New software idea for HM!!


 

LeeT

New member
Let me start off by saying I don’t have a Heatermeter. I have my own home brew Arduino based PID temp controller. Not as good as heater meter yet but it’s a work in progress. It gives me another project to tinker with and I'm too far down the road to give up and buy the HM.

Any way I was thinking wouldn’t it be awesome to have an output to tell you how long until your meat reaches the doneness temperature you have set.

Basically I put a cold pork shoulder on the smoker. Smoker is set at 225 meat is at 40 ish degrees. The meats heats up 1 degree every minute at the beginning and 1 degree every 10 minutes towards the end (these are just guesses). As the thermometer starts sensing the rate of rise (delta T over time) it could then output an approximate timer until your meat is done. IE pork shoulder will be done 12 hours and 42 minutes form now.

How difficult would it be to add some software code to the heater meter to do this? It looks like the graph of meat heating up is similar to a Y = LOG X type graph. I figured if I could get you guys with the Heatermeter helping out, I would be able to add it to my own temp controller also.

I did find this product that claims it can do it. I’m curious how well it works I haven’t seen this function on any other controllers or thermometers yet.
 
It already does this, but requires an hour of data to start calculating.

Look to the right of the temp of BeefShoulder

HMBeefShoulderRampDown.png
 
Last edited:
Perfect thank you. I will have to look at the code and see how its done on the HM.

Is it pretty accurate after that first hour? Or does it take a few hours until it becomes accurate?

I'm thinking if its linear it would not be very accurate early on in the smoke.
 
It does not take in effect "the Stall" and its not very accurate, and you are not going to find a way to do it accurately any way. At the beginning of a smoke the degree change per hour will be more then what you will see during the last hours of a smoke, so you are right its not linear. Although once the meat is past the "The Stall" and maybe a 2nd stall, the meat will have a more linear temperature rise.

But, most of the time there is not much learned from knowing when might something may be done, as most of the time its done when its done.

I have done plenty of pulled pork with the Heatermeter and Its only accurate in the last hour of the smoke, when the temperature is about 5 to 10 degrees to 195 or it could be longer if the meat is not tender enough.
 
Here is how I was hoping to use it.

1. I'm making a loin or shoulder or brisket or really anything that takes a few hours. I also want to throw on some sausages or something that only takes a short while to cook. But, I want them to all be done and come off the smoker at the same time. I could use this "time till done" as an indicator to put the sausage, bacon, ABT's on the smoker.

2. We're having people over for dinner at 6PM I'm smoking something all day long. I look at the "time till done" an it says things won't be done till 8PM. I can bump up my temp to try and have the meal ready at the designated time.

I wonder if the one form Acu-rite has a better algorithm which may be more accurate than the one in the HM.

Also it's definitely possible to get an accurate done time from the beginning. There have been plenty of CFD (computational fluid dynamics) models done on cooking and freezing meets by major universities. So, there are super smart people out there with the know how on exactly how meat cooks that could probably do this. I'm just not one of them. Google meat cooking CFD.

Here is an excellent paper using a CFD model to model the cooking hamburger... There are some equations on pages 5-6 that may prove useful at some point during my quest.

Although you do have a good point the stall's are going to be difficult to predict. For me they come at different temps and last different durations. I think it depending on the fat content of the meat. More marbled fat = longer duration stall from what I have observed.
 
With Shoulder, I just don't know that we will ever be able to predict this. There are too many variables including fat content, probe placement, location in smoker, etc. I normally do at least 2 butts at a time with probes in each and quite often the one that looks like it will be done first in HM finishes later.
 
Also it's definitely possible to get an accurate done time from the beginning.
This is not true at all. You neet calculate the heat permeation into the meat given temperature difference, surface area, weight, humidity, fat and moisture content, if it has a bone in it, etc. There are wayyyyy to many variables you're not going to be able to measure accurately, and definitely it isn't going to be something you'd analyze before each cook. You would probably be able to accurately model a piece of meat climbing to like 150F, but the stall messes up everything.

The code assumes a straight line interpolation based on a least squares fitting of the last hour's data. It's not right, but it gives you some idea what's going on. You can use the degrees per hour to determine when to foil and you'd be able to characterize the behavior pretty well. Like as soon as the DPH drops below 4F/hr, foil. You know that it takes 45 minutes from when you foil until the DPH starts to rise again. Then you know from experience what the DPH is going to max out at post-foil, and you can estimate time yourself based on that plus the estimate in HeaterMeter. (e.g. you know the post-foil cap is 15F/hr, heatermeter says 8F/hr and target in 3h = actual target about 1.5-2h)
 
Bryan, Thank you for answering this post. I appreciate everything you have done for the group with the HM project.

As I said above, I agree the stall is the most difficult thing to predict. I guess when I said accurate I meant maybe a range of +/-15 minutes near the beginning of the cook would be nice to know. I just looked at that burger cooking CFD model and they used a best fit straight line Y=mX+b graph as well. bummer....

Here is how I was thinking about it. If I'm in a car and a start off going 50mph but I decelerate steadily at 2mph/h. At The end of the first hour I would be going 48mph then the next hour would be 46mph and so on. The graph of my speed VS time would be a straight line going down form left to right. However the graph of my total distance traveled vs time would look similar to the graph of our meat cooking. I would always cover more distance the previous hour than the current hour.

