Running a vodka column with a Danfoss valve

[Golden Beaver Distillery]

4 hours ago, Southernhighlander said:

Golden Beaver,

I believe that Silk and Whiskey Tango's advice is spot on, especially Silks advice concerning your high pressure and backing off the coolant flow.  With that kind of pressure and the size of your lines you have far more flow than you need in this situation.   I rarely run stills anymore so the insight of people that run stills everyday can be really helpful in a situation like this.

We have run it just as suggested above several time to date.  We can balance the plates and start a flow at 180 proof. The minute you engage the bypasses/Danfoss  and try to increase the proof, the top plate floods or production is diminished to a trickle with no increase in proof.

We stepped on the input water pressure today and got better control but proof never got higher then 184.  We're starting with doubled distilled distillate at 100 proof. 

Going to put on a pressure regulator with gauge on the cooling water this afternoon so we have an actual PSI measurement. This should help allow the gate valve to have a higher range of adjustment. 

I think the output volume of an 8-inch column is too low to use the Danfoss. Just not enough BTU moving through the system to justify the functionality of the Danfoss.

[whiskeytango]

4 hours ago, Golden Beaver Distillery said:

We have run it just as suggested above several time to date.  We can balance the plates and start a flow at 180 proof. The minute you engage the bypasses/Danfoss  and try to increase the proof, the top plate floods or production is diminished to a trickle with no increase in proof.

 

when you do engage the danfoss by closing the bypass do you do it nice and slow?  I mean like slow in steps shut the valve 1/4 way let the still run for 5 minutes close it to 1/2 run for 5 more then 1/4 then just cracked open?   

Its hard to really trouble shoot without seeing it in action so some of our comments might be way off base but just tossing out all the posable options.   

 

 

[Southernhighlander]

5 hours ago, Golden Beaver Distillery said:

We have run it just as suggested above several time to date.  We can balance the plates and start a flow at 180 proof. The minute you engage the bypasses/Danfoss  and try to increase the proof, the top plate floods or production is diminished to a trickle with no increase in proof.

We stepped on the input water pressure today and got better control but proof never got higher then 184.  We're starting with doubled distilled distillate at 100 proof. 

Going to put on a pressure regulator with gauge on the cooling water this afternoon so we have an actual PSI measurement. This should help allow the gate valve to have a higher range of adjustment. 

I think the output volume of an 8-inch column is too low to use the Danfoss. Just not enough BTU moving through the system to justify the functionality of the Danfoss.

Golden Beaver,  I'm very sorry that you are having this issue.  I think you said your still is 150 gallons.  I'm thinking that with your high water pressure and the small size of your still, we may need to trade out your current thermostatic valve with a smaller one.  I believe that your still is the smallest that we have sold that is using the Danfoss valve that we normally use.  I will have Susan pull your order for me first thing in the morning and send it over to me. I will have one of my engineers take a look at things in the morning to see if we need to send you a smaller thermostatic valve with less throughput. I will get back to you by email and then we can talk on the phone.  I think I and my staff can help you better that way.

Concerning your proof.  How many plates are you running in how many columns?  If you are running multiple columns you will need to pump the condensate from the bottom of the last column in line to the top of the column before it to get full functionality.  It takes at least 16 vertical plates to maintain 190 proof and above or the equivalent number of theoretical plates in a packed column, or multiple columns with the same number of plates using condensate pumps, However even with 4 plates you can usually reach 190 at the beginning of the run.  If you have never ran plate columns before, or if you have never ran plated columns with this coolant plumbing design we should set up face time with my distiller so that he can show you how to run the system.   You get unlimited phone and email support with your purchase.   I can also set you up with a free distilling workshop at one of my customers distilleries, if you like.  Also my people are here to help you, so you should take advantage of that.  You can call me directly at 417-778-6908.

Thanks.  

 

 

[Golden Beaver Distillery]

18 hours ago, Southernhighlander said:

Golden Beaver,  I'm very sorry that you are having this issue.  I think you said your still is 150 gallons.  I'm thinking that with your high water pressure and the small size of your still, we may need to trade out your current thermostatic valve with a smaller one.  I believe that your still is the smallest that we have sold that is using the Danfoss valve that we normally use.  I will have Susan pull your order for me first thing in the morning and send it over to me. I will have one of my engineers take a look at things in the morning to see if we need to send you a smaller thermostatic valve with less throughput. I will get back to you by email and then we can talk on the phone.  I think I and my staff can help you better that way.

Concerning your proof.  How many plates are you running in how many columns?  If you are running multiple columns you will need to pump the condensate from the bottom of the last column in line to the top of the column before it to get full functionality.  It takes at least 16 vertical plates to maintain 190 proof and above or the equivalent number of theoretical plates in a packed column, or multiple columns with the same number of plates using condensate pumps, However even with 4 plates you can usually reach 190 at the beginning of the run.  If you have never ran plate columns before, or if you have never ran plated columns with this coolant plumbing design we should set up face time with my distiller so that he can show you how to run the system.   You get unlimited phone and email support with your purchase.   I can also set you up with a free distilling workshop at one of my customers distilleries, if you like.  Also my people are here to help you, so you should take advantage of that.  You can call me directly at 417-778-6908.

Thanks.  

 

 

I tried calling you but no answer. 

Making adjustments to our cooling circuit to slow the flow down.

I have run a column before - you spec'd this system off our need to upgrade and increase our vodka production.

Will try calling you again tomorrow.  Thanks for being proactive on the Danfoss swap out.

