BruControl: Brewery control & automation software

[aeviaanah]

Did you say you have a schematic for this? What you are saying sounds ok but I don't want to give any advice without being exactly clear what you are building.

I had a schematic of my interpretation of cranes recommendation. I’ll make one based on the new info revealed today.

 

[BrunDog]

Let me say that @crane know's what he is doing... so perhaps it comes down to interpretation like you are saying.

Question: How are your contactor coils powered... AC or DC?

 

[aeviaanah]

Let me say that @crane know's what he is doing... so perhaps it comes down to interpretation like you are saying.

Question: How are your contactor coils powered... AC or DC?

i was hoping he’d chine in for clarification.
I currently own (6) 24v DC coil contactors 50amp 3phase.

 

[BrunDog]

OK... hopefully just need flyback diodes across those input terminals then.

 

[aeviaanah]

I feel like contactors are more suited for the high amp draw and are more cost effective at 3phase. the cons- noise and lack of finer control. Currently battling with this pro/con comparison. Playing with the phasing circuits adds a layer of comfort and complexity that boggles my mind.

 

[BrunDog]

I think that since you have them, we should try to make them work. Most relays are rated for hundreds of thousands, if not millions of cycles assuming they are used within spec.

 

[aeviaanah]

I think that since you have them, we should try to make them work. Most relays are rated for hundreds of thousands, if not millions of cycles assuming they are used within spec.

My electrical friend says not to split up the phases through contactor and SSR. He says if there is a short circuit or phase imbalance the SSR could catch fire. I guess this leaves either use the contactors or source 50amp 3phase SSR along with the contactors(or 3 single phase contactors)

https://www.amazon.com/twidec-MT3-40DA-Temperature-Aluminum-Electric/dp/B0746G1W3H/ref=mp_s_a_1_3?dchild=1&keywords=3+phase+ssr&qid=1586188872&sr=8-3&th=1&psc=1

 

[Gruel]

One thing to keep in mind is that boiling water self-regulates its temperature via the boil off rate. Since the latent heat of water is so high - it takes about seven times the energy to boil off a quantity of water that it took to heat it from room temperature to boiling -, you have a lot of time to dial down your heating power once you reach the boiling point, without losing too much liquid.
Let's say you dimensioned your heaters such that you get to boiling in half an hour. If you keep going at the same power setting, you will boil off one seventh of the water (or wort) in the next half hour.

Additionally, the specific heat of water, even though small compared to the latent heat, is pretty large compared to all the other stuff around, so heat losses are not all that important in boiling (as opposed to cooling, where ambient heat gain can dominate).

I guess what I'm saying is you could use a simple linear equation to have your controller set the heating power only based on the volume your are boiling (with a fixed offset for the heat loss of your boiling vessel), and then another linear equation for the reduced steady boil power to use after a fixed delay. If you are off a bit, you are just changing your boil-off rate by a small amount.

Example: 15 kW takes about 26 minutes to heat 20 gallon of water from 25C to 100C (77F to 212F). If you left the power at 15 kW, you'd boil off 3 gallon in the next half hour. If you reduce to 5 kW, you'll boil off 2 gallon in the next hour, which is probably where you want to be at.
Adding 500 W heat loss for your kettle (assuming about 1 square meter surface area and 5 W per square meter per Kelvin heat loss, which is a good number for still air) changes time to boiling by almost nothing, and reduces the boil off rate by 10%, which is also nothing.

So, I don't think you need a PID or anything, nor could you really use it: it would be difficult to give it a good input, since you can't use temperature - the boil kettle temperature will be constant until almost the last drop of wort is boiled away.
You are simply starting with one power setting for a fixed amount of time to reach boil, and then turn some of the heaters off and run with that second power level for the rest of the boiling time.

 

[BrunDog]

Correct - never split up current pathways. Maybe I didn't understand that's what you were looking to do - as I mentioned above, you would want parallel circuits. This means all the way through to the consumer.

