Posted Sun Aug 2, 2009, 1:43am Subject: Re: (Another) Poppery PID controller project
One more thing. In case you are tempted to want to build the same controller unit, I am happy to help by providing plans for the front panel and back panel, including cutout placements. I printed the cutout diagram, and traced the cutout holes onto the aluminum, then cut them out.
You'll need: Electric drill A "nibbler" tool - get the Klein Tools #76011B tool. I got mine through an Amazon seller for $20. Great for cutting straight edges and squared off corners. A Dremel tool with a mandible and cutter wheels. NOTE: This is NOT optional -- it was INVALUABLE for fine-tuning the cutout dimensions, and getting things to fit right. Ron's advice on this was a Godsend.
I also am posting here link to a spreadsheet including the parts I used, and suppliers and cost: Click here (Excel spreadsheet)
WARNING: this is not a project for those wanting to build a controller on the cheap. I built a lot of extra functionality into this I probably won't use (e.g. the probe screwed into the chamber is too slow responding to use for MET temperature measurements - as Jim Gallt suggested it would be). But, I think the controller as built is a nice solid, relatively compact unit that is portable for roasting out on my porch.
This was a labor of love, and I used it as my vacation, instead of traveling for the summer. The materials alone (without shipping) were about $400. More than I was going to spend to start, but once I got started, I kept adding on to the functionality and cost along the way (e.g. powder coating instead of bare aluminum).
I may still add a voltmeter for the heater & fan voltage -- but do not think that really will help me run the roast more effectively, and the meter would be another $100. The 2 cheap meters I got could not handle the chopped up sine wave that the traics in the fan dimmer/router speed controller create, so that was a waste of $40. I also bought a few other parts that wound up not being used. If you use the attached bill of materials, you should not buy any unneccessary components (although I would suggest eliminating the TC probe screw mounted in the chamber, and replacing the DP3T switches with DPDT switches, but make sure the circuit outputs both TCs to one set of leads).
OK I am rambling. Any questions I would be happy to answer.
Dana Leighton - Espresso hack and CoffeeGeek moderator
Posted Sun Oct 11, 2009, 4:40pm Subject: Re: (Another) Poppery PID controller project
Updating the thread: Current status
I have been using the controller now for a couple months. It has generally performed well. The controls are easy to work, and it has not malfunctioned at all.
The Delta PID interface software has a bug in it where it uploads an invalid parameter to the PID occasionally. When it did, the PID would become nonfunctional, and has to have the power cycled to reset it. Since it's wired into the main power, the whole box needed to be switched off and on. This isn't a problem, except that my laptop (which does not have a functional battery) would have to be powered down also which was a pain. So I installed a momentary switch to reset the PID.
I have blew a fuse today, which may indicate I've got too much on the one circuit. The Poppery, transformer, a cooling fan, and the laptop power brick was just too much for it. It happened after the roast was complete, and I hadn't switched the PID to stop its program. So, the Poppery ran the heater up to full power and it blew the fuse. I do not dare put a fuse over 15 amps in it, so I may need to run the laptop off another circuit as I was before.
Poppery (1500 watts/125V) = 12 amps Laptop brick = 3.8 amps Fan = 2.6 amps Total = 18.4 amps! -- No wonder it blew the fuse.
Usually, the PID has the heater cycling at under full power, so I guess that's why it was working for the full duration of the roast before the fuse blew. No big deal - I should just get a battery for the laptop anyway. :)
Here's an updated back panel diagram with the reset and PID interface port. I also changed the labels for the thermocouples - one is mounted in a cooler part of the chamber, the other in the hottest part. Details of why here.
(Click for larger image)
Dana Leighton - Espresso hack and CoffeeGeek moderator
Very nice! Looks professional. My thermocouple placement is right where you drilled the hole in the chamber(the flat spot) and I've had no problems at all, the readings are very fast and accurate as far as I can tell. I think I ground down the metal a bit farther so the thermocouple extends into the roaster a bit more. Originally, I tried extending the roast time to more than 10 minutes but I found the coffee very flat tasting. My roasts last 9 to 10 minutes and I'm very happy with the taste. Good luck with it.
randytsuch Senior Member Joined: 11 Jun 2009 Posts: 578 Location: LA, Ca Expertise: I like coffee
Espresso: Expobar Office with... Grinder: Baratza Vario Roaster: Customized Alpenrost,...
Posted Sat Nov 7, 2009, 12:10am Subject: Re: (Another) Poppery PID controller project
Hi Dana I just finished a PID project, and I think I used the same series of PID, but from a different source. I found a open box one cheap on ebay, but the link is here Click Here (www.automationdirect.com)(SL4848_-z-_PM_-z-_TC_-z-_PC_Series)/SL4848-VR I tried to link to it, but I think the link is too long. It's the 4848-vr model at automation direct.
