The big TEC driver thread!

[dnar]

Excellent..



Chris (LaNeK779) sent you the boards, I’ve got boards here so if you can E-Mail me the code I can do some testing on smaller TEC’s 3.9-30W.


I’m currently redesigning the board into 3 versions..


1.) An SMD version still based on the attiny13 with some of the changes mentioned above.
2.) Same as 1 but no SMD’s (haven’t found a sub for the irf7413 as non SMD yet)
3.) And a ATmega8 Version


Cheers!

Cool. I'll email you the hex file and details for the programming (fuse settings). Is 22c ok? I can do several compiles say at 20, 22 and 24c? Is this ok?

The BEER controller uses the 5V VCC for the NTC divider and reference also, it appears to be ok, the regulator does drift over time/temp I have observed 5.01 to 5.03 swings.

EDIT: I placed 3 hex files in a zip on the ftp server. ftp://ftp.photonlexicon.com/Software...narPI_1.00.zip

Frank, just about any common muck N channel MOSFET will do, Vcs 30V, Id approx 10 amps, Pd >=2.5

[dave]

This sounds pretty sweet.

You could also add a diode driver inhibit output if the temperature is outside of a predefined window.

Cool. On the list for a new design are:

1. Precision op-amp NTC front-end to opimise the ADC range with a small window.
2. Dedicated ADC Voltage Reference.
3. Output inductor LPF to produce true analog output for the TEC.
4. Re-assignment of ATtiny13 I/O pins (TEC output on port B0).
5. Possible change to another MCU, ATmega8 for example to provide adjustable set point trim pot.
6. Mounting holes on the PCB.
7. Screw terminals for power, TEC, fan and sensor.
8. Possibly current limiting also.
9. Possibly a trough-hole PCB to aid hand assembly.
10. Analog monitoring output to read in either 100mV/C or 100mV/F.
11. Standard 0.1" pitch 6-pin in circuit programmer connector.

[dnar]

This sounds pretty sweet.

You could also add a diode driver inhibit output if the temperature is outside of a predefined window.

Good idea! Added to the list above.

I wonder if your keen to test one of these boards Dave? I could shoot one over to you.

Tonight in the shower I also thought about something I raised long ago in this thread, using a H-bridge instead of MOSFET, it is trivial to have a controller that will both cool and heat to obtain the setpoint...

[LaNeK779]

wayne, i think you should start thinking of a name for this qute little thing

one term only, it has to have something to do with beer!!!

[Solarfire]

Tonight in the shower I also thought about something I raised long ago in this thread, using a H-bridge instead of MOSFET, it is trivial to have a controller that will both cool and heat to obtain the setpoint...

I’m already checking out some H-bridge candidates..

[dnar]

wayne, i think you should start thinking of a name for this qute little thing

one term only, it has to have something to do with beer!!!

OrangeWhipCrackaTinnySinkThePissDidYaBringaBeerOpe nSourceClosedLoopPeltierController?

I’m already checking out some H-bridge candidates..

The Old L298N is a goody up to 4A, I have a drawer full of them.

More ideas, the ATtin13 MCU contains an eeprom. I could define the setpoint in eeprom, allowing those with programmers to simply alter the setpoint. Maybe even the PI tuning params.

[LaNeK779]

OrangeWhipCrackaTinnySinkThePissDidYaBringaBeerOpe nSourceClosedLoopPeltierController?

nice and easy

[laser_freak]

More ideas, the ATtin13 MCU contains an eeprom. I could define the setpoint in eeprom, allowing those with programmers to simply alter the setpoint. Maybe even the PI tuning params.

Very good idea. I'd much rather write to the eeprom than recompile the code.

[mophead]

Nice work dnar. Kickass

[dnar]

I have added setting the set point (in degrees Celcius) to the EEPROM. The valid range is 18-26C in 1 degree steps. If an invalid value is detected, the default of 22C will be used. If anyone thinks this range should be larger, state you case.

I also converted the code to WinAVR/AVR Studio as the code now exceeds the CodeVisionAVR evaluation size limit. I prefer WinAVR personally.

AVR Memory Usage
----------------
Device: attiny13

Program: 410 bytes (40.0% Full)
(.text + .data + .bootloader)

Data: 15 bytes (23.4% Full)
(.data + .bss + .noinit)


Build succeeded with 0 Warnings...

I'll post the hex file up once I have tested, it compiles but I have other things to do today like get the Barbecue cranking and get drunk! (Australia Day Holiday).