the BOM)

    *** Also ****

    It looks like it would be fairly easy to drop in the 7.0" TFT versus the 5.0", same resolution 800x480, but gives you a bit larger display..

    I may indeed grab one of these and try it out, would require a simple change out of the two resistors that setup the voltage for the LCD display driver, as the 5.0" requires approx. 19.0VDC for the backlight, whereas the 7.0" only takes 9.6V..

    I'll give it a try after I get an initial version of code working for this project

View all 23 project logs

Enjoy this project?



Sam wrote 07/11/2019 at 13:09 point

Good day You mentioned that you posted the sources and binaries, but for some reason I cannot find them.  Can someone provide a link?


  Are you sure? yes | no

PhillyFlyers wrote 07/12/2019 at 14:41 point

It's on the left-hand side when viewing the project, where the links to the PCB are, the SEGGER embedded studio, etc....

it's labeled as 'T962A Controller - Software'

i.e. this is the github repo:

  Are you sure? yes | no

Nicolas Chaslot wrote 06/05/2019 at 17:03 point

Good to see all the work done here!
I have received the boards, still have to order the components. From a first "external quick fit" i think it should more or less fit in the T962 oven too. I'll try to open my oven and throw it in to see more precisely where it (if it)  hits :)

  Are you sure? yes | no

Nicolas Chaslot wrote 05/28/2019 at 20:08 point

Awesome thank you for the hard work!

I'll order the parts as soon as I get the boards from china! I will also be able to confirm the fit in the T962 oven as i have the small one :)

Did you do some reflow testing yet?

  Are you sure? yes | no

PhillyFlyers wrote 05/29/2019 at 09:24 point

Still working on the code/GUI, little by little getting all the menus tied I added an 'ini' parser code in, so I could put a 'config.ini' onto the microsd as well, this is what stores configuration LCD brightness, sensor enables, etc...

Hopefully in the next week or two I'll get to that point where I work on enabling the reflow, but I'm still working on the menus for the profiles, reading/writing profiles to the microsd, etc....

  Are you sure? yes | no

Nicolas Chaslot wrote 05/16/2019 at 08:08 point

Good Job!
I've had a look at the gerbers, it looks awesome! I just wanted to  know, did you confirm the fit on the T962 oven?

I'll order 5pcbs in china now (17$ for all of them) so if anyone needs one... :)


  Are you sure? yes | no

PhillyFlyers wrote 05/16/2019 at 09:31 point

Holy crap, $17?  What place did you order from?  that is insanely cheap, I can't believe the prices I pay on OshPark sometimes :)

I don't have a T962, I only have the T962A... someone recently on my other Pic32 project had mention that board would not fit in the T962, so I really doubt this one would either..

  Are you sure? yes | no

Nicolas Chaslot wrote 05/16/2019 at 09:40 point

I quoted for that price on JLCPCB in china. 16,XX for 5 pieces.

Ok. I'll figure out a way to make it work then haha

  Are you sure? yes | no


[this comment has been deleted]

Jarrett wrote 05/15/2019 at 02:29 point

You should post these as project logs :)

Makes it easier to follow

  Are you sure? yes | no

PhillyFlyers wrote 05/15/2019 at 20:57 point

Great Idea!  did not even notice that was there!  I moved all these updates into logs, going to try and clean these comments up..

  Are you sure? yes | no


[this comment has been deleted]

Nicolas Chaslot wrote 05/05/2019 at 09:50 point


Definitely not cheap but if that can make that little oven a little bit more suitable for work that's worth it!

  Are you sure? yes | no


[this comment has been deleted]

Nicolas Chaslot wrote 04/16/2019 at 08:12 point

Great news!
If hardware is validated i'll order and build it then :)

  Are you sure? yes | no

PhillyFlyers wrote 03/11/2019 at 00:48 point

...moved to project log...

  Are you sure? yes | no

Nicolas Chaslot wrote 03/22/2019 at 10:32 point

Awesome work as always!

Maybe the sd card could store the reflow profiles and user parameters too?

