BruControl: Brewery control & automation software

Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum

Help Support Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
I actually saw this on my new glycol controller. But it’s when a relay shuts off. THe controller loses connection until reset.
yes Ive had this issue on my homebrewing rig causing my one wire sensor and some of my pt100 sensors to go whacky when my chiller or chiller pump turns on .. I have removed the one wire sensor and turned the chiller off when brewing to get around it... I have two separate control panels but the noise/ spikes still effects them both in my home brewery.
 
DC relays want the flyback diode type snubber, and yes, most relay board have them.. the AC driven ones, such as 24VAC, 120VAC or 240VAC coil contactors, will benefit.

Across the coil, and not inline with it is due to how the RC snubber works on an AC circuit: The RC circuit just is designed to ignore 50-60hz , that is they look like an open circuit (a capacitor passes high frequency, but not DC, and a larger capacitor passes lower frequencies, so a small capacitor is used), and do not interfere, but when the coil is de-energized by the opening of the switch on the left side, the coil, when de-energized, shoots out a spike of current due to the collapse of the magnetic field, and that snubber looks like a short to higher frequencies, so it dampens it. *note, if the relay powering the coil breaks contact at zero crossing, no spike is made... it may take a nearly full voltage contact break to make the energy needed to screw with the Arduino, so this may be why it is intermittent.
 
but when the coil is de-energized by the opening of the switch on the left side, the coil, when de-energized, shoots out a spike of current due to the collapse of the magnetic field, and that snubber looks like a short to higher frequencies, so it dampens it. *note, if the relay powering the coil breaks contact at zero crossing, no spike is made... it may take a nearly full voltage contact break to make the energy needed to screw with the Arduino, so this may be why it is intermittent.
Yes this is what Ive read also and it makes sense to me... although many arduino threads suggested putting the snubber across the actual relay contacts.. I'll admit in many of these cases the smaller relays werent driving a coil for a larger relay further down the line. I also have these
https://www.ebay.com/itm/MOV-V320LA40-V320LA40B-320-Volt-Littlefuse-Varistor-See-Data-Sheet-Quantity-of-5/321084946599?ssPageName=STRK:MEBIDX:IT&_trksid=p2057872.m2749.l2649

and 50 of these

https://www.ebay.com/itm/50-1uF-250VAC-250V-10-X2-Safety-Thomson-metalized-polyester-film-capacitors/290978723452?ssPageName=STRK:MEBIDX:IT&_trksid=p2057872.m2749.l2649
Would either of these be better to use across the relay coil or even main contacts?
 
Honestly the arch suppressors from auberins worked well to eliminate the noise for the 20 mins or so I was testing. I did not get one flicker on the sensor readouts when switching the relays on and off.

@augiedoggy I ordered five of the RC filters from Auber. The only place I will not have one is across the coil for the contactor that starts the glycol chiller compressor. The five filters will cover both chugger pumps attached to the brewstand enclosure, the two 4" muffin fans, the glycol circulation pump, and the glycol chiller contactor attached to the fermentation controller enclosure. I may need another one for the contactor coil, but I'm going to start with the five and get the sixth if necessary.
 
We haven’t tested Pt1000 amplifiers. I would like to tell you the firmware and software are compatible but I can’t guarantee it without verification.

Is there a reason you are contemplating PT1000 over Pt100? I don’t know of any improvement and would think the 100 ohm nominal sensors are more impervious to electrical noise.

Okay, I think then I will go ahead and order a single Pt1000 amplifier and test it out. Will report back the results for everyone.

I don't have a specific reason as to why I would use a Pt1000 over a Pt100. It just so happens that I work in a lab where we solely use Pt1000 probes due to the increase in accuracy and from a previous project I happen to already possess a few very nice Honeywell HEL-705 probes which I was hoping to utilize. I am not very familiar with arduinos measurement capabilities but I figured that since Pt1000 have a change in resistance of 3.91Ohms/C vs. the 0.391Ohms/C for a Pt100 I would probably be able to obtain more reliable and accurate measurements with the PT1000. I hear you on the shielding and noise and have already planned accordingly by using paired shielded cable. (I.e. I will setting up as a 4-wire rtd with each pair of wires shielded together and an overall shield. Is there anything else you think I could add to improve the system?
 
The fewer the connectors in between the RTD amplifier and the probe, the better. Quality cable which is as short as possible are better too. That said, if you are brewing beer, this will all give you much better results than you really need, but hey - bragging rights count!!
 
I am doing network setup on a W5500 and Mega after doing firmware install sucessfully, and I get this:

BruControl v43.F Setup 10s (y)?
BruControl v43.F Setup 9s (y)?y
Config net (y/n)? Y
MAC (##:##:##:##:##:##)?​

I do not see anything listed in the 'BruControl-Manual-v1.0-12-8-2017.pdf' manual, am I supposed to make up a mac-address and insert it? Is there really not a Burned In Address?

