Zizzle's Brewbot

[Martin]

Hi Zizzle, I've loved reading this topic

I've been wondering about a few things: how does the brewbot prevent dough-balls when mashing in? Is it a simple underlet, and does the brewbot stir the mash? If no stirring, are temperature gradients in the mash a problem? Thanks!

 

[Zizzle]

I usually just remember to give the mash a stir, but if I'm not around or forget I'm not too worried. I have the code setup to give a short recirc every few minutes.

But an automated mash stirrer is in the works. Stay tuned.

 

[chovain]

Hey Zizzle,

I dream of putting together a setup like this one day. I want to start out with some microcontrolled power switching, and I have a couple of technical questions about your triac setup:
* I assume you connect the opto to the triac via about a 340R? (Based on Vpeak of 340V)
* What power rating are your resistors?
* What frequency do you switch at for your power control?
* How have you connected the hot-side of the triacs to your high-current wiring? I'm assuming you don't run 10A through vero tracks .
* Have you ever toyed with voltage control (i.e. only partially opening the triac) instead of duty cycling?
* Any idea where you got your power triacs? Jaycar have a terrible range, and people like Farnell are just plain expensive.

 

[mesa]

* What frequency do you switch at for your power control?

A while back I was running an electric boiler with a single kettle element switched by a sawtooth oscillator and comparator at about 0.5Hz (i.e. 2 seconds). I guess I assumed that with the huge thermal mass of 20+L of wort and relatively low thermal resistivity between the element and the wort that there wouldn't be particularly large variation in the temperature of the element. Then again I have no idea how hot the element gets relative to the wort or what effect the formation of bubbles has on the thermal resistivity.

I think what people usually do is use the mains power itself as a reference oscillator. so it would be operating at 50Hz (or kind of 100Hz since you could do the comparision on both half cycles)

I eventually stopped using the switching running the element at full power instead (not enough insulation on the kettle to run at lower power). Then I decided kettle elements at full power scorch the wort too much and switched to a nice gas system with a thick bottomed pot.

* Have you ever toyed with voltage control (i.e. only partially opening the triac) instead of duty cycling?

I don't think a triac can be partly opened (if I remember what a triac is correctly) and if you did have some sort of power FET that could be partly opened, it would dissipate huge amounts on energy in the FET (necessitating huge heat sinking and wasted power)

 

[chovain]

* Any idea where you got your power triacs? Jaycar have a terrible range, and people like Farnell are just plain expensive.

Oh, wait - I just found Jaycar's 15A triac...

 

[chovain]

A while back I was running an electric boiler with a single kettle element switched by a sawtooth oscillator and comparator at about 0.5Hz (i.e. 2 seconds). [...]
I think what people usually do is use the mains power itself as a reference oscillator. so it would be operating at 50Hz (or kind of 100Hz since you could do the comparision on both half cycles)

I suppose as long as I don't run faster than 100, it'll work. Thinking aloud here: One factor I suppose I should consider, is how fine control do I want? Since the Triac will latch until the next zero crossing, the on-off cycles can be modified only by multiples of 10ms. To keep the PWM generation simple, I probably want to pick a power-resolution that I'm happy with, and go for a fixed frequency based it on that. If I want to be able to tweak by 10% granularity -> 10Hz; 1% granularity -> 1Hz.

I don't think a triac can be partly opened (if I remember what a triac is correctly) and if you did have some sort of power FET that could be partly opened, it would dissipate huge amounts on energy in the FET (necessitating huge heat sinking and wasted power)

You're right. I was a bit confused (never used triacs before, either).

 

[Doogiechap]

I suppose as long as I don't run faster than 100, it'll work. Thinking aloud here: One factor I suppose I should consider, is how fine control do I want? Since the Triac will latch until the next zero crossing, the on-off cycles can be modified only by multiples of 10ms. To keep the PWM generation simple, I probably want to pick a power-resolution that I'm happy with, and go for a fixed frequency based it on that. If I want to be able to tweak by 10% granularity -> 10Hz; 1% granularity -> 1Hz.


You're right. I was a bit confused (never used triacs before, either).

A partially informed comment. Triacs are used in my job of dimming lights in performances and do so by 'hacking' the leading or trailing edge of the sine wave. How this is done I can't remember as I wasn't paying enought attention in my electronics course in 1987 but it certainly happens.
The dimmers I have are rated to 2400 watts but 4800 watt units are readily available.
Cheers
Doug

 

[newguy]

Your mains voltage is a sine wave with one positive and one negative excursion. The period, or total time for one complete cycle is 1/50 = 20ms. Each excursion is 10ms long. A triac can be triggered to conduct sometime "in the middle" of one of these sine peaks - positive or negative. In the absence of the trigger signal, the triac will naturally extinguish at the zero crossing of the sine wave.

Triggering in the middle of the sine wave is certainly possible, but the drawback is EMI (electromagnetic interference). In essence, the sharp vertical spike in the voltage applied to the load when it is turned on in the middle of one of these sine peaks is produced by the sum of a very large number (infinite, actually) of harmonics - of very high frequency. These harmonics will happily radiate from the power cord that connects the heater to the triac. And from the upstream cords as well.

EMI isn't that big of a deal for you, but it is for your neighbours or anyone else who may experience radio or tv interference. The power companies around here frown upon it because this type of interference is absorbed almost 100% by their transformers which can cause them to overheat. I'd suspect the power companies there feel the same way.

The easiest approach is to trigger the triac for x 1/2 full cycles - in essence "letting through" only integer multiples of the sine wave peaks. When I designed my HERMS, I allowed for 64 power levels - from 0 to 63 full (one positive one negative) sine wave periods. With my 60Hz mains, this full cycle takes just over a second, which corresponds to how often I take my temperature measurements.

You can control the firing angle on each and every sine peak, but you'll have a lot of EMI unless you incorporate a lot of power line filtering. Given the very slow reaction between power application and temperature change, doing things this quickly is overkill.

 

[mesa]

I suppose as long as I don't run faster than 100, it'll work.

Correct. As long as your not faster than 100 you'll get some sort of control. How accurate that control is when your close to 100Hz but not quite 100 is another matter.

Since the Triac will latch until the next zero crossing, the on-off cycles can be modified only by multiples of 10ms.

Not quite. You appear to be saying that you want to run for a whole number of cycles turned on then a whole number turned off. That would work but its not necessary to do that. You can only turn off at the end of every 10ms cycle, but you can turn on at any point in that cycle.

If you simply use the mains voltage as a reference (i.e. use a resistor divider and compare to the voltage on a potentiometer) to decide when to turn on its very easy to get control betwen 50% to 100% duty cycle (i.e. you can turn on whenever the voltage is going more positive or more negative but not when its returning to 0). If you want to get between 0 and 50% duty cycle you need to build your own oscillator which is either much slower so that you work roughly in whole numbers of cycles as above, or run it at 25Hz and lock it in phase with the mains voltage (you would also need an edge detector so that your triac would switch on when your comparator went either on or off)

Personally I would just run the boiler at full power all the time. If I was worried about scorching, I'd run 4 elements instead of 1 (i.e. 2 in series then two series groups in parallel) that would maintain the total power but quarter the power density (I wish I'd though of this before investing in an LPG system). If I was doing a small batch on a system designed for larger batches I'd probably just unplug one of the two series groups entirely (halve the total power).

Ultimately though, returning to the question of brewing rather than electronics ;-) I don't really know what the advantage of controlling your boil rate is? If you've got a big enough kettle to avoid boil over can you boil to hard? Or is it all about not scorching the wort in an electric system?

 

[chovain]

I've actually seen references a kind of optocoupler called a "zero-crossing" coupler. These have a bit of trickery in the output stage that delays opening of the triac until the next zero crossing. Maybe I'll look at using the MOC3041: It's got a 15mA trigger (suits the 25mA IO on my PIC), and 400uA hold current. That'll eliminate most of the noise.

 

[chovain]

Correct. As long as your not faster than 100 you'll get some sort of control. How accurate that control is when your close to 100Hz but not quite 100 is another matter.

Not quite. You appear to be saying that you want to run for a whole number of cycles turned on then a whole number turned off. That would work but its not necessary to do that. You can only turn off at the end of every 10ms cycle, but you can turn on at any point in that cycle.

If you simply use the mains voltage as a reference (i.e. use a resistor divider and compare to the voltage on a potentiometer) to decide when to turn on its very easy to get control betwen 50% to 100% duty cycle (i.e. you can turn on whenever the voltage is going more positive or more negative but not when its returning to 0). If you want to get between 0 and 50% duty cycle you need to build your own oscillator which is either much slower so that you work roughly in whole numbers of cycles as above, or run it at 25Hz and lock it in phase with the mains voltage (you would also need an edge detector so that your triac would switch on when your comparator went either on or off)

I should probably have pointed out beforehand that I'm going to be using an MCU with a crystal clock and PWM module for this, so accurate timings are not a problem. I also want to keep the DC stuff isolated from the AC (for sanity reasons as much as anything, but also so it can be tested easily without connecting it to mains), so I want to avoid using mains as a reference.

Ultimately though, returning to the question of brewing rather than electronics ;-) I don't really know what the advantage of controlling your boil rate is? If you've got a big enough kettle to avoid boil over can you boil to hard? Or is it all about not scorching the wort in an electric system?

My main reason is because it's interesting, and I love tinkering .

 

[Zizzle]

I dream of putting together a setup like this one day. I want to start out with some microcontrolled power switching, and I have a couple of technical questions about your triac setup:
* I assume you connect the opto to the triac via about a 340R? (Based on Vpeak of 340V)
* What power rating are your resistors?
* What frequency do you switch at for your power control?
* How have you connected the hot-side of the triacs to your high-current wiring? I'm assuming you don't run 10A through vero tracks .
* Have you ever toyed with voltage control (i.e. only partially opening the triac) instead of duty cycling?
* Any idea where you got your power triacs? Jaycar have a terrible range, and people like Farnell are just plain expensive.

The data sheets for most triacs have a reference circuit that you can use.

Basically for my triac setup I canabalised a board already setup to do it.

See these boards:
http://secure.oatleyelectronics.com//produ...products_id=166
http://secure.oatleyelectronics.com//produ...products_id=668

I used the first board in the brewbot, but would recommend the second, which is better rated, and what I gave Brad to use in his bot. I think they are both opto isolated.

You will need a little skill to follow the tracks and work out the circuit diagram for the board. I hacked off the micro part of the board and used the power electronics part only, but you don't need to do that on the second board.

Both use nice wide tracks to handle the current, and I used soldered to the places designation to take main to/from the board.

At the moment brewbot uses fairly course control. Cycles times of a second or more. There is that much thermal mass that you could nearly get away will a cycle time of a minute or so.

You might also want to consider Solid State Relays. They would be much less stuffing around, but more expensive.

http://secure.oatleyelectronics.com//produ...products_id=526

Stay motivated, keep asking questions, and you'll learn a heap and have a lot of fun.

 

[browndog]

I've been following this thread as I'd like to build an electric boiler one day as they are much more economical than gas and you don't have to worry about running out of electricity mid boil. I'd like to but in here and say boilovers are a real threat and having a big kettle (not ginormous) is not the answer. I have a 50L kettle and my preboil is 33L, my nasa is set to give me a robust boil and I never alter the regulator. If I am brewing a mid range beer, a boilover is the last thing one my mind, however if I am making a high gravity beer then a boilover could happen any time, it is even worse if there are a lot of hops involved. I think this is why you need to be able to vary the power going to the element. You can't design your kettle around your element, however, I guess there are elements more suited to a 50L boiler.

cheers

Browndog

 

[newguy]