I got the 4848 series, which is the next size up.
My PID is in a much smaller case, with just the 4848 and a SSR, with a input AC line, and an output AC line, for the heater.
You might check out the manual from automation direct, it is much longer, with a lot more detail, but you probably figured out everything by now anyway.
I didn't know/think about a step up transformer and speed controller for the fan, or I might have done the same. But, I already have a variac, so I am going to try that to speed up my fan, and do bigger roasts.
I have used my PID 4 times now, I'm still getting the hang of it, but it does a better job than I can do manually with a thermocouple and variac. Maybe if I had more experience, I would be able to better it, but lately my manual roasts were bitter, and my PID roasts came out much better.
Based on this thread, I just added another thermocouple, to measure ET, or at least something close to ET. I put it inside the roasting area, but right next to the fins. I had just been using a thermocouple in the middle of the beans, running the PID using BT, and it seemed to work fine. I was always lagging behind target though, so I was underdriving things, which may be why I didn't have problems using BT. This may not be optimum, but I will see what I get like this. I can always drill another hole if I don't like the results.
So, I had a few questions. Did you autotune, or did you set the P, I and D parameters yourself? What did you set them to?
I am using the original Popcorn pumper, which is very similiar to the Poppery, but little different, and since my ET probe is in a different place, not sure if I can use your parameters anyway, but I was curious to see what you have.
I only tweaked my PID values during the first roast, and I am sure they are not optimum, but I'm not sure how I should tweak them.
I will have to check the PID and get back to you on the values. Although diferent TC placements will need different tuning parameters, because the environment and how the TC responds to temperature changes will be different.
I did use autotune, and then detuned according to the ideas in one post on this thread. Doubled the P, halved the T -- I was zero anyway.
One parameter I am not sure about is the "I offset" value that you can set via the software interface.
You can play around with the P I & D values, increasing and decreasing them, and see the results on stability. I went for the settings that minimized overshooting yet still kept the temperature from dropping too much. At first crack time, it stays +/- about 3°F within set point.
Here's the graph of my most recent roast - some Brazil Serrado. Not too happy with it so far (only on 36 hrs rest) but I'll et it rest some more. Notice that with the hottest temp TC, the MET exceeds the BT by around 100°
(Click for larger image)
Dana Leighton - Espresso hack and CoffeeGeek moderator
Wow Randy - you're not kidding! The Delta manual is pretty skimpy - this one is way better. The software also looks better - I will try it out tomorrow. It sets more parameters, including having multiple PID parameters, each at different target temperatures, which I suspect might be useful to better tune the PID to various temperatures during the roast. For example, when the roaster is moving the temperature at 300° it has to make changes differently than when moving the temperature at 500° -- does that make sense?
Dana Leighton - Espresso hack and CoffeeGeek moderator
randytsuch Senior Member Joined: 11 Jun 2009 Posts: 578 Location: LA, Ca Expertise: I like coffee
Espresso: Expobar Office with... Grinder: Baratza Vario Roaster: Customized Alpenrost,...
Posted Sat Nov 7, 2009, 11:08am Subject: Re: (Another) Poppery PID controller project
dana_leighton Said:
For example, when the roaster is moving the temperature at 300° it has to make changes differently than when moving the temperature at 500° -- does that make sense?
Yeah, it makes sense, but it will be a while before I'm that sophisticated. I just starting roasting this summer, and still have a lot to learn.
First thing I am going to do is compare my ET to BT's, and see how they track with my setup, so I can try controlling the roast with ET. Then I will start tweaking the P, I and D's stuff.
And, I want to try putting a variac on the fan, to do larger roasts.
BTW, does changing the fan speed, and bean load effect the PID?
There are lots of variables here, which will make it hard to optimize. One very nice thing is being able to save the setup information with the PC interface, will make it a lot easier to tweak things, and then get back to where you started.
The less expensive solution, as long as you only want a 15 VAC boost is the transformer I got, paired with a fan dimmer. I put the fan dimmer in the circuit prior to the transformer. The transformer is $9 plus about $12 shipping; the dimmer is less than 10 bucks. Add in a project box and you're up to about $35, way less than a variac. You can run the transformers in series and get a 30v increase if you want. Ron is doing this and says the fan on the Poppery handles it just fine.
BTW, does changing the fan speed, and bean load effect the PID?
Since you're controlling the heater via the ET then bean load does not have much effect, although it might affect the tuning parameters to get it more stable. When I turn the fan speed down as the beans swell and get lighter, the PID readjusts fine after a bit of an initial overshoot. If I were using the BT as input I am not sure what these variables would do. I may just try switching it over to BT one day and see what happens.