The HM already measures the delta T over time (speed). We need to know rate at which the Delta T over time is decreasing (deceleration rate). After knowing this variable you could plug that value into the Temperature vs time graph (total distance traveled) and get a better prediction of the estimated done time. You always know where you are currently at on the total distance traced via the probe.

This would not require any measuring of stuff at the beginning and no extra prep time as suggested above. I agree this still does not account for the stall. However, it will be a much more accurate prediction than using a straight line.
 
Sorry, I respectfully must disagree. I have purchased same size butts that began with identical curves but ended up being over an hour apart to reach the target temperature. There are too many variables.
 
Sorry, I respectfully must disagree. I have purchased same size butts that began with identical curves but ended up being over an hour apart to reach the target temperature. There are too many variables.

Just a couple weeks ago, I bought 2 3 pound chuck roast for some pulled beef. Put both in at 1030am one finished at 630pm(195) and was nice and tender. The other chucky was still tough as nails and still sitting at 140 and it stayed on the smoker until 1030pm until it reached 195 and was finally tender.

Smoking takes practice and after a while you develop a sense when something will be done, and when you think you have figured it out, a cut of meat throws you a curve ball and you are way off on when you think it would be finished.

But, the great thing about the Heatermeter, is that you can lower the pit to a holding temperature until you can remove the meat from the smoker.
 
Last edited:
Yeah I understand the concept of decaying "speed", which can be modeled no problem. But look at this graph (I wish I had a heatermeter graph but I keep forgetting to pull one from stash)


Ok. First the temperature "accelerates" from 0 degrees per hour at point 0 to max by the third point. Then it decelerates until it hits point 9 where it is back to 0. You need to know acceleration, how long it accelerates, deceleration, how long it decelerates. One of these values you could calculate if you knew the others and the stall temperature. Then it goes into stall for how long? It varies depending on surface area, humidity, temperature difference, and most importantly moisture content of the meat. Finally it comes out of the stall and you've got another S curve just like the first section, except this time we usually pull the BBQ before we see that deceleration (as it approaches the smoker temperature).

I'm guessing you could figure out ballpark figures for the acceleration and deceleration, maybe the max speed too (which would tell you time). But they'd be specific to a certain piece of meat and pit setpoint, say an 8lb bone-in butt in a 225F smoker. You'd still not know the stall duration though, which can throw your estimate off by an hour or more.
 
Just a couple weeks ago, I bought 2 3 pound chuck roast for some pulled beef. Put both in at 1030am one finished at 630pm(195) and was nice and tender. The other chucky was still tough nails and still sitting at 140 and it stayed on the smoker until 1030pm until it reached 195 and was finally tender.

Thanks for posting that, I had a pork yesterday that was sitting at 140 for 3 hours, I was questioning my probes, the weather, everything, I cranked up the smoker to 500 just to see the needle move, it didn't. I wrapped the meat and put it in my convection oven on full blast, it took 30 minutes before it got above 140, then it soared right up to 195 in about an hour and tasted horrible and tough!
 
I agree that graph would be impossible to estimate...I'm still going to try to get something more accurate than a straight line estimate. If i get something that works I'll post it. Ill have to look up the kinematics equations.
 
Ok, you guys are right… I tried… I used data form 3 different cooks. 1 brisket flat, and 1 pork loin and pork 1 loin roast.

I used the 4 basic physics kinematics equations. I know these are only good for constant acceleration. But I was looking for the “good enough” scenario not a perfect scenario. My definition of good enough was an estimate of within +/- 15 minutes of the actual done time.

I was able to get the rate of rise (delta T / time) for 2 different periods of time. Using these numbers I was able to determine the rate of acceleration or deceleration. As we all know the meat first accelerates in temperature then starts to decelerate until it is done.

I used the data form the loin and the loin roast each of these took 3.5 to 4 hours until finished. To my surprise I was able to estimate within 15 minutes when they would be done at about the half way mark of the cook. 62% of my calculated done time estimates were within 15 minutes. The beginning half was all wrong due to the acceleration of the rate of rise. But when the rate of rise started to decelerate the formula worked. The kicker is each of these chunks of meat were only cooked to 145.

The down fall… I then tried the formula on a Brisket flat. Only 3% of my estimated done time numbers were within 15 minutes. A small portion in the middle and a small portion at the end. As many of you suggested the magic of the smoker happens between 145 and 190. Between the stall and the big difference in the rate of rise for temps under 140 vs temps above 140, a better calculation method than a straight line may not exist.
 
Thanks for posting that, I had a pork yesterday that was sitting at 140 for 3 hours, I was questioning my probes, the weather, everything, I cranked up the smoker to 500 just to see the needle move, it didn't. I wrapped the meat and put it in my convection oven on full blast, it took 30 minutes before it got above 140, then it soared right up to 195 in about an hour and tasted horrible and tough!

Big chunks of meat like a shoulder or brisket will hit a plateau mid way through the cook where it will look like the cooking has stalled. I think this is because it hits the point where the juices and fat start flowing which takes away some of the heat with it. After a bit of a stall they generally pick up and the temps start to rise again at a more expected rate. The plateau varies a bit from roast to roast, and seems to be longer on the lower and slower cooks and on larger pieces of meat from my experience.
 

 

Back
Top