I finally have what I consider an answer to the non-support of the full line of PSOC 62 in the PSOC Creator. The answer may surprise you. The PSOC 62 “245” part that came out in March of 2020 is NOT a PSOC.
The PSOC 62 2×5 series of parts is an SOC! Think Atmel. Think Microchip. Think PIC. Do not think PSOC. It has NO UDB’s (Universal Digital Blocks.) I say again, it is NOT a PSOC, it is an SOC. You cannot create custom logic within the device.
So, the reason PSOC Creator does not support that board is because there are no logic devices to support. There appears to be a sea change in Cypress thinking. They appear to need to get higher volumes with cheaper parts. That precludes being state of the art. It requires that Cypress follows the crowd, probably somewhere near the back of the bus (in my opinion).
From a purely mercenary point of view this is a requirement. It is driven by Cypress’ parent company. I don’t blame them. They have to survive. Infineon has to get their money back. Cypress has to exist in the future in some form, or we won’t have access to the PSOC family in the future.
Unique Market Hole Being Established
So, there will be a hole in the market in the future. (Think 5 to 10 years) That hole is slowly being created by Infineon now.
Unfortunately for weekend engineers, it appears to have the distinct features of being a low volume, high piece cost per device hole. This market vacuum will not attract the attention of very many chip manufacturers. They depend upon volume to get their parts costs down and their profits up.
One possibility, which is a fairly recent trend in FPGA’s is to embed ARM Processors. So, a small FPGA with an ARM processor, if tied together with a GUI design environment, would provide a viable alternative to much the PSOC family, *if* it also included some analog. I think this will eventually happen, just not sure when. Whether or not it is available to the weekend engineer market, I can’t say.
Currently, Xilinx and Altera, which have been purchased by major companies (think AMD and Intel), have lines of FPGA’s with embedded processors. I think their economic drives will be to put together more and more expensive IC’s with more and more Logical Elements in each package. That will put a boundary on the cost at the low end, probably moving them out of the hobby market.
With some “baby” FPGA’s out there, think https://tinyfpga.com/, and Field Programmable Analog Arrays from https://www.anadigm.com/fpaa.asp, and https://www.okikatechnologies.com/solutions/field-programmable-analog-array-fpaa/, there are new possibilities opening up. Unfortunately, FPAA’s have not caught on like FPGA’s. Maybe that will change in the future.
If you look at the Anadigm part, there is the AN231E04-E2-QFNTY, it is only one flavor available on Arrow (Arrow.com) website. Minimum order 1,300 at around $11.50 per part. Part cost is reasonable, minimum order is not. Someone with more moxie than I have could order and resell, doubling their money, but I am not sure how long that would take. Probably not worth it unless the purchasers were lined up in advance.
If you look at the Okika Technologies part, it appears to be around $12 each, but you have to order 10 at a time. That is probably more doable with enough “friends.” Their development environment looks fairly nice. It is reminiscent of LabView. They have an expensive ($200) development board.
So, there are not any “FPAA” units out there for the weekend engineer, except for the PSoC 3,4,5 family, and some variants of the PSoC 6x family. Luckily, the PSOC 5 development board is available from Mouser, and DigiKey, and direct from Cypress website, for *very* reasonable prices.
I’ve been promising to get back to the 3D printer project and explain and show the heater code. Hopefully next time.
I have ordered the Cy8CKit-059 to RAMPS 1.4 adaptor boards (1v1) and will be building one. I tested the 1V0 board with “green wires” that fixed my mistakes. It currently is running a 3D printer for basic testing. I have X, Y, Z, and Extruder working, along with Extruder and Bed Heaters on a very old Mosaic 1v0 printer. I am not fully happy with the results yet, but it is getting there (slowly). Using FreeRTOS allows me to be straightforward in the code implementation, but there are other considerations. A fly can land in your soup at any time.
If the 1v1 059 to Ramps 1.4 board passes my tests, it will go out on GitHub, as an Eagle project, under GPL V3 license.
No UDB’s in ModusToolbox: Cypress Community to the Rescue!
At the current time, I recommend you stick with the PSOC parts that include UDB’s when doing weekend projects. If you use ModusToolbox, you *cannot* use UDB’s. If you *really* need to use ModusToolbox and can scrape together a Windows environment, then there is a way to use UDB’s in Modus Toolbox.
The solution was created by user “RONO” and he posted it in the Cypress Community here. The porting tool talked about in the thread is here. This gentleman is a brilliant engineer and seems dedicated to helping the Cypress Community use the Cypress PSOC parts.
Unfortunately, due to this Blog being focused on the PSOC family (as opposed to the SOC family of parts from Cypress), I will not at the current time finish covering the PSOC 62, 2×5 or the PSOC 6, 24A parts, which are in the CY8CPROTO-062 series of development hardware from Cypress.
This is rather sad, since those parts make it very difficult to use when prototyping dense projects. I can think of some very good wireless projects for units that have good UDB support. I would pay a few dollars more to get a prototype board hosting a PSOC that allowed the use of UDB’s, but that is not currently available from Cypress.
I can hear the question, Dense Projects? Huh? I design and program electronics that work up to 350 degrees Fahrenheit. The PSOC 5 series works reliably at that temperature, for a limited number of hours; but that is an understood and accepted limitation for that environment. I use logic gates, counters, PWM, DACS, A/D’s, comparators, and Op-amps in my designs inside a PSOC 5LP. All within a 68 pin QFN device, working at 350F, with known drift characteristics. Marvelous.
The width of the circuit boards is around 1/2 inch. The voltages interfaced to and powering these boards is around 100 to 200 volts, at temperature. Needless to say, a lot of trial and testing is involved. My analog skills have improved markedly over the last few years.