topbanner.gif
Login
Username

Password




Not a member yet?
Click here to register.

Forgotten your password?
Request a new one here.
Shoutbox
You must login to post a message.

04/18/2021 3:00 PM
mtsattemt welcome cup

04/18/2021 9:07 AM
Welcome, Ankafe!

04/16/2021 2:04 PM
SkipG welcome cup

04/16/2021 2:03 PM
meeuw2000 Welcome welcome cup

04/16/2021 8:29 AM
welcome cupfrankvw

Users Online
Guests Online: 10

Members Online: 0

Total Members: 7,161
Newest Member: mtsattemt
In Memory Of Ginny
Donations

Latest Donations
JackH - 25.00
snwcmpr - 10.00
Anonymous - 2.00
Anonymous - 5.00
Anonymous - 5.00

View Thread

Who is here? 1 guest(s)
 Print Thread
Barebones PIC Roaster Controller?
allenb
I'm creating this thread to investigate possibilities for a super low cost, simple, mains powered (dropping resistor) PIC based roaster controller.

The reason for wanting this is I'm in the beginning stages of a low cost roaster design for an open source project that will be a cross between a Gene Cafe and Turbo Roaster but will resemble neither of the two physically. Details about the concept will follow soon and will be posted within the open source build projects forum.

There will be options for controls starting with a non user interfacable (except for initiation of post 1C RoR segment) single profile model with no LCD readout panel (Barebones) or move up to TC4 or TC4C.

For the barebones controller concept:

- Using thermocouple or thermistor for bean temperature sensing, utilize a series of rate of rise ramps up through first crack. Example: start through minute 2 = X rate of rise, from minute 2 through minute 5 = X rate of rise, from minute 5 through beginning of first crack = X rate of rise. User will push a momentary push button as soon as the onset of first crack is heard which will initiate the final rate of rise slope which will continue till end of roast.

The initiation of each of the ramp segments will not be dependent on actual bean temperature but will be time dependent only.

The barebones board will include a triac with heatsink capable of handling 16 amps for it's output to a resistive heating element. Convection fan will be switched on manually and only when in the on position allow heater to be activated.

Questions

-Is it possible to utilize a mains powered PIC chip using a dropping resistor for power with minimum accompanying components on a board to read a thermistor and perform the minimum functions described above?

Without being a "packaged" system as with an Arduino, how would one program it?

If it was possible to do such a "barebones", no frills board would it necessarily be cheaper than buying a TC4C using thermocouple input?

If the cost difference was negligible than I can see no reason to bother with designing a less complex controller and would just stick with the TC4/TC4C.

Allen
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
oldgearhead
When you say 'mains powered' I have two thoughts:
1) England, Germany, or whose 'mains'?
2) Making it safe is probably more costly than the TC4C.
No oil on my beans...
oldgearhead
..but you might use a battery, a dial, an ESC, and an SSR..
Also, these wall-rat power supplies really aren't very expensive, so why were you thinking 'mains powered' anyway?
No oil on my beans...
Dan
Wall-warts aren't expensive? Heck, they are free, at least for me. I have been keeping mine from old devices and now have two boxes full. I've always found one that would work for a replacement or a new project, and have never bought a new one!
1 pound electric sample roaster, 3 pound direct-flame roaster, both handmade; modified Mazzer Mini grinder, LaSpaziale Vivaldi II automatic espresso machine. When the electricity goes out I make vacpot coffee from beans ground on my Zassenhaus hand grinder, and heat the water with a teakettle on the gas range.
allenb

Quote

oldgearhead wrote:

..but you might use a battery, a dial, an ESC, and an SSR..
Also, these wall-rat power supplies really aren't very expensive, so why were you thinking 'mains powered' anyway?


Some makers of PIC microcontrollers have seen the need for fitting a PIC microcontroller into ultra small packages and the need for reducing parts count for affordability when using in domestic household appliances as in electric fry pans, rice cookers etc. The cost of even a simple rectifier power supply can kill the deal. So they've come up with a PIC chip that has it's own on-board rectifier and all that is necessary is to use a "dropping resistor" and in some cases a dropping resistor in series with a diode for a half wave 5 to 7 volt input to the chip. The use of mains powered IC chips is not new and has been around since the mid 70's and isn't anymore unsafe than circuits using conventional power supplies as long as established safety protocols are followed as in proper grounding and over current protection fuses are included.

