BruControl: Brewery control & automation software

[aeviaanah]

Everything ran fine for approx 24hr period wired 12v through jack. Lost connection shortly after. Only component wired was the RP-3 board (with 4 amplifiers) powered from 5v pin. Basically the Ethernet board lights change, it blinks it's lights rapidly different from a healthy connection.

Any known issues with RP3 and or amplifiers? Any suggested tests/ diagnosis?

 

[RiverCityBrewer]

I found a screw terminal for an ESP32, but its the DevKitC version, so I picked up one of those boards as well.



When I flashed it, it appears to have flashed normally, but in the debug terminal it looks like its stuck in a boot loop. Does a new firmware need to be cut for the DevKitC variants? I can get more logs if you need them.

Edit: I flashed attached and detached from the shield, same result. I tried both 45L/N with the same result.

esptool.py v2.6-beta1
Serial port COM3
Connecting........_
Chip is ESP32D0WDQ5 (revision 1)
Features: WiFi, BT, Dual Core, 240MHz, VRef calibration in efuse, Coding Scheme None
MAC: fc:f5:c4:3c:4f:24
Uploading stub...
Running stub...
Stub running...
Changing baud rate to 921600
Changed.
Configuring flash size...
Auto-detected Flash size: 4MB
Compressed 8192 bytes to 47...
Wrote 8192 bytes (47 compressed) at 0x0000e000 in 0.0 seconds (effective 3120.8 kbit/s)...
Hash of data verified.
Compressed 17664 bytes to 11528...
Wrote 17664 bytes (11528 compressed) at 0x00001000 in 0.2 seconds (effective 929.7 kbit/s)...
Hash of data verified.
Compressed 1423664 bytes to 837158...
Wrote 1423664 bytes (837158 compressed) at 0x00010000 in 12.6 seconds (effective 901.1 kbit/s)...
Hash of data verified.
Compressed 3072 bytes to 143...
Wrote 3072 bytes (143 compressed) at 0x00008000 in 0.0 seconds (effective 1170.3 kbit/s)...
Hash of data verified.

Leaving...
Hard resetting via RTS pin...

rst:0x3 (SW_RESET),boot:0x13 (SPI_FAST_FLASH_BOOT)
configsip: 0, SPIWP:0xee
clk_drv:0x00,q_drv:0x00,d_drv:0x00,cs0_drv:0x00,hd_drv:0x00,wp_drv:0x00
mode:DIO, clock div:1
load:0x3fff0018,len:4
load:0x3fff001c,len:1100
load:0x40078000,len:10088
load:0x40080400,len:6380
entry 0x400806a4
ets Jul 29 2019 12:21:46

rst:0x3 (SW_RESET),boot:0x13 (SPI_FAST_FLASH_BOOT)
configsip: 0, SPIWP:0xee
clk_drv:0x00,q_drv:0x00,d_drv:0x00,cs0_drv:0x00,hd_drv:0x00,wp_drv:0x00
mode:DIO, clock div:1
load:0x3fff0018,len:4
load:0x3fff001c,len:1100
load:0x40078000,len:10088
load:0x40080400,len:6380
entry 0x400806a4
ets Jul 29 2019 12:21:46

rst:0x3 (SW_RESET),boot:0x13 (SPI_FAST_FLASH_BOOT)
configsip: 0, SPIWP:0xee
clk_drv:0x00,q_drv:0x00,d_drv:0x00,cs0_drv:0x00,hd_drv:0x00,wp_drv:0x00
mode:DIO, clock div:1
load:0x3fff0018,len:4
load:0x3fff001c,len:1100
load:0x40078000,len:10088
load:0x40080400,len:6380
entry 0x400806a4
ets Jul 29 2019 12:21:46

 

[BrunDog]

Everything ran fine for approx 24hr period wired 12v through jack. Lost connection shortly after. Only component wired was the RP-3 board (with 4 amplifiers) powered from 5v pin. Basically the Ethernet board lights change, it blinks it's lights rapidly different from a healthy connection.

Any known issues with RP3 and or amplifiers? Any suggested tests/ diagnosis?

Ethernet and the RP-3 both use the SPI bus for communications. If there is noise on the line, for example induced by the wiring to the RP-3, it could disrupt the Ethernet communications. I don't recall your build directly, but if your interface and RP-3 are not right next to each other, this could be happening. I'll go back and look for a picture.

