Who is here? 1 guest(s)
 Print Thread
Skywalker roaster mods
renatoa
When powered from USB, the regulator is bypassed, thus board is running at 5V.
You don't want board powered at 3.3V. this means you need level shifters on D2/3, as the pulses coming from skywalker are 5V.
So there is no absolute need to power the board from skywalker, but... with a warn !
You should always connect the USB first, THEN power skywalker.
And DO NOT disconnect USB from laptop when roast finishes, without first unplugging skywalker from the mains.
If the later happens, you can experience the so called phantom powering of the board, via data lines, which cold damage board, not instantly but in time.
This issue can be easily fixed placing a 2.2...4.7k resistor in series on each data line, i.e. D2/3. As sketched in the image attached.
Diodes aren't required, they are already existing, embedded into the microprocessor.
renatoa attached the following image:
image_2024-07-06_102614603.png

Edited by renatoa on 07/06/2024 11:28 AM
 
Dan N
There are 2 versions of Pro Micro: 5V/16MHz and 3.3V/8Mhz. Make sure you have the 5V/16Mhz version.

I just realized the Pro Micro power hookup is different from the others. Unlike the Nano, the Pro Micro can't have a connection to external 5V and USB 5V at the same time, so do not connect anything (like 5V from the roaster USB) to VCC pin. That also confirms renatoa's recommendation in the post above to always leave the laptop connected while the Skywalker is plugged in.

Pro Micro also has a jumper (J1) that may need to be shorted to bypass the regulator when using USB power only. It could be different for the different clones so check the docs for your board.

Sorry, if I'd known about all this complexity I wouldn't have recommended the Pro Micro. Definitely not beginner friendly. I do recommend the Nano for a small build.
 
billsey
The ones I got are 5V, so that isn't an issue. I believe I can just use three wires to connect, D2, D3 and GND. I'm going to connect that up and see what happens, I expect it will work correctly. I am going to have to use wires to make that connection, the three holes in the Pro Micro do line up with the three pins on the USB, but at right angles to where I want to USB connection to the computer. I think Ill use double sticky foam tape to hold the Pro Micro in place until someone designs a box for this all. It's just a slight bit larger than the USB Y configuration I started with, because the Pro Micro is just a bit longer than the USB female board. I'll see if I can post a pic once it's wired up...
 
renatoa
The phantom powering IS a known issue for ANY Arduino board, is not a specific case for the micro board only. Not related to 3.3V powering, happens for 5V boards too.
It happens when any Arduino data pins are connected to another board powered from other source.
 
billsey

Quote

renatoa wrote:

The phantom powering IS a known issue for ANY Arduino board, is not a specific case for the micro board only. Not related to 3.3V powering, happens for 5V boards too.
It happens when any Arduino data pins are connected to another board powered from other source.

OK, I promised pictures a few days ago, but didn't get around to soldering anything until today.
Let's see if the board will let me attach them today.
billsey attached the following image:
pxl_20240713_235027926.jpg
 
billsey

Quote

billsey wrote:
OK, I promised pictures a few days ago, but didn't get around to soldering anything until today.
Let's see if the board will let me attach them today.

That gave me one picture, let's try the other... Nope, same error "attachment type not allowed" even though it's a jpg just like the last one. :(
 
renatoa
Smaller than 10M ?

Any functional test so far ?
 
billsey
215KB vs. 218KB. I was going to try a batch where I just monitored using Artisan today, but ended up playing pool instead. Maybe tomorrow...
 
billsey
OK, first try was a rousing ... less than success. Artisan could talk with the Pro Micro and the roaster display came up fine, but when I told Artisan to Start the drum started spinning with a 100% heat setting and the roaster didn't actually heat. After a bit of fiddling I got the roaster to give me an E3 to say it wasn't communicating with the temperature sensor. My guess is I have the D+ and D- lines swapped, since the documentation doesn't really tell you which is supposed to connect to D2 and which to D3, except through wiring colors. I opened two different USB cables over the last couple of week and they both had different colored wires than illustrated, so I ended up guessing. I'll do a rewire tomorrow or the next day with those swapped to see if it makes a difference.
 
renatoa
E3 is drum not working/rotating, unrelated to sensor/temperature. That is E1.

Supposed you will try the Spy version first, before trying control from Artisan, which is tricky.
 
billsey
I was hoping I wouldn't have to flash a different version in order to spy... Thinking of the main version just watching without doing any control.
 
billsey
OK, I flashed the spy version and swapped those two digital lines, then ran a roast. Nothing came through at all. Either I messed up one of the connections or I had the wiring right before. Back to the drawing board to test the wiring...
 
renatoa

Quote

billsey wrote:

...then ran a roast...


