Sunday, September 26, 2010

Smashed

O HAI

funny pictures of cats with captions
see more Lolcats and funny pictures

A LOLcat is better than an excuse.

Kegged: 100801 Smash Mouth Volume 1


You may recall that last month's only batch was a SMaSH - Single Malt and Single Hop - that used domestic two-row and Cascade hops. I brewed it on August 13 and finally got around to kegging it today. I think the wait was worth it.

I had hit it with gelatin after two weeks, fully intending to keg it and have it around for Labor Day weekend. Well, I traveled the week before Labor Day so I didn't keg then. Instead, I just cranked the thermostat on the fermentation freezer to lower the temp to 38 F as a cold-crashing exercise, with the intention of kegging the following weekend.

I was hoping to get a really clean, really clear brew, and I think cold conditioning it for three weeks has actually done that. The taste of the gravity sample was really crisp, and I could definitely appreciate the flavor nuances that the Cascade brought. Plus, it's easily the clearest beer I have made to date.

The final gravity was 1.009, as planned, giving an ABV of 4.43%, a little higher than the estimated 4.16% due to the .002 higher OG.

I have this carbonating right now and am really looking forward to drinking and sharing it. Maybe this is the one that the casual drinkers will like. Or maybe not - doesn't matter, I like it.

Sunday, August 15, 2010

Hey now, you're an all star

I hate reading blog postings where the author makes excuses about the infrequency of his work and promises to do better. You know why? They sound like I wrote them. Instead of a pro forma mea culpa, today I'll follow up with the results of the most recent two kegged batches and describe a new recipe I made.

Kegged and tapped: 100701 Honey Half-Wit and 100702 Geordie-Boy

In the last post I talked about some newfound awareness on my part regarding how to hit the right mash temperature using my process and equipment. I have moved two batches through the process to consumption since then, and they are both pretty good. The Geordie-Boy is nowhere near as malty tasting as its recent forerunners have been, and the Half-Wit tastes as good as ever.

The Half-Wit was scheduled for a 154 F mash and it held that the whole time, but the OG came in several points short (1.036 measured vs. 1.041 predicted). I still don't know if this is due to the honey not being sufficiently dissolved at the time I take the sample, but the number makes the efficiency appear to be around 51%. The FG was 1.008 so the yeast at least took care of what little it could find. As I was racking this to the keg, it almost appeared that there was a layer of honey at the bottom of the carboy, but that might have been my imagination.

The Geordie-Boy is better than it has been. I had the mash adjusted down to 152 F and managed to start at 153. The OG into the boiler was 1.039, somewhat above the predicted 1.035, and the FG showed a corresponding overage of 1.044 vs the predicted 1.041. Since the resulting beer is a little hoppier than I perceive to be normal, I guess that overabundance of sugar was offset by the change to the hop schedule I had to adopt. During the brew day, I realized that I didn't have the right bittering hop and I had to make an adjustment on the fly, using Cascade instead of Target. I tried to adjust the IBU by adding more Cascade (at its 7% alpha acid rating) to take the place of the 11% AA Target. It worked out mathematically, but it might not have worked in the wort. It's not overbearingly hoppy, just more than I expected.

I ran the Geordie-Boy through the filter in an attempt to clean it up a little as well. I probably would have been better off to have used gelatin in it, because the amount of yeast residue that still showed up in the first few pints seems to indicate that the filter didn't do a whole lot of cleaning. Still, it's drinkable.

Didn't The Offspring have an album about this?

For the first batch of August, I wanted to try something new. Looking back on what I've made to date, one thing stands out: the recipes all have a fairly complicated grain bill, and none of them would pass muster under the German Reinheitsgebot beer purity law because of all the adjuncts I have used. (Fortunately the law doesn't seem to apply in Alabama outside of a few mile radius of the Mercedes plant in Vance.) With the recent process issues, I felt like I needed to get (back) to basics and see about creating a beer that was simple enough to help me troubleshoot further.

I have read in various places (including homebrewtalk.com) about a concept called SMaSH - Single Malt and Single Hop. The premise behind SMaSH is that if you really want to know how a particular grain or hop affects the flavor of your beer, you have to isolate them so you know what they're bringing to the recipe. Many people on HBT say that the SMaSH recipes they have made are among their best tasting beers.

For this batch I decided to build a SMaSH directly within BeerSmith. To keep it simple, I opted to use domestic 2-row for my base malt, even though many people have said it's almost too mild for this purpose and the resulting beers are very light tasting. I kind of see that as a virtue compared to what I've been making, and it might get some of the fence-sitters to try one of my beers. For the hops I chose Cascade, mostly because I had some.

When modeling the recipe I wanted to aim for a balanced beer that was neither malty nor hoppy. Not being a real beer scientist, I turned to an expert's advice to see how to do this. Homebrewtalk.com's Biermuncher posts an interesting chart with each of his recipes showing the relationship between OG and IBU (the international bittering unit, a measure of hop content), annotated with sensory bands indicating the degree of maltiness or hoppiness a given OG/IBU ratio has. For an example, look at the chart in the thread for the brown ale recipe that was the original basis for Geordie-Boy. I can't tell where the chart originally came from to provide proper attribution, but I found a copy of the chart hosted at BrewSupplies.com:

I wanted the recipe to be balanced, neither malty nor hoppy. I didn't want it to have a real high ABV, but preferably somewhere in the 4% range. I also figured I wanted to stick with a simple hop addition schedule, so I started out by planning to add equal quantities at 60 minutes for bittering and at 15 minutes for flavoring.

My first action was to add enough 2-row to the recipe to raise the OG to the point where the ABV would be in the right area assuming 70% efficiency. I settled on 8 1/2 pounds, for a predicted OG of 1.041 and an ABV of 4.16%. Next, I began adjusting the hop additions so that the resulting IBU prediction would be around 21-22, a number I selected from the chart as being in the center of the "evenly balanced" range. After several iterations I settled on 0.7 oz of Cascade at 60 minutes and 0.6 oz at 15 minutes, which leads to a predicted IBU of 20.7. I decided to use Safale US-05 yeast, as it's a pretty clean yeast and should allow the flavors of the ingredients to dominate the taste of the finished product. The most important decision was what to call this recipe, of course. I settled on Smash Mouth Volume 1, since I doubt it will be the only SMaSH I make and it will save me from having to think up clever names for Volumes 2, 3, and so forth.

Brewed: 100801 Smash Mouth Volume 1

I got the 2-row already crushed and measured out from Alabrew, and got a pleasant surprise in that it was only 99 cents a pound. With the yeast and the hops, this whole batch came in at under $15.00, making it by far the least expensive one I've made. Chalk that up as another potential plus for this idiom.

I picked 150 F as the mash temperature and almost hit it, coming in at 151 F instead, still well within the beta amylase range. The SG into the boiler was 1.038, which gave me an actual efficiency of 76% into the boiler, and with an OG of 1.043 I got 73% efficiency overall. The wort was exceptionally clear coming out of the mash tun, and it maintained its clarity through the boil (except for the hot break proteins floating around, of course). I tried to whirlpool it after it cooled and transferred it to the carboy with a siphon, and I succeeded in leaving a lot more trub than usual in the kettle.

Even with the slightly higher than planned OG, I'm "in the zone" for an evenly balanced beer. The Cascade hops smelled great when I added them. I am really interested to see how this turns out.

Sunday, August 1, 2010

Wish I could remember the awesome title I had for this post

I guess it's a sign of the times. I was getting in the car to go run (and no, the irony of that is not lost on me) and I had a great idea for the title for a blog entry for Fork and Hay. Now I can't recall what it was. I suppose I'll just have to wing it.

Today we're going to talk about basics. I've written before about a strange flavor change I have detected in my batches of late. I don't know how to classify the taste, but whatever it is it is wrong for the styles I have been making. As I was pondering the latest unusual situation that resulted in an unexpected flavor component in a batch, some pieces of information I had encountered in my seemingly endless web browsing over the last couple of years started to congeal into a possible explanation.

At first, I was all set to blame my 10-gallon batch setup for the problem, because all the faulty brews had been 10 gallon batches. I have noted issues with getting the correct mash temperature before, which has prompted me to pursue designing a RIMS system (which I will hopefully finish this fall). My initial thoughts were that I either had trouble measuring the temperature correctly in the first place, or that the mash tun was doing a poor job of holding the temperature required.

In an attempt to eliminate variables, I made the last four batches (including two I'm going to talk about later which are still fermenting) using the five gallon setup. Two of these were Honey Half-Wit recipes, and I have already chronicled the success of the one that's been kegged and consumed. The other finished batch deviated from expectations, and it was the manner of this deviation that finally helped the light bulb come on, because it told me the problem was not limited to the 10 gallon equipment.