This is why a schematic is needed (sorry!). Impossible for different people to be on the same page, each making unique assumptions, and get the same results.

 

[aeviaanah]

Correct - never split up current pathways. Maybe I didn't understand that's what you were looking to do - as I mentioned above, you would want parallel circuits. This means all the way through to the consumer.

This is why a schematic is needed (sorry!). Impossible for different people to be on the same page, each making unique assumptions, and get the same results.

I think that since you have them, we should try to make them work. Most relays are rated for hundreds of thousands, if not millions of cycles assuming they are used within spec.

I fully agree on schematic, I’ll provide one prior to wiring- I’m still in information gathering stage. I need to attempt to understand what to draw before I can draw it! I appreciate the help to date.

One thing to keep in mind is that boiling water self-regulates its temperature via the boil off rate. Since the latent heat of water is so high - it takes about seven times the energy to boil off a quantity of water that it took to heat it from room temperature to boiling -, you have a lot of time to dial down your heating power once you reach the boiling point, without losing too much liquid.
Let's say you dimensioned your heaters such that you get to boiling in half an hour. If you keep going at the same power setting, you will boil off one seventh of the water (or wort) in the next half hour.

Additionally, the specific heat of water, even though small compared to the latent heat, is pretty large compared to all the other stuff around, so heat losses are not all that important in boiling (as opposed to cooling, where ambient heat gain can dominate).

I guess what I'm saying is you could use a simple linear equation to have your controller set the heating power only based on the volume your are boiling (with a fixed offset for the heat loss of your boiling vessel), and then another linear equation for the reduced steady boil power to use after a fixed delay. If you are off a bit, you are just changing your boil-off rate by a small amount.

Example: 15 kW takes about 26 minutes to heat 20 gallon of water from 25C to 100C (77F to 212F). If you left the power at 15 kW, you'd boil off 3 gallon in the next half hour. If you reduce to 5 kW, you'll boil off 2 gallon in the next hour, which is probably where you want to be at.
Adding 500 W heat loss for your kettle (assuming about 1 square meter surface area and 5 W per square meter per Kelvin heat loss, which is a good number for still air) changes time to boiling by almost nothing, and reduces the boil off rate by 10%, which is also nothing.

So, I don't think you need a PID or anything, nor could you really use it: it would be difficult to give it a good input, since you can't use temperature - the boil kettle temperature will be constant until almost the last drop of wort is boiled away.
You are simply starting with one power setting for a fixed amount of time to reach boil, and then turn some of the heaters off and run with that second power level for the rest of the boiling time.

Thankyou for the information. I plan on taking this math and plugging it in to the values I have to validate using one of the three elements I have to maintain boil in kettle. Your comment on boil makes sense because once at 212, there isn’t much to control from there. However I am thinking of control in HLT which it would be nice to nail temps for strike etc. We are single infusion so I imagine we will put more weight on nailing strike temp since there is no herms/rims to correct the situation.

 

[clearwaterbrewer]

Can you provide a link to the 9kw 3 phase elements?

Just look for Dernord 9kw on aliexpress https://www.aliexpress.com/item/32851714074.html

 

[clearwaterbrewer]

Has anyone tried other MAX31865 PT100 amplifier boards for use with the RP-3 boards that BrunDog sells. With Adafruit out of stock on their board and shipping limited to health care only I was curious if the one below would work or anyone had tried it. Got burned with a different LCD previously so thought I better ask first. Thanks.

HiLetgo-MAX31865 - Amazon

yes, I have used the aliexpress ones, and frankly they seem the same quality, but in *my* experience, the RTD's, while having a *much* better resolution than the 1-wire, are FAR more off of any true value from the factory... I have a post 10 pages back or so on the RTD/RP3/MAX/ESP32 combo.. FYI - it was not fixed by getting a more expensive RTD probe... I also bought a pack of 1% resistors to test the adafruit vs ali boards, but have not gotten around to creating a test rig... My setup is on a distillery, and I want to monitor vapor temp very accurately to the tenth of a degree in the 170F range... really hoping a more accurate version of the 1-wire sensor becomes commonplace...

