Stepping Up A Starter

[glasseye]

I noticed that on brew strong Doc said to mix the DME with hot water. When making my starter I generally mix the DME with cold water then add to the flask. Is there a reason to boil the water first?

The DME will dissolve easier in hot water.

Obviously you want to boil the DME+water+nutrients and let cool before adding the yeast for sanitation purposes...

[Thirsty Boy]

I am still having problems trying to figure this out. I am going to attempt my first lager (5.5 Gal at about 1.050) and would like to not run into any problems with under attenuation. If I use the Mr. Malty calc and Thirsty's recommendation it had me start with a 1L starter and then step up to a 3L starter. What do I do if my flask is only 2L? I am guessing splitting the 2nd step into 2 1.5L starters will not give me the same growth.

You could muck about - but really, you kind of need to get a 4L (1G) flask/jug at least if you plan of brewing lagers regularly. I mainly use my "step up" method for big batches or big lagers where you'd need a stupid 6 or 8 Litre starter or something like that. Just trying to keep things down to 4 or 5 litres.

If you are going to change the calculator Jamil... I'd like to see it work for "smaller" volumes too, so I can get a bit more of a handle on cell counts when I am ramping up from 10ml of wort inoculated from a slant.

ATM I am basically guesstimating that going from a 10ml innoculum into 250ml on a stirplate, will give me roughly 100billion cells once its fermented out. I then calculate as though I had a perfectly fresh smackpack/vial.

[glasseye]

If you are going to change the calculator Jamil... I'd like to see it work for "smaller" volumes too, so I can get a bit more of a handle on cell counts when I am ramping up from 10ml of wort inoculated from a slant.

ATM I am basically guesstimating that going from a 10ml innoculum into 250ml on a stirplate, will give me roughly 100billion cells once its fermented out. I then calculate as though I had a perfectly fresh smackpack/vial.

You could just adjust the viability percentage in the calculator...

[Nyakavt]

If you are going to change the calculator Jamil... I'd like to see it work for "smaller" volumes too, so I can get a bit more of a handle on cell counts when I am ramping up from 10ml of wort inoculated from a slant.

ATM I am basically guesstimating that going from a 10ml innoculum into 250ml on a stirplate, will give me roughly 100billion cells once its fermented out. I then calculate as though I had a perfectly fresh smackpack/vial.

It's hard to know what the starting cell count is from a 10ml step off a slant, this is going to vary with the amount of yeast you put in there from the slant. A more reliable method is to estimate the volume of slurry when you get to a sufficiently large starter. I haven't kept great notes on stepup sizes, but I usually had to go at least to 2L (sometimes 1L->2L) on a stir plate to get what I thought was enough yeast for the batch. Kai did one experiment that went from 10->80->300->2000 mL and it got him 70 mL of slurry, (very) roughly 200 billion cells. These numbers aren't definitive, so measure your own slurry volume to account for the variances you see with the innoculum, then you can start with the calculator.

The curve fit formula provided on the graph a couple pages ago works with any size starter (provided you are within 1 and 8 L / 100 billion cells), you just need the starter wort volume and starting cell count. But once you go through a few steps the potential error is so high that it's far more accurate to just estimate the slurry volume.

[Thirsty Boy]

It's hard to know what the starting cell count is from a 10ml step off a slant, this is going to vary with the amount of yeast you put in there from the slant. . . .

The curve fit formula provided on the graph a couple pages ago works with any size starter (provided you are within 1 and 8 L / 100 billion cells) . . .

Yep, thats the problem - BUT - the graph you posted is why I think it might not be too bad an issue.

If you look at how it trends when you are adding small amounts of yeast to large (relative) amounts of wort, the significance of the starting number of cells diminishes as the relative difference increases.

So I scrape my inoculation loop over my slant - stir it into 5-10ml of wort - which is going to give me an amount of cells. But I am putting a tiny amount of yeast into a large amount of wort... I think that the final number of cells in that 10ml.. is going to be reasonably close whether I scooped 50million or 500million cells on the loop. The you repeat the ironing out of the difference by doing another "large step" by pitching the 10ml into 250ml.

I strongly suspect that pitching into such high wort ratios is going to result in the yeast pretty much breeding to the resource limit - which will result in a fairly similar number of cells in the final wort, regardless of how many you started with.