100703 Hook Me Up finished high, a full .007 above the predicted level, and it had a sweeter, thicker taste than I expected it to have. Recalling that the Por Favor and Geordie-Boy batches that weren't successful also tasted thicker and a touch sweeter, I noted that a couple of them had higher than expected finishes as well. As I read more, I learned that what I've been characterizing as "sweet" is really "malty," which is a way of stating that there were a lot non-maltose sugars in the wort, sugars that the yeast could not convert to alcohol. The presence of unfermented sugar in the finished beer helps explain the sweet, thick taste, and it also explains the high finish gravities (because sugar that the yeast can't eat continues to affect specific gravity). What could cause this excess maltiness as a shared characteristic of batches that span yeasts, grain bills, and equipment?

Sacch-re bleu!


There are numerous sources around that describe how mashing works, including a highly informative wiki entry at HomebrewTalk.com and this chapter of John Palmer's outstanding book "How To Brew." The essence of the process, as far as the maltiness problem is concerned, is the relative activity of two key enzymes during "saccharification" (the period during the mash where starch converts to sugar): beta amylase and alpha amylase. Beta amylase is the most effective agent in the conversion of mash starches to maltose. Alpha amylase converts some starch to maltose, but it also produces other sugars from starch as well.

Palmer's discussion of saccharification is particularly understandable and suggests the cause of my problem: if  my worts have excess non-fermentable sugar, there is too much alpha amylase action going on and not enough beta amylase activity. The two enzymes work at different temperature ranges, with beta amylase ramping down at about 152 F and alpha amylase picking up at 154 F. The conclusion, therefore, is that if there's too much non-fermentable sugar in the wort, the mash temperature was higher than it should be, because it left the saccharification mostly in the hands of alpha amylase.

So why am I mashing too high? In every batch so far, I have hit the strike water temperatures required by BeerSmith, and I am pretty sure I can eliminate problems with the software as a factor. To investigate further, I opened up BeerSmith to see what components influence the way the equation is solved, but before I describe what I found maybe I should backtrack a little and explain exactly what BeerSmith is doing for me.

Warning: physics content

The basic problem is simple in concept: heat a quantity of water and grain to a specific temperature. If you could do it in a way as simple as it sounds, for instance adding the grain to the water and then heating the result, everything would be ready to go as soon as the whole mash volume hit the temperature target. Doing it that way, you don't have to worry about how much water you're using and how much grain is involved - as long as the mash is fluid enough to allow you to stir it, it will all get to the right temperature at the same time. Of course the problem isn't that simple. I don't have a powerful enough agitator to be able to roll over a suspension of 8-plus pounds of grain in 3 or so gallons of water so that it doesn't scorch.

It's far more practical to heat the water separately and then add the grain into it, stirring to make sure all the grain is well saturated and in suspension. However, if you've ever cooked pasta  (or grits, but of course never instant) you know what happens when you add pasta to boiling water: the boil stops. That's because adding the room-temperature solids to the hot water lowered the overall temperature of the suspension as the pasta took heat out of the water until the whole mass reached thermal equilibrium.

What we're doing in the mash tun is essentially the same thing, but with a slightly different approach. With a single-infusion mash (one where there's only one charge of strike water), we need to know what temperature the water has to be in order to source enough heat to bring the whole volume into equilibrium right at our desired mash temperature. Mathematically, it's simple - all you need to know is the volume of the water you're using, the mass of the grain you're adding, and the grain's starting temperature so you'll know how much heat it's already harboring. (You'd think you could do the same thing with pasta, but there's a catch: you can't heat water to over 212 F because the result isn't water, it's steam, and it's not helping cook your penne to al dente when it rises out of the pot.)

That's pretty basic high-school level physics and chemistry, and as I said before it seemed unlikely to me that BeerSmith could screw the calculations up. As it turns out, BeerSmith has a far more sophisticated model in place that factors in several variables, including:

  • The material used in the mash tun and its weight - because the mash tun itself is not a perfect insulator and will suck heat from the mash both initially and over time. Different materials will have different specific heat values, where specific heat is a measure of how much energy is required to raise the temperature of a gram of the material 1 C. Knowing the material gives you a reasonable ballpark on its specific heat; knowing its mass lets you then calculate how much heat it's going to need as it joins the mash in reaching thermal equilibrium.
  • The starting temperature of the mash tun.
  • The volume of water to be used in the mash (which is internally converted to mass for another specific heat calculation).
  • The weight of the grain bill to be used (but not its specific heat: BeerSmith uses a constant).
  • The starting temperature of the grain.
When you create a batch sheet in BeerSmith it automatically pulls in the grain bill weight from the recipe and the strike water volume from the recipe's mash profile. (It can actually do a lot more than that, but I only use single infusion mashing so there are no other steps.) The mash tun material and weight are defined in the equipment profile for your recipe.

The starting temperature of the mash tun and the starting temperature of the grain are given default values of 72 F by the software.

Now we're getting somewhere...I think

My problem batches were created over a fairly long span of time (from November through May), and the grain storage area in the garage varied from being in the 40's to the 70's during that time. That means that the grain and mash tun temperatures varied that much as well, and you know what? I never changed the values in the software to reflect the current conditions.

Aha! I thought. Let me adjust the grain temperature of a Geordie-Boy batch and see what effect that has on the strike water temperature BeerSmith calculates to hit the target 154 F:

  • 50 F grain: 173 F water
  • 72 F grain: 169.8 F water
These numbers seem to indicate that I should have had lower mash temperatures when the grain was colder than the calculation allowed for, because I used 169 F water instead of the 173 F that was required. Clearly that's not the case, because lower mashing would favor beta amylase activity (as long is it wasn't too low, like below 131 F) and the production of maltose instead of unfermentable sugars. When you add in the fact that I didn't adjust the mash tun temperature either, the "too cold" disparity becomes even greater, with the strike water requirement moving up to 174.1 F when the grain and tun were at 50 F.

The analysis left me puzzled for a while, and then it hit me: it was about the time of the first weird batch that I started heating the strike water directly in the mash tun with my heat sticks. I first did that all the way back in November with 091103 Por Favor, and I've continued that way ever since. You've doubtless spotted the issue by now, and it goes all the way back to the beginning of the physics lesson above. If you heat the water in the mash tun, the mash tun and water are the same temperature all the way along.

Rather than starting with a 50 F or 72 F mash tun, I was starting with a tun at the strike water temperature of 169 F. There's no doubt that having all that extra heat available influenced the mash temperature upward, which could explain the apparent preference for alpha amylase, but what was the real effect? Could it really have been enough to push the mash clear out of the beta amylase range and lead to an excess of unfermentable sugar?

Going back to BeerSmith's calculator, I started doing a little iterative solving, holding the grain temperature constant while changing the mash tun temperature until it and the calculated strike water temperature converged. What I found was startling. The strike water temperature I had been using was up to 5 F too hot, because the number I converged on was around 164 F.

To figure out the effect on the mash temperature, I did a little more iterative solving by changing the mash temperature desired and the tun temperature until the strike water and tun temperatures were both at 169 F. The calculated mash temperature under those conditions was 157.5 F - pretty dramatically above the top of the beta amylase range. No wonder those beers were too malty!

The moral of the story

If anything, this exercise has taught me the value of paying attention to details (again, but if you're a regular reader here you know I'm a slow learner). By not fully understanding what all the factors were that went into the strike water temperature calculation, I wasn't able to adjust for a process change that affected its computational basis. This translated into a probable error of up to 4 F in the mash temperature of several batches, which I now know results in an increase in the ratio of unfermentables to maltose in the wort. Unplanned non-fermentable sugars lead to higher than expected finishing gravities and a maltier taste which, if not properly balanced, might not be desirable.

In short, if you don't know what you're doing, you'll screw up. And that truly is a lesson I keep learning.

Next time (soon, I promise) we'll meet the first post-revelation batch of Geordie-Boy and check in on another batch of Honey Half-Wit.


Monday, July 5, 2010

The biggest holiday disaster since the "Star Wars Holiday Special"

There were so many things that went wrong this weekend I don't even know where to start. I guess I should begin with the one thing that actually seemed to work.

Filtered and kegged: 100503 Hook Me Up


This batch had been in fermentation for about 5 weeks before I finally got to a place where I could keg it. I had already decided that this would be the test case for my beer filtration setup. I had shopped around and listened to some podcasts on The Brewing Network to finalize the design for a keg-to-keg filtering system, and what I finally decided on was an adaptation of the same water filter I described last October.

The filter idea is pretty straightforward. I took another Omni U25 whole house water filter and traded out the 3/4" NPT connections for a pair of reducing bushings and 1/4" barb fittings. I then attached about a foot or so of beverage tubing to the barb with a hose clamp, and at the other ends I added swivel flare fittings. To those I attached a pair of liquid ball-lock quick disconnects.