I hit y and enter and get:
BruControl v43.F Setup 10s (y)?
BruControl v43.F Setup 9s (y)?y
Config net (y/n)? Y
MAC (##:##:##:##:##:##)?MAC invalid
DHCP (y/n)?​

now I am locked out of it... I restarted the PC it is connected to via RDP to reset the USB, and now it does not come back up... to be continued tomorrow when I am at the shop..
 
Most W5100 boards do not come with a burned-in MAC address. What do you expect for a $19 card delivered? Most W5500 boards do have a MAC address provided on a sticker somewhere on the board.

Once you get into network config like you did, at the MAC stage, you enter the hex pairs separated by colons and hit enter. Example: AB:BC:CD:DE:EF:FA

If you have a problem, close the terminal editor, reset the MEGA, and re-enter with the terminal editor. You cannot get locked out. If there is no response after entering the configuration code, just wait a bit. It might be choking on a default network setting but should respond again shortly.
 
Understanding now you are not local to the board... I imagine that would make it tricky. But nonetheless you should be able to close and re-open the terminal editor and communicate with the board via serial comms.
 
I was able to measure the amp draw on 4 of my boil kettle ssrs.. 32ma total. I'm thinking the loose ground wire going from my shield to my ground terminal caused the overheating and my rtd amplifiers to all fail but one that works unreliable at this point. It's weird but when I plug in 12v ti the arduino the voltage regulator still gets very hot pretty quickly.. not sure how hot they should get but when feelng the back of the board it's hot but not when running off usb.. I swapped meats with a new one.. same results and 12v supply has the ground connected to same ground terminal.
 
So the 32mA is for all 4 SSR’s meaning 8 each? Are these powered simultaneously via one pin or individually via discrete pins? 32 would be too much for one pin by far.
yes and yes all on one pin they are all being turned on and off at the same time. sorry if I wasnt clear but thats what I meant by controlling groups of ssrs in post #686..

This is good to know because I only know of the max limit on the arduino I dont recall seeing limits on each pin. now I need to find an ssr that will switch such a small load since the din rail ssr I bought only works with 24 AC and I missed that when I purchased them.

My arduino gets hot even when all of the ssrs are off though and everything is disconnected or shut off so I'm assuming the voltage regulators just get hot on these.
 
Last edited:
Yes the regulator will get warm with 12 V input. It is soldered to the PC board as a heat sink but if you would like more cooling add a stick-on mini heat sink.

You can use a darlington array to drive higher loads. I will look one up for you.
 
Most W5100 boards do not come with a burned-in MAC address. What do you expect for a $19 card delivered? Most W5500 boards do have a MAC address provided on a sticker somewhere on the board.

Once you get into network config like you did, at the MAC stage, you enter the hex pairs separated by colons and hit enter. Example: AB:BC:CD:DE:EF:FA

If you have a problem, close the terminal editor, reset the MEGA, and re-enter with the terminal editor. You cannot get locked out. If there is no response after entering the configuration code, just wait a bit. It might be choking on a default network setting but should respond again shortly.

OK, Rebooting the PC must have had some of the long updates, a couple hours later I could RDP back in.. Problem one fixed... I was able to get back in to the device with InterfaceSetup.bat, problem two solved... This W5500 does not have a MAC, or a sticker, and I did not want to screw stuff up before I knew exactly what the device had... I took my PC MAC and incremented it by one to make sure I did not step on any reserved MAC address space and created a static binding and host mapping in my Cisco router. I am now able to ping the device from the network. Problem 3 now solved also. I think a couple lines added to the next version of the manual would help a lot of folks.
 
Last edited by a moderator:
OK, Rebooting the PC must have had some of the long updates, a couple hours later I could RDP back in.. Problem one fixed... I was able to get back in to the device with InterfaceSetup.bat, problem two solved... This W5500 does not have a MAC, or a sticker, and I did not want to screw stuff up before I knew exactly what the device had... I took my PC MAC and incremented it by one to make sure I did not step on any reserved MAC address space and created a static binding and host mapping in my Cisco router. I am now able to ping the device from the network. Problem 3 now solved also. I think a couple lines added to the next version of the manual would help a lot of folks.

I thought we addressed this in the appendix but I will look. Always a work in progress and I appreciate the feedback!

I am surprised the W5500 board did not come with a MAC listed. The W5100’s are typically Chinese knock-offs so they certainly do not pay for the reservations. Maybe the same is starting with the W5500’s.
 
I thought we addressed this in the appendix but I will look. Always a work in progress and I appreciate the feedback!

