Controling Elements and Reflux Condensor

Hi guys,

I'm configuring my 5'' Dash column next week. It's heated with 3 x 5500W elements. I have 3 PID controllers (Elitech STC1000) that were used in a brewing process before that I'm going to use to control the heating elements. Is this a better option than controlling two manually and using the DIY Controller on the third element?

In terms of the reflux condenser I would like to add some automation to it. I have attached the diagram of the process flow. I would be very grateful for any feedback. Essentially I'm recirculating water from a 1300 litre buffer tank for both the reflux condenser and and product condenser. The product condenser will be always open. The reflux condenser water flow will be controlled by a PID controller, PT100 thermocouple and a solenoid valve on the water out hose from the reflux condenser.

image

Comments

  • edited March 2016

    You can use PIDs on the elements, but only if you are using them for start-up automation. For example, you set the set points to 180f or so, to shut down the elements at 180f, leaving only the single manual control element operating. Otherwise, I'm sure you know, temperature control of the boiling liquid won't be useful. I'll say,it's nice to see the wash temperature during heat up too.

    Why not use the extra one for your recirculation tank? Setup an overflow to the drain, and use a solenoid on your water supply to feed the tank. It will provide a bit of automation for you on a hot day where maybe your tank starts getting too hot. It's not necessary, but if you have an extra PID and solenoid, why not?

  • Grim. Thank you for the advice. I will use the DIY controller with one element and set the other two to switch off at say 180f. When you mention using a solenoid on the water supply, is that still to control the temperature in the reflux condenser? Is the solenoid, PID and thermocouple reliable for keeping the temperature of the reflux condenser constant?

  • No no, just to dump water straight into your holding tank. Let's say you set your tank to 80F - once you have 81F - the PID opens the solenoid - dumps colder water into your holding tank, warmer water overflows, and brings your tank temp down to 80F. Will it trigger every time? Nah, probably not.

  • I would use a PID with alarm contacts to turn off the other two elements (at 195 for wash and 170 for low wines)

    I would use a single centrifugal pump like a march 809/815 to pump the cooling water to lower port of both the condensers, RC and PC.

    I would restrict the PC with a danfoss thermostatic valve and have it spit out 50-60C water. Since you will likely not have a way to get the probe IN the condenser, you will have to put it on the outflow, and therefore will need a small bypass needle valve or orifice.

    Thermostatic Valve Installation on Product Condenser

    I would restrict the RC with a proportional motorized 2-way valve that runs on 2-10v or 4-20mA and has the sensor IN the RC (a three way valve may work better, but a centrifugal pump is designed to have flow restricted)

    Automating Reflux Condenser Water Flow

    I would get 'digitemp' meters, 1/4" for PC and 1/2" or 3/8" for the RC to monitor flow.

    I use this on a production still in a production distillery, a 380L, 5 element, 27.5KW, 8" system with a 1000L tank, and I can get 1.5 stripping runs out of the tank before it gets hot, and it cools in 2 days (I have several tanks and can run 100% on city water if needed by replacing the pump with a solenoid valve.. (yes the pressure changes and the cV value is way out, but it WORKS, I do not care if it should or shouldn't, but it WORKS with both 10 and 60PSI)

    I will add a radiator and fan to cool the 50-60C water to 30C with outside air...

  • Just a note of caution the STC 1000's we see in Australia are rated for 10 amps normally, i did once see one that was 16 amps but they are uncommon, some are even 5 amps. Not something i would use to control a 2400 watt element really, let alone a 5500 watt one, they are meant for controlling fridges and light heaters.

    StillDragon Australia & New Zealand - Your StillDragon® Distributor for Australia & New Zealand

  • Never control a load with a controller directly, always relay.

  • I'd put that return flow line around the pump and include a manual throttling valve on it.

    I assume it's purpose is so you don't 'deadhead' the pump?
    Without the manual valve to throttle back you have no usable pressure and a lot more cold bypass than you need just to protect the pump. You would only set it up so there is enough bypass flow to keep the pump happy when the auto valve is closed then pull the handle off and forget about it.
    Is it a mag drive pump? You might not need it at all?

    With the line feeding into the return to the tank the way it is now you'll have a lot of cold water mixing with the hot.
    So shit loads of warm water returning to the tank.
    I'd aim to a small volume of very hot water returning to the top of the tank so you set up a nice thermocline and your feed water temp stays nice and consistent until the tank is spent, heat wise.
    Some sort of hot supply water alarm would be a good idea in these setups too but is often overlooked.

    I'd do the probe and valve differently as well but what you have seems to be the SD standard so I won't enter that one again.

    For the elements I'd have two on a switch and control the third. So all three on heat up and a single 5.5kW through the run would be fine IMO.
    You could control all three of the same controller easy enough but 100% would give you 16.5kW on a 5". Too much?
    How you go about ti is up to you. There are heaps of options, just find the one that suits you.

    If you going to use a controller to look after an unattended heat up stage I would include some sort of double redundancy. There is huge potential for a colossal fucup on that.
    Something as simple as a miscalculation on the BP could result in coming to work to a smoldering mess.

  • Great information @CothermanDistilling. Thank you very much @punkin and @jacksonbrown for the advice. I will be using the water that was cooling the RC and PC to brew so if the temperature goes up it's not a problem, I will empty the cooling tank to the hot liquor tank and refill with municipal water.

    I'm based in Ireland so warm weather isn't usually a problem even in the summer.