Once the assembly was completed, there were some setup tasks for the filter. First, I took an already clean keg and put about a gallon of water into it. I then used my compressor and a gas ball-lock QD I already had, and pressurized the water keg. I then attached the filter to the "out" posts of the water keg and a separate corny. This allowed the water to flow from the water keg through the filter. I did this part of the setup with plain water because the filter cartridge was new and I wanted to get any activated charcoal dust out of it before I started running sanitizer or beer. After I had flushed all the water through, I went back to the water keg (bleeding the pressure first of course) and added about a gallon of Star-San solution. I then pressurized the keg with CO2 and bled the air out of the headspace, so that there was mostly only CO2 remaining. I hooked it back up to the filter and ran Star-San through the assembly for about three minutes. When I had gotten all of the sanitizer out of the keg, the filter emptied (mostly) and the remaining space was filled with CO2. At this point I disconnected everything - the filter was ready to go.

In operation, the process went as follows. I racked the Hook Me Up batch to a clean, sanitized corny keg in the usual fashion, taking a gravity sample along the way (more on this later). Once the keg was filled, I sealed it and bled the air out by filling the headspace with CO2. I then prepared another clean, sanitized corny and purged its interior with CO2. I connected the "in" side of the filter to the liquid post of the full keg, the "out" side to the liquid post of the empty keg, and pressurized the full keg to about 4 PSI (to the limit of my ability to read the gauge on the regulator).

Slowly but surely, beer started filling the filter assembly and moving into the empty keg. I kept the lid cracked open on the destination keg so that the CO2 inside would escape as it was displaced with beer from the bottom of the dip tube. I didn't keep a close eye on the clock, but I would guess it took between 20 and 30 minutes to push the beer through the filter and fill the destination keg. At the point where it started pushing foam through the line I turned the filter off, even though there was about a pint or so of beer left in the filter. It appears that that's the dead loss for the device. I sealed the keg and put some CO2 in to set the lid.

Cleaning up was fairly simple. Basically I backflushed the filter with sanitizer from a keg that I pressurized with air. I guess it's a reflection of how much sediment had already dropped out before I racked, but it didn't look like there was much material in the effluent from the backflushing. (There was some, so the filtering wasn't a totally wasted effort, but it wasn't like the effluent from, say, the pool filter, which is probably a good thing after all.) I opened the filter up and drained the dead-loss sanitizer but left everything wet, then sealed it up. I don't think there will be a mold issue but only time will tell. I'm wondering if I should have left it full of Star-San instead. I guess I could just go fill it up.

And now for the bad news

The 100503 Hook Me Up finished at 1.020. That's really high compared to what was predicted, which was 1.013. I don't have any idea why after four plus weeks it didn't attenuate any more than that. The flavor is OK, but the finish is a touch thick and a little sweet. I was expecting it to be drier.  That wasn't the worst news for this batch however. In my rush to get it ready for the Fourth of July, I managed to overcarb it pretty badly. I think I have it bled down now, but I really can't tell. Why I can't tell will become evident soon.

So I have a lot of family around the house on the third, just hanging out and enjoying a strangely cool day for this time of year. The Honey Half-Wit, as expected, has proven to be the crowd favorite, and at some point during the day I am dispatched downstairs to draw a few pints for the card-playing crowd. I crack open the tap and...nothing comes out. Not foam, not CO2, nothing.

I quickly check some of the other taps and beer flows from them, so I know there's CO2 in the system. I change out the beer line from the tap to another keg and beer flows, so I know there's no blockage in the tap or line. Not knowing what else to do, I bleed pressure from the keg and open the lid, at which point I notice that the beer is frozen. WTF? A further examination shows that the bottled beer filling in the spaces in the floor of the keezer has suffered the same fate. I check the setpoint on the temperature controller, and it shows 34 F, same as usual, with the current reading being 31 F. Since the setpoint hasn't changed in months I don't know why the keezer picked Saturday to decide to freeze beer, but it did. I pulled the keg out and set it in a bucket of water to thaw, and in about an hour it had re-liquefied and I was able to return it to the keezer (after raising the setpoint several degrees).

Thinking all was well, I went about my business and we had a reasonably good remainder of the day, capped off by a trip to the local Double-A team's game and fireworks show. I discovered that they were serving Gordon Biersch brands at one of the beer stands and had about half a Maibock, but was underwhelmed.

Independence Day dawned and we were caught up in the whirlwind of finishing all the entries for the dessert contest at the annual Indian Springs Village Fourth of July barbecue. Returning home, everything seemed OK and the day wore on well, until I was again dispatched to draw a pint or two. Once again, I opened the tap, and once again, I got nothing. Fearing the worst, I went straight to the "beer is frozen" diagnostic and started to bleed the keg preparatory to thawing it again. However, it turned out there was nothing to bleed. A quick glance at the CO2 regulator told me the tale: the gas was out. How in the hell did that happen? I just exchanged that cylinder!

Needless to say, I wasn't very popular when I went back upstairs with the news.

Sunday, June 6, 2010

Zestfully clean

After brewing 20 gallons over the last two weekends, I didn't plan much for this week. About all I accomplished was the kegging of 15 of those 20 gallons and a little cleanup.

Kegged: 100501 Geordie-Boy Ale


Interestingly, the gravity samples of the two carboys tasted different even though they both used the same yeast. It could just be that the first sample taste was my first beer of the day, and by the time I tasted the second sample my taste buds were already aligned. We'll see when it comes to serving time. Both of them fit within my expectations for the house brown ale.

The second sample came from the first carboy to take off in fermentation. Its FG was 1.014, for an ABV of  4.43%. The other sample, with the delayed fermentation, came in at 1.013 for an ABV of  4.56%. Isn't it interesting how a single gravity point moves that number so much. I'm going to have to look into the math for that calculation.

Kegged: 100502 Honey Half-Wit


I was a little apprehensive about kegging this one because it looked like there was still a foamy surface on the beer in the carboy, but as I pulled it up to rack it, I noticed that the foam was mostly CO2 bubbles that had aggregated against the floating orange zest. The sample had a pretty good flavor to it, and I definitely tasted a citrusy overtone from the orange. I think this one will be a winner. As it happened, this is the first batch I have had that finished below 1.010, coming in at 1.009 for an ABV of 4.03%.

Good thing I took three semesters of calculus in college


Going back to the ABV math for a second: The Geordie-Boys started at 1.048 and attenuated 34 and 35 points, respectively, to finish at 1.014 and 1.013, giving them ABVs of 4.43% and 4.56%. The Honey Half-Wit started at 1.040 and attenuated 31 points to 1.009 for an ABV of 4.03%. If one point of attenuation is worth .13% ABV, as the Geordie-Boys indicate, then linear interpolation would predict the Honey Half-Wit to be 4 x .13% = .52% lower than the 4.56% of the 1.013 Geordie-Boy, or 4.04%. Pretty close. I will have to dig around in some references to see if that's how it really works or if there's something else involved.

Monday, May 31, 2010

Resisting the Urge for a Headline Pun

It's Monday, the day we observe Memorial Day in the US in honor of those who have served the cause of freedom and given all. I brewed today, exercising one of the rights that the fallen have preserved for me. It seems fitting, especially when it's time to start filling the supply pipeline for Independence Day.

Brewed: 100503 Hook Me Up

Another commissioned batch, Hook Me Up is intended to be an English extra-special bitters based on the commercial beer Red Hook ESB. I started with a base recipe from HBT's Biermuncher and manipulated it to match what I can do and what I have to work with. The original recipe called for Wyeast 1098 but I'm using Nottingham, and I made a substitution on the hop schedule, putting in Saaz instead of Tettnang.

This recipe had a new wrinkle for me - it called for some "toasted malt." Toasted malt is just regular two-row malt that is heated at 350 F for 12 to 15 minutes. It darkens the malt (from 2 SRM to 22 SRM according to BeerSmith), and it's supposed to impart a bit of flavor as well.

I spread the malt onto cookie sheets in as close to a single layer as I could manage, then popped the cookie sheets into a preheated, 350 F oven. The smell of the malt as it toasted was quite pleasant, and surprisingly strong. In fact, at one point during the first toasting go-round, I panicked because the scent was so powerful I thought I was burning it.

Here's a picture of some before and after - the malt on the trays is toasted, while the malt in the round pan is not. It's hard to see the color difference in this picture, but it was more obvious in person. I'm not sure it was 20 SRM different, though.


Other than that, and the four separate hop additions, this was a pretty standard brew session. I did it all-electric again, relying mostly on my 2000W heatstick to provide the energy. I heated the strike water in the mash tun so it was adequately pre-heated. I didn't obsess over the mash temperature this time, in fact I didn't even measure it after I hit the initial 158 F I was aiming for. With an OG of 1.052 it hit the software prediction so that's a plus given my recent problems in that regard.

RIP 100401 Honey Half-Wit


As predicted, this one went fast. It got the Brewmistress Seal of Approval, which is a first for Fork and Hay. (Could the Five Diamond award be next?) Luckily another 5 gallons is already in the pipeline, and I may go ahead and lay in a third batch before July 4th comes.

