biodzldan

Typically, the more you increase the frequency, the quieter it will become. This is for a few reasons: you're hopefully well above the resonant frequencies of anything you're attached to and therefore not exciting them, as well as moving most of the noise above the range where human hearing is most sensitive (2-5kHz). This comes at the cost of lower efficiency - most of the losses in the silicon occur during the switching as opposed to the 'ON' state. If your motor is rated for use with a VFD, I don't think you will experience much trouble by raising the frequency. The voltage transients and internal heating effects should *hopefully* have been anticipated by the manufacturer. I just had to deal with a similar problem on a machine that my company manufactures.

Jon, I live in Vine. We've met before, briefly, and have several mutual friends. The building I'm looking at is about a mile away from you, if the inspectors are in favor of me having an H-3 occupancy there. They've been hard to pin down. Incidentally, I think I managed to point out to them that you guys didn't need an ASME stamp for your still because B31 doesn't kick in until 15 psig (they were hinting they wanted me to get stamps too and I argued). Although obviously that doesn't help you much now. If you end up running any small test batches, something like a sugar wash will give you an indication of whether infection or enzymes are at play.

Jon, Kalamazoo uses gaseous chlorine year-round, meaning simple aeration should allow it to dissipate. The concentration varies throughout the year - more in the summer when the water tower is warm (although to me, it didn't smell like it got as high as other years, possibly because we had a cool summer). The point that Silk City made is worth paying attention to. There is construction near you right now that I believe involves water mains. A quick phone call might prove helpful. Personally, I would not continue to waste resources attempting production-size fermentations unless you're absolutely starved for output. I would consider running a series of experiments in the range of 5 gallons or less. I would say the first question you want to answer is whether an infection is present or not. A microscope and some stain should give you an idea fairly quickly. I think I know a couple people in town that may have something appropriate if you don't have one (I was waiting to purchase one until early next year). If there is an infection, you can start working backward to the cause. Aerate some water to dissipate gaseous chlorine, carbon filter some to remove chlorine, maybe get 5-gallons of RO water from the co-op, you get the idea. If you do all of those with careful attention attention to avoiding an infection, and none of them become infected, they may indicate something in your building.

Have you tried plating it to see if you have anything else growing, or checked cell density?

I had the impression that the problem with keeping a mostly-empty bottle around was not so much with oxidation, but the fact that so much aroma is lost to the headspace.

Agreed.

At best, they could make GNS. Where the heck is it all going to go?

Read through the CFR referenced above and realized that the TTB won't care about how I weigh my fermenters, since that is not where I'm calculating tax. In that application, you frequently have the opportunity to re-zero. It's also helpful that what I really care about is differential weight and not absolute weight, and being within 1-2% of absolute is sufficient. For permanent connections to tanks, you just try use flexible, horizontal connections where possible to reduce reaction forces. I doubt they would account for more than 1% of the total load in my my application.

Do they not allow the tank and load cells to be calibrated as a system? For large tanks and hoppers it is not uncommon to measure an amount of material on a smaller, calibrated scale (say, a tote in our case), and then pump that liquid into the tank being calibrated. Repeat several times. As long as you start with sufficiently accurate calibration weights and are mindful of things like creep, you can maintain a good level of accuracy in the final system. It has the advantage of taking into account reactions forces and deflection that you weren't able to fully prevent. This is how I am planning to calibrate my fermenters.

As a general rule, never distill until dryness. I know it's not likely, but if you had tried that with ether, your equipment might have exploded. It's one of those things like adding acid to water. Do it every time. I referenced the TTB videos as the standard technique I use for experiments.

That's a clever way to safely make a pressurized system. However, I will have pressures ranging from 7psia-15psia, depending on the part of the system so it probably isn't a great fit for my application. My water columns would be clumsily tall and I would run the risk of back siphoning during process upsets and startup conditions.

I am considering that as an option as well. The cost per piece will be a big factor. I will not be using a conventional pot still or kettle. The design I am using will require about a dozen pressure relief points. My main reason for not using rupture disks is the one-time use. During software tuning, I could accidentally overpressure several times. It's not that I am seriously worried about my safety while doing this, the components are more than capable of handling the pressures, but I am planning to stay below 15 psig in operation to avoid ASME B31 compliance and let my building and plumbing inspectors rest easy.

