Heating and cooling error why?

[ericmark]

When one sets something up one tries to work out what is likely to happen. I bought two controllers as STC-1000 with double relay and a MH1210A with single relay, both rather simple devices they have a differential which can be set minimum for STC-1000 is 0.3°C and can be used for heating and cooling, The MH1210A it is 0.1°C minimum and can be used for heating or cooling. There is no pulse width modulation simply on/off also with a programmable delay on start.



So I use these units to keep my brewing beer at the right temperature using an old fridge/freezer a demo under floor heating tile 18W and a CFL of 8W and to start with only used to heat, as only one controller the MH1210A.



Once the weather started to improve I found it would over shoot with the 18W tile, and once too hot 40 pints takes a lot of cooling down again, the sensor was held tight to fermentor under a sponge to insulate it from the air. The 8W bulb was a lot better, and I assumed due to size, using an energy meter it showed only really needed 5W in the heart of winter to maintain a temperature of 20°C.



Now the refrigeration motor is a fixed size of around 80W, so my thoughts were, if 18W causes over shoot on heating then 80W on cooling would result in a massive over shoot on cooling, so idea was to use one controller to heat and the other to cool to increase the gap between heating and cooling to stop any yo-yo effect.



However when it came time to cool, there was no over shooting, now on second brew to use cooling and even on first switch on with 2°C to cool set with 0.3°C differential the freezer air did cool to 5.4°C but the fermentor came down to 18.5°C set and the only time it went over the 18.8°C when the fridge should be turned back on was due to the time delay set on controller and the time the freezer takes to do a system check. So on first switch off, it rose again to 19.1°C.



It is a frost free freezer so there is of course a circulation fan when freezer is on, also the fermentation does release some heat. But even set to 0.1°C on heating it over shot by 2°C yet on cooling it is really spot on.



I can’t explain why it works so well on cooling but not on heating, using the 8W instead of the 18W heater did improve the heating so over shoot was only 0.5°C, but 18W v 60W the cooling should have been worse not better.



So any ideas why?

 

[big-all]

i used to use a desk knee well covered with a thick towel in a warm but unheated room
30pint plus a 40 pint open tubs
2 light fittings about 1 ft above
start with a 1x100w bulb check 8 hrs later too cold then another 40/60/100w bulb dependent on how cold
from recollection from about 30 years ago 140-160 on the coldest days and 60-80w on a cool day but no to occasional heat in the summer
checking once a day was more than enough once it settled down
now it still brews and to the same strength but takes perhaps 2 days longer if the margin is a bit out[to warm /too cold 3deg]
my home brew was 1120 starting gravity
a beer kit and 3kg for the 30 pint and 4kg for the 40 pint and yes it finished about 13%

 

[ericmark]

I also use to use manual control to set brew temperature, stick a body warmer on the fermentor or switch on the heater for an hour. I had no way to cool, and beer brewed in the summer needed a very long conditioning time before drinking and was never as good as a winter brew.

When the fridge/freezer was condemned and insurance paid out, I moved it to garage. Idea was at first to use it purely as an insulated box, but this year I ran short of beer down to last 80 pints. So decided to try summer brewing and use fridge/freezer to keep it cool. I expected problems, however it went well, so now brewing all year around no excuse for a beer festival to celebrate first brew of the year!

But question remains why does the cooling work so much better than the heating?

I love peoples reaction when you let drop you have 80 pints in the fridge and freezer.

 

[ban-all-sheds]

If it's 13%, and they have any sense, they run away...

 

[ericmark]

I have done a brew at 21% ABV but the method used is very different to beer brewing. Over 6% you start to get a bitter after taste, by brewing just sugar and then adding charcoal to absorb the after taste and then finally adding flavour it does work, but that's not really brewing beer.

However what I am looking for is why does heating over shoot but cooling does not when there is far more energy going into cooling than heating. Or how do I stop the over shooting on the heat cycle.

With the old Cooper heat the controller would pulse the heat and we got a near straight line graph, but the controllers were expensive. Heart of winter not a problem I only need to heat, hight of summer not a problem only need to cool. But in between those two extremes I have a problem the heating over shoots so the cooling then has to come on. Ideas on how to stop it.

 

[ban-all-sheds]

I had some homebrew once, student days, where the guys who made it used the fairly unsophisticated technique of using a high-alcohol tolerant wine yeast (CWE 33, IIRC), more sugar, and just kept going until the yeast died.

It was one of the most vile brews I have ever tasted - really strained the notion that "there are no bad beers, it's just that some are better than others".

Although, by the time I'd finished the first pint (which took a while) it was tasting better, and pint #2 went down reasonably quickly. At which point they looked at each other, and to resolve their private argument about how best to share an odd number of bottles between them, said "bet you couldn't drink another one".

"Bet you I could", I said, and I did, in about 10 minutes.

That was a Wednesday evening. I woke up on Thursday evening with only patchy memories of the previous 24 hours. I still had a hangover on Saturday.