 

[clearwaterbrewer]

My electrical friend says not to split up the phases through contactor and SSR. He says if there is a short circuit or phase imbalance the SSR could catch fire. I guess this leaves either use the contactors or source 50amp 3phase SSR along with the contactors(or 3 single phase contactors)

https://www.amazon.com/twidec-MT3-40DA-Temperature-Aluminum-Electric/dp/B0746G1W3H/ref=mp_s_a_1_3?dchild=1&keywords=3+phase+ssr&qid=1586188872&sr=8-3&th=1&psc=1

the SSR can short and catch fire in normal use, it is no different if you split up phases...

SSR's generate heat, so my preference is to use just contactors and create steps by dissecting the operation of a single three phase element. I have one 6-pole/3-phase triple element have 2 contactors controlling it, one contactor controls one individual loop of the element with 2 wires, and the other contactor controls power to 2 of the individual elements with 3 wires. this requires that you remove one of the jumpers on the element and run 5 wires to it. this gives you 3kw (for 9kw element) or 4KW(for a 12kw element) when the single loop contactor is on, and 6/8kw when the other is on. when both are on you get full element power.

with 4 other elements on their own 3-phase contactors, for 9kw, with a little logic in a script, you can get 3/6/9/12/15/18/21/24/27/30/33/36/39/42/45kw

 

[aeviaanah]

Just look for Dernord 9kw on aliexpress https://www.aliexpress.com/item/32851714074.html

Man I looked all over for these. Looks like it’s time I use AliExpress. You run these in your kettle- you feel the watt density is ok?

 

[clearwaterbrewer]

I only use the 9kw in the water jacket portion of my Bain-Marie kettle, I use the 6kw and 4.5kw in a direct contact still boiler but only with clear spirit, not wash. You can get even lower watt density by either going to a WYE wiring of the element(wire one side of each loop together, then L1, L2, L3 to the remaining three terminals) or order the 380V version and run on 240V 3-phase...

 

[aeviaanah]

I only use the 9kw in the water jacket portion of my Bain-Marie kettle, I use the 6kw and 4.5kw in a direct contact still boiler but only with clear spirit, not wash. You can get even lower watt density by either going to a WYE wiring of the element(wire one side of each loop together, then L1, L2, L3 to the remaining three terminals) or order the 380V version and run on 240V 3-phase...

There’s always these guys that are listed at 30 watt density
https://www.glaciertanks.com/elemt-r12kw-208-delta-low.html

 

[crane]

I've Been busy the last couple of days. So much discussion over one of my silly ideas. Also, the mobile browser sucks compared to the defunct app.

If i understand this correctly, you are saying it is ok to use output from Mega direct to 24v coil side of contactors. Specs below...

View attachment 674363
I thought the Mega is 5v output and I purchased 24v contactors. If my understanding is correct, how does 5v provide enough volts to power a 24v coil? Also, will the 9w draw (x3 contactors = 18w) be too much draw from the single PID output on Mega- thought these were rated around 40mA? What am I missing here?
View attachment 674366

The idea was that if someone was convinced the needed ultimate control, they would use SSR's with PID control. Also, it was under the ASSumption that 3 elements controlled by contactors can only provide 3 different power values(1/3, 2/3, full). The circuit you drew out was correct, except that you would use SSR's instead of relays. Each SSR would power a separate element. The second part of that thought would be to compromise and control one element through PID/SSR and the other 2 with contactors/hysteresis. This would allow you to most likely have just as precise of control with less SSRs. However, using SSRs for the power and volume that you are describing would most likely be overkill.

I think what @clearwaterbrewer is doing is able to achieve good results on that scale.

Let me say that @crane know's what he is doing...

Well, I have fooled at least one of you.