I've been following this thread as I'd like to build an electric boiler one day as they are much more economical than gas and you don't have to worry about running out of electricity mid boil. I'd like to but in here and say boilovers are a real threat and having a big kettle (not ginormous) is not the answer. I have a 50L kettle and my preboil is 33L, my nasa is set to give me a robust boil and I never alter the regulator. If I am brewing a mid range beer, a boilover is the last thing one my mind, however if I am making a high gravity beer then a boilover could happen any time, it is even worse if there are a lot of hops involved. I think this is why you need to be able to vary the power going to the element. You can't design your kettle around your element, however, I guess there are elements more suited to a 50L boiler.

A friend of mine built himself an electric brewery in his basement. He did 10 gallon batches. He had 2 x 4500W elements in his kettle and didn't have any fancy driving circuit either. Just two switches. Both on to bring it up to a boil. Once it started to boil, one off. He didn't have any problems with scorching either. Long story, but the larger the element is (physically), the less chance of scorching as there is a larger surface area which brings down the power density (watts/cm^2) in contact with the wort. You can buy elements with different rated power densities (link here and another here).

 

[chovain]

Long story, but the larger the element is (physically), the less chance of scorching as there is a larger surface area which brings down the power density (watts/cm^2) in contact with the wort.

Or as mesa points out, put a couple of elements in series, then duplicate in parallel. Elements are relatively ohmic, so by having two on a circuit, you'll halve each elements voltage drop, while leaving the resistance about the same. P=V^2/R, so each element quarters in power. Duplicating the circuit brings you up to the original power.

That arrangement has quarter the power density of a normal element, and would work well if you had a source of kettle elements.

 

[randyrob]

hmm.. first automation thread i could find sorry zizzle!

has anyone thought about or used "Arduino" for brewery automation

http://www.arduino.cc/

Arduino is an open-source electronics prototyping platform based on flexible, easy-to-use hardware and software. It's intended for artists, designers, hobbyists, and anyone interested in creating interactive objects or environments.

Arduino can sense the environment by receiving input from a variety of sensors and can affect its surroundings by controlling lights, motors, and other actuators.

there very cheap if they suit the purpose, starting from $17 locally.


dreaming of the next big thing
Rob.

 

[bonj]

Hey Randyrob,

I have an arduino diecimila here that I have grand plans for. Step one was to get the DS18S20s reading temperatures, but it seems that all three samples I've got are duds. :angry: I have been troubleshooting them with a DS1820 guru on one of the arduino forums, but they're definitely cactus... I'm pretty mad about that because they were free samples from Maxim. I also have an ex-Nokia mobile phone LCD, that I'd love to incorporate if I have enough pins free. Just watch the prices from some of the local suppliers as they don't include GST (which I thought was illegal for retail) and you only find this out during the ordering.

 

[the_fuzz]

Hey Randyrob,

I have an arduino diecimila here that I have grand plans for. Step one was to get the DS18S20s reading temperatures, but it seems that all three samples I've got are duds. :angry: I have been troubleshooting them with a DS1820 guru on one of the arduino forums, but they're definitely cactus... I'm pretty mad about that because they were free samples from Maxim. I also have an ex-Nokia mobile phone LCD, that I'd love to incorporate if I have enough pins free. Just watch the prices from some of the local suppliers as they don't include GST (which I thought was illegal for retail) and you only find this out during the ordering.

I've got 2 DS1820s 's here you can have?

 

[randyrob]

i put in my order for 2 x samples of DS18S20+ back in November last year and they still haven't rocked up,
where would be the best place to buy them?

 

[bonj]

DS18B20 (parasite power only) AUS$4.15ea : http://www.modtronix.com.au/product_info.p...products_id=285
DS18B20 AUS$6.50ea : http://www.microzed.com.au/chips.html
DS18S20 (powered or parasite) AUS$12ea : http://www.oceancontrols.com.au/data_acqui...logger_k145.htm

The + means it's certified lead-free, I believe.