Reflow Controllers delayed
by MOAM Industries
We have taken the decision to put the reflow controller through one last revision before we make them available. This has been a ahrd decision to make as we felt we were so close with the current revision and we possibly could have made do with the current version. However following our motto of make great things we felt we could not release a product that we were not 100% happy with and felt it should be better.
The troublesome area was our thermocouple amplifier, unfortunatly we could not create an accurate op amp stage to reliably provide feedback on the oven’s temperature. Considering the accuracy required when reflowing PCBs we felt this was not acceptable. The problem arrises from our now clearly too simply differential amplifier circuit, we put a little too much faith into this application note from micro-chip which we based our design upon. Realistically at a minimum we should have utilised an instrumentation amplifier configuration as opposed to a single stage differential amplifier. With the current design we were susceptible to too many variables for a reliable 1 degree repeatability, we required higher than 1% accurate resistors, emf shielding and additional amplifier stages to improve uppon the current design.
A complete redesign of the anolgue circuit and the inevitable increase in complexity associated with this would have delayed us months and added considerable cost in high accuracy components. The first alternative is the AD595Q from nalog devices, a through hole mounted chip which amplifies, linearises and cold junction compensates a thermocouple with little additional circuitry. Unfortunatly these chips are $15 each! from digikey, a considerable step up from the $1 amplifier and passive components of the previous design. The other alternatives to this chip are all surface mounted components of which the assembly techniques are too complex for a kit.
Therefore our final solution has been to develop a breakout for the MAX31855 surface mounted thermocouple amplifier and to provide it with the kit. The design is based on that developed by Adafruit Industries (we really liked there level shifting technique) and will also be available seperatly from our store. There are a few minor diferences in our version, filtering capacitors on the supply input to the regulator and the input as well as 3.5mm mounting holes for compatability with the M3 hardware we typically use, we also plan to bring out an alternate version including a “propper” thermocouple connection as oppesed to the common screw terminals.
The advantage of using the MAX31855 is we no longer require complicated analogue circuitry or the risk of signal noise. Temperature information is retrieved by the microcontroller using a serial interface to the breakout board where the signal is level shifted to communicate with the 3.3V MAX31855. Both the thermocouple temperature and cold junction temperature can be retrieved from the amplifier. The only unfortunate side effect of this alteration is the removal of the running indication LED from above the LCD. As the removal of the previous analogue circuitry only made two pins available we needed to sacrifice the LED pin to provide the third signal pin required for the amplifier.
So over the next few weeks we will be ordering the additional components, and testing our schematic design before ordering the next production batch of PCBs. Unfortunatly last time we were a little too over eager and we took a gamble that did not pay off, however you live and learn and we shall be more prudent in the future.