SSR's generate heat, so my preference is to use just contactors and create steps by dissecting the operation of a single three phase element. I have one 6-pole/3-phase triple element have 2 contactors controlling it, one contactor controls one individual loop of the element with 2 wires, and the other contactor controls power to 2 of the individual elements with 3 wires. this requires that you remove one of the jumpers on the element and run 5 wires to it. this gives you 3kw (for 9kw element) or 4KW(for a 12kw element) when the single loop contactor is on, and 6/8kw when the other is on. when both are on you get full element power.

with 4 other elements on their own 3-phase contactors, for 9kw, with a little logic in a script, you can get 3/6/9/12/15/18/21/24/27/30/33/36/39/42/45kw

With this many discrete power steps you don't need PID or other control to vary the boil rate. Everywhere else you can use on/off control on this scale.

Looks like it’s time I use AliExpress.

You are late to the party my friend.

 

[oakbarn]

Use of K Type Probe sensors? You can get ceramic ones up to 1300 C. I was thinking of using one for a flame sensor. Other ideas for a flame sensor? Right now I just have tickler to tell me to light the burner.

hese are High Pressure Burners so a standard furnace valve cannot be used.

 

[RiverCityBrewer]

Do you want to get an actual temperature, or simply a flame/no flame input?

I see the converters to make K type thermocouples interface with arduino, but their output is serial.

Edit: Nevermind, here is an analog converter that might work for you:
https://learn.adafruit.com/ad8495-thermocouple-amplifier/overview
Looks like it would handle up to 750C on a 5v input MAX.

 

[oakbarn]

I am not trying to get the Temp but just that it is hot ( above a certain temp) so the gas would continue to flow. I use high pressure burners that operate around 5 psi which is way high for 11 inch water column Furnace valves. I now turn on my Burners manually (although the have a plate ignitor that works well). I could do it automatically, but if the burner did not light? I have though of IR sensors but not sure if they would work as they might get too hot. With the BCS 482 I used Alexa. If I get that working with BruControl that is fine as well as I was used to that method.

 

[BrunDog]

Not really sure what you are

I am not trying to get the Temp but just that it is hot ( above a certain temp) so the gas would continue to flow. I use high pressure burners that operate around 5 psi which is way high for 11 inch water column Furnace valves. I now turn on my Burners manually (although the have a plate ignitor that works well). I could do it automatically, but if the burner did not light? I have though of IR sensors but not sure if they would work as they might get too hot. With the BCS 482 I used Alexa. If I get that working with BruControl that is fine as well as I was used to that method.

Not really sure what you are looking for here. Measure heat? Pressure? Ignition? Communication with Alexa?

If its that the flame is lit... this is simple enough. For example: https://www.digikey.com/short/zpjbhb

 

[day_trippr]

fwiw, some Honeywell gas controllers use flame detectors integrated with HSI igniters to stage operation and back out of unsafe conditions. The detectors rely on plasma ions to provide conduction between detector poles - basically, using the flame itself as a conductor. Some type of narrow-band amplifier is likely involved within the controller to translate to a binary state...

Cheers!

 

[oakbarn]

Not really sure what you are


Not really sure what you are looking for here. Measure heat? Pressure? Ignition? Communication with Alexa?

If its that the flame is lit... this is simple enough. For example: https://www.digikey.com/short/zpjbhb

I tried a Flame Sensor with the BCS. In order for it to detect the flame reliably, it needs to be too close to the flame and would melt if left there. Maybe I can come up with a "protective metal case" that has a small hole for the flame to be detected while protecting the Electronics. I was hoping that someone had a different solution. These small modules seem to be so that you can tell if a candle is lit for a learning experience but are not practical in operation of a burner. An obvious answer is to switch to low pressure and get Furnace Valves, but I like the High Pressure burners.