 

[SFK]

Why is the hysteresis of your cooling cycle worse than the hysteresis of your heating cycle.

My quick thoughts were that:
- Specific Heating Capacity of the beer coming into play >>> but I think that should not matter as in this case will be the same in both directions (no phase change)
- Heating System (at say 100% for heating plate) being more efficient than Cooling System (at say 70% for a fridge)>>> but this simply changes your 18W v 60W to say 18W v 42W.
- Speed of reaction of Heating System being faster than speed of reaction of Cooling System.

My gut feel is that this last reason, is the issue. But I have no proof. My guess is that :
- the heater gets hot fast and might have some thermal mass that continues to give out heat after electricity is removed (making it fast and continue to provide heat after turning off).
- the fridge with its compressor needing to warm up, gets cold air in slowly, and as soon as it switches off the cooling stops (making it slow and provides no cooling after turning off).

How to stop it???...??? Sorry but I do not have a simple answer to this. My suggestion means you have two different hysteresis for your controllers to address, and I am not sure that your controllers can easlity be modified to control that. Is it as simple as having two diffrent set points on the two controlers?

SFK

 

[big-all]

i used to let my home brew clear naturally but no finings no secondary fermentation would take about a week and needed drinking within about 4 weeks
it was decanted into 1-3L pop bottles with no head space
and what did it taste like ---- barley wine an absolute dream
i used to drink 50% diet lemonade and 50% homebrew giving a very palatable 6.5% very close to the 6.1 cider i drink now

 

[ericmark]

I think SFK might be working towards the problem. Not quite worked out cure, but idea that the under floor heating tile stores more heat than the freezer is something to consider.

I will guess the evaporator unit behind the shield even if it is cold is not still removing heat from the freezer once the fan stops. Where the under floor heating tile will have some heat after the controller switches off.

I have time on my side, until the weather cools down I don't need to heat, so I can think about it. Quick thoughts are a USB power supply and USB fan both cheap and a simple bulb. Thinking about two things, the faster the heat gets into the fermentor the less residual heat is left, and the smaller the heater the less residual heat is left. I think the under floor heating tile just stores too much heat.

 

[stillp]

Could it just be that the condemned F/F is horribly inefficient? With a heater, all the energy going in is available as heat. With a fridge compressor...

 

[ericmark]

I am sure it does not match the replacement inverter drive unit. But to get fermentor from 20.8°C to set 18.5°C the compressor was running 40 minutes and the freezer air temperature dropped to 5.4°C. However with the heating tile when in error the sensor was pulled out so it was running for days without any control it reached 40°C lucky the beer had already finished brewing so some new yeast before bottling resulted in it still being drinkable.

So working on that the motor has more cooling power than the heater has heating power.

I looked at the 5.4°C and thought that must over shoot, but no, the fermentor did not go below the 18.5°C set.

 

[SFK]

Eric,
I have been thinking about whether modifying your 'deadband' would improve (or worsen) your control system (so that heating cycle stops before setpoint earlier than cooling cycle), and I found this. Have you seen this brewers discussion on hysteresis effecting the setpoint as they seem to be having some similar issues?:

http://discourse.brewbit.com/t/hysteresis-above-and-below-the-setpoint/78

sfk

 

[ericmark]

I have found many who brew actually measure the air temperature rather than the brew temperature to stop the hysteresis, I think there are two reasons why this is done. One is the STC-1000 temperature controller has a minimum setting of 0.3°C which is common differential between heat on and heat off, and heat off and cool off and cool off and cool on. The second is use of very large heaters. As said mine is too large at 18W yet many use 60W tube heaters which are clearly from my measurements really OTT.

Because they measure air temperature rather than fermentor temperature they then have to compensate so will start off at 16°C and slowly lift it to 20°C measuring on the fermentor this is not required.

I had read many reports on making brew fridges and the problems of over shooting so I had decided easy way was to use two temperature controllers one for heating and one for cooling so I could increase the gap between heating and cooling. However using the MH1210A with cooling I found to my surprise it stayed within the 0.1°C set other then while waiting the 15 minutes set as re-start inhibit. So since the STC-1000 has a minimum of 0.3°C I retested with it set to 0.3°C and still stayed within the 0.3°C set except for the 15 minutes set as re-start inhibit.

So with the cooling there is no problem putting the sensor on the fermentor. Hence the question why is there a problem with heating?

SFK is clearly using some logic and likely myself trying to work out why. My thanks. I can't test until it cools down, today I don't need heat, I think two things may help, one a smaller heater and two a heater which has a lower mass. I think most likely using a simple bulb will cure the problem. Less mass and smaller wattage. But likely September before I can put it to the test.

It is very easy to adjust the dead band, my thoughts are it would make it worse, however thought the same changing from 0.1°C to 0.3°C on the cooling, but in fact it made not a scrap of difference. As on first switch on I saw the air temperature in the freezer plummet to 5.6°C I was sure the fermentor would over shoot, but no, it kept to exactly what it was set to. There is a lot of mass in 40 pints of beer, and since fermenting it is giving off heat.