  Are you sure? yes | no

PhillyFlyers wrote 03/22/2019 at 14:05 point

Yes indeed, I was thinking this would be the way to go once things settle in, as we'll have almost endless space on a microSD card..

So I got the microSD card all working, just hacked up a socket on my current test board, since my newly spun board is showing up today (with the socket actually part of the board design :)

That was fun trying to figure out all the public microSD card code, using chan's FatFS lib (which is freakin bad-ass)... So got it all going, nice and clean running the SPI interface to it at 10Mhz using a 32gb microSD..

So once I start populating the new board this weekend, I'll test out that everything is up and working, and I didn't forget any of the many cuts/jumps I had to add to my current board.. lol... plus keeping my fingers crossed the SRAM works... in that case I can start testing displaying images from the SD card to the LCD...

  Are you sure? yes | no

Nicolas Chaslot wrote 03/22/2019 at 14:08 point

Sounds good!
Really looking forward to ugrading my oven !

  Are you sure? yes | no

PhillyFlyers wrote 02/21/2019 at 14:04 point

...moved to project log...

  Are you sure? yes | no

Nicolas Chaslot wrote 01/16/2019 at 09:21 point

I can see there have been some awesome work done here! Don't know why hackaday doesn't keep me notified...

Anyway glad to see the thing going along and can't wait to get my hands on one!

  Are you sure? yes | no

PhillyFlyers wrote 01/13/2019 at 21:20 point

...moved to project log...

  Are you sure? yes | no

Elasia Squishie wrote 01/09/2019 at 16:34 point

I mean just the circuit as is on figure 4 and powering from 120V the application circuit, that chip looks like it needs to float on the ac line to work proper and you get that by the ground which is really connected to hot via the fets diode and then VS is connected to neutral via a rectifier diode and related resistors/transistor

so then you use an isolator connected to the fl5160 ground and power pins to drive a pwm into a low pass or feed a dac chip that is powered by the fl5160 as well

i plan to build the above once my voltera shows up next week, this is right up its ally

  Are you sure? yes | no

PhillyFlyers wrote 01/09/2019 at 17:06 point

Yep, you got it... this is why it get's tricky, because you are dealing with AC neutral/gnd and then the digital gnd for the rest of the logic.  

It made more sense to me to just keep the AC section as isolated as possible, and just power everything from my DC supply.  So this way, everything including the 5160 is powered by the DC supply, AC remains on the FET side, and nothing else.

but yeah, you can do it whatever way you want really... just gets complicated as you power more things from the 5160, and then try to interface with other logic on earth ground.

  Are you sure? yes | no

Elasia Squishie wrote 01/09/2019 at 20:10 point

Yeah, ideally only need to bang the serial pins through an isolator and leave the entire rest of it as an isolated mains power circuit

I just got done baking the oven for almost a full day... even after removing all that ridiculous masking tape before hand is still had this nasty smell i couldn't quite place other than being some kind of organic voc

In manual mode on the chinese controller it leveled off nicely at around 160C or so with the fan and lamp on out of the box.. only thing i can figure is whatever garbage mineral wool they used must have had the chemical in it because i had it completely torn down to inspect for putting an internal air stiring fan too.. that or maybe the adhesive from the heat tape they used?

Either way can barely smell it now... i'll still be placing it by my window in the lab though

  Are you sure? yes | no

PhillyFlyers wrote 01/09/2019 at 14:17 point

...moved to project log...

  Are you sure? yes | no

PhillyFlyers wrote 01/05/2019 at 20:16 point

...moved to project log...

  Are you sure? yes | no

Elasia Squishie wrote 01/09/2019 at 05:54 point

This peaked my interest so i picked up some parts to build figure 4 in the datasheet and then feed the dim control doing

pic square wave -> isolator -> low pass -> dim control

have you done anything of the like? The FL5160 has even its own built in power regulator to supply 5v

  Are you sure? yes | no

PhillyFlyers wrote 01/09/2019 at 14:14 point

yeah, you can do that as well, the dim control in the chip has it's own internal ADC to determine what 'level' your output AC will be at... so whether you feed it a straight-up analog voltage level from a DAC (as I'm doing), or a smoothed out pwm signal, or whatever, it doesn't care... 

