Jump to content

Pressure Release Valves for Stills


John McKee

Recommended Posts

Pete,

Nothing on the condenser, no need its open to the parrot. If there is a reason that you might think i need one im all ears. I just dont see why since the whole column would need to clog.

See above for the deflag, i use valves for water adjustments and if one gets shut off it allows the vapor to escape.

I don't see any reason for a PRV on the condenser,. but there should be one high up in the vapor on the pot.

You said "I just dont see why since the whole column would need to clog.", I think the poor guy who died recently may have thought the same thing. It appears as if his column clogged and the still split covering him with boiling mash.

  • Thumbs up 1
Link to comment
Share on other sites

Pete,

Doc is correct, PRVs are usually used on cooling water heat exchangers in industrial applications. There are a few reasons, but especially in copper and bronze shell and tube design, the test pressures are usually only 75-125 psi.

Yes (you are correct) the water supply to those heat exchangers isn't that high and probably won't be the direct reason for a heat exchanger failure, but if the water supply gets closed in, the water can boil in the heat exchanger and the hammer condition that occurs can produce enough pressure to crack a tube weld or lead to another source of failure.

I can totally see your point, but shell and tube heat exchangers (at least those built for our industry) are pretty weak with regard to overall design pressure....certainly not 3000 psi....maybe 300, but not 3000.

...and thank you Pete for reasonably replying to Williamsburg's comment, a column does appear to have clogged and inappropriate PRV sizing killed a person. This is important stuff. Thanks for being a part of consistently raising the bar for our industry.

Cheers,

McKee

Link to comment
Share on other sites

Pete, John

If you were to place a pressure relief valve on the column would you place it on the very bottom of the column just above the pot?

If someone packed their column to much or a puking situation developed it would have a chance to rupture for sure.

Our water supply for the deflag comes from one source, i can reroute it in different directions, mash, condenser ect..or shut off the deflag. Having the relief valve on the outlet side of the deflag allows for pressure to escape just in case i forget to open the other end to allow the water to go to the holding tank.

It reminds and works 2 ways, reminds me if i don't have the valves open properly the relieve will open and the cooling water will spill on the floor, or allowing the water in the deflag to escape once the system is closed off.

Link to comment
Share on other sites

Williamsburg,

(Sorry I don't know your name, so using your username)....

On a pot still, you'd want the PRV on the body of the pot itself, since that is where the energy is being input...and therefore where the pressure will build up. If the column were to become clogged (as it appears it did at Silver Trails), the pressure build up in the pot can relieve. Silver Trails had a PRV, but it was grossly oversized, meaning that it allowed the pressure in the still pot to build to a point where it did explode. That was the core of this entire thread, that anything over a 15 psi PRV is not appropriate for the systems this industry run.

Again, you need to direct the output of the PRV, via metal piping, to outside of your distillery, via an external wall or roof penetration. Simply exhausting the still into the still environment is not safe.

Some of the discussion above is very helpful as to other appropriate locations for PRV's and I would watch the posts from Doc Porter very closely. He's the most knowledgeable of anyone on the forums at this time regarding PRV applications.

Cheers,

McKee

Link to comment
Share on other sites

Alright I finally got some free time. As always Spirax Sarco has great data on their website. It mostly refers to stea systems since that's what they do, but it still has a lot of good info:

http://www2.spiraxsarco.com/resources/steam-engineering-tutorials/safety-valves.asp

Some manufacturers (such as Kunkle) put their sizing charts online, but those are better for simple relief scenarios:

http://valves.pentair.com/valves//resources/KUKMC-0392-US_Model_900-EN.pdf

What I did was take equation 3 of this document:

http://www.pdhonline.org/courses/m112/Selection%20and%20Sizing%20of%20Pressure%20Relief%20Valves.pdf

And greatly simplified it to be specifically applied to heating ethanol water solutions by taking into account their thermal properties. The equation I came up with is:

A = Q/[84.8*(960-596X)*(P+14.7)]

A is orifice area in (sq in)

Q is heat input (BTU/hr)

X is mass fraction of ethanol in liquid form in the kettle

P is set pressure (psig)

The tricky part is knowing what Q to use. I for instance have a 10 HP boiler (~335,000 BTU/hr), so this is the maximum I can put into my kettle. Let's assume the maximum alcohol conc. I distill is 30wt% and my kettle is designed for 15 psig. Then using the equation above I require a 0.17 in2 relief valve. Looking at the chart in the article above this equates to an E orifice which varies from manufacturer to manufacturer, but using a Kunkle 911 it would be a 3/4" x 1-1/4" valve.

