StillDragon® Community Forum

Welcome!

Be part of our community & join our international next generation forum now!

In this Discussion

BCS Dephlegmator Control

From scanning this forum and others I've heard mention of using the BCS 460/462 as a controller.

Specifically I'm planning on using one to control my dephlegmator. Before I jump in and try to make one on my own, has anyone been able to accomplish this successfully and would like to share their experience?

Comments

  • I have one, it operates a 1/4" solenoid valve on city water, I have a 3/8" ball valve on the outlet side of the dephleg wide open at power up, and I think the SD push-fit valve about 2 turns out. When the vapor temp (I forget if it is the one below the column or below the dephleg) gets to 120F, it opens the solenoid valve and advances to the next stage which is a lower power setting and timer for 30 minutes of heads compression.

    check out this thread.

  • Thanks. I must have missed that thread. Is that just the heat up curve? Do you have one during the run showing the solenoid opening/closing to maintain the vapor temp?

    I'm thinking of using a Jonson Controls modulating (proportional) valve to control the cooling water flowrate based on the outlet vapor temperature. I know they are slow responding, but I hope I can use the PID in the BCS to smooth out the response.

    I'll keep you posted as I start the build, but it probably won't be for a while. I definitely owe it to the site. I feel like I've taken so much and contributed so little...

  • edited October 2014

    I am currently using just a PID and solenoid valve system to control reflux which works well for me, see the attached video, I was taking heads at the time with a reflux temp set at 50c. I get a variance of around half a degree +/-

    Sorry guys the video does not seem to want to upload, tried it twice now and left it uploading for 5 mins each time

  • I've got a BCS 460 as part of my build (nearing completion), but I'm monitoring vapor temp above the dephlegmator and have the option for either DIY controller or BCS controller for the main element in the pot.

    I was going to try using it for neutral runs with varying the power to maximize the take of hearts, but it if doesn't help by monitoring vapor temps, I'll try the dephlegmator output liquid temp monitoring with solenoid valve control... I hadn't thought of that.

    I do love using the BCS to control mash temps as part of my HERMS setup for brewing/mashing.

  • edited October 2014

    Looks like my idea of using proportional valves is out. BCS doesn't have analog outputs. See last item on their FAQ.

    Guess I'll be using solenoid valves as well...

  • edited October 2014

    Solenoid valves will work just fine. Nice thing about solenoids is you can use low cost 2-way solenoid valves to simulate a much more expensive 3-way solenoid valve (one open when the other is closed, and vice versa). This makes managing pump back pressure very easy. Even easier if the solenoid you pick comes in both normally open and normally closed options, just connect them both up in parallel, zero circuitry required. Use the normally open on your condenser line, and the normally closed on the bypass line. If your controller should fail or power down, you'll flood your condensers and not have a catastrophe.

    There are options to take the pulsed DC output (pwm) and convert it to a process voltage, 0-10v, however, the converters are expensive and complicated. Frankly, if the cash was burning in my pocket, I'd just go with a higher quality solenoid valve for the money, or nicer recirc pumps, or a chiller.

    Really only two downsides, if you are flowing a lot of liquid, under high pressure, through a narrow line, you'll have a little bit of water hammer, especially if you have loosey-goosey plumbing, and the clicking of the action (will depend on your duty cycle).

    You can actually combat the water hammer with an arrestor, they make them for household plumbing and other uses. I've never tried it in this use, but I'd wager a bet they'd really cut down on any hammer you might have (it might not even be a problem).

  • edited October 2014

    Would this work for a PWM to 10V?

    Has anyone used one of these before? Do they work well? I'm still feeling that proportional valves are more trouble than their worth...

    Another thing I was thinking was having 4 small solenoid valves (1/4" or 3/8") to allow for 0%, 25%, 50%, 75%, 100%. But it would require a total of 8 valves to allow for the by-pass. There would also be a manual globe valve to allow for the minimum flow at all times.

    I should also mention that this is for a 12" Column.

  • I do not 'control' the vapor temp or dephleg water temp... I just provide auto turn on of dephleg water for heads compression, that is what I was trying to convey... I have found this level of automation to be a huge time and water saver. If you want to have a reasonably fast acting valve control the dephlegmator water, you are getting into some serious process control. This is why suggest using solenoid for full on coolant, and a manual needle valve for 10 runs or so while you design and build your automated RC water flow system... I did buy the Johnson valve and 2-10V/4020mA PID as in another thread, but have not implemented yet. If you want the system to follow an exact temp of either vapor or coolant temp, I feel you have to keep an eye on the system until you are really comfortable that the swings don't give you product that is of lower quality. Not saying it is impossible, just that it could take a lot of time and expense to do something that does not take that much time to do manually.

    Other options for BCS users to experiment with:

    1 - Enter a state where you vary the boiler power controlled by the vapor temp (highly suggest gaining knowledge of tuning PID, as that could be huge cause of frustration when using this method, and if you do get it working, post your PID numbers!)

    2 - use n number of solenoid valves with each valve x, 2x, 4x, flow, and you have 2^n flow rates, you have the process move from state to state to increase or decrease flow...