My reason for wanting to dig into this is to see if enough cost savings could be had by cutting the parts count.

It may well be that the only mains powered PIC's available are not capable of handling the functions needed even for the somewhat simple controller I'm proposing here.

Adding wallwarts or any additional baggage just increases cost and clumsiness when using the roaster.

Who's mains? As with any appliance, one would need to specify what their local household voltage is and order parts accordingly.

Like many open source design/build projects, people will have the ability to choose from various options when assembling their bag of parts to customize to their liking. If saving $ 20 to $50 dollars by going "barebones" isn't of interest then they can opt for using the TC4/TC4C or what ever tickles their fancy.

Allen
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
Dan
When prototyping a one-off controller of any sort, the controller itself is generally the cheapest part. Labor is usually the big bite, but as hobbyists we tend to ignore this investment. Generally, and this is my point, you need to buy programming software. For instance, you need $300 to buy software to program a $100 PLC, plus cabling. Sure, there are some entry-level trial software around. But my point is that the whole idea behind the Arduino is an open-source system from hardware to software using USB cables.
1 pound electric sample roaster, 3 pound direct-flame roaster, both handmade; modified Mazzer Mini grinder, LaSpaziale Vivaldi II automatic espresso machine. When the electricity goes out I make vacpot coffee from beans ground on my Zassenhaus hand grinder, and heat the water with a teakettle on the gas range.
allenb
You bring up some valid points with microcontroller programming interfaces and the software and because of this could we:

1. With members expertise, determine if a PIC microcontroller and other necessary hardware exists to accomplish the goal of a much reduced cost barebones controller.

2. If yes to the above, find volunteers who would be willing to set up a breadboard testing package.

3. Find volunteers who already have PIC microcontroller programming interfaces to share in creating the software.

4. Test the controller by connecting to a volunteer members turbo oven roaster.

Obviously, different turbo oven roasters with different thermal properties and element wattage will present a challenge to a one size fits all software program.

I'd like to emphasize that the goal of creating this controller is to give people a lower cost, low complexity, easy to install option for controlling the open source roaster mentioned above. I'm sure that down the road, many would be interested in upgrading to the TC4/TC4C and enjoy being able to utilize all the various programs available for it.

Allen
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
JimG
There's this application note from microchip:
http://ww1.microc...00954a.pdf

Jim
allenb

Quote

JimG wrote:

There's this application note from microchip:
http://ww1.microc...00954a.pdf

Jim


Thanks Jim, this is good information on alternatives to transformer based power supplies.

Question for you. With your experience with microcontroller circuits, would building a microcontroller package without a typical transformer based power supply (by using a resistive or capacitive dropping system) and by limiting the inputs and outputs to one each, would this necessarily be much less expensive than a TC4C or only by a few bucks?

This is assuming a chip even exists that would allow forgoing a typical transformer based power supply.

The reason I ask is if your guestimate is not much savings then I'm going to drop putting any more time into this and just concentrate on the open source roaster instead.

Allen
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
JimG
Allen,

I don't feel qualified to comment on the safety of a transformerless power supply. A lot of people warn against using them, but I'm not sure I understand all of the reasons.

Since the TC4 design allows the use of grounded thermocouples, it would seem that a power source that is not isolated from mains voltage could be a problem? I would need to satisfy myself that the sensors would not create a shock hazard.

Power issues aside, I think a much less expensive TC4C variant is possible for a dedicated application. The present design is intentionally "general purpose" and has some whistles and bells that wouldn't be needed. The FTDI chip for USB communication costs more than $4 and could be eliminated, for instance.

I would probably still go the AVR route, but possibly using a less expensive processor chip. Perhaps one of the ATtiny series.

Other high cost items are the ADC chip and the 8 position terminal block. Both of these would be reduced if only two sensor inputs were required. The list goes on...

Jim
bvwelch
Greetings,

My two cents --

I have been down this road -- transformerless picmicro controllers -- and after a year or so, rejected it.

I sleep much better having the transformer as a "given", and it works universally too. I also sleep better knowing folks are using off-the-shelf SSR devices rather than TRIACs. Especially since one of my "early adopters" smoked his kitchen counter.

I recommend that you look into a cost-reduced TC4C instead. What are you willing to throw out, what are your minimum requirements?