The starters will perhaps take different times to ferment out... but should finish pretty close to each other in total cell numbers - not the same I know, but pretty close.

Practically - I notice no significant difference in the volume of slurry I collect stepping up a starter from loop - 10ml - 250ml - 2L, than I do if I start by adding 10ml of wort to the entire slant.

So - Given Jamil's experience in translating this sort of stuff into a calculator - I wonder if that curve can be pushed with any accuracy down to the smaller sizes. Is there a point where the starting number of cells into a given volume of growth medium becomes unimportant? Or perhaps just less important? The graph says there is... but does the math translate into achievable reality? A point where you can say -- "no matter what you start with... if you make Xml of starter @ 1.040 - you will have approximately Y billion cells of yeast when it has fermented out" Start here!

In the meantime - I will just borrow a haemocytometer and some Methylene Blue from work and count the little buggers -- work it out from there. I'll translate that into mm of settled slurry in a tube and use that for future guesstimates.

A nice calculator that worked at small volumes would be so much nicer though

TB

[bspisak]

Stepping back to the original process, is it better to chill and decant the first step before pitching to the next or can you just pitch the whole thing? If you chill and decant, you'd have more room in the second step for more fresh wort because you don't have to leave room for the first starter.

But, if you want to save time, and will get the cell count you need, is there any real disadvantage? Does it affect the viability of the next step to pitch a spent wort? Seems like it's better to pitch active yeast than to shock them into a slurry.

Example: I have a 2L flask and a 5L flask. I need a 4L starter. My first step is 1.7L and my second step uses 2.3L of fresh wort. Will I get the cell growth equivalent to that of a fresh 4L starter (assuming otherwise pitch a chilled and decanted slurry from the first step?)

Brian

[Thirsty Boy]

I think - that the "best" way to do it is to ferment out the starter, decant and pitch just the yeast into fresh wort. The yeast is then growing for more of the time in an environment that is ethanol and C02 free or at least at lower levels. If you transfer the whole volume, you are transferring in all the fermentation products - all of which inhibit yeast growth.

There is no need to worry about "shocking" the yeast - yeast in stationary phase (flocced out) is very resistant to shock of all kinds

Mind you - there is also evidence out there that says that yeast shouldn't be allowed to reach stationary phase during propagation - so... basically there is an argument both ways.

Decanting each phase takes a lot of time though, so I usually dont do it. I add the whole thing to the next step. I usually make the wort I am adding into - a little stronger, so that when "dluted" by the previous step it will still make the whole volume about 1.040 - otherwise each time you step up you are diluting by more and more.

I am not a fan of pitching the entire starter volume into my wort - I frequently make lagers though, so often a starter might be 25% of the size of my main fermentation... I might be (but usually aren't) less fussy with a 2L starter for an ale.


I suspect the difference between the two methods would be minor - certainly not enough to worry about. Pluses and minuses in each direction... so take your pick based on what suits your system and process really.


TB

[Nyakavt]

If you look at how it trends when you are adding small amounts of yeast to large (relative) amounts of wort, the significance of the starting number of cells diminishes as the relative difference increases.

So I scrape my inoculation loop over my slant - stir it into 5-10ml of wort - which is going to give me an amount of cells. But I am putting a tiny amount of yeast into a large amount of wort... I think that the final number of cells in that 10ml.. is going to be reasonably close whether I scooped 50million or 500million cells on the loop. The you repeat the ironing out of the difference by doing another "large step" by pitching the 10ml into 250ml.

I strongly suspect that pitching into such high wort ratios is going to result in the yeast pretty much breeding to the resource limit - which will result in a fairly similar number of cells in the final wort, regardless of how many you started with.

The starters will perhaps take different times to ferment out... but should finish pretty close to each other in total cell numbers - not the same I know, but pretty close.

Practically - I notice no significant difference in the volume of slurry I collect stepping up a starter from loop - 10ml - 250ml - 2L, than I do if I start by adding 10ml of wort to the entire slant.

This is an interesting thought, if your first cell count is small enough relative to the starter size, you may end up with roughly the same amount of yeast after the first step regardless of how much you get on the loop. I've got an excel sheet that I made for a multi step calculator using the data from the mr. malty calculator, see link below. Keep in mind it won't spit out accurate numbers if the ratio is outside of 1-8mL / billion cells.

http://rapidshare.com/files/357468416/m ... p.xls.html