Thursday, May 27, 2010

Quick notes

Just a couple of things:
  • In an attempt to cut down on the Chinese 'blog spam' that I get in comments to this page, I have turned on comment moderation. If you post a comment I will have to approve it before it shows up. 
  • All three carboys from last weekend are fermenting away in the freezer. One of the Geordie-Boy carboys was slow to take off, which I attribute to it being in actual contact with the side of the freezer, causing it to be too cool for the ale yeast. I moved it away and the next day the krausen was in full form.
  • The next 'commissioned batch' will be a Red Hook ESB clone - "Hook Me Up" - based on another well-regarded recipe by Biermuncher at HBT.

Sunday, May 23, 2010

Brew morning, brew day

It's a mark of how old I am getting that I'm reduced to making bad headline puns based on song lyrics from bands I don't even like. At least it's from one of their least objectionable albums. The headline refers to this weekend's brew sessions - yes, that's a plural - wherein I laid down 15 gallons in an attempt to refill the pipeline after a long brewing drought.

Brewed: 100501 Geordie-Boy Ale

First up was a 10 gallon batch of the house favorite. I've been disappointed with the last couple of batches of this beer, so I decided to go mostly back to basics. I went back to plain tap water (unfiltered), I didn't use the hop sack, and I cut way down on the agitation of the immersion chiller.

I used the rectangular cooler mash tun, because it was the only one I have that is big enough for the grain bill for this recipe. As before, I heated the strike water in the tun, and after it got to about 140 F I added the grits to get them hydrated before the mash got started in earnest. That allowed me to keep them stirred up and loose instead of having them get all clumpy. I also stirred a lot when adding in the rest of the grain to make sure that there were no obvious dough balls or layers. I got a respectable 1.048 OG out of it, for an efficiency of 79% or so, so it must have been OK. The temperature held just fine as well, so I think the lesson to be had is don't use the big tun for a 5 gallon batch - but with a 10 gallon batch and sufficient preconditioning, it's fine.

When cooling the wort, I realized it was summertime again and my tap water wasn't as cold as it had been. I managed to get the volume down to about 78 F in 35 minutes, at which point I went ahead and put it into the carboys and put them in the cooler to ramp the rest of the way down overnight, with just some aluminum foil to seal the opening. The next morning, I added Nottingham yeast to each carboy and put on airlocks.

Brewed: 100502 Honey Half-Wit


The Honey Half-Wit is proving to be popular with more casual drinkers, so in anticipation of summer visitors I went ahead and put down another 5 gallon batch. I actually considered doing 10 gallons, but as it turns out, getting 4 carboys into the cooler is a bit of a challenge and I didn't want to mess with it. I had logistics challenges enough with this batch without the aggravation.

Alabrew moving closer to the Fork and Hay worldwide headquarters proved to be as useful as I expected it to be. When I inventoried raw materials I found I was short the yeast and all the malted wheat required for Honey Half-Wit, so I sent the Head Apprentice over to Alabrew to pick it up. I learned an important lesson - write out specific product names. It took me going over there to help straighten things out before I had 4 pounds of crushed wheat malt. On the shopping list I had written "4# wheat malt crushed." Turns out the Head Apprentice asked for "4 pounds of wheat malt," which was reasonably interpreted as a request for wheat malt extract. Unfortunately there were two problems with that: first, 4 pounds of malt extract is a lot more expensive than 4 pounds of crushed grain; and second, I probably won't use 4 pounds of wheat malt extract in the entire future of Fork and Hay. I still have an unopened one pound bag of DME I bought a year ago. In the end, Kim worked it out and I got the grain I needed to brew, along with some WLP400 Belgian Wit yeast. I have been using the Wyeast smack pack equivalent (3944) but Alabrew carries White Labs, and this table at Jamil Zainasheff's website says they're the same thing, both being derived from Hoegaarden.

The original Honey Half-Wit batch didn't have much sweet orange character after the boil, so I decided to add an additional ounce of orange zest into the carboy to see if it added some flavor tones. I also dumped the honey in before the wort in hopes that they would mix and I would get a more accurate OG reading. Either the honey didn't mix correctly or my recipe calculations are off dramatically, because I still came in at .010 lower than the prediction of 1.050. The first batch turned out OK even so, so I'm not all that worried about it - just a little puzzled.

Edit: It turns out that part of the issue was that I had the BeerSmith template set to expect 75% brewhouse efficiency. If I ramp it back to 65% the prediction becomes 1.044 which is a lot more reasonable. I calculated 58% actual efficiency for this batch so I need to figure out where I'm losing my sugar.

The first batch fermented so vigorously it blew out through the airlock, so I decided to go with a blow-off tube this time. It looks like three carboys and a blow-off bucket are about all that I can get in the fermentation cooler at a time, so now I know what the real production bottleneck is. I suppose I could bring carboys inside and do the T-shirt trick to keep them cool, so there is a workaround available if I need the throughput. I can only serve 5 at a time anyway, three on real taps and two on picnic taps, so there's an upper limit on demand for ready product.

Sunday, May 9, 2010

Nose to the Grindstone

Yesterday was a big day around Fork and Hay. The eldest apprentice brewmaster graduated from college with a BS in biology and we had quite a celebration in the hospitality area of the brewery (the portion that everyone else refers to as "the house"). The apprentice will be continuing his biology studies in pursuit of a PhD, but in a career-limiting move for a brewer he is concentrating on algal phycology instead of the mycology of Saccharomyces cerevisiae. He will be spending a lot of time in the lab, though, and I hope he'll be able to culture out some commercial yeast strains for the brewery's use.

Kegged: 100401 Honey Half-Wit

Just prior to heading down to the apprentice's commencement ceremonies, I kegged the Honey Half-Wit batch I made two weeks ago. I had cold-crashed it in the carboy overnight, so the sampling and kegging took place at about 37 F. The temperature-adjusted FG for this batch came in at 1.011, which seems like a reasonable finish, but given the uncertainty of the OG reading I can't figure the exact ABV. The beer is a nice bright color and has the characteristic cloudiness of the style. The Brewmistress tasted a portion of the SG sample and gave her conditional approval (while noting that it was flat and not cold enough).

I started to carb it but I didn't quite get enough CO2 dissolved before we had to depart for the ceremony. I'll zap it again today and bring it up close to 2.4 volumes, because it's supposed to be a reasonably fizzy beer. The Brewmistress likes fizz. However, I'm going to have a bit of a serving dilemma, because I only have three taps - yes, taps - from which to serve.

Grinding teeth

I am edging ever closer to finishing the keezer after only 10 months. The three taps are installed, tested leak-free, and functioning. About all it lacks now are casters for mobility and a drip tray.

Getting the taps installed and leak-free was a challenge. As I mentioned before, I got a great deal on some Perlick 525SS stainless steel faucets. Faucets, of course, are only part of the entire tap system, and they are attached to the all-important beer line with a shank and a tail piece.

Instead of getting stainless steel shanks, I opted to save a few bucks and go with chrome plated brass ones. After all, nobody can see the shank (unlike the faucet), and while stainless is more durable and less subject to degradation through abrasion it's not like there will be a lot of abrasion in this application. I picked tail pieces with flare fittings, so they can be disconnected from the beer line for cleaning without disassembly.

In the last post I mentioned that I found some issues with this arrangement due to what appeared to be manufacturing tolerance variations among the shanks, and between the spec conformance of the shanks as compared to the faucets. Some real-life pictures would help to explain the issues, but I don't have any, so you'll have to use your imagination and look at this diagram that I shamelessly lifted from chicompany.net.



The faucet (1) attaches to the end of the shank, where its threads are engaged by the collar (2). The faucet has an internal o-ring that is supposed to press up against the end of the shank as you tighten the collar, forming a seal.  To keep the faucet from rotating on the shank, there are some teeth (like a gear) formed on the inside of the faucet and on the outside of the shank.

It was the teeth that were creating my leak issues. I was only able to find one shank-faucet combo where the teeth mated well enough that I could get the faucet pushed onto the shank far enough to allow the seal to be formed correctly. On every other set, I could only get the faucet onto the shank a little, and no amount of tightening of the collar would drive it down to set the o-ring.

I used (and wore out) two steel brush attachments for my Dremel trying to clean up the shank teeth enough to allow the faucet to fit correctly. Eventually, I switched to a grinding stone and simply wore the teeth down to a degree that I could get the faucet connected. Once I did that I was able to get a solid seal that's leak free, and now all three taps are hooked up and freely flowing.

The problem, of course, is that I now have four active kegs. Time for a bigger keezer? Edit: it may be time for a bigger keezer but I solved the immediate problem by dumping the last gallon of 090901 Gayle Bait. It just wasn't doing it for me any more. It's nice to have Honey Half-Wit on tap, and with the right carbonation and an orange slice, it really hits the spot!

Tuesday, April 27, 2010

Who's the half wit now?

