A simple Bubble Lock vs Tinfoil experiment.

In the thread "Yeast starter - airlock or tinfoil" the general consensus is that tinfoil on starters is superior to airlocks/bubble locks because tinfoil lets oxygen into the container and thus the yeast will multiply and perform better than a container with a bubble lock.

I'm not convinced. I'm not in any way saying that tinfoil doesn't work; it does and it works well. But I'm not sure that the claim that "because oxygen gets into the container the yeast multiply more and one can get a bigger starter" is true. So I've decided to conduct an experiment to test my hypothesis that yeast starters with bubble locks perform as good or better than than starters covered with tinfoil.

The Experiment to test my hypothesis:

1. Make two (2) one liter starters. The preparation of each starter will be created by making 8 cups of water and 2 cups of DME. After the wort has boiled and cooled I will mix thoroughly and divide the wort evenly into two jugs. (thanks StoneHands and kh54s10)

2. Each starter will be transferred to sanitized half gallon jugs that are identical in size and shape.

3. When the temperature in both starters is below 80F I will pitch the exact strand of yeast in the exact quantity into each of the containers.

4. I will swirl/shake each starter the same number of times i.e., if I swirl one starter 5 times then I will also swirl the other starter 5 times, no more no less.

5. Each starter will work for 72 hours.

6. Before I seal each starter and after 72 hours the starters will be measured using a hemocytometer (thanks StoneHands and mewithstewpid)


Variables.

One jug will be covered with tinfoil.
The other jug will have a stopper and a 3 piece bubble lock.

Hypothesis - The starter with the 3 piece bubble lock will produce equal amounts of yeast (or more) as the starter with tinfoil.


Limitations:
1. I do not have 2 stir plates and therefore they cannot be used in this experiment.
2. I have to learn how to use a hemocytometer properly and accurately.


So now I have to ask you all how to make this experiment better. I need a way to measure yeast counts but I'm not sure how to do it in the best way possible. Any kind of suggestion would be helpful and will greatly influence how I measure the yeast count. Anything else that you see to make this experiment better is welcomed!

Reserved for Results of Experiment

I use the foam stopper that came with my flask, maybe you could throw that into the mix as well test out all 3 options

I'd mix up your starter wort in the same batch, boil it, etc. all in one, and then transfer the exact volume into two separate containers. But, that being said, I think you need a scope and a hemocytometer (wow, that spelling seems awful) to get an accurate count. Without that, the two amounts of yeast in your separate containers will seem the same, no matter the method you use. Also, how will you make sure you're pitching the same in each? Even if you do a full vial in each, I would think you'd have a lot of variability from one vial to the next. I think you'd need your scope and hemo to count your cells pre pitch (and pitch by volume here too).

Not trying to discourage, I just don't think you'll get good results without a hemocytometer.

To get an accurate wort sample I would create it as one batch then split it into equal parts. If you don't you have 2 variables - how densely packed is the cup of DME and did you get equal boil off and thus samples of equal gravity?

Swirling 5 times seems a bit inaccurate... How much variable for the force of each swirl?

As to measuring the yeast I guess if you were able to chill and make them settle in graduated cylinders you might get a close idea of the yeast amounts. This would be in volume not cell counts.

For me, I'll just take the word of more experienced people, including yeast lab scientists, that it is better to use foil than an airlock.

I am afraid the experimental design is not valid... too many uncontrolled variables and no way to actually measure your final yeast count.

StoneHands said:

I'd mix up your starter wort in the same batch, boil it, etc. all in one, and then transfer the exact volume into two separate containers. But, that being said, I think you need a scope and a hemocytometer (wow, that spelling seems awful) to get an accurate count. Without that, the two amounts of yeast in your separate containers will seem the same, no matter the method you use. Also, how will you make sure you're pitching the same in each? Even if you do a full vial in each, I would think you'd have a lot of variability from one vial to the next. I think you'd need your scope and hemo to count your cells pre pitch (and pitch by volume here too).

Not trying to discourage, I just don't think you'll get good results without a hemocytometer.

Click to expand...

Not discouraging at all! That's exactly what I'm looking for. But since a hemocytometer is for blood cells (hematic - pertaining to blood) is this still the best tool for measuring?

To get an accurate wort sample I would create it as one batch then split it into equal parts. If you don't you have 2 variables - how densely packed is the cup of DME and did you get equal boil off and thus samples of equal gravity?

Click to expand...

Excellent point, which is exactly why I asked for advise. I will certainly do this. Thanks for the input.

I am afraid the experimental design is not valid... too many uncontrolled variables and no way to actually measure your final yeast count.

Click to expand...

No offense, but this is a very unhelpful statement and one that seems to be rather careless. It's fine to be critical of the experiment, but at least offer some advise (or at least tell me where the problem are so that I can try to fix them) like the others above you. I'm extremely receptive and open to thoughtful comments.

hungry4hops said:

I use the foam stopper that came with my flask, maybe you could throw that into the mix as well test out all 3 options

Click to expand...

That's also an excellent idea, I just need a stopper!

If I can get this experiment to work (after I have worked out the bugs a bit first) then I'll do it again and add more variables. later I'll make two or three stir plates and see what that does!

Hemocytometers are used to count cells under a microscope. I use them all the time. In order for this experiment to work you need to precisely count the number of cells before and after fermentation. If you don't count before you may end up thinking there is a difference when it might just be different pitch rates.

Without counting you might just be wasting your time, sorry!

Without the use of two stir plates your experiment will have no bearing on how I prepare a yeast starter. Having a stir plate changes everything in my opinion.

aubiecat said:

Without the use of two stir plates your experiment will have no bearing on how I prepare a yeast starter. Having a stir plate changes everything in my opinion.

Click to expand...

Depending on how long it will take me to begin this experiment I may have the time to acquire two stir plates. I also agree that I will more accurate results than the swirling method, but I also know that lots of homebrewers don't use stir plates and therefore this experiment can still be useful for those who don't use stirplates for whatever reason.

bizarrojosh said:

Depending on how long it will take me to begin this experiment I may have the time to acquire two stir plates. I also agree that I will more accurate results than the swirling method, but I also know that lots of homebrewers don't use stir plates and therefore this experiment can still be useful for those who don't use stirplates for whatever reason.

Click to expand...

But then you need to make sure they spin at same rate or else that could be a confound.

mewithstewpid said:

But then you need to make sure they spin at same rate or else that could be a confound.

Click to expand...

Mental note taken. Science is complicated

This is awesome. Very excited to seethe results. I am skeptical too, but always use foil and hopefully won't change unless you prove me otherwise buy a large margin. I feel that the c o 2 will always push the o 2 out of the container and therefore no air is going to get in, regardless of method used. there's not much exchange to me, but rather a push of co2 during the replication stage....

I think the variables in this is too much to test. With putting on an airlock you are basically cutting off the introduction of any new o2. If you do this experiment with a 1l starter in a 2l flask, then the results you get would only be specific to the amount of headspace available. So if I use a 1l flask with the same volume, the amount of o2 available will be much smaller, which would most likely result in lower growth. Yeast need o2 to reproduce, so once it is used up growth will decline, if you have a continuous flow of o2 then the yeast should reach its full growth potential, where if it runs out it will stop.

Viability will be a big variable as well. You would need to test viability of the yeast before pitching, do make sure they are the same. Less yeast in more wort, depending on the volumes compared, will grow more than more yeast in the same amount of wort.

My guess: a 1l flask will have less growth with an airlock, a 2l flask will have will have less growth, but you will get close to your pitching rate. I would be willing to bet that if you did this experiment with the airlock and swirled 5 times and the one with the foil never swirled the one with the foil would still get more growth, because at some point you would just be introducing mostly co2 into the one with the airlock.

Im not sure if either method would hurt a 5 gallon batch, you don't even typically need to double the yeast count for MOST ales.

A way to get rid of the intial viabilty/cell count issue would be to use a single vial of yeast homogenize and volumetrically measure out equal volumes into each starter.

ccudc9 said:

I think the variables in this is too much to test. With putting on an airlock you are basically cutting off the introduction of any new o2. If you do this experiment with a 1l starter in a 2l flask, then the results you get would only be specific to the amount of headspace available. So if I use a 1l flask with the same volume, the amount of o2 available will be much smaller, which would most likely result in lower growth. Yeast need o2 to reproduce, so once it is used up growth will decline, if you have a continuous flow of o2 then the yeast should reach its full growth potential, where if it runs out it will stop.

Viability will be a big variable as well. You would need to test viability of the yeast before pitching, do make sure they are the same. Less yeast in more wort, depending on the volumes compared, will grow more than more yeast in the same amount of wort.

My guess: a 1l flask will have less growth with an airlock, a 2l flask will have will have less growth, but you will get close to your pitching rate. I would be willing to bet that if you did this experiment with the airlock and swirled 5 times and the one with the foil never swirled the one with the foil would still get more growth, because at some point you would just be introducing mostly co2 into the one with the airlock.

Im not sure if either method would hurt a 5 gallon batch, you don't even typically need to double the yeast count for MOST ales.

Click to expand...

Right, I think most of this is good stuff. However much of this conjecture is untested and that's what I plan to do!

Since many people make 1L starters regularly this will be a great experiment. You are right in that the results will be limited to 1L and may not represent anything bigger or smaller accurately. However, this is just to get some data for us all to at least have some verifiable evidence on whether the current favor for aluminum foil is as good or better than bubble locks. Everyone may be right, but I've asked for evidence in other threads an no one can give me any scientific data. I don't know if that means there isn't any or if that just means that no one posted it but either way I hope to get some repeatable data for those of us who make 1L starters.


Also, the viability issue is one that I too am thinking about. Besides buying the same yeast and pitching it in the two jugs, how else can I determine that they yeast source has the same viability? Or would it be better to make a preliminary yeast starter and then use that for the experiment?

Brad2287 said:

A way to get rid of the intial viabilty/cell count issue would be to use a single vial of yeast homogenize and volumetrically measure out equal volumes into each starter.

Click to expand...

Ok, this sounds like a plan! Where can I find instructions on how to do this?

bizarrojosh said:

Ok, this sounds like a plan! Where can I find instructions on how to do this?

Click to expand...

i believe he is saying to take one vial of yeast, mix it until it seems evenly suspended, and then take a specific volume for each starter, for example 10mL in each.

The problem with this is you still do not know whether you have the same number of cells, and you dont know the viability in each sample. therefore, you would need to do a cell count and cell viability assay (using a stain)


Another thing to think about is the OG of the starting wort, you would need to make sure they are exact, and that the same volume of wort was in each starter container. Honestly, with all of the potential variables, you really need a way to count everything, and i mean everything, or else your final result will not be reliable!

I cant see the results varying SO much in either container (airlock vs no airlock) that it will overcome the potential variability.

trust me, reviewers rip other peoples experiments to shreds. You have to think of everything before you start, and think of the questions they will ask, so your experiment is designed properly. if it is not the results cannot be trusted.

-rich

Here is a test that was done similar to what you are proposing.

http://seanterrill.com/2010/01/14/aeration-and-yeast-starters/

Removed double post

While cell count and viability are great, I would forgo both of these metrics in my evaluation. The observations I would want to compare in my analysis is how quickly does fermentation start, how vigorous the fermentation is, and both the actual & time to FG.

To determine superiority of foam, foil or airlock, this is how I would set up my experiment (i.e. I'm not saying this is how you should set it up):

1) Prepare 2250-2500 ml of wort.
2) Divide wort equally into three identical 1000ml flasks.
3) Take a single yeast vial, shake well, divide into three equal portions, and pitch one portion into each flask.
4) Cover one flask with foil, airlock one, and foam stopper the final.
5) Place flasks on three stirplates... I would build three "PC fan" stirplates with speed sensing/control for best results. In a pinch, I think you could power three identical stirplates in parallel from the same power source, but I'd want to hold the RPMs at a constant value in all three.
6) After 24-48 hours, prepare three gallons of wort & divide equally into three identical transparent primaries.
7) Pitch one flask into each primary and place a hydrometer into each (make sure they're calibrated the same)

This is where it gets tricky... somehow you want to be able to observe when fermentation starts, the rate of fermentation, and when fermentation ends. I would probably do this by measuring the CO2 production via some sort of liquid displacement method, and recording both those results and the readings from the hydrometers at regular intervals until fermentation has ended in all three primaries. I haven't quite thought it through how I would do this, but it would probably involve equal lengths of tubing, a bunch of hermetically-sealed quart-sized Mason jars and a few graduated cylinders.

After completing the experiment, I would repeat two or three times.

If the nay-sayers are really convinced that the experiment has to be this closely controlled to see any discernible definitive difference between foil and an airlock, then awesome! I'm using an airlock for its superior infection resistance!

I made a 2L start of a 3 year old pack of Windsor yeast. When I pitched it into the wort 46 hours later, I had a bubble a second within 3 hours. I did the same with a pack of Nottingham (same age) and when I pitched it into the wort 49 hours later, it took 2.5 hours to start bubbling away.

Ok. Maybe I can make better starters... but at this point, I really think I'm splitting hairs...

And yet, I'm still building a stirplate. I like the nerdy geek aspect of it.

mvanwie said:

While cell count and viability are great, I would forgo both of these metrics in my evaluation. The observations I would want to compare in my analysis is how quickly does fermentation start, how vigorous the fermentation is, and both the actual & time to FG.

To determine superiority of foam, foil or airlock, this is how I would set up my experiment (i.e. I'm not saying this is how you should set it up):

1) Prepare 2250-2500 ml of wort.
2) Divide wort equally into three identical 1000ml flasks.
3) Take a single yeast vial, shake well, divide into three equal portions, and pitch one portion into each flask.
4) Cover one flask with foil, airlock one, and foam stopper the final.
5) Place flasks on three stirplates... I would build three "PC fan" stirplates with speed sensing/control for best results. In a pinch, I think you could power three identical stirplates in parallel from the same power source, but I'd want to hold the RPMs at a constant value in all three.
6) After 24-48 hours, prepare three gallons of wort & divide equally into three identical transparent primaries.
7) Pitch one flask into each primary and place a hydrometer into each (make sure they're calibrated the same)

This is where it gets tricky... somehow you want to be able to observe when fermentation starts, the rate of fermentation, and when fermentation ends. I would probably do this by measuring the CO2 production via some sort of liquid displacement method, and recording both those results and the readings from the hydrometers at regular intervals until fermentation has ended in all three primaries. I haven't quite thought it through how I would do this, but it would probably involve equal lengths of tubing, a bunch of hermetically-sealed quart-sized Mason jars and a few graduated cylinders.

After completing the experiment, I would repeat two or three times.

Click to expand...

While reading this thread, I was thinking along these lines. I thought if you put an uninflated balloon over the top of each batch, you could easily determine which was fermenting faster.

kh54s10 said:

To get an accurate wort sample I would create it as one batch then split it into equal parts. If you don't you have 2 variables - how densely packed is the cup of DME and did you get equal boil off and thus samples of equal gravity?

Swirling 5 times seems a bit inaccurate... How much variable for the force of each swirl?

As to measuring the yeast I guess if you were able to chill and make them settle in graduated cylinders you might get a close idea of the yeast amounts. This would be in volume not cell counts.

For me, I'll just take the word of more experienced people, including yeast lab scientists, that it is better to use foil than an airlock.

Click to expand...

Making beer in a home environment will never lead to lab quality statistically supported analysis or research. I like the simple, cold crash in graduated cylinders idea. It could easily be repeated each brew day and would give another piece of brew day data for yeast measurement rather than using just a batch starter formula to reach a target pitching rate. How do I know my smack pack has the percent yeast viability the calculator says it has? How do I know that my starter grew to the desired pitch count? I know. Too many thoughts, just have another beer.

Thanks for the graduated cylinder idea. I will be trying it on my brew days!

I would agree with the yeast distribution method. It seems that in a situation like this, the best results (and cheaply) are going to be gained from using the same volume of yeast slurry from a single vile. The accuracy of yeast count is then pushed off onto the volume reading itself. Do you happen to have a pipette, or anything that resembles a pipette? the smaller the diameter the tube the better, as the hieght of liquid in the tube becomes more sensitive. Using a turkey baster with some tape would get you caveman close. Using the long part of the bottling wand would get you bronze age accuracy, a coffee stirrer full (small diameter) would be even better. If you can find one, a small graduated cylinder would work too.

I can't wait for the results! However, I would like to comment on all the naysayers here. The test should probably be carried out with some very precise equipment, but lets be realistic; the results gained here are pretty much anecdotal unless he were to do this procedure at least 30 times over. It would be hard to do this with controlled accuracy in a lab setting, let alone a kitchen. I think that if he can get close enough to the same number of yeast, and close enough to the same stir plate activity and the results are minimal then we can say there is no difference. But, if by doing the best he can to stay close, and a huge difference is apparant (pictures posted would be nice) then he can think about getting more accurate.

It would be cool if you could take a series of pictures, spaced about an hour apart with the stir plate off for a moment, and post them so that we can have a look at the action.

mewithstewpid said:

Hemocytometers are used to count cells under a microscope. I use them all the time. In order for this experiment to work you need to precisely count the number of cells before and after fermentation. If you don't count before you may end up thinking there is a difference when it might just be different pitch rates.

Without counting you might just be wasting your time, sorry!

Click to expand...

THIS^^^^!!! You need to count cells and total volume of both what you pitch and what you end up with.

wow, thanks for the feed back everyone! Since this experiment is going to be a bit more complicated than I anticipated it's going to take me a bit of time to do it, especially since my graduate classes start this coming Monday. But I will take detailed pictures and I will keep a thorough log of the data. As one poster said, this is only a home experiment and it can't and shouldn't be to be as conclusive as as a lab experiment, but I do hope to get some data that could be useful to homebrewers and I'm very happy that other on here are giving such thoughtful input!

Also, the yeast distribution method seems to be the most logical and the most exact so I'll acquire a pipette. My father is a MD and so when his office is closed I can use his equipment and I'm nearly positive that they have a pipette and all the other "lab" things that I need.

Denny said:

THIS^^^^!!! You need to count cells and total volume of both what you pitch and what you end up with.

Click to expand...

Perhaps I'm missing the point of preparing a yeast starter, but I see massive potential failure in basing and analysis strictly on yeast cell count/volume. Unless there's a proven correlation near 1 between number of yeast cells and the effectiveness of a starter, it would seem to me that yeast - being a living organism - can have varying levels of viability as well as rates of reporduction and metabolism from cell to cell. Therefore, your "foam" starter may produce 5x the number of cells as your "airlock" starter, but your airlock starter may outperform your foam starter in fermentation.

To measure the effectiveness of a yeast starter, you are need to observe the effectiveness of fermentation resulting from said starter. Counting yeast cells may seem like a good "predictor" of which way a starter would go in fermentation, but it doesn't seem like something upon which you could draw a conclusion.

Experiment died?

Kialya said:

Experiment died?

Click to expand...

Thanks for asking! Not dead, but I have to put it on hold because I've started an internship on top of my normal class load so I have zero time for it right now. But if others want to do it before me that would be awesome! It will be done once things cool off but it might be many months.

subscribed, awating an experiment