    3 - use a solenoid in conjunction with a danfoss thermostatic valve or a johnson electric valve

  • edited October 2014

    I wouldn't bother, if you've got PWM, just use PWM to control the solenoid. Yes, the solenoid needs to run either on or off, but with the right solenoid, on and off can be very, very fast, which approximates a proportional flow.

    Are you recirculating coolant or will you be using municipal water supply? This is by far the most important question for sizing your valves. Using "city water", you can generally get by with smaller valves, especially if you have high supply pressure. If you are recirculating coolant from a reservoir, you want larger valves, otherwise they'll be a bottleneck and choke your pumps. I wouldn't want to pump into 1/4" hose, that's like trying to breath through a mixing straw.

    For recirc, I'd go at least 3/4" solenoid valves, this is also the point after which the valves start to get very expensive.

  • edited October 2014

    I don't know BCS at all, so I'm not at all sure how easy or hard it is to tune the PWM output to control a solenoid in a proportional fashion.

    At a minimum you'd need to control for at least the frequency, and the minimum open duration. Most of the solenoids we use can run in 100% duty cycle so you don't need to worry about maximum open.

    Here is an oversimplified example. If your frequency is set for 1hz, 1 cycle per second, it means each second it will adjust the time open (duty cycle) based on the demand being called for.

    So, if you only need 20% flow rate from your valve, each second the valve will be open for .2 seconds, and closed for .8 seconds. If you need 50% flow, it will be open for .5 seconds, closed for .5 seconds. In these cases you'll have two "clicks" each second, one to open, and one to close. If the controller calls for 100% flow, it'll turn the solenoid on for the full second, each second, and you'll near no more clicking until lower demand is called for.

    Sounds great, right? You can go from 0% flow to 100% flow in less than a second. Given the solenoid, you can approximate flow anywhere along the curve. Problem is bigger solenoids can't switch as fast as small ones do, and will have a minimum open time. So in the example above, it might be .2 seconds, which means there is no control between 0-20%. You'd need to use a longer frequency, like 1 cycle per 2 seconds, and then 20% open flow is .4 open, 1.6 closed. Now the temp control goes from being smooth to stepped, and if you had an accurate temp monitor, you'd see some wavering. Now, let's say you need 10% flow, no problem, .2 open, 1.8 closed. But what if you need 5% flow? Sorry Charlie, you hit that darn minimum again, now you've got to slow down the cycling to 1 cycle per 4 seconds, so open for .2 and closed for 3.8. Now you can see, not ideal at all, the speed of the valve is starting to impact the low end, and the slower cycle time is going to start causing more fluctuation in temp. Imagine 50% flow in this scenario, you'll have 2 seconds full blast, 2 seconds full closed. If you've got a big ass condenser, probably not a big problem, but your temp will look like a sine wave. Smaller condenser, bigger sine wave. You can't get away from a scenario where you don't have to do some work to select the appropriate valve, and there is no one-size-fits-all valve. A small 2" still is going to need a very different valve from a 8", or a 12". The proportional valves I'm playing around with now are 1.25". They are big - 12" dephlegmator and 4" x 1meter product condenser.

    The second downside is noise, they click, big ones click loud, if you are pushing the limits looking for the fastest frequency that can be supported by the solenoid, the thing is going to be chattering. Now imagine 4 of them (one per condenser, 1 bypass per condenser). Maybe you don't care about the chattering, in which case no problem. Ever see an old video of ticker tape machines? Or an old newsroom with teletype machines? Very similar noise.

    Lastly, these things may have a finite life being switched as fast and as frequently as we're switching, especially some of the larger solenoids, they just aren't built intended for this use.

  • @DocPorter said: Looks like my idea of using proportional valves is out. BCS doesn't have analog outputs. See last item on their FAQ.

    Guess I'll be using solenoid valves as well...

    @DocPorter said: Would this work for a PWM to 10V?

    Has anyone used one of these before? Do they work well? I'm still feeling that proportional valves are more trouble than their worth...

    Another thing I was thinking was having 4 small solenoid valves (1/4" or 3/8") to allow for 0%, 25%, 50%, 75%, 100%. But it would require a total of 8 valves to allow for the by-pass. There would also be a manual globe valve to allow for the minimum flow at all times.

    I should also mention that this is for a 12" Column.

    The PWM to Voltage/Current converter will work. The accuracy of the converter will play a marginal role in the control of the process variable if you are using a PID controller. If the PID is asking for an output of 40% and the PWM converter is only giving 3VDC (30%), the PID will see the temperature is still rising and simply ask for more output (more cooling), PWM and hence voltage/current will continue to increase its output until the input is satisfied. The PID controller is actually a great little object in correcting inaccuracies on the output side of the control process. That's it's job, to constantly monitor the input and adjust the output accordingly. The same goes if the PID settles on say an output of 60% and the cooling water starts to increase in temperature. The process variable will start to rise and the PID will simply ask for more output. It's more important to get the process output control (valve action) linearised than it is to get an accurate position of PID output to valve position.

Sign In or Register to comment.