Edit: Went back to look for a picture. I would suggest sliding your RP-3's to the right to be as close to the MEGA as possible, and trim the wires down as short as possible. Also consider twisting them to help reduce noise pickup.

 

[BrunDog]

I found a screw terminal for an ESP32, but its the DevKitC version, so I picked up one of those boards as well.

View attachment 697745

When I flashed it, it appears to have flashed normally, but in the debug terminal it looks like its stuck in a boot loop. Does a new firmware need to be cut for the DevKitC variants? I can get more logs if you need them.

Edit: I flashed attached and detached from the shield, same result. I tried both 45L/N with the same result.

No, DevKitC is what we use quite a lot. Make sure the buttons aren't stuck closed. Otherwise I'm not sure why its looping. Do you have another one on hand?

Nice to see a shield for this.

 

[RiverCityBrewer]

Ok, I'll mess with it a bit more. I don't have another on hand that isn't already flashed and the DevKitC variants seem to be a bit wider than the other boards so they wouldn't fit the shield anyway.

 

[RiverCityBrewer]

No dice on flashing, buttons are not closed and they respond as expected when pressed. I tried manually putting the board in flash mode but it didn't make a difference. The variant I have is a ESP32-WROOM-32E, these were released this year, possibly as recently at June. This appears to be a new chip (ESP32-D0WD-V3) from previous gen ESP32-WROOM-32D. https://www.espressif.com/sites/default/files/documentation/ESP32_ECO_V3_User_Guide__EN.pdf

Do you have any 32E boards? I don't think this board is a dud since I can flash other sketches to it.

 

[BrunDog]

Doesn't seem there is a software change. When you flash over other sketches, what target are you using?

Do you have a link to this board - we may need to source one to debug it?

Edit: I found one... will order.

 

[RiverCityBrewer]

Well, it installed a sketch correctly one time, now its doing the same thing with the constant boot loop... I'll send it back and get a new one and we'll see what happens.

 

[RiverCityBrewer]

Now I'm at a loss... using the Arduino IDE 1.8.13 in Ubuntu, I changed the flash mode from the default QIO to DIO and now it will load sketches properly. Using the esptool.exe program in Windows, it gives the boot error. I'll look further into the flags set in the InterfaceSetup script and see if anything jumps out.

 

[aeviaanah]

Ethernet and the RP-3 both use the SPI bus for communications. If there is noise on the line, for example induced by the wiring to the RP-3, it could disrupt the Ethernet communications. I don't recall your build directly, but if your interface and RP-3 are not right next to each other, this could be happening. I'll go back and look for a picture.

Edit: Went back to look for a picture. I would suggest sliding your RP-3's to the right to be as close to the MEGA as possible, and trim the wires down as short as possible. Also consider twisting them to help reduce noise pickup.

What picture you looking at? My RP3s are directly above Mega pretty close actually

 

[BrunDog]

What picture you looking at? My RP3s are directly above Mega pretty close actually

Ok. Might have been a dated pic. Make sure the wires are short as possible

 

[RiverCityBrewer]

I assumed the issue was with the bootloader file (bootloader_qio_80m.bin), so I pulled down the latest bits from Github for the ESP32. I swapped the latest of bootloader_qio_80m.bin into the resources folder but had no luck with it. For grins, I then swapped the bootloader_dio_80m.bin in, modified the script, and now it seems to have loaded correctly. I'm not sure what the long term ramifications are of that, and I haven't tried flashing any of my older ESP32 boards with the current configuration. I'll be curious to see if you have the same issue with the 32E version as well.

 

[aeviaanah]

Ethernet and the RP-3 both use the SPI bus for communications. If there is noise on the line, for example induced by the wiring to the RP-3, it could disrupt the Ethernet communications. I don't recall your build directly, but if your interface and RP-3 are not right next to each other, this could be happening. I'll go back and look for a picture.

Edit: Went back to look for a picture. I would suggest sliding your RP-3's to the right to be as close to the MEGA as possible, and trim the wires down as short as possible. Also consider twisting them to help reduce noise pickup.

I'm assuming your talking about the SDO, SDI and SCLK wires? I am currently using Arduino breadboard wires with male pins. Wires are about 10" long, does this sound too long? Should I use shielded wire? What gauge wire should be used at VCC, GND, SDO, SDI, SCLK and CS1/4 at RP3?

 

[BrunDog]

Short as possible. Twisted if possible. 22 or 24 AWG is ok.