It has been a couple years -- is there anything cheaper than the mcp3424 for reading the TC ? I haven't seen anything myself. But if there *was* a picmicro or any other low-end processor that had the 18-bit ADC function built-in, that might might cause me to switch from AVR. Otherwise I'd stick with AVR.
bvwelch
ps: for single channel, the MCP3421 costs less than $2.
allenb
Originally, looking for lowest possible cost I was thinking of a single input and single output controller with no LCD that would sense bean temp only and would run a single generic rate of rise profile producing a 5 m to yellow, around 4 min from there to first crack and have a user initiated momentary button to manually initiate a 7 degree/m RoR at start of first crack.

To accomplish this I had envisioned the controller running a very high RoR up to minute 2, a much reduced RoR from minute 2 to minute 5 to bring the coffee to around 300-310F, then run a 25F/m RoR up to around 400F where 1st crack should be close at hand and at the start of first crack the user hits the momentary button initiating the final 7F/m RoR till dumping the batch.

One problem to overcome with this control scenario is how to deal with roasting outdoors in lower ambient temps where the KW is insufficient to run the profile without lagging behind? Add a auto/manual switch with power adjust pot for fully manual control? If this is needed then youd have to display RoR and bean temp on an LCD.

By doing this are we saving anything worthwhile?
After hearing about your difficulties with on-board triacs Ill drop that idea and stick with user supplied SSR. The mains powered avenue sounds like more trouble than its worth as well.

Id like to keep this conversation going as I think it would be good to brainstorm a low cost, simple, profile following TC4C at some point down the road for using with any roaster.

Allen
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
Dan
I think you want a PID with ramp/soak feature like the popular Fuji PXR3 or PXR4.
1 pound electric sample roaster, 3 pound direct-flame roaster, both handmade; modified Mazzer Mini grinder, LaSpaziale Vivaldi II automatic espresso machine. When the electricity goes out I make vacpot coffee from beans ground on my Zassenhaus hand grinder, and heat the water with a teakettle on the gas range.
oldgearhead
These days seventy-five U.S. dollars will buy a nice little temp controller with % output (or PID if you want to mess with it), a 25 amp S.S.R., and a real nice TC probe. Furthermore,
you don't need a PC or SBC to see the temperature..
No oil on my beans...
allenb
Kick this around and let me know your thoughts.

For anyone taking on the open source roaster build who have the desire and skill set to work with controls will be able to select their desired control setup whether it be variac, dimmer, PID with ramp/soak or TC4/TC4C with their choice of programming.

What is hard for me to get my hands around is what kind of plug and play, simple control could be offered to the ones who want little to do with playing control geek and just want to get to roasting? Something where they can just plug the cord coming from the control box into the jack on the side of the roaster and power into the wall outlet, flip a couple of switches and be off and roasting? And at the same time one that will take the coffee through a basic 4-4-4 profile and not require the person to twist a dial throughout the roast while doing mental math gymnastics calculating rate of rise.

To me, this type of controller should be available to use with any home built roaster project as well.

If you were designing one controller to accomplish the above need, what would it be like? Features?

Again, simple, low cost (less than $100), easy to use for anyone.

Allen
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
JimG

Quote

allenb wrote:
What is hard for me to get my hands around is what kind of plug and play, simple control could be offered to the ones who want little to do with playing control geek and just want to get to roasting? Something where they can just plug the cord coming from the control box into the jack on the side of the roaster and power into the wall outlet, flip a couple of switches and be off and roasting? And at the same time one that will take the coffee through a basic 4-4-4 profile and not require the person to twist a dial throughout the roast while doing mental math gymnastics calculating rate of rise.

To me, this type of controller should be available to use with any home built roaster project as well.


For a specific roaster, I think an inexpensive solution could be developed. I would hard-code a generic profile into the firmware that works on that roaster. Maybe a few simple buttons like "go", "mark FC", and "stop" for profile control?

Where things get difficult is trying to create a device that is plug-n-play across many different roasters.

The TC4C offers a basic foundation for controlling a wide variety of roasters. But custom firmware and interface hardware is needed for each particular roaster, i.e., you have the foundation but now you have to design and build the house.

This development has been done for the Hottop with the HTRI system. And greencardigan and others have undertaken similar development for their own roasters.

But there is such a wide variety of home built roasters out there that I get woozy thinking about a plug-n-play solution that would work on most of them.