And yes, the 5160/5150 puts out a 5VDC supply from that pin, you can use that for other things if you want..

All of my issues with this chip were configuring as you were above (using figure 4 or similar), and getting the thing to run with the entire circuit connected, and proper 3-wire grounding, etc.  It was not working right in that scenario, so it made more sense to supply the chip from a more stable source.

  Are you sure? yes | no

Elasia Squishie wrote 01/09/2019 at 14:53 point

Did you use an isolated ground between the ground pin and in the middle of the fets?

  Are you sure? yes | no

PhillyFlyers wrote 01/09/2019 at 16:03 point

Yep,  I have an opto-isolator being driven by the FL5160 gate drive, which is in turn driving the FETs.  (gnd of FETs is gnd of opto)

You can look at the PIC32 project schematic to see how I set it up, as I'm using the same setup here, except for the extra boost circuit I'll be using in this version to drive a stable VS for the 5160..

In the PIC32 version I was driving VS from the 9VDC coming from the oven 9VAC transformer...but 9-10VDC is right on the borderline of POR margin for the VS, so for stable running, you need it to really be around 11-12VDC (at least).  5160 clamps it at 17VDC of course...

  Are you sure? yes | no

PhillyFlyers wrote 01/03/2019 at 01:26 point

...moved to project log...

  Are you sure? yes | no

Nicolas Chaslot wrote 01/03/2019 at 16:31 point

Yeah that's a pain when the datasheet schematics don't work as they should...
I'm sure having a 12v reg inside the oven on top of the ssr would not cause any problem to most of us wanting to go the custom pcb route.

It might simply be the simplest/quickest option to get the best results here, leaving time after to figure out what the problem is... Doesn't Fairchild have a forum or contact line for technical issues like this?

  Are you sure? yes | no

ghulands wrote 12/08/2018 at 07:09 point

Are you thinking of doing a kickstarter or small board runs for this board?

  Are you sure? yes | no

PhillyFlyers wrote 12/19/2018 at 00:24 point

...moved to project log...

  Are you sure? yes | no

ghulands wrote 12/19/2018 at 00:34 point

How's your progress going lately?

  Are you sure? yes | no

Nicolas Chaslot wrote 10/22/2018 at 08:02 point

Joining here to follow the progress of this version!
I'll try to be patient and not order the pic version for the moment haha :)

Keep up the good work :)

  Are you sure? yes | no

PhillyFlyers wrote 11/11/2018 at 20:33 point

...moved to project log...

  Are you sure? yes | no

Nicolas Chaslot wrote 11/12/2018 at 08:14 point

Haha yeah i know the feeeling of being overwelmed by projects ^^
Anyway, that's awesome!

  Are you sure? yes | no

PhillyFlyers wrote 12/19/2018 at 00:22 point

...moved to project log...

  Are you sure? yes | no

Nicolas Chaslot wrote 12/19/2018 at 08:35 point


No big deal, there's no rush, you're doing an awesome job :)
Good that all the magic smoke stayed where it belongs!

I'll probably just wait for the stm32 board anyway, i don't really see the point of making two successive updates, considering the oven kinda works already using some low-temp soldering paste and I didn't even have the time to run as many boards as i wanted to :)

I started looking at stm32 development too a few days ago and i must say that it's way more complicated than I initially thought, with a lot of different libs and files to gather from all around to make the thing to build!

  Are you sure? yes | no

PhillyFlyers wrote 10/09/2018 at 21:25 point

FYI, please see the *details*, I've only just started this project, I'm just populating my 1st revision of a PCB now, then will have to start testing the hardware to see how many screw ups I made on the PCB...

So gonna be quite some time before this is doing anything cool to show :)

  Are you sure? yes | no

Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates