Yeast Harvesting

[FunkedOut]

I have been overbuilding starters and reserving a calculated 100B cells for the next brew's starter for a few years now.
Never done a cell count, just using the BrewUnited calculator to estimate, but results are consistent and good for me.
http://www.brewunited.com/yeast_calculator.php
Recently got into a conical and wanted to harvest yeast from the actual batch and use for subsequent pitching.
Using the volume of compacted slurry that was calculated at 100B cells, I estimated what the compacted slurry would be from a batch of beer and ended up with way less than estimated.
Estimations were based on the calculator's no agitation setting.

From my first batch (6 gallons) I had estimated 800mL of yeast cake and only ended up with 400mL (German Bock).
From my second batch (16 gallons) I had estimated 1100mL of yeast cake on only ended up with 500mL.

I haven't kegged this second batch yet, but the fermenter walls were clean of yeast on the first batch.
I fermented with the dump valve open and a mason jar attached to the bottom.

Are these harvests in line with what you guys are getting from your 5 and 15 gallon batches?
If I harvest 500mL from a 15 gallon batch, would pitching 1/3 of that be enough for an identical batch?
Not sure how to proceed from here with this new territory.
Thanks for your time and help.

 

[renstyle]

Was your calculation for 100B cells using a compacted slurry based on previous pre-fermentation starter volume? or with the post-fermentation volume of cake/trub?

Post-fermentation yeast cakes have layers. The lowest contains the majority of the dead yeast cells and trub that has settled out.

The "good stuff" is generally in a middle layer, with some of the top layer also containing viable cells.

This is where yeast rinsing comes into play. The amount of viable yeast you can utilize after decanting the excess water after swirling will be smaller than the solids left at the bottom of the vessel.

400mL of cake could get you 50-100mL of yeast slurry. Can't say what the concentration of cells would be with this. The yeast calc you referenced above has a +/- 15% variance in it's assumptions, which will get you in the ballpark.

Use this slurry in a new starter for the next batch and build it up. Personally I'd assume a smaller number of viable cells in the slurry when plugging in values to a build/over-build calculator. It's good CYA to ensure you have enough healthy cells for your next batch.

 

[FunkedOut]

Every time I overbuild a starter by 100B cells (by calculator), I collect it in a jar. I used that consistent, compacted volume of cake to equal 100B cells.

my plan was to take the mid 1/3rd and pitch it directly with no starter.
no rinsing either.
just spoon off a 1/3rd and discard.
spoon off the next 1/3 an pitch that.

looking for a sanity check on harvested volumes.
my post ferment harvest is trub free.
i dump all the trub before pitching.

 

[renstyle]

my plan was to take the mid 1/3rd and pitch it directly with no starter.
no rinsing either.
just spoon off a 1/3rd and discard.
spoon off the next 1/3 an pitch that.

looking for a sanity check on harvested volumes.
my post ferment harvest is trub free.
i dump all the trub before pitching.

Taking the middle third as you describe is basically the definition of "yeast rinsing". Most folks just hold that middle third back for a starter down the road, rather than direct-pitch. Without a microscope and some patience counting cells, it's difficult to know for certain.

Is the 500mL you're quoting for harvested volume before or after the top and bottom third are removed (i.e. only the middle third)? If 500mL after is the case, it may be enough for a 15gal re-pitch. To me it seems a bit low based on my 5 gal batch history.

It's only a feeling tho, as I've only harvested to make a new starter... I've always underestimated my yield in the calculators, usually 60-70B cells, to allow more wiggle room (basically an overbuild) in the calculator and avoid an under-pitch situation.

So, I answer your sanity check question with a "definitely maybe".

I'd love to hear how it works out for you on the next brew!

 

[CascadesBrewer]

I am not sure if I can help with your specifics. I have been harvesting yeast for a few years with generally good luck. I don't have a conical. I don't have a way to dump trub and I tend to transfer quite a bit of trub into my fermenter (I will try to filter out the majority of hop matter). Since my harvested slurry had a lot of trub, I knew many of the "X B cells in Y ml slurry" did not apply to me.

I am sure it is not exact, a few sources hinted at the idea that if you pitch 200B cells into an average 5 gal batch, you will end up with 800B cells at the end of fermentation. I took that to imply that pitching 1/4 of my slurry into another batch would be around pitching 200B cells. I tended to find that I could fill around four 16 oz (~0.5L) jars of slurry from a typical 5 gal batch. I have had good luck with direct pitching one of those 16 oz jars of slurry if done within 2 to 4 weeks from harvesting.

So my take on your "would pitching 1/3 of that be enough for an identical batch?" question is: yeah, that seems like a pretty reasonable pitch. I am not sure about your 1/3 + 1/3 + 1/3 strategy. It seems like your harvested slurry will tend to be well mixed with lots of viable yeast throughout. You would probably have to try to capture these layers as you dump the yeast.

I am still trying to figure out my best strategy when yeast starts to reach beyond 4 weeks...maybe 3 months. I will make a starter but I am not sure how big of a starter is ideal or how much slurry to add to the starter.

 

[Culinarytracker]

I am still trying to figure out my best strategy when yeast starts to reach beyond 4 weeks...maybe 3 months. I will make a starter but I am not sure how big of a starter is idea or how much slurry to add to the starter.

That's the thing for me. I just don't brew enough I guess.

I went on a deep dive into repitching a few years back when I was brewing for a friends wedding, and ended up making 5 or 6 batches with rinsed yeast slurry. It worked really well!

I am thinking about building myself a little yeast lab so I can do some long term storage stuff and play around with keeping several strains going.

 

[Off Balance Brewing]

I'll give the lazy man's opinion. I've been repitching for the last few years. Mainly repitch 3 to 5 times on 1050ish brews. I harvest the top 2/3 of the cake in a 24 oz mason jar. With some beer in there I gently rotate to mix slurry and try to get the trub to settle. I keep these for 1 to 3 months, decant the beer off the top and then pitch about 10 oz of slurry in the next beer. Seems to work out well. If I need a higher gravity beer I would pitch more fresher but this has been working for me. I usually try to avoid building a starter. I've read pros and cons of washing versus storing in alcohol beer slurry.

 

[FunkedOut]

Is the 500mL you're quoting for harvested volume before or after the top and bottom third are removed (i.e. only the middle third)? If 500mL after is the case, it may be enough for a 15gal re-pitch. To me it seems a bit low based on my 5 gal batch history.

I collected 500mL of yeast cake in total from this 15 gallon batch.
It seemed low to me too, that's why I came here.

So, I answer your sanity check question with a "definitely maybe".

I'd love to hear how it works out for you on the next brew!

Well, par for the course.
I'll share my results here.

 

[Brooothru]

So this thread has gotten me to do some serious thinking about my yeast harvesting and propagation (I know, that's a pretty dangerous undertaking). I've been playing around with several different calculators for pitch rate, but all I've achieved is some self-doubt about my methodology. The latest calculator I've been playing with is BrewUnited's pitch calculator, and it's taught me a lot about assumptions and entering argument values. I like the way it's calculations flow and how the final numbers change radically based on initial entries. That's where my conundrum starts.

It would be easy if I were just basing my assumptions on the manufacture dates and the manufacturer's stated initial cell counts. But where things start getting murky is when I estimate the initial cell count value for yeasts I have previously harvested and either stored in the refrigerator for several months or frozen with glycol/sterile water for several years.

So I'm asking for some educated 'guesstimates' as to what my starting values are for over-built starters, harvested and washed starters, and revived starters from slants or frozen samples. For instance, if I start with a White Labs (100 billion cell count) or an Imperial (200B cell count) and harvest and wash a clean 300ml slurry from a ferment 1.050 O.G. beer, what range of cell count is logical to assume? I realize that the co-dependent factors are huge influencers on the actual number, but short of using a microscope and performing an actual cell count, what are some average values you use as entering arguments in these calculators?

Age, handling, initial viability, temperature control, etc., all play determinative roles. But the initial assumed value of cell count is where I'm getting hung up. Garbage in, garbage out. For any of the other values to be anywhere close to valid, there has to be at least a semi-accurate range of values for initial cell count. What do the rest of you use?

 

[Brooothru]

I am not sure if I can help with your specifics. I have been harvesting yeast for a few years with generally good luck. I don't have a conical. I don't have a way to dump trub and I tend to transfer quite a bit of trub into my fermenter (I will try to filter out the majority of hop matter). Since my harvested slurry had a lot of trub, I knew many of the "X B cells in Y ml slurry" did not apply to me.

I am sure it is not exact, a few sources hinted at the idea that if you pitch 200B cells into an average 5 gal batch, you will end up with 800B cells at the end of fermentation. I took that to imply that pitching 1/4 of my slurry into another batch would be around pitching 200B cells. I tended to find that I could fill around four 16 oz (~0.5L) jars of slurry from a typical 5 gal batch. I have had good luck with direct pitching one of those 16 oz jars of slurry if done within 2 to 4 weeks from harvesting.

So my take on your "would pitching 1/3 of that be enough for an identical batch?" question is: yeah, that seems like a pretty reasonable pitch. I am not sure about your 1/3 + 1/3 + 1/3 strategy. It seems like your harvested slurry will tend to be well mixed with lots of viable yeast throughout. You would probably have to try to capture these layers as you dump the yeast.

I am still trying to figure out my best strategy when yeast starts to reach beyond 4 weeks...maybe 3 months. I will make a starter but I am not sure how big of a starter is ideal or how much slurry to add to the starter.

Same question I keep asking myself. ^^^

 

[FunkedOut]

Let me see if I can recreate my estimations and calculations here for posterity...

Data Point 1:
For years now, when I overbuild a starter by 100B cells, using the calculator linked to above, the 100B cells compacts to ~37.5mL. So that math works out to ~2.67 billion cells per mL.


Data Point 2:
In preparation for a 6 gallon batch in the conical, I built up a starter.
6 gallons of 1.072 SG called for ~597B cells with that calculator, which actually compacted to ~250mL. So that math works out to ~2.39 billion cells per mL.


Data Point 3:
Using the calculator to model the 6 gallon batch...
5.0625 gallons after dumping trub (no hop socks used), no agitation, 597B cell pitch, 1.072 SG...
Expected a calculated 2,102B cells. Using an average of 2.5 billion cells per mL, that would mean the 2,102B cells would yield ~840mL of yeast cake.
I only collected 400mL.
So that math works out to ~5.25 billion cells per mL.

At this point, the calculated number of cells is suspect, as I've never propagated yeast without a stir plate.
Or, it seems that as the volume of yeast cake increases, so does the density.

Data Point 4:
Using the calculator to model the 16 gallon batch...
14.75 gallons after dumping trub (no hop socks used), no agitation, 1,190B cell pitch, 1.053 SG...
Expected a calculated 4,418B cells.
Using an average of 2.5 billion cells per mL, that would mean the 4,418B cells would yield ~1,770mL of yeast cake.
Using an average of 5.25 billion cells per mL, that would mean the 4,418B cells would yield ~840mL of yeast cake.
I only collected 500mL.
So that math works out to ~8.8 billion cells per mL.

The trend continues, as the batches get larger, either the cell counts do not multiply as much, or the density increases.
I tend to believe the cell counts do not multiply as much, with no basis or rationale.
I am thinking of taking this 500mL of yeast cake and pitching it entirely into the next 16 gallon batch. Makes life easy.

 

[FunkedOut]

Regarding the initial values on the calculator, I always assume that the calculated overbuild of 100B cells contains exactly 100B cells.
I trust the calculator's aging math and build up from there.
For years now, every time I overbuild a starter by 100B cells, whether from a fresh vial or from a previously overbuilt harvest, I end up with the same amount of yeast cake and results in my fermentations.
No idea what the actual cell counts are, but consistency is king.

Moving away from overbuilt starters, harvested and washed, I have almost zero experience and have shared it all in this thread.

I have always read that a batch will yield enough yeast to pitch 3 more batches, with middle third of the yeast being the most desirable, but the data I have collected thus far does not line up as cleanly as I would have hoped.

 

[Brooothru]

Excellent. Thanks for taking the time to outline your methodology. I've been pretty much flying blind with regard to whether I was over-pitching or under pitching. Most of my yeast ranching has either been harvested and washed samples stored in the beer fridge or samples from overbuilds that got frozen with glycerin/water for long term storage. The problem with fridge storage obviously is the loss of viability during storage. With freezer storage the losses occur mostly upfront and later upon thawing, though not so much during storage assuming they stay frozen. With a 100B overbuild you pretty much know what the count is going into the freezer.

I've arbitrarily settled on 1 month viability loss = 1 year frozen, but it's nothing more than a WAG and probably is a totally worthless number based on nothing more than an unfounded guess. After thawing I decant the slurry separated from the glycerin/water (about 20ml) and propagate it un-agitated on 200ml of 1.020 wort for 3~4 days at room temperature. Then I add that to .75L of 1.040 wort for 48 hours on a stir plate, followed by another 1.5L of 1.040 wort for 48 hours of spin time. That's something over 2~2.5L that results in 200~300ml of dense slurry after crashing and decanting. I have a known volume but don't have a clue about cell count. I will say that my starters almost always take off and produce beer though I always have a nagging fear that I'm making gross under-pitches some times. Your methodology gives me some basis for making a bit more accurate estimate.

The harvesting and washing presents more of a challenge for accurately "guesstimating" cell counts. Usually I can recover more than a liter of clean, dense slurry from a batch. I take the date of the start of fermentation as the age for the "best by" date loss of viability value. Once again, pure speculation as to what the true cell count is. As near as I can see from research it could be anywhere from 100 Million to 600 Billion. That's an unacceptable margin. Once again, I have to guess that a 1~2 month old 2-step up will produce 200 billion at least with the margin of error increasing exponentially as the age increases.

Increasingly I'm of the opinion that the only workable method to getting anything even approaching an accurate estimate is to overbuild rather than harvest, and go from there. When I propagate a starter I can save several 50ml samples of slurry to be refrigerated and used to re-propagate within 4~5 months, or to freeze and use within 2~3 years. I hate to dump all that good yeast down the drain (although my septic tank loves me for it), but if I want accurate and predictable outcomes as well as extremely clean, untainted yeast without genetic drift, that's what I need to start doing.