There are lots of variables here, which will make it hard to optimize. One very nice thing is being able to save the setup information with the PC interface, will make it a lot easier to tweak things, and then get back to where you started.
YES!!! That is a real help. I use a spreadsheet to help me calculate estimated ETs from desired BTs and then load the parameters by hand. But I can save the parameters and reload them at will. Very nice.
Your project is coming along well -- and your curiosity serves you well in this task. It's really a try-it-and-see-what-happens kind of thing.
Dana Leighton - Espresso hack and CoffeeGeek moderator
randytsuch Senior Member Joined: 11 Jun 2009 Posts: 578 Location: LA, Ca Expertise: I like coffee
Espresso: Expobar Office with... Grinder: Baratza Vario Roaster: Customized Alpenrost,...
Posted Sat Nov 7, 2009, 4:15pm Subject: Re: (Another) Poppery PID controller project
dana_leighton Said:
The less expensive solution, as long as you only want a 15 VAC boost is the transformer I got, paired with a fan dimmer. I put the fan dimmer in the circuit prior to the transformer. The transformer is $9 plus about $12 shipping; the dimmer is less than 10 bucks. Add in a project box and you're up to about $35, way less than a variac. You can run the transformers in series and get a 30v increase if you want. Ron is doing this and says the fan on the Poppery handles it just fine.
I already have a variac, so cost is $0 for that method, otherwise I would trip a step up transformer and a dimmer. I may still try it at some point, since I have a couple dimmers, and the transformer is cheap enough, it would be nice to put everything in one box, like you have, but for now I will start with the variac. I bought the variac from ebay a while ago, for another project, and it sure has come in handy for home roasting.
Thanks for the advice and encouragement. I thought I was the only one using this controller, I searched for solo and automation direct, and did not find anything here, funny that the Delta is the same thing, just from a different company. Must be a place in China making a ton of these things.
randytsuch Senior Member Joined: 11 Jun 2009 Posts: 578 Location: LA, Ca Expertise: I like coffee
Espresso: Expobar Office with... Grinder: Baratza Vario Roaster: Customized Alpenrost,...
Posted Sun Nov 8, 2009, 10:34pm Subject: Re: (Another) Poppery PID controller project
So I tried to put a thermocouple to measure ET to roast with, but have not had much luck with it. First, I put the thermcouple so it was just inside the roasting chamber, and I did a couple test roasts like this. Seemed to work OK, started off about 50C higher then BT, and was about 20C higher then BT by the end of the roast. But, with the wire sticking in the chamber, it trapped a bean, and burnt it, like someone else in this thread.
Today, I decided to put the thermocouple in differently, and I completly disassebled my popper, so I could drill a hole from inside the chamber. I started at one of the openings in the fins, where it is the largest, and drilled at a angle. With it like this, the probe end is just barely inside the chamber, at the mouth of one of the openings.
I tried roasting like this today, expecting the readings to be similar from yesterday, but they are way different. With the new thermocouple placement, I am getting a over a 100C degree difference between BT and ET, or about 200F.
I did use a thermocouple, with the popper running, to see what the hottest spot was, but decided to put the thermocouple in a "average" spot instead. After I opened my roaster, I decided the one spot was hotter because the heating element was a little closer to the opening there. Figured the air gets mixed together in the chamber anyway, so average shoud be fine.
I wasted two test roasts, and started to get frustrated, so I decided to go back to roasting with BT for now. That roast appeared to come out fine, if a little slow. I have not given up on ET yet, just taking a break from it for a little while, until I have some more time to tweak things.
Symbols: = New Posts since your last visit = No New Posts since last visit = Newest post
Forum Rules: No profanity, illegal acts or personal attacks will be tolerated in these discussion boards. No commercial posting of any nature will be tolerated; only private sales by private individuals, in the "Buy and Sell" forum. No cross posting allowed - do not post your topic to more than one forum, nor repost a topic to the same forum. Who Can Read The Forum? Anyone can read posts in these discussion boards. Who Can Post New Topics? Any registered CoffeeGeek member can post new topics. Who Can Post Replies? Any registered CoffeeGeek member can post replies. Can Photos be posted? Anyone can post photos in their new topics or replies. Who can change or delete posts? Any CoffeeGeek member can edit their own posts. Only moderators can delete posts. Probationary Period: If you are a new signup for CoffeeGeek, you cannot promote, endorse, criticise or otherwise post an unsolicited endorsement for any company, product or service in your first five postings.
![dana_leighton: bp.png](javascript:ImagePopup("/images/45526/bp.png", 576, 432, "dana_leighton: bp.png");)
![dana_leighton: graph.jpg](javascript:ImagePopup("/images/45808/graph.jpg", 368, 295, "dana_leighton: graph.jpg");)
= No New Posts since last visit 