Kris

 

[Southernhighlander]

Golden Beaver.

I left you a phone message in the early afternoon.  I apologize for missing your call.  I also apologize for not understanding who you were on the forum and what you ordered at first.  I remember people names better than distillery names.  

Your still will do Vodka.  I look forward to your call in the morning.  

 

Thanks.

[Golden Beaver Distillery]

 

Huge thanks to all that gave advice - special thanks to @Silk City Distillers for his time and @Southernhighlander for making it right with some additional parts. 

More tweaking will have to be done with the new parts coming in but I'm confident we can get better output rate and even higher proof.

[jocko]

I know this problem appears to resolved, but I'm putting together the same thing right now.  I have the Danfoss valve and and assembling necessary fittings. 

It occurs to me that, based on Paul's diagram, the Danfoss probe may not be entirely submerged in water coming off of the condenser, particularly if the flow is low.  It seems that there is a strong possibility for air being trapped near the probe, and water simply flowing by/on the probe, not submerging it.  The Danfoss instructions say that the probe needs to be submerge as much as possible in the liquid.

I would think that a trap just after the probe that would force the water up on the probe before leaving would gaurantee excellent constant contact of the probe with the water no matter how small the flow is.  Unfortunately, I am unable to upload photos for some reason, or I'd post my diagram of what I am thinking of. 

Make sense?  Thoughts?

[Southernhighlander]

Jocko,

The problems that Golden Beaver had were due to the wrong size Danfoss valve.  As long as the Danfoss valve and probe are plumbed according to the diagram and the bypass at the Danfoss stays cracked just a little, there are zero issues because the danfoss valve's probe is always in constant contact with the coolant flow.  This being the case there is no reason to change anything.  We have over 100 customers using Danfoss valves plumbed according to the diagram that we supply and they work great.  Also this same plumbing design is used by our competitors with good operational functionality.    

[Golden Beaver Distillery]

1 hour ago, jocko said:

I know this problem appears to resolved, but I'm putting together the same thing right now.  I have the Danfoss valve and and assembling necessary fittings. 

It occurs to me that, based on Paul's diagram, the Danfoss probe may not be entirely submerged in water coming off of the condenser, particularly if the flow is low.  It seems that there is a strong possibility for air being trapped near the probe, and water simply flowing by/on the probe, not submerging it.  The Danfoss instructions say that the probe needs to be submerge as much as possible in the liquid.

I would think that a trap just after the probe that would force the water up on the probe before leaving would gaurantee excellent constant contact of the probe with the water no matter how small the flow is.  Unfortunately, I am unable to upload photos for some reason, or I'd post my diagram of what I am thinking of. 

Make sense?  Thoughts?

Jocko,

It works. Took a leap of faith on my part, but it works as designed.  Sometimes we can over think this stuff before we actual see it in action.

Kris

[jocko]

I can now see from your photos that your setup does force contact because after the probe the line rises above the probe.  

[Silk City Distillers]

Trapped air is more of a function of flow rate and air bubble entrainment on the cooling side.  For most of us running cooling loops, the fact of the matter is that most "trapped air" simply gets carried out by the liquid flow.  As long as your not pumping through liquid filled with bubbles, it's not likely that those areas will stay air-filled for very long.  This isn't necessarily something that needs to be "trapped", even if you are running a system that lets the coolant naturally drain out after use.

What you can't do, is just put a hose on the end of that Danfoss adapter tube, and just let it drain to the ground (I don't think this is what you are suggesting at all, but would be a scenario where you *would* need to trap).

 

[jocko]

That is the exact area that I was concerned about. Especially if you are really cutting back on flow.

[Silk City Distillers]

Honestly, just open that bypass and introduce some higher, more turbulent flow, and it'll take care of that.  As long as you have no plumbing leaks (where air can be sucked in when not in use), it's unlikely to build up much air over time.  Higher flow rates will grab that air as little bubbles and pull it out.  Absolutely nothing to worry about re-engineering.

[Silk City Distillers]

I know a couple of guys that have mounted that Danfoss adapter in the opposite direction, with that bottom end attached to the condenser, and using the current input port as the output, facing straight up.

In every case, they did it that way because their reflux condenser was higher than the product condenser, and lining it up that way saved a bunch of plumbing fittings.  Position isn't universal though, for example, the most common probe type, you can't install it with the line going downwards, so do keep that in mind.  It can be vertical (line upwards) or horizontal.

[Southernhighlander]

4 hours ago, jocko said:

That is the exact area that I was concerned about. Especially if you are really cutting back on flow.

There is no reason for concern. As long as the bypass at the valve body is open a little, there are no issues. If there had been any issues, we would have changed the design years ago.  Redesigning won't gain you anything. 

[Alex_Sor]

16 hours ago, Southernhighlander said:

There is no reason for concern. As long as the bypass at the valve body is open a little, there are no issues. If there had been any issues, we would have changed the design years ago.  Redesigning won't gain you anything. 

I agree with Paul

In this design (in the picture), you need to worry more about the passage sections so that the water flow does not receive a lot of hydraulic resistance. It's cooling water, it doesn't go fast enough to create a lot of turbulence.

The Danfoss sensor itself is just a tube filled with gas inside, the expansion of the gas leads to a deflection of the pressure gauge needle. If the flask of the sensor itself is iron and is washed with water at 1/2 or 2/3 of the height, then I don’t see any problems. The thermal conductivity of the sensor tube will distribute heat throughout the tube.