    A number of questions to clarify some bits. I already have a booster pump on the tank with cooling water. Could I 'T' the flow pipe from that and use it as a feed for the RC and PC? I would have a ball valve or gate valve on the flow to each condenser to stop the flow when not in use. The pump is fitted with an automatic cut off when it reaches 4 bar pressure.

    Is it better to place a thermocouple within the RC or on the outflow pipe? I have read somewhere it's better on the outflow pipe.

    I actually have a used actuator with 1/2 bsp fittings and as I mentioned previously I have 3 PID controllers. Would one of the PID controllers mentioned be capable of controlling the actuator? Should I plumb the actuator on the in flow or the outflow of the RC?

    Regarding controlling the elements I may go with on/off switches for two and the manual controller on the third.

    I will post pictures of when it's all setup.

  • @pogriallais said: Great information CothermanDistilling. Thank you very much punkin and jacksonbrown for the advice. I will be using the water that was cooling the RC and PC to brew so if the temperature goes up it's not a problem, I will empty the cooling tank to the hot liquor tank and refill with municipal water.

    that is a great idea, and you will be able to get nearly to your strike temperature and save a lot of time and energy!

    @pogriallais said: A number of questions to clarify some bits. I already have a booster pump on the tank with cooling water. Could I 'T' the flow pipe from that and use it as a feed for the RC and PC? I would have a ball valve or gate valve on the flow to each condenser to stop the flow when not in use. The pump is fitted with an automatic cut off when it reaches 4 bar pressure.

    yes, I feed both, and you should be able to do so. So that is NOT a centrifugal pump, but with the 4 Bar pressure switch, you can 'deadhead' it. True, you may run into some controlability if you pressure switch turns the pump off at 4 Bar and back on at 3 Bar, but maybe an expansion tank and possibly a regulator would fix that..

    @pogriallais said: Is it better to place a thermocouple within the RC or on the outflow pipe? I have read somewhere it's better on the outflow pipe.

    If the probe is not within the RC, it will not be able to detect changes as well with low or no flow, I elude to this in my description of the RC and PC implementation.

    @pogriallais said: I actually have a used actuator with 1/2 bsp fittings and as I mentioned previously I have 3 PID controllers. Would one of the PID controllers mentioned be capable of controlling the actuator? Should I plumb the actuator on the in flow or the outflow of the RC?

    unless that PID can do 4-20mA or 2-10v or 0-10V output, No, you will need a new PID.

  • Thanks for this. Great ideas

  • edited March 2016

    just my opinion

    image

    column.jpg
    800 x 517 - 40K
  • I think it makes sense Frodo. Are there any advantages to this method versus the previous?

  • Just to keep you updated. I spoke to a guy that does automation for the pharmaceutical industry. He has done some beverage plants too. He recommended an actuated lift and lay valve- it opens proportionally. He also recommended a PID controller with an analog output. Cotherman you were right on the money with the PID controller you mentioned. 4 -20 mA output. Frodo - would the thermocouple not be better placed within the RC?

  • That won't work if there is zero flow but it is a poor solution to the problem IMO. Your mate is on the right track with the prop valve.
    Set the limits on the valve so it never shuts completely or set up a trickle bypass.
    Best solution is the hi flow recirculating systems.

  • not much, i think this answers his questions on probe placement and actuator placement.

    pump needs to pump into tank, not away from tank

    the probe needs to be on the return line,

    theory is. the supply line to column is X temp.. it will heat to Y temp as it circulates thru column. take the reading on the return side at Y. to tell the pump to circulate , changing Y to X temp as water circ's

  • Having all the pump work on the suction is a bit odd. I'd return hot to the top of the tank too, not the bottom.

  • I get what the advantage of the constant flow is - the temperature of the RC jacket is more consistent. Therefore it's quicker to react to temperature changes as it's not relying on convection and/ or a low flow rate.

  • @jacksonbrown said: Having all the pump work on the suction is a bit odd. I'd return hot to the top of the tank too, not the bottom.

    I do it that way, because being a plumber/hvac tech we install the circulating pumps in that configuration. as per mechanical engineer drawings all the time. what is the math/calculus behind the reason ? I really do not know. LOL..good question, jacksonbrown. Maybe I should fined out. Have never been asked WHY?

  • Ask your mate about NPSH. Pulling though a control valve and what else will cause cavitation. Always throttle the outlet.
    To suck the max deltaP is limited by vapour pressure but the other way is practicaly infinite. Something like that anyway. Especially with hot water, the reduced pressure will boil in the pump and stop it working.

  • @pogriallais said: I get what the advantage of the constant flow is - the temperature of the RC jacket is more consistent. Therefore it's quicker to react to temperature changes as it's not relying on convection and/ or a low flow rate.

    that is why we put the probe in the RC, not on the outlet flow...

    and like others say, always throttle on the outlet...

    My personal preference is to use a 2-way valve... let me explain why... With a 3-way as shown, and @ 4gpm, I would have a 240 gallon tank of uniform 120 degree water in an hour and have to stop.... my PC output would be 120... no good! With a 2-way valve on each the PC(mechanical danfoss) and RC(sestos w/ johnson), I am running .8-1.2 gpm total between the RC and the PC and after 1 hour I have a tank with a layer of 150-160 degree water on top and room temp water on the bottom...

    sure, I could use a 2nd tank and a 2nd pump... but why? I have over 50 runs on this setup since I last changed anything on it, and it is idiot proof... my BCS can shut the system off if the product temp is over 90, which means my cooling tank is depleted, and I move the hoses to another tank..

Sign In or Register to comment.