I am surprised the W5500 board did not come with a MAC listed. The W5100’s are typically Chinese knock-offs so they certainly do not pay for the reservations. Maybe the same is starting with the W5500’s.

and I assume that mouser would sell genuine...

Also, while in the manual, maybe update the virtual addresses being numbered 200+ for 1-wire, it is not in there..

I am chugging away and have relay board and one-wire working, and it is nearly ready to install.
IMG_20180704_104721s.jpg
 
so while I wait for parts to arrive I decided to do so research on arduinos and supposedly the I/O pins on a mega can handle up to 40ma unless im missing something? again just trying to figure out what took out my one of my rtd amplifiers which caused them all to go to -412 on the interface board until I isolated and removed the bad one. I suppose it could have just been coincidence or my relay board which started acting up at the same time.

Also on an unrelated note, I swapped my 12v high/low switching relay board with a supposedly isolated 5v low switching board from another manufacturer and my temp spikes are way worse with the new relay board, I can actually see the voltage spike on the digital out pin on the arduino and they are fairly consistent. I tried powering both from the arduino and from a separate meanwell din supply which made no difference.
I have ordered a sunfounder high level trigger board which im hoping will be better qualityand have better isolation than what I have.
I'm using the same relay board as clearwaterbrewer in my homebrewing panel but I only need 3 relays for my bigger panel and I need the NO/NC state to wire them so the 4th element kicks off while the rims is on and to prevent having my bk and hlt elements on accidently or I would just use SSR relay boards
 
Last edited:
I went into the Edit Interface page (gear icon) and changed the refresh interval from 1 sec. to 5 sec. It really reduced the number of spikes I'm seeing. I can also report that all of my spikes are on the shutoff of the inductive load.
 
I went into the Edit Interface page (gear icon) and changed the refresh interval from 1 sec. to 5 sec. It really reduced the number of spikes I'm seeing. I can also report that all of my spikes are on the shutoff of the inductive load.
I actually thought about doing this, But I fear these spikes may cause damage whether I am seeing them in brucontrol or not. plus it wont stop the arduino from randomly resetting every now and then.
My ferrites should be arriving today or tomorrow which hopefully help along with the new relay board. I tried putting the arch suppressors on both the contactor coils as well as across the load and they had no effect. When I tried them on the relay board relay contacts the relays didnt reliably turn off so I suspect they may have caused the damage to my last relay board and possibly the rtd amplifier that failed.
 
ok, I was hoping for a 16 channel just like the regular sunfounder... luckily BC handles low level will, my concern is only if the mega dies and all relays energize...
The 16 channel sunfounder relay board I have in my home brewing panel looks just like yours and is high level so they do make them. looks like sainsmart and sunfounder both sell the same stuff in some cases.
 
Last edited:
ahhh, the next problem.... (well, besides the AC going out as I got home to play with this last night...) It seems certain Cisco devices do not like to give a link to the W5500... I wondered why it worked on the 3560 at my shop but not the NM-16ESW in my 2811... now at home, I program a new mega and W5500 and get no link, I thought I had a bad W5500 and reprogrammed it several times... looks like others have had similar problems, solved by changing speed/duplex/negotiation, but I had no such luck... Cisco stuff is my day job and this is not the first time I have seen this... link comes up physically (link LED) immediately when connected to cable modem or PC..... side note, it looks like itead.cc makes a POE version of the W5500

edit, it looks like some wiznet folks have experienced the same thing.. https://forum.wiznet.io/t/topic/2614/2
 
Last edited:
physical... basically has to do with negotiating the physical link, speed/duplex/etc... sometimes you can overcome this with setting the switch or router to manual instead of auto, but not in this case... I am trying upgrading IOS code on the Cisco device and crossing fingers, worst case is adding a 'dumb' switch in-between...
 
that wiznet forum link I gave has a lot of detail...

Update, updated to IOS 15.4 from 12.4 on my Cisco881W router, and thought still had the same issue, except that I locked myself out of the router via telnet (I guess it enforced ssh and I did nto have that set before upgrading, and do not have a serial cable at home for console access)... but anyway, my vpn came up on it's own, all else came up, but the W5500 did not come up... and then an hour later I looked at it and it had a link... and I can ping it... I will be damned...
 
@BrunDog @augiedoggy it looks like the RC filters have helped with my setup. Even with the refresh interval dialed back to 1 second, I'm not getting the heavy cycling on the glycol chiller that I was before. I'm also not seeing anywhere near the number of temp spikes. I saw one last night, and it was across only one sensor, not all three. Nothing in the attached screenshot, which is roughly 0600-0900 this morning.

I'm really liking the temperature driving the cycles, not the spikes.

Screen Shot 2018-07-07 at 9.08.27 AM.png
 
Back
Top