Ok, I had looked at a few other sources for stainless ones, but I'll see what you find. Personally, I am comfortable with a brass valve, knowing that I can locate it properly. My bigger fear was whether an inspector would have a problem with it.

I was specifically referring to the relief valve in isolation, not to any particular still design. Meaning, if the valve was placed in a location where it could lead to condensate collection, I was wondering if it could be a problem.

I was specifically referring to the pressure relief valve. The only risk I could foresee is if there was a drip path that allowed it to drain to the condensate collection.

Yes, I was thinking more along the lines of permanent inks directly on the glass. I did some homework on this a while ago, and then shelves to focus on more important matters.

@Southernhighlander We might be hijacking this thread a bit... Are you talking about using those relief valves on a steam supply or on the product side as well? I noticed that there are parts with B16 brass, which I believe contains up to 2% lead? Have you had any resistance from inspectors for using that on the product side, or is it not a problem since material is not routinely flowing through the valve?

Tossing this out for anyone that may be interested. I may use some products from these guys to replace some Johnson Controls motorized ball valves. Faster, and more linear range. http://www.hassmfg.com/search.pl/1244582085-65969?keyword=1161&submit=Search Specifically, I had a situation where the process dynamics were about the same speed as the Johnson Controls valve, and it became unstable with any aggressive PID tuning. I achieved acceptable performance using feed forward with PID, but a faster valve is really the best solution.

Yes, you could certainly just cool the water with ambient air. As stated above, just remember that the low delta-Ts will drive larger surface area requirements in both your condenser and the air cooler, and likely a larger pump. The place I work now makes nothing but glycol chillers. When you really get down to it, many people use them to reduce surface area requirements. It makes assembly lines and manufacturing spaces more compact, i.e. you place the chiller remotely, provide chilled glycol and need only a small heat exchanger at the process. This is meaningful in an automotive plant, for example, but may be something you won't consider important.

I'm curious why there were so many references to removing the labels? If you place the same product into the bottle again, wouldn't it be better to have writing or label that could survive the cleaning process?

"Rather" accurate isn't a great specification. Are you looking for 0.1%, 1.0%, 5.0%? What peak flow rate, how much turndown? Just taking a guess at your requirements, give this a look: https://www.instrumart.com/products/18337/blancett-1100-turbine-flow-meter https://www.instrumart.com/products/18340/blancett-magnetic-pickups Omega sells the same one, albeit more expensive. The meter has the entity parameters on it, for use in hazardous atmosphere (designed for oil and gas applications). The base unit has an unamplified hall effect sensor that detects when the turbine wheel passes by (around 10-20 mV output, maybe). You can either use it as is, or use one of the amplified magnetic pickups. You'll need to read through the datasheets, but I believe it can give you either 5 VDC pulses or a 4-20mA signal. You would obviously need to use zener barriers, sealed conduit, or whatever methods suit you if you will be using it in a classified area. The big thing that you should notice in the specs is the 0.1% repeatability. That means it will be accurate IF you calibrate it for your needs. They are calibrated for water at room temperature (20C), I believe. If you want accurate readings, you will need to calibrate for the water/ethanol concentration and temperature that you will be using. If you don't calibrate the meter for the fluid and temperature you are measuring, the stated accuracy of the meter is meaningless. Keep in mind that for most meters, you will probably need 10 pipe diameters upstream and 5 downstream to ensure fully developed flow, but this will be specified by the manufacturer.

Yes, you're right. I dropped a zero.

I don't think that is quite correct. A normal retailer is allowed to charge more than the state minimum price, if they think their market will tolerate it. The way I read the laws, and the answer that I got from the State when I asked, is that bottles from a tasting room will have to be sold AT that price. You could, however, make a product that you only intend for consumption in your tasting room, set the minimum retail price low, and then charge more for the drink, like a bar would. Obviously, there are limitations to how useful that strategy is.

Look at it this way: The lower flammability limit (LFL) for ethanol in air at 25C is 3300 ppm (3.3%). Assuming you do you do not have sensors installed to monitor vapor concentration, you should be trying to ensure that during normal operation you are staying under 25% of the LFL, which would be 825 ppm. If you persistently operated at the OSHA limit, not only would you be exposing yourself to a health hazard, you would also exceed the targets in the DISCUS and NFPA codes.

I don't have my copy with me, but I seem to recall that the DISCUS Fire Protection Practices covers this.