A lot of people have been asking “How is the chip shortage effecting the project? How does this impact V2? Can you help us find some LPC chips?”…it is a major issue. For those of you who don’t know what is happening there is a great writeup by long time community member Kliment.
A few months back we went about getting a batch of 5xc V1.1 boards produced by our usual supplier. Initially there was no concern. This was just before Chinese New Year…so we had our usual delay that one would expect during that time when dealing with anything from China. Then shortly after we started getting emails..
“The MCU you use has gone up in price by 4x, the PHY and stepper drivers as well…there will be a price adjustment which will need to be made” (greatly paraphrasing)
This is a massive issue when importing to the United States due to the 25% tariffs. When cost goes up it goes up by x+(.25x). This was annoying but acceptable. Then came 2 days later….
“The MCUs we secured were sold out from under us…we cannot even get the chips. (and since we already started purchasing we cannot refund)” (again…paraphrasing)
At this point…what do you do? But lucky for us before we figured that out…2 days later another email.
“We found the MCU…but now they are 5x what you normally pay for each unit”
So…at least there is that.
The initial plan was to continue purchasing batches of V1.1 boards as we transitioned to V2 to allow to keep a backstock for OEMs or those who still wanted to use V1.1. Due to the shortage we are limited to the boards we purchased in this batch and we do not expect more V1.1 Smoothieboards until there is stock of the LPC1769 at a “normal price” (last price estimate I saw was ~$100/MCU and it was not from a reputable vendor).
What is the influence on V2? What now?
Immediately when we became aware of the shortage the first thing we did was look up the LPC4330 and realized…yep. Sold out worldwide for the most part…just like the LPC1769.
So…the realization sets in. No more V1.1 boards. V2 which was weeks away from final prototype was now completely blown out of the water.
First step was to talk to Wolfmanjm. “so…um…you haven’t liked the MCU we were using a whole lot anyways…if you could have any chip what would it be?” to which he responded “something STM32xxx because the code I initially wrote was already based for that series..” (quotes are meaningless…this is very much not verbatim 🙂 )
Ok…so great. That is something at least. But now to see about the board.
At this time Kliment had volunteered to help with the remainder of the final revisions to the board and was just getting started on sorting files/docs/etc. We looked the board over and he saw quite a few improvements which he could make and everything was looking good. Then I dropped the question…
“What do you think about doing the core layout but leaving the center blank in case we need a new MCU?” To which he respectfully told me “no”. Then I mentioned the issue with the MCU shortage and how it was dead in the water essentially. His very valid argument was it would be way too much work to port the firmware. I let him know WJM said the FW was fairly easy to port and it wouldn’t be much work…his response was “I don’t believe you” 🙂 (totally valid statement). Lucky for me WJM was in the channel and the discussion started back and forth and we decided it could be done as well as “should” be done.
About this time Arthur realized what we were up to and came in to point out that “this isn’t what the backers of the Kickstarter were promised…and there would be some angry customers” but it did not take much discussion for him to realize the situation.
2 months of dev work…or 10+ months of no possibility of components.
So.
-Flash forward…we found the MCU we wanted.
-The port is largely completed based on what WJM could do with a dev board. Firmware seems to be going very well.
-Kliment is working on the board layout and there is quite a bit of redesign going on based on what is available as well as taking “lead time” into major consideration when component choice was done.
-We expect to have final protos shortly and once the build is verified and tested we will be able to produce.
But what about the MCU shortage? What MCU are you using? Is the sky falling anywhere else?
We chose to use the STM32H745 and as you may notice. It is nearly sold out worldwide as well. Isn’t that an issue? Well…in our case no. We managed to secure enough MCU to get the initial Kickstarter batch out as well as some retail boards which should give us a good start on the project that aligns nicely with the MCU coming back in stock through the upcoming months.
Chip comparison link: Here
As for the sky falling…Too busy with our noses to the grindstone to look up but I don’t hear the meteor whistling through the atmosphere yet. But who knows. Pandemics, shortages, wars….they make things a bit less stable than they used to be but it just gives a bit more challenge to an already difficult task. But I for one believe that no good story ends with “…and then we gave up…”. So I will likely be here for the long haul.
The Smoothie team seems to be pushing forward with V2 through the worst of this all…and without the superheros like WolfmanJM and Kliment we would be in a very, very difficult position. If anyone wants to know how to help the project feel free to go to Smoothieware.org and look into the information about how to help or volunteer.
Anyone who wishes to help out financially there are links to the developer donation which (as of this writing) goes directly to the software development side of things.
As for helping out with financials for the hardware side of things. The primary hardware development (aside from the Kickstarter campaigns) has largely come from Smoothieboard sales. Support authentic Smoothieboards and their resellers. A portion of the profits from sales go directly to further Smoothieware/Smoothieboard development.