Has it really been two months since I brewed a batch? It sure seemed like it when I went out to the garage Saturday afternoon and tried to remember what to do. At least it wasn't cold.

Brewed: 100401 Honey Half-Wit

When last I tried to make a honey-bearing beer I ended up with the disaster known as 090502 Honey-Brew List. It wasn't the honey's fault, it was my fault for over-adding the coriander and creating a spicy mess. Unfortunately I wasted a pound of good homemade honey on that batch and got nothing but a smelly bucket and a disappointed Brewmistress in return.

Not long after I created that lousy wheat beer, I brewed a different wheat beer batch, 090603 Half-Wit, which was very highly regarded among the members of the focus panel that convened in the garage on the Fourth of July. That recipe had no amendments, and it ended up with a nice clean taste. It clearly is among the top recipes I have brewed to date. (I'm not listening to you out there who are keeping score and saying "you've only brewed six recipes to date, that's not much of a distinction.")

It seemed logical to me that I could meet the Brewmistress' demand for a honey wheat by starting with a known good base recipe (Half-Wit) and adding honey and fruit amendments - and thus was born Honey Half-Wit. The modifications I made to the basic Half-Wit recipe were minor. I traded out (in a five gallon batch formulation) 1# of base malt in the mash for 1# of honey, which I added directly to the primary instead of to the boil. The reason for the late addition is that knowledgeable people have written that the risk of introducing contaminants or infections from unpasteurized honey is minimal, while introducing the honey in the boil kills off the components that provide the honey character. To the boil I added about an ounce of orange zest (sweet orange, not bitter orange), which smelled good at the time but I wonder if it will show up in the final product.

An electrifying experience

I decided to try and make this batch using only electric power. With a 2000W heatstick and a 1500W heatstick I figured I had enough power to manage a 5 gallon batch, and this turned out to be the case. I used a 1500W stick to heat the strike water and pre-condition the mash tun. Unfortunately, I also managed to melt a hole in the side of the cooler when I let the tip of the stick rise above the water level for a second. I plugged it with a high-tech solution (metallic tape) and made a note to be more careful.

I am still concerned about the ability of this tun to hold mash temperature. This mash was scheduled for 154F and it actually started just a tad higher, but an hour later it looked like I had lost 8 degrees. It might just be that the rectangular cooler has too much headspace for a small batch, but I have noticed comparable drops even with the 10 gallon Geordie-Boy and Por Favor batches. I either need to switch coolers (again) or speed up my plans to build the new recirculating mash rig.

I used the 1500W and 2000W sticks to run the boil and was able to get a gigantic rolling boil pretty quickly. The 2000W stick was sufficient to maintain the boil by itself, with a pretty reasonable wort turnover. When I put the immersion chiller in to sterilize I added the 1500W stick back in to compensate for the immediate cooling effect, then removed it once temperature stabilized.

When I moved the cool wort to the carboy, I simply opened the honey, pulled off the safety seal, and dumped it in. When I took the specific gravity it was much lower than predicted, coming in at 1.039 instead of the 1.052 that BeerSmith suggested, but at least part of that has to be the fact that none of the sugar in the honey was in solution at that time. However, even when I back the honey out of the BeerSmith model I still show about a 6 point deficit from the prediction. Clearly I'm not doing something right, and again I look to the cooler as a potential issue with efficiency. I guess if it tastes good though I'm not going to complain.

I'm tapping that

Get your mind out of the gutter, I'm referring to my serving system. I have started installing the new Perlick stainless steel forward-sealing faucets I got last month with mixed success. The first installation I did leaked like a sieve through the ring that fastens the faucet to the shank. The shanks are garden-variety chrome plated brass, not stainless, and I recall reading that there can be manufacturing tolerance issues with mating the stainless faucets with the plated shanks. With a little experimentation, I was able to find one faucet-shank pairing that would work correctly, and I now have one functioning tap on the keezer. For the other two I need to try and tidy up the end of the shank to get a better fit for the faucet. I think a little wire brush work may be in order.

Monday, April 12, 2010

No news is...no news

There's no useful news from the Fork and Hay production area. I have the materials all set to make a new batch of Geordie-Boy and a fresh run of Honey Half-Wit, but this weekend was consumed by the rebuilding of a retaining wall in the back yard. Maybe tonight I'll get to the Honey Half-Wit so it will be ready for the big graduation to-do in early May. The brown will just have to wait, but there's about half a keg left for now and I'm in training for a 5K so I have dialed back consumption estimates dramatically.

I have finally accumulated all the parts for the faucet assembly for my dispensing system, which I have been referring to as a 'kegerator' but in point of fact is a 'keezer' because it uses a freezer and not a refrigerator. I just don't like the word 'keezer' as it is uncomfortably close to 'geezer' and I don't need the reminder. I have some pictures that I'll post when I finish the assembly and draw the first ceremonial pint.

Outside the four walls of Fork and Hay there is mixed news. On the legislative front, Senate Bill 253, which would have formally legalized homebrewing in Alabama, died in committee because of politics having to do with petty, trivial issues - pissing matches over electronic bingo, some kind of ruckus around funding the pre-paid college tuition program, and (can you believe it?) school funding. It's too bad our legislators can't get their priorities straight.

On a more exciting note, Alabrew, our local homebrew shop, is moving to a location just minutes away from the world headquarters of Fork and Hay. I view this as being similar to all the auto component manufacturers who have moved into the Vance and Tuscaloosa areas to service the Mercedes plant. Fork and Hay's supply chain just got a lot closer to the just-in-time ideal, and I'll be able to reduce my on-hand inventory of grains and yeast accordingly.

Thursday, April 1, 2010

Suits me to a tea

Fork and Hay Brewing is turning over a new leaf, so to speak.

I've spent the better part of the last year trying to learn the art of brewing beer, with mixed success. As the time has worn on I have found that I derive less and less satisfaction from the effort - the thrill is gone. I certainly didn't expect that outcome. The effort I've put into this hobby just hasn't given me beer that is different enough from what I can get at the convenience store to make it worthwhile.

As I sat here this morning wondering what I could do to recapture the spark that pushed me into this whole exercise, I started to realize that the solution was in my very hand - a bottle of green tea.

When I started Fork and Hay I didn't know the first thing about how beer was made, what ingredients were required, what processes were followed...and now I do. It struck me that on the whole lately I've consumed a lot more green tea than beer due to annoyances like company bans on alcohol in the workplace and the fact that Breakfast Stout isn't on tap next to the cinnamon rolls at the Holiday Inn Express, and I don't know the first thing about it. I did a little Topeka-ing
and found that there was a vibrant, if small, home tea-ing community just beginning to explore the boundaries of the green tea revolution.

I've grown weary of the limited taste opportunities afforded by the BNL (Bigelow-Nestle-Lipton) megabrewer varieties I can get at the grocery store. I want to have new citrus flavor options - try new concentrations and blends - maybe even process my own decaf. There's a lot of effort going into fusions like "Earl Green" too - tea blends that cross the boundaries of the traditional styles. I want in on the ground floor.

The upshot of all this is that I'm dumping the beer when I get home and starting a basic Sencha. It's been a great ride but it's time to move on to the Next Big Thing.

Sunday, March 14, 2010

That's more like it, and it's portable too

The family took a little trip and I decided that rather than be away from Fork and Hay, I would take a little Fork and Hay with me. I could get used to this concept.

Kegged and tapped: 100201 Geordie-Boy Ale Keg 2

I still haven't finished the first keg yet but I went ahead and tapped the second keg, since I wanted to test the travel properties of a full keg. The second keg has two main differences from the first one. It was fermented with Danstar Nottingham yeast rather than Safale US-05, and I brought it inside to warm up for another week in the hope of getting a diacetyl rest (if you recall the last post).

Once again I have changed too many variables simultaneously to draw specific causalities, but the observed properties of this sub-batch are pretty clear. It finished at 1.012, which is a much better attenuation than the other carboy exhibited. It's also free of the off-flavor I tasted in the other keg and in the previous Por Favor batch. Subjectively, it's a better tasting beer than the other keg, and I have some ideas for the next batch to help nail it down.

I think I'm going to do two things in the short term. I will brew a Por Favor with a different yeast rather than US-05, like maybe White Labs WLP001 or Wyeast 1056. Those are both American ale type yeasts and maybe they will provide a different result. The second thing is that I will return to using Wyeast 1098 for the Geordie-Boy recipes. The Nottingham batch tastes good but it's not the same as the first batches of brown ale I made, and I need to eliminate some variables.

Is driving with a corny keg a violation of "open container" laws?

I don't bottle much. That presents a mobility challenge of a sort, because a corny keg isn't exactly the most packing-considerate form factor for carrying beer. (For one thing, they don't fit too well in ice chests.) There appear to be three challenges to overcome when traveling with a keg beer system, and I addressed them on this trip with varying results.

First and perhaps most obvious is the relentless pressure. Not the pressure of traveling or of being expected to share your beer, but the need to pressurize the keg for dispensing. At home, pressure is maintained by the CO2 system in the kegerator, with regulators and manifolds and all the various required accoutrements. On the road, you have to be able to apply pressure to serve. (A fundamental assumption is that the keg is already properly carbonated.)

The Cornelius Keg CO2 Charger seemed to be the simplest option for managing pressurization. It uses the little CO2 cartridges that are often used in pellet guns, and you use it to maintain sufficient head pressure in the keg to force the beer to dispense. It has no regulator; instead, it has a trigger mechanism that you use to shoot a burst of CO2 into the keg to get the beer flowing. It kind of works like the party taps you might have used in your college days. On the whole, I'm pretty pleased with this solution, as it's pretty easy to tell when you need to trigger the gas because the beer flow slows to a trickle.

In this picture you can see the CO2 charger attached to the keg. It connects to a regular 1/4" male flare fitting, like the one found on some gas-side ball lock quick disconnects. I didn't have a spare lying around, so I had to disassemble one of my kegerator gas hoses to get the fitting I needed. It worked just fine, and I'll order another QD just for the portable charger.

Once you have the head pressure on the keg you need to get the beer out. For this, I went all out and...well, nothing. As you can see I actually backrobbed one of my cobra tap hoses from the kegerator. Hey, it works. The problem I found with it is that it doesn't stay cold like it would in the kegerator, so the beer that runs out first is warmer than the rest of the pour. Since the unofficial motto of Fork and Hay is "the warmest beer you ever drank was 'just cold enough'" that doesn't present a real problem for me. I did find that it was best to rinse the hose out every day since warm beer left in it overnight gets sticky and could present an avenue for contamination.

The other challenge is temperature regulation. As I mentioned earler, the corny keg isn't exactly a great fit in a cooler, and the other folks I traveled with objected to me putting it into the fridge at the expense of things like food. What I did was jury rig an ice based cooler out of a garbage bag and a plastic bucket. It keeps the bottom of the keg ice cold and thermal transfer chills the upper portion somewhat. Since the keg dispenses from the pickup tube in the bottom, I got cold beer fills as long as I maintained the ice level. The keg is insulated with a blanket, but the bucket is not, so one bag of ice lasts about one day. More insulation would provide a better result I'm sure.

The expense of the ice is more than offset by the convenience of having draft beer on demand and the satisfaction of knowing it's something you brewed and enjoy. I'm chalking this up as a success.

Sunday, February 28, 2010

A pain in the diacetyl

Something is wrong with my process. I have now encountered the same off-flavor in two consecutive batches, which I think is excess diacetyl. While I can't really say it tastes like butter or buttered popcorn to me (as described in various Internet sources) I can see how it might taste that way to others.

The last batch of Por Favor has been hard to take because of this flavor. I have been describing it as "yeasty" because my taste sensor isn't calibrated very well. Yesterday, I kegged the first carboy of the 100201 Geordie-Boy Ale and carbed it. Imagine my surprise when I took that first big sip and found the same flavor there as well.

I'm looking into what could be causing this. As usual, the HomeBrewTalk forum has a wealth of knowledge on the subject, and based on what I'm reading, it appears that the main factor contributing to excess diacetyl is a combination of lazy yeast and lack of patience. I confess to the lack of patience part. As far as the lazy yeast is concerned, I am thinking that my poor winter temperature control has to be to blame. Both these batches were made when I really didn't try to control the temperature more than what was available in the environment.

The forum post states that having a diacetyl rest at 75 F will help the yeast clean up their mess. I don't think I've been able to come close to 65 F on a consistent basis recently, much less 75 F. Thinking back, I don't recall having had this issue with earlier batches, but for them the fermentation temperatures were a lot closer to 68 F in the first place than I've been able to manage recently, so it's possible that the yeast did more cleanup as they went along and I didn't need a diacetyl rest.

Now I'm in a quandary. I have five gallons of Geordie-Boy still in a carboy but I moved it to the cooler to cold crash yesterday. Should I get it out and let it ride at ambient for another week? If I do that I need to find a warmer spot in the house. What to do, what to do?

I opted to move it carboy and placed it into the warmest room in the house. I'll leave it there for a week and see how it tastes.

Monday, February 22, 2010

CAI WebControl - I Don't Understand You Anymore, We've Grown Apart

I didn't brew last weekend but I did spend some time in the garage dinking with the probes.

As you may recall, I had an issue with one of them seemingly getting stuck at 39.1 C during the last batch. I wanted to check whether it was the probe itself, or maybe something in the CAI WebControl board, so I took my 2000W heat stick and started to heat a few gallons of hot water from the tap. I moved the WebControl's probe to the Arduino and replaced it with a probe from the Arduino that I knew to be working.

When I dropped both probes into the bucket, it was already at 130 F or so. The probe attached to the Arduino immediately started indicating the correct temperature. (Well, after a couple of scan cycles anyway, as the stainless probe body heated up and transferred the temperature to the DS18B20.) The probe attached to the WebControl hit 130 F (54.4 C) and stopped changing.

Intrigued, I pulled the WebControl-attached probe out and cooled it. It started registering temperature drops almost immediately. When it read 99 F I put it back into the bucket, whereupon it decided to start tracking at the same rate as the other probe.

Speculation? The WebControl's firmware can't handle a too-rapid spiking of temperature, but when you let it ease into a change (i.e. it has a non-trivial ramp time) it does OK. Conclusion? None, except that the suspect probe from last weekend is good after all.

Thursday, February 18, 2010

An epic entry, this is not

An update on the Nottingham vs US-05 contest for 100201 Geordie-Boy Ale: While looking at the temp trends last night I noticed that the temperatures on the carboys had declined a couple of degrees. Since it was closing in on three days since fermentation started, I figured that the temperature decline meant that the most active phase was on the wane. I brought the batch upstairs into a slightly warmer environment (somewhere around 68-70F). As I am sitting here, the Better Bottle (with Safale US-05) continues to bubble at rate of about once every 30 seconds. I thought that the glass carboy with Nottingham was less active, but it's bubbling at about that rate as well - I just can't hear it bubble like the Better Bottle because it has a different airlock type. With the move the batch is out of reach of the temperature probes, so the trend no longer shows anything useful.

Monday, February 15, 2010

TMI

I'm beginning to wonder if having all these temperature probes is a curse, rather than a blessing. I mean, sure, it has been fun learning about them, prototyping, testing, and redesigning the support circuitry, and I have certainly appreciated the chance to rediscover lost talents in electronics and programming. The problem is, I'm not sure that the volume of information I now have at my disposal is helping me address the whole reason I'm doing all this, which is to make more consistent beer.

Sometimes I feel like I'm on the way to a sort of "paralysis by analysis" situation. Take yesterday for example. I brewed a 10 gallon batch of Geordie-Boy and I had temperature probes everywhere. Did they help me? Probably not as much as they confused me. But let's not jump to the finale, I have a lot of ground to cover first.

Stick it where the tun don't shine

While I was cleaning the downstairs utility closet and the garage, I happened to finally identify a "mystery" 20A circuit. It turns out that it runs a single outlet in the garage into which my house's central vac system is plugged. That outlet is in a corner of the garage whose very existence was just a rumor up until two weeks ago, and I had not really factored it into my electric brewing plans to this point.

As you may recall, I have been using two 1500W heat sticks recently to augment my propane burner. They run on separate 15A circuits. When I built the heat sticks, I also acquired the materials for a 2000W heat stick, but since I didn't know where I was going to be able to plug it in, I tabled its final assembly.

Finding the 20A circuit prompted me to finish that heat stick, which of course became a multiple Lowe's trips activity because the 20A 115V outlet was keyed differently from the 20A 240V plug I had originally bought. (Who knew? Except everyone who has ever done actual electrical work before, obviously. For future reference, while they both have the hot and neutral blades at 90 degree angles from one another, the 115V horizontal blade is the opposite side from the 240V one.) Once assembled, the 2000W heat stick was able to raise 4.5 gallons of water from 8 C to boiling in about 90 minutes, uncovered, in an ambient environment of 45 F. Not too shabby as an adjunct heater, but still not enough wattage to go all-electric.

I have been having trouble hitting the right mash temperature, so I decided to heat my strike water directly in the mash tun rather than using a separate propane-fired vessel and risk losing heat during the transfer. I also decided to skip the boiling grits preparation, reasoning that since they were already gelatinized I could get the same effect by just adding them to the strike water as I heated it. (You'll recall I have had problems with the near-boiling grits adversely affecting the overall mash temperature - this way, I figured, at least everything would be about the right temperature at the start. More on this later.) I used hot water from the tap so the actual temperature rise required was substantially less than usual. I used one of the 1500W sticks to get the final 40 or so degrees I needed to get to the strike temperature of 169.5 F.

RIMS is looking better all the time

Here's where the weirdness started. I dropped one of the new active mode temperature probes into the tun at the start of this process and its reading stuck at 39.1 C, which I knew to be lower than the starting temperature of the tap water. No amount of resetting the WebControl board would make that reading budge - in fact, as long as the probe was hot, resetting the WebControl only caused it to not read at all. Once the probe cooled below 39 C (about 102 F) it seemed to work normally.

OK, duly noted, move on to a different probe and mark that one as "refrigerator temps only" for the time being. I switched to one of the parasite mode probes and was rewarded with a more reasonable reading. I happily went on my way, weighing out the grain bill and conducting other normal pre-mash activities, glancing occasionally at the web page that refreshes the probe readings every 20 seconds or so.

After a while, I began to notice that the probe reading didn't seem to be changing very much. I suspected that the reason was a lack of circulation in the water - with the way the heat stick was oriented, it looked likely that a layer of hot water would form near the heating element at one end of the tun. The end of the probe hung down very nearly to the bottom of the tun, so I guess it's reasonable to think that the water down there was not warming as rapidly as the water up top. This is a known phenomenon in the ocean but I had no idea that the vastness of my beverage cooler mash tun would be subject to it as well.

So now I started to periodically stir the strike water. Every time I did this, I was rewarded by a 2-3 degree jump in the probe reading. That's not great, but it's something that I can plan for in the future, and in fact I could probably rig a little agitator to help keep the water homogenized during this process if I wanted to. That's what I need, another time-sucking project. I'll get right on that.

Now the TMI factor starts to weigh in. I noticed a few things that affected my confidence in my measurement setup. First, there's a pretty big latency in the measurements themselves, due to a combination of scan rates on the Arduino and update rates on the display page. I got to the point of trying to guess where the reading was going to be on the next update rather than just accepting it at face value. In reality the maximum drift could only be about 40 seconds. The probes are read in a loop that cycles about every 4-5 seconds. The PC scans the Arduino every 20 seconds, but it only services serial I/O at the end of each loop. The display reloads every 20 seconds, so at worst your reading is less than 40 seconds old, because the worst case timing happens when a display update happens right after a scan, then happens again before another scan finishes its I/O). That's hard to remember when you're intently staring at the screen waiting for the update, and when the screen refreshes you see other values change, but not the one you're monitoring.

Anyway, eventually the tun reached its target temperature of 169-170 F more or less, and I doughed in with the remainder of the grain, stirring it to make sure there were no obvious dough balls. (Side note - I need a longer spoon or a mash paddle - 170 F wort stings a little when your fingers inadvertently dip into it on the down stroke.) I pulled the heat stick out of the tun but left the probe. Turning to the readings, I saw that the probe was showing about 138 F, which was substantially below the 154 F I was shooting for. (And before you ask, yes, I compensated for the grain temperature being at garage ambient when the strike water temperature was set.) The problem was by this time I had zero confidence in the reading on the probe, and I sure wasn't going to sit there and stir that glop until the probe told me the temperature was good, so I yanked it out of the tun and put my trust in the math of thermodynamics. (And before you comment, yes, I'm aware that that is a juvenile reaction along the lines of "la la la I can't hear you.")

After getting the mash started, there really wasn't anything remarkable about the remainder of the batch. I split the sparge water into two volumes and heated each one with a separate heat stick. They were ready way before the mash ended (partially because I started them before the mash started) and I didn't figure a few degrees high was going to hurt so I didn't obsess over their readings as much as I did the mash tun. I did screw one thing up by not closing the valve on the mash tun for a few seconds when I did the second sparge, and because the tubing wasn't in the container I vorlauf to, I'm sure some extra grain particles ended up in the wort. I guess they'll settle out on their own eventually.

I used the 2000W heat stick, a 1500W heat stick, and the propane burner for the boil. I had to rig a wind shield by the burner because the day turned a little blustery, but I had a folding camp table that fit the bill nicely as a 90-degree wind barrier. I scooped off the majority of the foam at the beginning of the boil. I hope this reduces the amount of trub and helps clarify the wort. Chilling was obviously no problem with cooling water and ambient temperature at about 45 F.

Brewed: 100201 Geordie-Boy Ale

I filled two carboys and still had a substantial amount of wort left over. I'm not surprised: I spent more time worrying with and tweaking the temperature readings than I did paying attention to things like volumes. In fact, I didn't even measure the wort volume before the boil or afterward, not because I don't care, but because I was so rattled I forgot. (Where's that Workflow process? Did that get sidetracked by these other BS activities? Fire the developer.)

I used one Better Bottle and one glass carboy. I'm getting a little leery of the Better Bottles because I have a tendency to handle them in a way that causes them to "suck back" through the airlock. I don't think it's me, I think it's just a natural effect of having flexible sides. I put this BB in a milk crate and it seems to be a little more stable.

I am doing an experiment with yeast. In the glass carboy, I used one packet of Danstar Nottingham dry yeast, and in the Better Bottle I used one packet of Safale US-05 dry yeast. I didn't pre-hydrate either of them.

The gravity of this batch was a little low - I measured it at 1.039, down from the anticipated 1.043. Again, if I had spent more time doing useful measurements I might have a clue as to why, but I have a couple of ideas. I'm not sure that I got all the starch from the grits converted correctly, for one thing. When I dumped the spent grains, I noticed there were a few sizable homogeneous grit clods at the bottom of the tun. I doubt they pulled their weight in the mash. Also, with all the angst over the mash temperature, I have no idea if I actually hit 154 or went higher, and without measuring volumes I really don't know if I just oversparged or underboiled.

I guess we'll find out in a couple of weeks if this batch can overcome its humble beginnings to become a fine, upstanding ale. I have it in the inside portion of the basement, where the ambient temperature is in the 60-70 degree range. I have probes strapped to the outside of the bottles. You can see them on the charts as probes A8 and A9. This may be the most useful application of the probes yet - I should have a great strip chart showing the fermentation temperature profile for both yeasts.

Thursday, January 28, 2010

This one's for Richard

Richard! Look at the chart!

Fair warning: this is going to be one of those posts that rambles on and on with little to do with beer and a lot to do with technology.

My troubles with getting multiple Maxim DS18B20 temperature sensors working reliably with my Arduino Duemilanove microcontroller and my CAI WebControl board have been noted here many times. I think I may have finally figured out what was causing my problems. As with most things I resolve, it turns out to have been equal parts unreasonable expectations and learning curve. Rather than backtrack through all the previous postings I think it would be easier to describe what I have arrived at compared with what I had hoped to achieve, and then explain how the differences have arisen through this journey.

As of this moment, I have the following configuration working:
  • Arduino with 3 "parasite mode" probes connected via 1/8" mono phono plugs, two of which are using 50' headphone extension cables with the other simply on its 6' base lead, with one 50' probe on a separate digital input pin from the other two.
  • Arduino temperatures are displayed on the 2x16 LCD connected to the board and printed to the serial port. The temperatures are collected about every 6 seconds or so as the display loops (not interrupt driven).
  • WebControl with 1 "active mode" probe connected via 1/8" stereo phono plug and using a 50' headphone extension cable.
  • WebControl temperature reading collected remotely over Ethernet using its CGI capability, then stored and charted using RRDTool.

The graph to your left shows a thumbnail of the last hour of the WebControl-attached probe's reading of the ambient temperature in the garage. Clicking the chart will take you to a page showing a larger version of that graph and some graphs showing more historical timespans. The traces start about 10:30 PM on January 27, 2010.

What I had hoped to achieve, at various times during this escapade, included
  • Multiple probes on the Arduino with 50' extensions (at least 3 and as many as 7 or 8).
  • The Arduino providing on-demand updates via its serial port and a web service on another host.
  • The Arduino managing temperatures in the kegerator and fermentation chamber, activating an SSR to cut the compressors on and off as needed.
  • Multiple 50' probes on the WebControl (which is alleged to support as many as 8 DS18B20 devices).
  • The WebControl using its digital logic feature to manage the temperatures in the kegerator and fermentation chamber via SSR.

The principle sticking point in most of these scenarios has been getting the probes to work reliably, and that's where today's ramble will focus.

I selected the DS18B20 digital thermometer for the probes because I didn't want to mess with managing A/D conversions with something like the LM34. There are numerous working examples of DS18B20 to Arduino projects all over the Internet, and the Arduino home page even has links to a predefined library for handling the DS18B20's 1-Wire interface. The promise was simple: multiple DS18B20s, which can exist in a bus structure because they're independently addressed, without fear of signal degradation or need of op-amp amplification due to run lengths of sensor cable. Maxim (who purchased Dallas Semiconductor, the originator of the DS18B20 and 1-Wire) claims distances of 300m or more for the 1-Wire bus. How could this possibly go wrong?

Chasing Resistance is Futile

Well, the first place it could go wrong is with basic topology. The Maxim IC application note on reliable 1-Wire networks is very clear that the "star" (hub and spoke) network layout is the most difficult to arrange reliably. Maxim describes the issue as one of reflections and mismatched impedances causing clashes among different devices on the bus when a star topology is used.

I must have read that document a dozen times before it finally sunk in. One of the major contributors to my challenges has been my desire to have multiple probes attached via a single digital bus point, creating a star topology. This was much more evident (in hindsight) with the work I did with the WebControl board than the Arduino. I have written about some of that work already but the short version of my problem was lack of reliability. I could, at various times, get one, two, three, or no probes working with the exact same wiring configuration on the WebControl board. It didn't occur to me until very recently that the problem was the impedance issue introduced by the star topology I was wiring.

What finally tipped me off was something I read later in that application note regarding the use of stub drops off a single long line connecting the DQ of all DS18B20s in a long bus. The note talks about the optimum length of said stubs as being under 3m. Suddenly it occurred to me that the majority of time that I had anything working, all the run lengths were individually less than 3m long, so essentially I had a zero-length long line with a handful of stubs. Things went south when I added the 50' extensions to one or more lines.

I had spent hours chasing line resistance as the cause of this problem, thinking that the power supply from the board wasn't able to source enough current to overcome the resistivity over the length of the cable. I came to realize I was seeing the reflectivity issue instead. That's how I have ended up with the three probes on the Arduino on separate digital inputs and why I have cut down to a single probe on the WebControl board. I now have only one device on any given bus, so there's no cross-talk and reflection among devices. Each of the buses supports a 50' combined length of probe and extension, providing me with the maximum flexibility in configuration.

How does that fall short of expectations? The main issue is pinout. Three's about all the digital inputs I have to spare on the Arduino after all the pins that are used to support the LCD display are accounted for. (I know the analog pins can be used as digital inputs too, but I don't want to give them up just yet.) It would have been much better if I could have run the probes off one pin. The WebControl is more problematic, as it has a single port purpose-built to support DS18x20 devices. I am approaching this as meaning I can use one of my probes reliably, or set up a more conventional long-line with drops. The long-line strategy would be good for relatively fixed device measurements, so I may yet do that, but for now I'm going to concentrate on just the one probe and leave the other 7 the board supports as undiscovered country.

I promise to post some schematic drawings of the final configuration of this stuff soon. In the meantime, enjoy the chart showing the ambient temperature in my garage.

Sunday, January 17, 2010

Channeling Pedro Cerrano

Unlike Pedro, I don't hit the straight ball very much, because it's not just the bats that are afraid when I step into the batter's box. Like Pedro in the playoff game, however, I'm pissed off (albeit not at Jobu). I'm beginning to think there's some sort of cosmic conspiracy to undermine my confidence in my technical abilities.

I spent this weekend cleaning the garage so I would have a place other than the dining room in which to work on my automation projects. This afternoon, I finally cleared and lit a place on my workbench(!) where I can prototype circuits and solder and do all the stuff I need to do to finish the temperature monitoring and control system I have been working on. I proceeded to assemble a probe from 50' of Cat-5 cable in the expectation of moving the project forward.

Interestingly enough, once I had a place to work in the garage, my soldering skills improved (why, I don't know). I got the DS18B20 attached to the cable the first time and tested the new probe along with the two that were working previously.

Here's where the Cerrano situation comes into play. I cannot get the WebControl board to recognize all three probes every time it powers on. Sometimes it will only see the original two, sometimes it won't see any. If I remove the original two probes it recognizes the one on the Cat-5 cable. Sometimes it will see the Cat-5 probe and one of the two others.

I'm unable to figure out why things are so inconsistent.  I'm sure it's due to either (a) my own idiocy or (b) some issue with the Vdd power on the probes. Maybe I'll switch the probes back to the Arduino and see what happens.

Kegged: 091103 Por Favor Keg #2

So I finally got around to kegging the second carboy of the Por Favor batch I made back in November. It's been patiently biding its time for about 6 weeks. I put it into the cooler about two weeks ago to cold crash and clarify it. Today I kegged it, and I noticed from the start of the racking process that the beer was clearer than usual. I didn't take a gravity on it.

Since the beer was already cold I went ahead and carbed it. I can definitely tell the difference between this portion and the beer that came out of the other carboy a month ago. The flavor is much smoother, and I don't sense the yeast overtone that was apparent (to me) in the other keg.

I guess this is a lesson to be learned - beer tastes better if you let it condition for a while instead of rushing it to serving. I need to adjust my production schedule (and equipment) to allow for more aging for each batch. What that likely means is that I will need more kegs and carboys because they won't come available as frequently as they do now.

Sunday, January 10, 2010

2010 From the Consumer's Perspective

My last post outlined Fork and Hay's 2010 objectives from the supplier's point of view. As in real life, however, all those goals are useful only in the context of consumer satisfaction: you can make the most consistent, high-quality products in the world, but if they're not what the consumer wants, they're not going to sell. Well, to be fair there's no selling going on at Fork and Hay Brewing (DO YOU HEAR ME ATF AND IRS? I'M NOT SELLING BEER), but the idea is the same: if the people who are the potential consumers of my batches don't like them, I will have wasted my effort and money in producing them.

What does the Fork and Hay consumer want? I said before I don't like looking backwards, but as I see it there are only two ways to determine what products to make: see what was popular in the past and make more of that, and do market research to determine what the people in your target market want and figure out how to provide that. (I know that's a simplistic way to look at it but, as some of you have heard me say, the only time I ever spent in the business building in college was when the Brewmistress and I would use their dial-up terminals to play Adventure on the campus mainframe - at that time the engineering building only had card punches and readers. XYZZY FTW.)

If we take a quick click on the ads...sorry, did that come out in print?...I mean, a quick look back at what Fork and Hay produced and how long it took to get it consumed, we should have some idea of what was popular in the past:

"Geordie" family: 50 gallons brewed, consumed
Por Favor: 20 gallons brewed, 10 consumed, 10 serving
Half Wit: 5 gallons brewed, consumed
Gayle Bait: 5 gallons brewed, still serving since October
Honey-Brew List: 5 gallons brewed, scrapped
Coldwater 420: 5 gallons brewed, consumed
Shaun of the Dead: 5 gallons brewed, serving

Clearly this tells us a few things:

  • I like the Geordie recipes and Por Favor
  • Honey-Brew List was a disaster
  • Half Wit and Coldwater 420 were enjoyed by other folks
  • Gayle Bait isn't a disaster but it's not knocking the crowd's socks off

What does it really mean? It's obvious that I'm the primary consumer of my work, which isn't news. Given that, planning for the new year should be pretty easy - keep the keg pipeline full of Geordie Boy and Por Favor. The problem with that plan is that it doesn't account for the other significant use case for my beer, which is serving friends and family during various gatherings.

For this year, I need to be thinking pipeline for some significant events: Spring Break, St. Patrick's Day, Scotty's graduation, and the Fourth of July are those most impending. The people who will be around the house during that time are a mixture of "serious beer" drinkers (some of whom are also serious "beer drinkers") who have already exhibited an interest in the richer recipes like Por Favor and the two flavors that Tim commissioned, and casual beer drinkers who seem to gravitate toward lighter brews. The challenge before me for 2010 is to determine what I can make to satisfy the casual drinkers, because personally I'll be satisfied to keep drinking Geordie-Boy and Por Favor.

Note: at no point in this discussion do I intend to be demeaning to those whose tastes aren't like mine. I'm not trying to stand in judgment of people who like lighter beer than I do. It may come out sounding pejorative but I don't mean for it to.

I had intended for Gayle Bait to be the principle beer that I could offer the casual drinkers, but so far it hasn't really taken off. I'm not sure why. However, I think that it has some of the characteristics that are likely to appeal to this demographic:
  • It's light in color
  • It's fizzy
  • It's not too hoppy
  • It's clear
Not coincidentally, those features happen to describe the main characteristics of the most popular commercially available US lagers.Thinking back on last Fourth of July, when my available offerings were Half Wit, Coldwater 420, and Geordie Ale, the most popular beer among the casual crowd was Half Wit. Geordie Ale was too dark and Coldwater 420 was too hoppy. Half Wit was light, fizzy, and had a mild flavor, even though it was based on a witbier and therefore was somewhat cloudy. It seems clear (no pun intended) that this is an area I need to try and master in the upcoming year. I can repeat the Half Wit recipe, and I think I can do it in a way that adds some flavor but doesn't risk the unfortunate consequences that the Honey-Brew List batch encountered.

So what do I do? I think that I am on solid ground by planning a 10 gallon batch of Half Wit for readiness by St. Patrick's Day. That should carry my light and fizzy inventory at a sufficient level to hold me through Scotty's graduation in May, and give me time to replenish before the Brewmistress's annual domination of the Indian Springs Fourth of July Celebration dessert contest.  However, I'm really not satisfied with that response. Am I really going to have to start lagering before I can make everyone happy?