For some time now we have had the Microchip Microstick that I reviewed back in December 2010. Its a great little product and one I have used in anger over and over again when trying out new ideas. Then a few weeks back there was lots of hype and excitement about the new PIC32MX range of chips that come in a 28 pin package. Its no great surprise then that Microchip have upgraded the Microstick to now include the new processor. I was lucky enough to get one of these early kits to test drive and review, so you can see how its measures up.
So out of the box there is little difference from a standard Microchip development board as you get the board, USB cable and a quick start guide. The kit comes with a PIC32MX220F032B chip which I’m guess is what everyone wants to play with, but the kit also includes a PIC24HJ128 a PIC24FJ64 as well as a dsPIC33FJ128 for all your basic DSP requirements. Like before the kit comes with a on-board programmer that can programme and debug any of these chips as well as supply power via the USB. The board has the same 28 pin wide DIL connector on the bottom that allows you to plug this directly into a bread board or your own designs. The new design is also 90% pin compatible - they have added the PGEDA and PGECA pins that allow programming off board but also gives two more I/O lines than before. However for some reason they have taken off the reset / MCLR pin. So there is no way of syncing your reset on your design with that of the micro - totally daft idea in my view and I’ll be hacking mine to bring the pin out.
Other updates to the board include moving the LED that was on pin 26 to pin 2, however good news is that you can remove a jumper so if you don't want to use it for that purpose then you can remove it and reuse the pin as an input for example. We also now have a switch added that allows for switching between the two programming ports of the chip. This is good as it allows the developer to select which pins are used for programming / debugging and what is used for I/O.
Other than that its all pretty much the same. There is still the extra TTL serial port available via a end connector that is compatible with the Microchip Serial Analyzer and the board is a little longer than last time.
However I’m not certain this kit is all about the Microstick. Yes its easy to use and cuts down on the cost of having a programmer, standalone chip and then connecting it all up. But the power of these kits can also come from the chips that are on them. The PIC32MX220 and its fellow family members are definitely chips to consider. Free development software and cheap and easy tools help lots and this I think is the one that will make me consider the switch to 32 bit micros for small embedded systems.
Thanks Paul
( @monpjc )
Microchip Microstick II Home Page.
RS Microstick II Page
jamodio
It is a shame that Microchip does not provide the entire schematics for the Microstiks. IMHO for the DIP parts could be a better investment to buy a Pickit3 or equivalent and get a sample for the PIC32MX220, I'm also dissapointed that Microchip didn't release yet the 128K flash parts.
Before the Microstick existed I put together a small pcb for rapid prototyping of Microchip 28-pin SOIC parts, and I'm about to confirm that it will probably also work with the new PIC32MX family.
http://www.ljcv.net/pp28/index.html
If somebody is interested to review this board feel free to contact me.
Cheers
Jorge
MicrochipRTCfr
Hi Pete,
Maybe what is missing from this review is the fact that this kit includes also a debugger (2nd PIC micrco) on the bottom side of the microcstick II
It can be used in MPLAB environment to program/debug the user PIC/dsPIC.
It is as is you had a Pickit3 on-board ;=)
Therefore, the ICSP is connected between the debug PIC and the user PIC.
The full schematic of the microstick II can be seen in the information sheet : http://ww1.microchip.com/downloads/en/DeviceDoc/51951B.pdf
regards
saipan59
Thanks for your review Paul.
What do the 6 vias on the end of the board do? Does it support the PICkit style programmer hookup? I think when I looked at the schematic, it only showed 2 of the pins being connected to anything, but I was wondering if that was an error on the schematic.
Pete
michaelkellett
Why should the PIC32 tempt you to use a 32 bit micro for small embedded systems more than an ST or NXP ARM ? All are supported by free tools (for small sizes) and all are very cheap (at the low end: much less than £2 for prototype quantitites and much less than £1 for 1000+).
The PIC is available in DIP but this shouldn't make any difference - it's easy to solder TQFP or SOIC by hand and not much harder to solder the leadless HVQFN packages with solder paste and a domestic (cooker type) hot plate.
I would definately go for a 32 bit processor for any but mega volume (or other very special requirement) applications now, but I think the ARMS give you more choice (far more sources for Cortex M0,3,and 4 parts).
Michael Kellett
pauljclarke
I think the thing for me is that my industry (and our products) has not needed 32bits cores and it’s arguable that we still don’t. However having a chip that matches cost and foot print with more features is tempting as a route forward. I’m always hesitant about using a sledge hammer approach for small embedded applications. I laughed just a few day ago when I say a dev board that is running a demo using a 32bit ARM to launch a RTOS to blink a LED.
Your correct that ARM are offing some great choices, however there are still problems that people are experiencing with the tool chains. This shows though with kit like the mbed and the LeafLabs Maple boards which avoid these tools and use their own. Microchip for good or bad reasons have always offered a all in one IDE platform that works out the box. There are many reasons why people select a chip and this should always be part of the engineers processer selection process. Identify the right chip for the right application with the right number of bits with the right support and tools.
AtomSoft
Why did you laugh? Running RTOS to blink a LED shows one can implement custom hardware in a RTOS enviorment. For instance creating a built in COM port for serial communication and such.
I find it pretty cool. Also these dev sticks are a waste i think. If its a DIP package i dont think a Dev board is needed at all! Its its TQFP and LEADLESS parts then a dev board is a nice solution to get playing and testing quickly if one doesnt have the means to create there own PCB at home.