I don't know anything about direct fired stills, but we look at building fire cases all the time at my job and the equation is Q = 21,000*SF^0.82 (where SF is the exposed surface are in ft2). Not sure how this compares to forced draft burners.

There are tons of manufacturers out there as well and some of them will even size relief vlaves for you, but they are often expensive since they carry all the required certifications:

LESER
Farris
Crosby
Consolidated
Aquatrol
Taylor Valve
Fulflo
Kunkle
And as McKee said, direct mount it to the kettle and pipe the discharge away from harms way. Never reduce line sizes and make sure the discharge piping is short, has few elbows and has a weep hole for draining liquids. If the discharge fills with liquid there is a lot of back pressure on the valve. It can only handle a pressure drop of about 10% of the set pressure on the discharge side and 3% on the suction.
That's very simplified, but I tried to keep it short and sweet.
Link to comment
Share on other sites

Thanks for doing the research agporte

You have possibly sized relief valves for the ethanol vapor over-pressure, but you have not yet sized valves to avoid the main cause (in my opinion) of the still rupturing at Silver Trail.

How do you vent the puke ( boiling foam and mash) That stuff needs an orifice way bigger than ethanol + water vapor.

Link to comment
Share on other sites

Pete,

Doc is correct, PRVs are usually used on cooling water heat exchangers in industrial applications. There are a few reasons, but especially in copper and bronze shell and tube design, the test pressures are usually only 75-125 psi.

Yes (you are correct) the water supply to those heat exchangers isn't that high and probably won't be the direct reason for a heat exchanger failure, but if the water supply gets closed in, the water can boil in the heat exchanger and the hammer condition that occurs can produce enough pressure to crack a tube weld or lead to another source of failure.

I can totally see your point, but shell and tube heat exchangers (at least those built for our industry) are pretty weak with regard to overall design pressure....certainly not 3000 psi....maybe 300, but not 3000.

...and thank you Pete for reasonably replying to Williamsburg's comment, a column does appear to have clogged and inappropriate PRV sizing killed a person. This is important stuff. Thanks for being a part of consistently raising the bar for our industry.

Cheers,

McKee

Thanks John for pointing out that shell and tube heat exchangers are probably only low-ish pressure. I certainly see that in a high temperature industrial situation the water could boil inside those if cooling water stopped.

I don't have a steam calculator handy, but what would the water temperature need to get to in a closed short length of pipe, to raise the steam pressure so it would burst even a low grade pipe?

Whatever the answer is I bet the temperature of a deflag will not get any where near that. If there is any alcohol in the column it won't even get to 100C

If for some hypothetical reason the water in the deflag got way over 100 C then you would be in an extremely dangerous situation, the least of your worries would be the little puff of steam from the deflag if it burst.

The reason I am going on about this is that I don't want to see safety pushed to a ridiculous level.

If everyone is made to put a PRV on their deflg then everything in and around a distillery that is slightly more dangerous than that will need to safety devices as well.

Link to comment
Share on other sites

The relief valve on a deflag would be a tiny $15 water heater relief valve. Just for thermal expansion. If you block in the water and it has nowhere to expand to it could burst the pipe (again I don't know what your specific equipment is designed for). I have an expansion tank and a relief valve on my deflag cooling system since it is a closed loop.

You are correct my calc above does not address foam. I don't even want to try to address that. It would probably be 20 hr+ and I don't have the time. If someone want's to pay my engineering company to do it for them, then I can do it while I work at my full time job.

Link to comment
Share on other sites

  • 5 months later...

I don't think anyone is advocating for increased regulation, just increased safety. I'm confused as to why you can't just slap a 15 psi PRV onto your still and be done with it (forgive me if you've already done this). It is widely considered to be the basic safety standard in this industry, what every distiller I know would require, and if the still is never going to pressurize to even 15 psi (true under NORMAL conditions) then the most pointless thing about this discussion is arguing that you should be okay using the 60 psi PRV. Having a higher psi PRV in an unpressurized vessel serves no purpose other than not working as quickly in case of an emergency.

