calibration driving me (less) nuts (now)!

General discussion on BeerTools Pro Software.
Post Reply
erykmynn
Posts: 5
Joined: Tue Mar 02, 2010 5:02 pm

calibration driving me (less) nuts (now)!

Post by erykmynn »

I'm trying to get calibrated, after missing various targets on mashes several times.

Obviously sometimes I could just "preheat" and go for it, but that doesn't really help if there is more to do.

So I have a 10 gal rubbermaid:

width 12.5
length 12.5
height 18.8
capacity 10.0 gal
dead space 0.3 gal
heat capacity: 4646.0 j/kg k
heat xfer coefficient 2.98 w/m2 k

environment 65 F
infusion volume 5 gal
infusion temp 180.4 F
5 minutes 174.0 F
65 minutes 165.5


Now I know from experience that once equalized, this thing will only drop 1-2 degrees in an hour. People say they "tweak" their numbers, but how can I be comfortable mashing first at a temperature prescribed by something that has the thermodynamics so out of whack as to predict 7-8 degree drops in temperature?! And even if I can be comfortable as to hitting my strike temp, how do I begin to adjust it, so as to get in in the ballpark finally?
Last edited by erykmynn on Wed Mar 03, 2010 5:37 pm, edited 1 time in total.
User avatar
slothrob
Moderator
Moderator
Posts: 1831
Joined: Mon Apr 10, 2006 1:36 pm
Location: Greater Boston

Calibration

Post by slothrob »

It sounds like some condition was different for your calibration than your mash that is making the two values different. For example, perhaps the volume used to calibrate was much smaller than your mash volume so it lost heat faster. I'm not sure that's how the math works, but it sounds like you have real mash data that you can use to fill in the calibration window for your tun.

I'd go back and, in the calibration window, change your "heat after 65 minute" to a value closer to your 1-2 degree drop. You might need to change your "volume" to something close to your mash volume, also, I suppose. Then check this in a mash schedule to see if it comes close to predicting your heat loss.

Alternately, you could try repeating your calibration to see if you get different values than you did the first time.
BTP v2.0.* Windows XP
bobcat_brewer
Light Lager
Light Lager
Posts: 49
Joined: Mon Jan 18, 2010 9:15 pm

Post by bobcat_brewer »

The odds are pretty good, that something was out of sorts when you did your initial calibration. Maybe the mash tun was really colder than you initially thought and calibrated at?

Anyway, if you're confident that you took accurate temperatures during your mash, you can adjust the heat capacity and heat transfer numbers until the predictions match your results.

Or, you can start all over again ensuring you have accurate temperatures for the mash tun and water, use a larger amount of water and see if that helps. The think about a cooler, is that it will take a LONG time for the entire thing to reach room temperature if you keep it in your garage or a basement that is much cooler than the room you did your calibration at. When I calibrated my equipment, I had my tun at room temperature for a few days - only because of procrastination - but my temperature loss was only about a degree, maybe 2, than what I typically get doing an actual mash.

Good luck.
erykmynn
Posts: 5
Joined: Tue Mar 02, 2010 5:02 pm

Post by erykmynn »

Well, I've actually tried the calibration twice. Once with 3 gal and once with 5 gal and got similar numbers, but I need to remeasure the temp in the storage room I guess.

I can post those other numbers later if its helpful. I think I'd have to adjust after my next brew because I preheated most other times, and the time I forgot I don't think I realized how valuable data of the discrepancy would be.

EDIT: storage room temp: 65.5 F

ALSO: if it is so sensitive to tun temp for coolers, isn't the term 'ambient' a bit misleading? I'm guessing it would be wise once calibrated to always use tun temp as 'ambient', correct??


UPDATE:

I see if I duplicate the vessel and turn the Heat Transfer way down, the drop-off in a 60 minute rest comes more in line with experience, but the strike temp stays the same. Has anyone else "tweaked" their cooler's settings? Can I expect that the strike temp will be pretty close as-is? Looking at this thread make me think that might be the way to proceed: http://www.beertools.com/forum/viewtopic.php?t=4377 ... though it would be great to know why and whether its user error or a glitch in the software.

But
User avatar
slothrob
Moderator
Moderator
Posts: 1831
Joined: Mon Apr 10, 2006 1:36 pm
Location: Greater Boston

Post by slothrob »

erykmynn wrote:I see if I duplicate the vessel and turn the Heat Transfer way down, the drop-off in a 60 minute rest comes more in line with experience, but the strike temp stays the same.
That is the reason I suggested this:
slothrob wrote:change your "heat after 65 minute" to a value closer to your 1-2 degree drop.
If I understand the math correctly, it's the difference between the 5' and 65' temperature that sets the Heat Transfer.

Odds are, you entered that your tun lost something more than 1-2 degrees during this period, but that doesn't hold up to your current experience, so it probably needs to be changed.

I believe that the calibration, as it stands, will only be accurate if the tun is at room temperature during the calibration and you subsequently use the tun at that same room temperature. If you calibrate at 65F and use the tun at 50F, your numbers will be off. So, I think, at least with a cooler, you need to use the tun temperature as Ambient.
BTP v2.0.* Windows XP
erykmynn
Posts: 5
Joined: Tue Mar 02, 2010 5:02 pm

Post by erykmynn »

[quote="slothrob"][quote="erykmynn"]I see if I duplicate the vessel and turn the Heat Transfer way down, the drop-off in a 60 minute rest comes more in line with experience, but the strike temp stays the same.[/quote]
That is the reason I suggested this:[quote="slothrob"]change your "heat after 65 minute" to a value closer to your 1-2 degree drop.[/quote]
If I understand the math correctly, it's the difference between the 5' and 65' temperature that sets the Heat Transfer.

