This should help you out.
Disclaimer: I do not claim that this is original work on my part. It was copied from a post on HBT made by user "Bob" - a most intelligent guy.
About Yeast, yeast slurry and pitching rates
The amount of yeast pitched is crucial to a quality ferment. Too little or too much, problems develop.
Overpitching prematurely saturates the wort with active cells, skipping the aerobic phase of the yeast life cycle. This significantly reduces ester formation, but more importantly results in few new cells; this causes unhealthy yeast and low viability for subsequent ferments. In other words, if you repitch that yeast, the next ferment(s) will be well short of optimal. Overpitching may also result in yeast bite, which flavor is particularly nasty.
Underpitching results in longer lag times, which present contamination possibilities; excessive ester production, due to an excessively long aerobic phase; underattenuation, due to simply not enough cells to do the work; and stuck fermentations. Underpitching is especially a danger because homebrewers tend to insufficiently aerate their wort; the yeast need to work twice as hard to simply multiply to saturation, and they will probably run out of oxygen before then. That means saturation is never reached, which means you rely on a relatively small amount of yeast to attenuate your sugars. Not good.
Instead of talking about underpitching and overpitching, let's cut to the chase and talk about the proper amount of yeast to pitch!
Ray Daniels, in Designing Great Beers, says that good brewing practice demands 10 million active cells per mL of wort for a wort of "normal" OG (say, 1.050). That approximates to 200 billion cells in five gallons. Jamil Z, on Mr Malty, gives the industry-standard rule of thumb: 1 million active cells per ml of wort per °Plato.* These both amount to the same number, practically speaking; have a look:
1 million x nml x °P = cells needed
1,000,000 x 20,000ml x 12°P = 240 billion cells
I know what you're saying - "Bob, you're a ding dong! Homebrewers don't work in °Plato!" To which I reply, "Zip it, sparky! You should be using °P, but you're a benighted heathen. Until you come to the light, convert °Plato to Specific Gravity by dividing by four. That'll get you close enough for this exercise, for beers below ~1.075 OG." Happy?
Now let's examine how these numbers apply to our breweries.
As I've written elsewhere, manufacturers of liquid brewing yeasts do not include sufficient active cells to meet the industry standard. But that's all right - commercial brewers have to build starters, too! Here's a breakdown of yeast types commonly available to homebrewers.
Dry: An 11g packet of dry yeast contains about the amount of required yeast - 20 billion cells per gram - so that doesn't need a starter; just rehydrate according to the instructions on the packet and pitch. If you're pitching into a big wort, dry yeast is inexpensive enough that you should purchase multiple packets. Dry yeast is nowhere near as susceptible to temperature extremes as liquid cultures, and can be stored longer before viability becomes an issue. Never make a starter with dry yeast; the manufacturer has engineered nutrients and reserves into the powder, and making a starter does nothing but deplete those reserves.
White Labs: A vial of White Labs yeast contains, optimally, 140 billion cells. Unfortunately, that number is no longer reliable once the yeast is packed for shipment. Once it's shipped to the homebrew store, it's been shipped once, potentially exposing it to temperature extremes. Check the manufacture date on the package carefully. Significantly, the numbers, even when the culture is new, still require a starter for proper pitching. Thus, it's wise to always make a starter for beers over about OG 1.037 when using White Labs yeasts, or buy two vials.
Wyeast: The "Activator" packs contain, according to Wyeast, "a minimum of 100 billion cells". What they really say is that each package contains 1.2 x 109 cells/ml. (That's 1.2 billion per milliliter, by the by). No matter how you slice it, even the Activator pack is insufficient to inoculate a wort stronger than about OG 1.026. The Propagator pack is even worse, with ~25 billion cells in a pristine pack.
By far the best method is to pitch fresh, healthy yeast from a previous fermentation. Now we get into yeast cakes, and I'm probably going to get under some brewers' skins. Let me struggle into my asbestos suit before continuing. [pause] There; got it.
Pitching onto yeast cakes is overpitching, and it's dangerous.
In the first place, it's infinitely more yeast than is necessary for healthy fermentation. See the above effects of overpitching. In the second, it's lazy. Harvest an appropriate amount of yeast, clean your fermentor, and stop cutting corners; you're simply not going to brew the best beer possible with that method.
Assuming the yeast in your cake is 80-90% viable, you can harvest a goodly portion of the slurry for repitching. Slurry at 85% viability may be assumed to contain ~1.5 billion active cells per ml. There are endless threads about yeast washing on HBT; use the "Search" function. But you don't really need to wash it - just use a sanitized implement to scoop out some yeast. Here's a tip, especially for you conical owners: Don't use the first or last yeast you get from your cone for repitching; use the stuff in the middle. The first layer is a mix of trub and yeast that flocculated prematurely; both useless to our cause. By the same token, the top layer consists of powdery yeast that settled last. Natural selection dictates that, if you pitch that yeast, subsequent ferments will take forever to clear. Ale Pail brewers can harvest yeast also, by carefully skimming the cake until fresh yeast is seen. No matter what sort of fermentor you've got, you're looking for tan or cream colored slime that smells like fresh yeast. Carboy users, you're screwed. I got nothin' for you, other than stirring with a sanitized implement and mixing it all up in order to dump it into another container. Maybe there's a technique for that here on HBT.
Given that we assume 80-90% viability and given a known OG, we can calculate - well, it's a SWAG, really** - how much slurry to pitch. The old commercial-brewery rule of thumb is one pound of slurry per barrel of wort. According to Daniels, that equates to a half-ounce by weight per gallon, or 2.5 ounces per five-gallon brew length - or - one fluid ounce per gallon. Our formula says that we need about 160ml of slurry for a OG 1048 brew; since 1 fluid oz = 30ml, we need - you guessed it - 5.33333 fluid oz of slurry! Interesting how that works out, isn't it?
It isn't hard to properly pitch a wort of any given strength or size. All you need is a little bit of information and a calculator.
Cheers!
Bob
* I took the liberty of simplifying JZ's 0.75 million ales and 1.5 million lagers and split the difference. As long as we're talking about rules of thumb, I find it a decent compromise.
