Firmware for Parametric Profiles

[borland1]

Parametric Profiles. Here's a different approach to modifying Reflow Profiles, by changing parameters in a systematic way. Here's highlight of this firmware.

• Easy way to fine tune profiles for variations due to ambient temperatures, SMD components sizes, PCB size changes, varying solder paste characteristics, or other considerations.

Parametric Profile Features

• Seven fully editable reflow profiles
• 13 user editable profile parameters, plus title length up to 19 characters
• All changes are saved to EEPROM.
• Default profiles can be restored, if needed.

Other improvements

When running a reflow profile...

• History is maintained on display until user manually exits back to main menu or restarts another reflow run
• Reflow Phases are shown on display as a status message (preheat, soak, reflow, cooldown)
• Countdown timers for individual phases and for total reflow runtime.
• Improved PID Tuning, with linear interpolation between all data points over entire profile
• Some bug fixes included

Some screenshots

Edits enabled for all selectable Reflow Profiles

Profile Edit Options

Edit Parameters Scrolling Menu with scrollbar

Rename Profile Menu

Restore Defaults

Example Reflow Run

Reflow Run Completed, Cycling Message on Status line

Example RAMP SPEED TEST

---

Download Latest Firmware Release Version: PP.05.17.26 HEX file:

firmwarePP05_17_2026.zip

Release notes: This update provides for Setup/Calibration Menu support for an optional thermocouple interface boards. The interface board, when connected, replaces the built-in thermocouple interface normally provided by the T-162 Controller Board. Setup/Calibration Menu settings, for left and right thermocouple Gain and Offset values, now allow calibration of the thermocouples attached to the interface board, but only if the interface board and at least two thermocouples are detected by the firmware. See Wiki article "Better Dual Thermocouple Interface" for specific source code changes. No other source code changes were made.

Download Previous Firmware Release Version: PP.12.20.25 HEX File:

firmwarePP12_20_2025.zip

Firmware Installation Considerations

• No affect on Setup raw values saved to EEPROM, however the scaling for TC gains and offsets were changed to improve precision, so values displayed will be slightly different. You should save those calibration settings for readjustment after flashing. All other oven setup settings are retained.
• This firmware does overwrite any Custom 1 and Custom 2 temperature values previously stored in EEPROM.
• Recommend, immediately after flashing firmware, turning off power switch, disconnect ISP ground wire and FTDI serial programmer, then do a power-on boot.

Profile Support Info:

• Solder Paste Datasheets with manufacturer's guidance on adjusting reflow profiles...

Amtech Syntech_LF_42Sn58Bi.pdf

Amtech_4300 63Sn37Pb.pdf

Amtech_NC_31_42Sn57Bi.pdf

Final thoughts...

PID re-tuning was done for later model unmodified T-962 oven with exhaust vent duct. Tuning should be good for other models too. This firmware seems pretty bullet proof to me, but there might be something I missed or didn't fully test. Post comments if you have any problems. Enjoy.

[Maurizio66blu]

Hi! Please, I'd like to know what compiler (and the version) you used. Thanks.

[borland1]

I used Microsoft's Visual Studio Code with PlatformIO extension installed. Setting up PlatmormIO is easy, but requires help from Github user maxgerhardt. Here's a link to his fork for the T-962-improvements repository... maxgerhardt's Repository of Forked T-962-improvements

To make it happen, you only need to change/add files found in his main Github repository directory ( boards/lpc2134_01.json, Makefile, create_version.py, custom_upload.py, fix_linkflags.py, lpc21isp_linux, lpc21isp_win.exe, and platformio.ini ).

[GitLang]

Maurizio, I used NXP's own LPCXpresso which was free dirct from NXP. Others have used Keil, but it's not cheap.

[Maurizio66blu]

Thanks for the info, I didn't know Visual Studio could also compile for ARM.
LPCXpresso and the Keil compiler are also interesting.
But I have another question: has anyone used the IAR compiler?

[VincentFichetElectronique]

Hello, in the setup menu, it is impossible de set the TC gain under 0.8 with the firmwarePP12_20_2025.zip.
Please, is it possible to correct this for the next version. I need to set the gain at 0.506 for the right thermocouple.
Best regards, Vince.

[borland1]

VincentFichetElectronique,

Thanks for posting. I do have an update coming soon. This update will allow the Setup Menu's TC calibration settings to work with thermocouple interface boards similar to the board shown in the Wiki of this Repository (See: "A Better Thermocouple Interface"). I designed a similar interface which is compact, low cost, Dual MAX31850K PCB. It is suitable for mounting on the Main Controller PCB. In addition to updating the firmware for calibration with above interface, I'll also be adding the interface board details in the Wiki here ( Gerber PCB files, list of materials, parts sourcing, instructions, etc. ).

About your calibration request. I could make some adjustments to increase the range to say 0.500 to 1.500, with limits which you may not like. Because the Setup Menu settings are saved to EEPROM as unsigned 8-bit integers ( values of 0-255), there are limitations to what can be done. My prior thinking was that the original code setting range was too course, so my revised changes limited the Gain settings to values between 0.800 and 1.200, in increments of 0.004. Increment of 0.005 would have been better, but like I said, there are limitations. I did limit the range to minimize the time it takes to adjust from one end of the scale to the other. I could make changes to increase the range to between 0.500 and 1.500, with increments of 0.004, however your required Gain setting of 0.506 would be only adjustable to either 0.504 or 0.508.

Here's the original settings:
raw value range: 10 - 190 (180 steps), gain increment: 0.01, gain range: 0.1 - 1.9, less precision, wide range, but many steps

Current in PP firmware:
raw value range: 0 - 100 (101 steps), gain increment 0.004, gain range: 0.8 - 1.2, uses maximum available precision, less range, but fewer steps.

Some options:
raw value range: 0 - 250 (251 steps), gain increment: 0.004, gain range: 0.5 - 1.5 . max precision, but too many steps.
raw value range: 50 - 250 (200 steps), gain increment: 0.005, gain range: 0.5 - 1.5, slightly less precision, less steps.
raw value range: 0 - 100 (101 steps), gain increment: 0.005, gain range: 0.75 - 1.25, slightly more range, slightly less precision.

Because the two TC gains can roughly be set manually, by adjusting the two potentiometers on the oven's controller PCB, You might want to try this...

1. Reset the Setup Menu Right TC Gain to 1.0 (unity) and Offset back to zero, then adjust the appropriate Potentiometer to the desired Display temperature shown on the LCD display while running the Bake/Manual Mode at the high end of the oven temperature range.
2. Fine tune calibration for the Right TC Gain and Offset using the Setup Menu.

Let me know what you think.

Here's a couple photos of the populated interface PCB and also with it installed on the Controller Board. The MAX31850K chips include the cold junction temperature measurement, so the Dallas 18B20 Cold Junction Temp mod is unnecessary when this interface is implemented. Also, with the data line pullup resistor is on the interface PCB, making installation much easier to install than the Cold Junction mod.

[VincentFichetElectronique]

borland1,
hello. Thank you for your reply. After investigation, the best way is to adjust the gain potentiometers of the TC that have a problem. To do this, it works and I can adjust gain and offset in the range purposed in the firmware.
Also, is there a source code for the version 2025 ?
Best regards, Vince.