I agree that describing your equipment as a bomb was inflammatory, but to suggest that the manufacturers of our equipment aren't responsible for what "idiots" do is false and infuriatingly insensitive. First of all, by selling Silver Trails a still which included a 125 psi PRV, the manufacturer was implying that the still would not suffer catastrophic damage until it reached that degree of pressurization, whether it was explicitly stated or not. Otherwise why have the PRV at all? Manufacturers should have a firm grasp on the failure conditions of their equipment. We are distillers. It is a really good idea to have some basic engineering knowledge, as is evidenced by this tragedy, but that is not a prerequisite. It is, however, a requirement for the people designing our equipment.

If manufacturers do not tailor their equipment to the lowest common denominator, which we have already discussed in this case doesn't actually cost anything (other than maybe a little forum pride), there will be more accidents, injuries, and deaths. You say you are opposed to McKee's original objection because it could lead to greater regulation? Let's see what happens after a few more people die from avoidable accidents. As a manufacturer, the liability for something like the Silver Trails incident would not fall on you, but it could have been avoided at the manufacturing level. Now that you are aware of how some people may be misusing their equipment it is your obligation to adjust your standards. If you knowingly supply equipment that will fail in the same way as the Silver Trails still, after learning that the fix is as trivial as switching out PRVs, you ARE to blame for the next accident. Self regulation is the only way to avoid imposed governmental regulation. Darwinism will not work.

Link to comment
Share on other sites

This is an important conversation and I'm glad it is continuing.

I think we can all agree a 125 PSI relief valve is way to high of a relief set for a still that hasn't been pressure tested. I think we can all agree that distillation in most of our settings typically operates at atmospheric or maybe slightly about depending on operating conditions.

While I'm not an engineer, relieving pressure if it gets to 15 PSI is OK, but I think the situation could be dangerous well before that.

I previously worked at a brewery where we did everything under pressure after primary fermentation was complete. This included filtering using a balance line. We had to fill tanks with CO2 to match the pressure of the tank being filtered from. If the target tank was one or two PSI lower or higher when you opened that balance line the turbulence from one tank equalizing with the other was amazing. These were not large tanks either, 15 or 30 barrel tanks.

Now if you have a 250 gallon still with 15 PSI of alcohol vapor pressure is a serious problem that should have been relieved well before 15 PSI. If your still didn't come with a PRV on it, it probably wasn't pressure tested. If it wasn't pressure tested who knows when it will pop. We have no idea what the internal pressure of the still at Silver Trail was when it blew up, but it could have popped in the 5 to 10 PSI range and done all that damage.

If a still hasn't been pressure tested, I think it makes the most sense to have the lowest PSI setting you can find on a PRV. We recently installed a check-valve above the liquid line as a PRV on an old still we retrofitted as it will stay closed during normal operation, but will open at the first sign of pressure. Which is what I want.

Again, I'm no engineer, and if I'm wrong please enlighten me, I'd love to learn more.

Pressure is scary and I think if your still hasn't been pressure tested you should rethink a 15 PSI relief valve as 'safe' because it could be just as bad as a 125 PSI PRV. Even if it is pressure tested to 15 PSI, if you are typically operating at atmospheric pressure, wouldn't you want it to relieve earlier? The sooner you see or hear the relief the sooner you can stop the still and the less violent the relief will be.

Link to comment
Share on other sites

  • 4 weeks later...

Just my 2 cents but here is my setup and how it operates. My still is a steam jacketed (15 psi)1000L boiler with a 10 plate 8" column. I have a 5 psi PRV on the still (installed on one of the tri-clamp openings) and a 20 psi PRV on the steam jacket. I installed a 0-5 psi pressure gauge on the still(via a tri-clamp opening) and when running stable I'm typically seeing 0.5 psi(or less) in the still. If I run it too hard with full stack condenser cooling enabled and most trays flooded it might hit 0.9 psi which typically results in a 'puke' out the parrot and drop down to 0 psi (Obviously I try to avoid this as much as possible). I have the still PRV discharging into a 10 gallon milk can full of water so if it ever does discharge the vapor should condense immediately. I've never had either the still 5 psi or the steam jacket 20 psi PRV's ever discharge........