Jim
Dan
That's a tall design goal: off-the-shelf, universal profile controller, less than $100. The closest we have to that is the pre-loaded TC4C controller.
1 pound electric sample roaster, 3 pound direct-flame roaster, both handmade; modified Mazzer Mini grinder, LaSpaziale Vivaldi II automatic espresso machine. When the electricity goes out I make vacpot coffee from beans ground on my Zassenhaus hand grinder, and heat the water with a teakettle on the gas range.
allenb
I would translate any controller called a profile controller as one that is able to save and run multiple profiles which this would not do. I'm proposing that the "barebones", "starter" model have only one basic 4-4-4 profile which would do a pretty good job on most coffees in a decent roaster capable of good bean agitation and heat transfer. To me, fan control would have to be manual by the user with no connection to the controller which I can't see anyone having a problem with. The roaster concept I'm working on for the open source project will have a single speed blower.

As far as being able to work with all possible varieties of home built roasters? No way unless there's an easy way to incorporate an autotune function that the user would run with the push of a button when the person first fires it up with a batch of sacrificial beans.

Other than that, could we find a set of PID settings that would do at least a fair job of smooth tracking with some of the major types of convection roasters (turbo, fluidbed, breadmaker-heatgun)?

For drum, I've never had any luck with any PID settings that would work for both browning phase and post 1C phase so I'm thinking drums would be best served with just a manual power adjust knob and rate of rise readout for a "barebones" drum version. I would leave fan control out of it for the drum version due to all of the various fan types (C-frame, DC, permanent magnet etc...

Allen
1/2 lb and 1 lb drum, Siemens Sirocco fluidbed, presspot, chemex, cajun biggin brewer from the backwoods of Louisiana
JimG

Quote

allenb wrote:

I would translate any controller called a profile controller as one that is able to save and run multiple profiles which this would not do. I'm proposing that the "barebones", "starter" model have only one basic 4-4-4 profile which would do a pretty good job on most coffees in a decent roaster capable of good bean agitation and heat transfer. To me, fan control would have to be manual by the user with no connection to the controller which I can't see anyone having a problem with. The roaster concept I'm working on for the open source project will have a single speed blower.

As far as being able to work with all possible varieties of home built roasters? No way unless there's an easy way to incorporate an autotune function that the user would run with the push of a button when the person first fires it up with a batch of sacrificial beans.

Other than that, could we find a set of PID settings that would do at least a fair job of smooth tracking with some of the major types of convection roasters (turbo, fluidbed, breadmaker-heatgun)?

For drum, I've never had any luck with any PID settings that would work for both browning phase and post 1C phase so I'm thinking drums would be best served with just a manual power adjust knob and rate of rise readout for a "barebones" drum version. I would leave fan control out of it for the drum version due to all of the various fan types (C-frame, DC, permanent magnet etc...

Allen


Limiting the controller device to fit a relatively small subset of roasters makes the task more approachable, as does removing fan control from the equation.

My initial thinking is that you would choose a particular target convection roaster and develop/test a prototype controller for that roaster.

Perhaps a 2-button interface can work here. Functions assigned to buttons might be

1) start profile/stop roast
2) mark first crack

Might use LED's to indicate

1) profiling is active (i.e. start has been pressed)
2) heater on (this will pulse with PWM signal)

The biggest concern is that sensor type and placement greatly affects temperature readings, even in identical roasters. This makes a pre-programmed profile really tricky.

Jim
Jump to Forum:

Similar Threads

Thread Forum Replies Last Post
Android App: Coffee Roaster Free ROASTING SOFTWARE APPS 1 04/16/2021 4:03 AM
HearthWare Roaster Value? JAVA TRADING COMPANY 7 04/02/2021 2:35 AM
Roaster build forced air and agitator Building a Coffee Roaster 6 03/31/2021 2:50 AM
TC4+ Arduino coffee roaster shield (TC4-compatible) Dataloggers/Controllers/Rate of Rise Meters 229 03/24/2021 11:30 AM
Heat gun bread machine roaster--how can I change container? Heat Gun Roasting 8 03/05/2021 4:14 PM
Homeroasters Association Logo, and all Content, Images, and Icons © 2005-2016 Homeroasters Association - Logos are the property of their respective owners.
Powered by PHP-Fusion Copyright © 2021 PHP-Fusion Inc
Released as free software without warranties under GNU Affero GPL v3
Designed with by NetriX