 

[DoubleHaulBrewing]

Want to double check my thinking on something. I have multiple fermentation chambers all monitored with RTD probes. I have one workspace that shows all the chambers and then a separate workspace for each chamber that shows more detail, Tilt graph, Temp Graph, etc. Since an element can only be displayed on one workspace I have added the RTD probe elements on each individual chamber workspace and then added a global for each probe on the chamber "overview" workspace. I have a script that loops ever 5 seconds to update the value of each global with the value of it's respective RTD Probe. The graphs point to the RTD Probe itself rather than the global.

Is this the best way to handle displaying the same info on multiple workspaces? Anyone see anything wrong with doing it this way or a better way to accomplish the same thing?

 

[RiverCityBrewer]

Want to double check my thinking on something. I have multiple fermentation chambers all monitored with RTD probes. I have one workspace that shows all the chambers and then a separate workspace for each chamber that shows more detail, Tilt graph, Temp Graph, etc. Since an element can only be displayed on one workspace I have added the RTD probe elements on each individual chamber workspace and then added a global for each probe on the chamber "overview" workspace. I have a script that loops ever 5 seconds to update the value of each global with the value of it's respective RTD Probe. The graphs point to the RTD Probe itself rather than the global.

Is this the best way to handle displaying the same info on multiple workspaces? Anyone see anything wrong with doing it this way or a better way to accomplish the same thing?

That's what I came up with as well... globals on the overview screen with a script to update. It has worked well for me thus far, so I don't see a problem.

 

[DoubleHaulBrewing]

That's what I came up with as well... globals on the overview screen with a script to update. It has worked well for me thus far, so I don't see a problem.

Thanks RiverCity. Glad to hear I am on the right track and it works well.

 

[BrunDog]

We discussed having multiple elements which point to the same device but thought it would be too confusing to manage. I think the way you are doing it is best and makes sense.

Perhaps we re-visit a "mirror" element?

 

[BrunDog]

I tried a Flame Sensor with the BCS. In order for it to detect the flame reliably, it needs to be too close to the flame and would melt if left there. Maybe I can come up with a "protective metal case" that has a small hole for the flame to be detected while protecting the Electronics. I was hoping that someone had a different solution. These small modules seem to be so that you can tell if a candle is lit for a learning experience but are not practical in operation of a burner. An obvious answer is to switch to low pressure and get Furnace Valves, but I like the High Pressure burners.

The BCS could not accept analog inputs. And while I am not vouching for this unit as I have never tested it, my guess is it can see some distance. You would always put it below the kettle. This sensor says it's detection is up to 1 meter - that seems far enough even at 1/10th that range. I would not necessarily care about seeing the flame - I would look for the heat radiating back from the bottom of the kettle if possible. Also, do you really care if the flame is lit? You should be able to tell from the temperature trends, no?

Finally... I'm sorry man... I just do not understand what high pressure vs. furnace valves have to do with this. Admittedly I have no experience with gas, but what does the pressure matter when you are trying to assess whether the flame is lit or not?

 

[crane]

Furnace valves can only be used with low pressure burners. A furnace valve will provide everything needed for ignition, flame detection, etc. By going with high pressure burners, you can't use a furnace valve and therefore need to re-invent the wheel WRT auto ignition and flame detection. So @oakbarn is making his life harder than it needs to be, because he doesn't want to switch to low pressure burners.

 

[DoubleHaulBrewing]

Has anyone tried other MAX31865 PT100 amplifier boards for use with the RP-3 boards that BrunDog sells. With Adafruit out of stock on their board and shipping limited to health care only I was curious if the one below would work or anyone had tried it. Got burned with a different LCD previously so thought I better ask first. Thanks.

HiLetgo-MAX31865 - Amazon

I wanted to update the thread and report that I tested the HiLetgo - MAX31865 and they worked no problem with the RP-3 boards and a Mega controller. The only difference that I noticed from the Adafruit was that the reference resistor is not as consistent as the Adafruit model. They can range anywhere from 430 to 435. Enough to change my temp readings 7 degrees. Once I measured the resister value and updated the value calibration entry appropriately for each probe they were spot on and held consistent.