For the first builds, to accommodate with the new routine of roasting tethered, without wasting beans, I simply simulated a roast with empty machine, trying to follow a profile, just to train my senses with machine response to different commands.
Attached is a sample of such empty machine "roast cycle".
Except for the lower power levels, anything else is same as for a real roast.
Also, lower charge temperature was required, to be able to simulate a TP as close to machine loaded.
For this reason I cut the power and maxed up the fan in the first minute.
renatoa attached the following image:
image_2024-07-18_090754797.png
 
billsey
Ah, but since so far I am not seeing any of the data, my curve is completely flat at the bottom of the scale. I should be able to turn the roaster on, so the display is showing ambient temperature (say 23C), hit the start button on Roaster Scope and see that 23C as a flat line in the display, correct?
 
billsey
OK, more testing... For some reason when I plug the roaster in it lights the LED on the Arduino even without the Vcc connected to anything. Likely it's seeing power on those two data pins. No big deal, just surprising.
When I plug it into my laptop with SkywalkerSpy flashed it doesn't see ambient temperature even when I tell Artisan to start. When I flash SkyCommand instead I do see ambient temperature, but it wants to start the drum for me when I start. So, I'm not getting what I hoped for which was to be able to monitor with Artisan or control with Artisan without reflashing the Arduino. I'll see if I can delve into the code to try and figure out why the spy software doesn't seem to be working at all...
With the roaster plugged in, the Arduino connected but nothing turned on I'm reading 2.5V or so with my VOM between either D2 and GND or D3 and GND. I believe that is why the LED lights. There is 5V between VBUS and GND on the 'notUSB' port as expected, but I don't have VBUS connected to anything. It makes perfect sense that there be power available there since you have to have power to be able to turn the system on with the controller. My guess is the data pins are getting data signals even before you turn everything on, which explains the ~2.5V my meter shows.
What I really want is to monitor and graph using Artisan by default, and either take over control in the middle of a roast or use full control during the roast without having to either switch out the Arduino or flash new software to it. Hopefully that will be doable eventually.
My initial thoughts regarding the VBUS connection on the 'notUSB' port seem to be born out, it's not needed to receive or send data to the Arduino and from there to the roaster, though it could be used to run the Arduino without a laptop connected. Unfortunately, without a separate computer connected it's kind of tough to run Artisan for control, so you will likely be always using the USB connection to power the Arduino. At least, I don't think there's a version of Artisan that will run on the Arduino... That's more of a Pi thing, or a dedicated computer.
 
billsey
OK, first glance at the code... SkyCommand sets the serial port at 115200, SkywalkerSpy sets it to 9600 instead. That could easily be why I'm not seeing anything.
 
DancingDog

Quote

billsey wrote:
OK, more testing... For some reason when I plug the roaster in it lights the LED on the Arduino even without the Vcc connected to anything. Likely it's seeing power on those two data pins. No big deal, just surprising.

This is a known issue, and will eventually fry the Arduino. Best practice is to power the Arduino first, by plugging it into the computer's USB before plugging the modified data cable into the roaster; this prevents the backfeed through the data pins.
 
renatoa

Quote

billsey wrote:

OK, first glance at the code... SkyCommand sets the serial port at 115200, SkywalkerSpy sets it to 9600 instead. That could easily be why I'm not seeing anything.


Perhaps do you mean josh original spy... I thought you will be using mine, which is 115200, just checked.
 
renatoa

Quote

billsey wrote:

OK, more testing... For some reason when I plug the roaster in it lights the LED on the Arduino even without the Vcc connected to anything. Likely it's seeing power on those two data pins. No big deal, just surprising.
....


Really? Wasn't this beaten to dead last page... the ghost powering issue... and seems not read...
As DancingDog said...
 
billsey

Quote

renatoa wrote:
Perhaps do you mean josh original spy... I thought you will be using mine, which is 115200, just checked.

I'm using the one on the jmoore52 github page. Where is yours?
 
billsey

Quote

DancingDog wrote:This is a known issue, and will eventually fry the Arduino. Best practice is to power the Arduino first, by plugging it into the computer's USB before plugging the modified data cable into the roaster; this prevents the backfeed through the data pins.

Good to know, I likely skipped over that part of the discussion back when I initially read this thread. I only plug the roaster in when I'm ready to start a roast session, since it takes up so much room in my small kitchen, so that shouldn't be a problem over the long term. I'll just connect the laptop first, then plug the roaster in.
 
renatoa

Quote

billsey wrote:

Quote

renatoa wrote:
Perhaps do you mean josh original spy... I thought you will be using mine, which is 115200, just checked.

I'm using the one on the jmoore52 github page. Where is yours?


post #220 https://homeroast...post_78398
 
Jump to Forum: