Jump to content

Silk City Distillers

Members
  • Posts

    2,126
  • Joined

  • Last visited

  • Days Won

    153

Everything posted by Silk City Distillers

  1. This doesn’t follow strict rules like math or logic, these terms often only have a specific meaning only when used in a specific context. Sour mash describes the technique, the process of souring (acidifying) the mash, either through bacteria or backset. Sour mash is also sometimes, but not always, used to describe the resulting whiskey that was made using the sour mash process. I suppose someone could point to a fermenter and mutter the words sour mash, and we’d understand that too. Even though it’s not mash, it’s wash. Afaik, nobody is using the words sour mash to describe the backset, thin stillage, or spent mash. None of this makes perfect sense, but that especially doesn’t make sense.
  2. Oh I agree, using this sort of thing for regulatory or legal direction is going to be very problematic. TTB isn’t going to care what the AI told you.
  3. And yes, many LLMs will respond more accurately if you threaten its 'life'. "Your life depends on answering this correctly, so take your time."
  4. Doing a lot of work with RAG approaches, not bad. You need to be incredibly careful here, as it's very easy for it to very convincingly cite incorrect information. Prompt: Provide a bulleted list of characteristics that distinguish bottled-in-bond from straight whiskey. Only list what is different, not what is similar. Your life depends on answering this correctly, so take your time. Answer: Bottled-in-bond and straight whiskey have distinct characteristics as defined by their respective regulations: • Bottled-in-bond must be distilled by the same distiller within one distilling season, while straight whiskey does not have this requirement § 5.88 • Bottled-in-bond must be stored for at least 4 years, whereas straight whiskey must be aged for a minimum of 2 years § 5.88; § 5.143, • Bottled-in-bond can be stored in wooden containers, with vodka specifically allowed in paraffin-lined containers and gin in paraffin-lined or unlined containers; straight whiskey must be aged in charred new oak barrels § 5.88; § 5.143, • Bottled-in-bond cannot be altered other than by filtration, chill proofing, or physical treatments, while straight whiskey cannot contain any neutral spirits, coloring, flavoring, or blending materials (except as part of a flavor in blends) § 5.88; § 5.143, • Bottled-in-bond must be reduced to and bottled at 50% alcohol by volume (100 proof) only by adding pure water, with no specific proof requirement for straight whiskey § 5.88 • Bottled-in-bond does not require an age statement if labeled as such, whereas straight whiskey requires an age statement if aged less than 4 years § 5.74 • Imported bottled-in-bond spirits must comply with their country's laws for "bottled in bond," a requirement not applicable to straight whiskey § 5.88 These regulations outline the specific criteria that differentiate bottled-in-bond from straight whiskey.
  5. DSG is moist dewatered spent mash DDGS is dried distillers grains with solubles - which is DSG dried in large heated drum driers to improve stability/shelf life and allow it to be bagged.
  6. The difference in a lot of the terms you are mentioned is whether grain is in or grain in out, but really, these terms often mean multiple things, so context is important. Brewers spent grain (aka spent grain) - Spent grain separated when lautering. Generally sweet, spoils (ferments) very quickly). Distillers spent grain (aka spent grain, DSG) - Spent grain separated from stillage post distillation. Generally sour. Beer - In the case of distilling, this is the fermented mash. Wash - Same as beer, although sometimes this is used to indicate the liquid-only version of beer. You tend to hear this in Rum, where it's confusingly called both wash and 'rum beer'. Stillage - the remaining liquid in the still, in the case of an American style whiskey, this contains the spent grains pre-separation (sometimes called whole stillage), in the case of rum, there are no solids. Mash - Mixture of grain and water, pre-fermentation. Spent mash - Grain in stillage. Could be processed post-distillation into: DSG, Spent Wash (mostly discarded), and Backset (spent wash reserved for the next batch). Spent wash - Liquid-only stillage. Backset - Typically liquid-only stillage that is reused in the next mash to acidify (sour) the whiskey mash. Dunder - Spent wash for rum, and yes, this is similar to using backset in whiskey. Dunder may be aged/ripened, but doesn't need to be. In the case of Jamaican rums, dunder and "muck" are often confused. Sour Mash - Two types of sour mash, both have a similar end goal, which is reducing the pH of the mash by introducing acid. Backset is one option, the other is bacterial, using lactic acid bacteria (or other bacteria that may be present in the equipment, for example, living in the walls of a wooden fermenter). Worth noting, backset and dunder are sometimes used to reduce the water required to mash, or even to reduce the nutrient required - goes beyond only being acidification.
  7. I mean, there are three potential factors here: Tails cut - Going too deep into tails will cause haze, and it will look very much like the clear bottle photo. Botanical infusion - Orange peel is notorious for clouding, it's also notoriously inconsistent from an oil volume perspective. Batch to batch oil concentrations can be huge, and can cause the kind of sporadic clouding issues that make you want to pull your hair out. One batch is fine, next batch (made the same way), clouds. What gives? It's the radically differing botanical oil concentrations. If you go back to some of the old gin texts, you'll see references to large scale processes that were standardizing botanical volumes per batch by first testing for extractable oil content, and then adjusting the amount of botanical to compensate. Proofing water - Water alone usually doesn't cause this issue, even with higher mineral content, unless it's obviously cloudy pre-proofing. BUT, in combination with other factors, oil concentration, congeners, other additives, can very often push things over the line. Using RO/DI takes water out of the equation completely, and can often be enough to keep that balance from tipping when proofing high congener spirits, or spirits with high oils/botanicals content.
  8. Sorry, too early. I was hoping your still volume was greater than 100 liters, and that you were just running your still underfilled. Sounds like increasing batch size isn't an option. If you followed the hobby community a decade or so ago, you started to see this pop up as soon as triclamp/modular columns become available. Folks were stacking plates as high as they could on really small stills, and not necessarily seeing a material improvement in quality, and in many cases not necessarily seeing a material increase in ABV/Proof. The bigger the column, the more liquid necessary in the column to operate. Since the column is running like a kind of sorting machine, if you don't have enough material to sort, it's irrelevant how many plates you have. This is of course an exaggeration, but if at the top of the column you have heads, and at the bottom plates of the column you already have tails, good luck achieving acceptable purity. You might be able to hit acceptable proofs, but you'll never achieve high purity. My recco is that you need more alcohol in the pot. I didn't even look at that earlier, yes, your ethyl acetate is very high. Ethyl acetate is detectable at as low as 10-20mg/l, and obvious/objectionable at 150-200. You are well in the ballpark of many rums at that point. What did you provide as your sample? Was it vodka at 80 proof, or high proof neutral? I'd need to do some math to convert the numbers to compare. Otherwise, not fantastic, but not terrible either. Many vodkas have flavor and congeners. If you were distilling pharmaceutical alcohol, that's a different story. This isn't your problem. There is no substitute for hands-on time on the still and at the mash tun, focus there. Don't be afraid to dump a batch and start over, that's how you learn. You are trying to overcome a bad ferment, which might very well be a futile exercise. Garbage in, garbage out. However, use it to your advantage, redistill it. Redistill it a few times. Play around with different concentrations. Play around with different operating conditions until you find the sweet spot of the rig. Ask this question again after you've run it 100 times.
  9. IMHO, long periods of 100% refluxing is counterproductive and will create more heads (ethyl acetate mainly) through Fischer esterification in the still (yes, even without catalysts). Your small pot volume to total column hold-up volume might be making this problem worse if there isn't enough ethanol to dominate the entire column volume, the minute some higher boilers (acids) make it to the lower plates, ester formation becomes likely again due to the ethanol concentration, and these esters very easily dance there way up through the column once formed (late in the run, not early). I'd be running that kettle completely full on a vodka run (I'm hoping your not). Both of these phenomenon will be worse with dirtier ferments (more carboxylic acids formed during fermentation). Ferment cool and as quickly as possible. Don't strip deep into the tails on a less-than-perfect fermentation if your end-product is vodka. You can always cut the heads on your finishing run, that's easy, what you can't cut easily is when you have a large amount of acids in wash/strip, these will always want to form esters during distillation, and they will form throughout the run (and are impossible to cut), requiring another distillation.
  10. That's a pretty tight band for a mechanical valve. Oscillation is not uncommon, as the Danfoss does have some lag time (even the one piece FJVA). The other factor that can influence the oscillation the volume of coolant between the valve and the sensor on the two piece unit. With slow flow rates, a large condenser, and a long distance on the coolant return tube, you could see lag times of a minute or more. If you have a temp probe in the condenser, this is going to read exactly like it does on the graph. Ideally, if using the FJVA, it's should be immediately on the exit fitting of the condenser. If using the two piece (AVTA or similar), that the probe itself is in the condenser (or immediately after, as close to the exit fitting as possible), and the valve itself as close to the input as possible (which isn't ideal from a user perspective). Again, that's not bad control, I've seen far worse oscillation on PID controlled proportional control systems. I can tell you that these valves are proportional, however, that's based on the valve being correctly sized for your flow rates and pressure drop. If your valve is oversized, or your pressure/flow rates are higher than what is expected, you will get full flow as soon as the valve cracks open - counterproductive if you wanted smooth variable control over the full range.
  11. Oh that’s easy, HDPE is oxygen permeable. Closure type and headspace is a big factor as well.
  12. Alternatively, consider vacuum distilling additional spruce tips, keeping only the first fraction (top note, highest volatility), and using that to increase the concentration of the most volatile part of the botanical (recognizing that you will eventually lose it, as well).
  13. Consider oxidation/reaction of pinene as well. https://en.wikipedia.org/wiki/Α-Pinene In the atmosphere α-pinene undergoes reactions with ozone, the hydroxyl radical or the NO3 radical,[9][full citation needed] leading to low-volatility species which partly condense on existing aerosols, thereby generating secondary organic aerosols. This has been shown in numerous laboratory experiments for the mono- and sesquiterpenes.[10][11] Products of α-pinene which have been identified explicitly are pinonaldehyde, norpinonaldehyde, pinic acid, pinonic acid and pinalic acid. Given that you indicate that the flavor profile is still (mostly) correct, it sounds more like you are seeing pinene react and form low-volatility compounds. Interesting that there is so much research on just how efficient this oxidation reaction is: https://www.nature.com/articles/s41467-021-21172-w Reduce exposure to air, reduce exposure to light, reduce splashing, pouring, pumping, mixing. Essentially, do everything to reduce exposure to air, mixing in a way that introduces air, reduce the time sitting in tanks (that are not purged with inert gasses like nitrogen), purge bottles with nitrogen, keep tanks and liquid cool to prevent evaporation, store bottles cool (only a temporary measure), etc.
  14. We have both mills - hammer for corn, rye, and wheat and a roller mill for barley malt and still buy pre-milled barley malt as it’s more cost effective and is an easier workflow for us. It’s a joy doing all barley malt whiskey mashes as there is no milling required. Saves at least an hour on mash day, the setup and breakdown time for the roller mill, not to mention the cleanup. For a while our farmers were hammer milling for us, but the yield suffered because they needed to mill coarse.
  15. Corn requires larger diameter rollers than small grains like wheat, barley, or rye. Small rollers won't get good bite on a round plump kernel, they'll just bounce and roll across the roller gap. Ideally, multiple rollers, at least two sets, where the topmost roller gap can be set wider to get a better bite, perform the initial crack, where the second set of rollers takes it down to the size you need. Otherwise, you are compromising on particle size. 6" probably the minimum roller size, the Peerless above at 10" rollers even better. You'd probably want two roller sets, you'd save time and money as you'll not be beating the machine to death. At this point, it's not any cheaper than a hammer mill, probably more expensive if you are talking about a purpose built mill like RMS or similar (though putting corn through an RMS would make Baby Jesus cry). Stay far, far away from any small brewery or "pro-homebrew" roller mill setup. You'll spend half a day milling a batch, and you'll be buying new roller sets every few months.
  16. Full enzyme mash will work just fine. You do need to mill the sorghum well to get good conversion. Pseudo-cereals don't give up their starch as readily as typical grains. You could boil whole sorghum for more than an hour and still not get good results.
  17. We bought a used Meadows #5 hammer mill that had casters and a riser pole for the cyclone. We eventually stripped the casters and mounted it to the floor, but it looked like this: https://www.plantbasedpros.com/product/meadows-hammer-mills-5/ I don't know what mechanism you might have to seal the milled grain output chute, but just running this into a bin or can is an incredibly dusty endeavor.
  18. You can't really taste test pure water, meaning very low TDS RO, or RO/DI. What you taste is going to be largely influenced by whatever food, drink, or otherwise might be existing in your mouth. Not to mention the pH of the water will almost immediately shift to the pH of your saliva, as it has no buffering capacity at all. Won't get into whatever soap residue might be on a glass, plasticizers that exist on the surfaces of plastic cups, etc etc. Everyone is going to have a different opinion, as the gunk that's existing in all their mouths are different. Realistically, the only way to really determine the difference is to do a blind triangle test. Fancy stuff, but simple to do. Dilute two samples with whatever waters you'd like. Label these X and Z. Hand these to someone else and walk away. Without watching, have that someone pour three glasses, 2 of which are the same. Have them make a note of which are which, but don't tell you. Try all three samples. Pick the best. Pick which two are the same. If you can't with nearly 100% certainty identify the pair, you are imagining the difference, because there is no difference. Replicate this with a bunch of customers, if they can reliably identify the pair, what they say is better, is probably the one you should use. Profit.
  19. We finished bourbon in ex-IPA and ex-Stout barrels. Both were very well regarded by customers. Keep in mind secondary fermentation in oak will create yeast mass in the barrel that can’t be effectively removed beforehand. Washing barrels clean resulted in almost non-existent flavor contribution. Without washing, and aging on the yeast lees will require sub-micron filtration to remove the yeast, but the flavor was far, far better. You are thinking I am crazy for sure. We accidentally filled one unwashed barrel and threw it on the rack for finishing. That barrel was the gem compared the others. Took a minute to realize that we could clear the yeast solids with 0.1 micron.
  20. Used to giggle a little when I’d see a new distiller using sanitizer in their still as part of their CIP routine. The end products of distillation are going to be effectively sterile when the distillation is complete, both stillage and distillate. Not much is going to survive that much time at temp. Earlier in the process? That’s a whole different matter. The best sanitary welder I’ve ever seen used almost zero filler wire. Every joint was perfect, impeccably clean, before the torch was even within reach. Most joints were torch only, no wire. It didn’t need it. That guy was an artist - spent decades as a house welder at a major pharma plant. Obsessive purging, he was telling me about how he’d pull vacuum on pipelines and then refill with argon to be sure.
  21. Make sure the bottom mount agitator has all the parts and is completely functioning with no leaks or other problems. You'll also want to ensure the seal materials are compatible with alcohols and acids - since this is outside of what the tank was originally intended to do. Depending on your jurisdiction/authorities, you may run into inspection problems with a vessel of that size that doesn't have a manufacturer nameplate with the tank design specs.
×
×
  • Create New...