I would highly recommend installing a low pressure (0-5 psi) gauge on your still, its cheap, easy and helps you understand how your system operates. I got mine on Amazon for around $30.

I haven't had a chance to read the results from the Silver Trails accident investigation but will....

Link to comment
Share on other sites

Pressurized water is much safer than steam which is what we are dealing with. with pressurized water a leak immediately reduces the pressure. Quite safe. Look up hydro forming. It's forming steel to molds simply using water pressure. Safe since water does not compress.

From working with using air pressure to pump flammable fluids, 55 gallon drums won't handle much pressure, even 5 psi will start to distort the shape. I think even a 15 psi relief is probably too high on most stills.

Link to comment
Share on other sites

Sounds like a good setup with the 5 psi valve.

............. it might hit 0.9 psi which typically results in a 'puke' out the parrot and drop down to 0 psi................). I have the still PRV discharging into a 10 gallon milk can full of water so if it ever does discharge the vapor should condense immediately. ......................

What are you calling "puke", maybe I have the wrong term. I consider it foam containing solids boiling over, but reading your thread I think you mean vapor coming off so fast that it blows the liquid out of your parrot. ??

I think the prv venting into a can of water is a good idea. The steam would make quite a noise as it hit the cold water = audible alarm

Some people have a hose from the still to a deep water container and no PRV. 4 feet water = 2 psi release "valve"

Link to comment
Share on other sites

I would highly recommend installing a low pressure (0-5 psi) gauge on your still, its cheap, easy and helps you understand how your system operates. I got mine on Amazon for around $30.

Got a URL for that? And for the PRV?

Link to comment
Share on other sites

Here is the 0-5 psi gauge, $31, I purchased on Amazon, Kodiak Controls KC25-5# Low Pressure Gauge 5 PSI

http://www.amazon.com/Kodiak-Controls-KC25-5-Pressure-Gauge/dp/B00H9ZW7F0?ie=UTF8&psc=1&redirect=true&ref_=oh_aui_search_detailpage

Here is the 5 psi PRV, $41 Its a basic PRV (typcially used for Steam) from Apollo with a PTFE seal http://www.amazon.com/Apollo-Valve-13-202-Bronze-Pressure/dp/B008837VGW?ie=UTF8&psc=1&redirect=true&ref_=oh_aui_search_detailpage

Also what I have termed a 'puke' is a big increase in distillate output. Its never gotten to the point where its just vapor but it ruins whatever is in my collection can and I have to set that can aside for the next spirit run. The first time it happened it overwhelmed my parrot and I had alcohol coming out the parrot surge vent(=scary). A larger parrot solved that issue.... Now I have a pretty good handle on the issue and it rarely happens. When I finally get my automation system setup I'll set the steam regulator to be automatically dialed back when the inside of the still is roughly >0.7 psi.

Link to comment
Share on other sites

I wrap the threads with PTFE tape and also a thin swab of PTFE pipe dope. Has not leaked in 2 years. Can you clarify what the issue is at less than 20 psi (or any psi)? thanks.

Link to comment
Share on other sites

Hi Smoky Quartz, I am talking about the actual internal seal of the PRV not the threaded portion where it meets the still. EP is a rubbery compound so I worry about fast degradation under ethanol vapor. PTFE seals are known to require more pressure than softer compounds like EP or Silicone, hence the internal PTFE seal can only be used on springs with ratings above 20 PSI on the generant units. I don't want a PRV above 15 PSI because according to the ASME anything above 15 PSI is considered a pressure vessel and should be inspected accordingly.

Link to comment
Share on other sites

Thanks for clarifying, I wasn't aware of the issue. The 5psi Apollo PRV I've been using has a flat PTFE sealing disk and I haven't had an issue with it sealing in the two years its been on the still but I'll be sure to inspect it more often.

Link to comment
Share on other sites

I was looking at the Apollo units as well but no one answered my question as to the material of the body. If it is lead free or not since most of the bodies they sell are brass. The Generant ones come with stainless bodies, which is nice, but the seal seems to be the issue.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...