Odds are, you entered that your tun lost something more than 1-2 degrees during this period, but that doesn't hold up to your current experience, so it probably needs to be changed.

I believe that the calibration, as it stands, will only be accurate if the tun is at room temperature during the calibration and you subsequently use the tun at that same room temperature. If you calibrate at 65F and use the tun at 50F, your numbers will be off. So, I think, at least with a cooler, you need to use the tun temperature as Ambient.[/quote]

right, thanks, that does give the same effect. my efforts were also confounded by the fact that I had a sample schedule open and the "recipe" copy of the mash tun was no longer the same as the "database" copy and I was making changes and not seeing the effects.

As far as my test goes, ambient in the kitchen was almost the same as ambient in the gear room, so the cooler calib. should be OK (besides the previously mentioned issue). It's been my recent practice to use the MLT as a holding tank (degassing etc) in the kitchen before I mete my water out to my pots. Since it's usually sitting like that for 8-24 hours, the water should come up to ambient from tap, but I will check on that next time. I suppose I could use a different vessel if its a problem.
User avatar
slothrob
Moderator
Moderator
Posts: 1831
Joined: Mon Apr 10, 2006 1:36 pm
Location: Greater Boston

temp

Post by slothrob »

It doesn't seem that storing your water in the tun would be a problem. In fact it seems like it would give you pretty stable starting temperatures.

If your temperatures are reliably low, I'd just correct the calibration values. There is clearly a variable that we're not accounting for, somehow, but you seem to have the data that will allow you to correct for that variable and I don't see any downside to that.
BTP v2.0.* Windows XP
User avatar
jeff
Imperial Stout
Imperial Stout
Posts: 1602
Joined: Sun Jan 09, 2000 8:16 pm
Location: Hollywood, SC
Contact:

Post by jeff »

This thread demonstrates the great need for a feature I have been planning to implement in version 2.0. It is clear that the best way to calibrate a vessel or heat source is during real-world usage. As pointed out, the temperatures that are predicted can be changed to reflect real-world results. BTP adds a correction factor internally to account for these discrepancies.

What I intend to do is alter that behavior. Instead of adding the internal correction factor, BTP will ask the user if they would like to update the calibration of the vessel itself to reflect the new temperature. Of course, doing so would affect all of the calculations where the vessel is involved but this may be desired. The idea is to eliminate the need for a separate calibration procedure, as well as improve the accuracy of the calibration.

Thoughts?
Jeff
BeerTools.com Staff
User avatar
slothrob
Moderator
Moderator
Posts: 1831
Joined: Mon Apr 10, 2006 1:36 pm
Location: Greater Boston

Post by slothrob »

jeff wrote: What I intend to do is alter that behavior. Instead of adding the internal correction factor, BTP will ask the user if they would like to update the calibration of the vessel itself to reflect the new temperature. Of course, doing so would affect all of the calculations where the vessel is involved but this may be desired. The idea is to eliminate the need for a separate calibration procedure, as well as improve the accuracy of the calibration.

Thoughts?
This sounds like a great idea.

Perhaps this action could create a new vessel that has the calibration date addended to the name, so that we could revert to a previous tun if the new information turned out to be aberrant.

Would there be a way to accumulate data over multiple runs to determine a median value?
BTP v2.0.* Windows XP
erykmynn
Posts: 5
Joined: Tue Mar 02, 2010 5:02 pm

Post by erykmynn »

Once I figured out that my database equipment numbers weren't going straight to my recipe, I was able to fudge my calibration and get more real-world numbers...

I brewed an oatmeal porter this weekend, and my mash-in temp couldn't have been more dead-on. (though I did accidentally introduce some extra thermal mass to the cooler, but that is a whole other story)... my temp drops in the rest were in the ball park, and after another mash I will really have them dialed in.

what a relief!
jeff wrote:This thread demonstrates the great need for a feature I have been planning to implement in version 2.0. It is clear that the best way to calibrate a vessel or heat source is during real-world usage. As pointed out, the temperatures that are predicted can be changed to reflect real-world results. BTP adds a correction factor internally to account for these discrepancies.

What I intend to do is alter that behavior. Instead of adding the internal correction factor, BTP will ask the user if they would like to update the calibration of the vessel itself to reflect the new temperature. Of course, doing so would affect all of the calculations where the vessel is involved but this may be desired. The idea is to eliminate the need for a separate calibration procedure, as well as improve the accuracy of the calibration.

Thoughts?

Well, I liked the idea of a dry-run calibration. And it was really helpful for figuring out the heat output of my burners and heatstick, and nailing down my boil-off (which isn't actually in your program but it could be calculated from the power or the heat source I believe.... which I did after BTP helped me calc the kW and think I finally got right this time).

I also used that dry-run data to better estimate my timing.... I made a 'fake' recipe that has only a fraction of a percent of grain and use that to estimate how long each heating step will be for my convoluted stove-top all-grain method I've adopted. It would be great if you could input that kind of thing in the recipe itself, but there's a whole other thread about that so I'll stop being off-topic and describe it there.

Anyways, I don't think I would want to calibrate my burners or kettles in a "real-world" situation, and I don't know that it would make much difference.... But as far as the Mash Tun goes, there is obviously something that is not equivalent between the real-world and the calibration procedure. I'm not sure what would be best. Maybe calibrate the heat it absorbs in a dry-run still, but then calibrate the temp drop based on real world stuff? or just have a "wizard" that helps you adjust after you do a batch. The reason I suggest that is because it was really really helpful to have BTP give me an accurate mash-in even though the weird end-of-rest temperatures it predicted made me panic a bit.
Post Reply