 

[aeviaanah]

Short as possible. Twisted if possible. 22 or 24 AWG is ok.

Which wires pick up the noise?

 

[BrunDog]

All can. Also route them away from HV devices.


Do you have MOVs and/or snubbers in your contactor coils? These only induce noise when switched, especially off.

 

[Paulaner]

Which wires pick up the noise?

All this information is already in this thread, if you read the manual in post 1475 I can all but guarantee half of the problems you keep having will go away.

 

[aeviaanah]

All can. Also route them away from HV devices.


Do you have MOVs and/or snubbers in your contactor coils? These only induce noise when switched, especially off.

Yes I plan to wire MOVs to contactor coils. Correct me if I'm wrong but the issue is likely beyond that since the only components currently wired are Mega, Ethernet, RP3 and power supply. I have all recommendations (including routing away from high voltage devices) already covered with the exception of twisting the wires.

I tried a different RP3 board perhaps the issue is in one of those. It's been running all night with no issues but I'll give it a few days since I got this far with the previous RP3. I'll go to a higher quality shielded wire like the stuff I bought for RTDs.

 

[aeviaanah]

All this information is already in this thread, if you read the manual in post 1475 I can all but guarantee half of the problems you keep having will go away.

Thankyou for the link, I'll check it out.

 

[BrunDog]

Correct me if I'm wrong but the issue is likely beyond that since the only components currently wired are Mega, Ethernet, RP3 and power supply. I have all recommendations (including routing away from high voltage devices) already covered with the exception of twisting the wires.

I tried a different RP3 board perhaps the issue is in one of those. It's been running all night with no issues but I'll give it a few days since I got this far with the previous RP3. I'll go to a higher quality shielded wire like the stuff I bought for RTDs.

Definitely something odd going on in this panel. The RP3 is passive, so shouldn't be the issue.

 

[aeviaanah]

Definitely something odd going on in this panel. The RP3 is passive, so shouldn't be the issue.

Maybe I'll order another screw shield from you.

 

[BrunDog]

It’s also unlikely that is the issue as it is also passive. Let’s try my suggestions above first.

 

[aeviaanah]

It’s also unlikely that is the issue as it is also passive. Let’s try my suggestions above first.

Shielded wire would be superior to twisting right?

 

[BrunDog]

Yes for sure. Ground the shield back at the MEGA side only.

 

[aeviaanah]

Yes for sure. Ground the shield back at the MEGA side only.

Can you elaborate a bit more on what you mean?

 

[RiverCityBrewer]

Attach the drain/shield wire to a ground on one end of the wire only, preferably the panel/MEGA side. Chances are higher of creating a ground loop if you attach the drain wire at both ends, and that will just cause more problems.

 

[aeviaanah]

Attach the drain/shield wire to a ground on one end of the wire only, preferably the panel/MEGA side. Chances are higher of creating a ground loop if you attach the drain wire at both ends, and that will just cause more problems.

Ah took my a second, your takling about the shield inside the wire. My head went to shield like screw shield/ethernet shield. Got it thanks

 

[aeviaanah]

Edit

 

[aeviaanah]

Still having issues after adding the shielded and twisted wire. Could the Electronics Salon MD-D1125CT-1 be contributing to this issue?

https://www.amazon.com/gp/product/B01FCVZ3JE/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&psc=1
If not, perhaps its something in my wiring? A few PT100s are wired (not shown) to ensure I have connection to amplifiers. THIS IS ALL THAT IS WIRED- NO SOLENOIDS, ELEMENTS, BALL VALVES ETC AND THE ISSUE PERSISTS.

Things seem fairly stable when only one RP-3 is wired and others disconnected. One board lasted 24hrs then observed a disconnect from Wiznet 5100. The other lasted 48hrs and I manually disconnected to continue diagnosis. Issue seems to be magnified as I introduce more than one RP-3 board. I observe ethernet losing connection within a few minutes whereas I get 24hrs+ with a single board.

I'm getting desperate as our startup brewery hinges on this system quiet heavily. (10) 7-10bbl fermenters glycol control, cold storagel monitoring and hot side operations. I ordered a WINC1500 in an attempt to get this to work. Does this work on SPI BUS as well?

 

[oakbarn]

I think this third point is REALLY solid. There is no doubt in my mind this is where most noise problems come from. Do you have any recommended inline noise filters?

Is this what you mean about an in line filter?

https://rcdude.com/hyperion-fpv-power-filter-12v.html