Thanks guys. Lloyd you have hit on how I am going to do mine when I get it. The reason I am doing it this way is I have the valves, controllers, fittings etc already. And this is how condensers have always been piped in systems I have worked on. But I am not disagreeing with you. I have plenty of components, controllers, plc's, sensors and valves to muck around for ever and see what works best for me, the other reason I am doing it. I also have some stainless steel air bleed valves (I might mount on top of dephleg) and some non return valves so I can try both top and bottom fed, vapour and condenser water control and see how it goes. Thanks for your help guys. I have to get punkin to put sensor points everywhere for me.
Ahhh, sorry when you said bypass I thought you were talking about a 3 way valve B-AB bypass like in my drawings. I have some stainless steel weld in air bleed valves, I might put one in anyway, it can't hurt.
I'm currently experimenting with a proportional valve and controller. The mixing valve approach is interesting, since it would keep flow rates in the reflux condenser high.
I'm using a Johnson Controls valve (VA9104-GGA-2S) and a matching Omega controller (CN4316-F1-R2).
A couple of findings - system works well in maintaining reflux condenser temp regardless of conditions, as coolant reservoir temp progressively increases, the valve opens wider to compensate.
Nice thing is on startup, when coolant temps are low, the flow rate needed to maintain temps is very slow - which might be an interesting option for those using public water supply and looking to reduce consumption. This is why I want to monitor the coolant in the reflux condenser, versus measuring temps at the outlet. When the supply water is very cold, the valve can actually shut down, resulting in a situation where there isn't sufficient flow to move the effluent past the sensor. What would happen is natural convection would move the water, up the the sensor, it would trip, valve open, but then overshoot. I suspect another easy fix would be to adjust the controller such that the valve never fully closed - but this would use more coolant.
I know a few others are using solenoids with pulse modulation - the clicking would drive me insane, but the impact is similar.
Also have a Danfoss, I couldn't help but pick one up to play with, but mounting that "sensor" is a PITA unless you build a separate housing (like you did). Or, you are a large German still manufacturer and you can engineer your condensers to accept it through a fitting.
I'm just using a tiny NPT thread stainless RTD sensor, I think the probe itself only extends out 3/4 past the threads. I don't have my hands on the big boy 8" condenser yet, but finding a 1/2 NPT that will thread into one of the top (or bottom) fittings shouldn't be any problem at all.
The AVTA would take the cake if the condensers had an appropriate fitting and the tubes were spaced in such a way that it could accept it horizontally.
Agree 100% when we get 12" condensers, we will make sure @Lloyd puts in ports for AVTA's ;-)
I just happened to look for the above mentioned VA9104-GGA-2S on eBay, and found 2 for $40 each with free shipping... comes with a 2-way valve, cannot beat the price until Lloyd sources them for us... ($200 on grainger with no valve)
Oh, @grim - I have to make another AVTA sensor housing or two in the next few weeks, if I have enough parts and material laying around, I will make an extra...
The 2 drawings I put up are diverting circuits, they divert water away from the condenser and would be better suited to controlling by vapour temp, when the controller is not calling for any cooling no coolant flows through the condenser. A mixed flow is below, constant flow, constant temperature in the circuit and better suited to condenser water control. Its a bit trickier to pipe up though as the pump must be in the circuit to maintain the constant flow. I am going to try both ways and see what happens. Also with a constant flow it does not really matter if you have flow to the top or the bottom since the amount of flow through the condenser is equal to pump flow, which is probably going to be heaps.
You can also use a 2 way modulating valve but the circuit would require some pressure relief at the pump. That's the good thing about 3 way modulating valves, you maintain a constant flow through the circuit regardless of whether you pipe it for mixing or diverting.
The D1S are great little controllers and very cheap at $31. They have the added ability of being able to control the valve manually. I have used them for years but I am going to use the ones @olddog sourced at work for a while so I have a few of these to play with, unfortunately no manual control on them.
OK, received the Johnson valves($40) and the sestos PIDs($29.99), looking at enclosures... thinking about just keeping the reflux condenser coolant operation separate from everything else in it's own box on the wall since it is 4-20mA, and runs on 24vAC where the product condenser is on the danfoss valve and the boiler power is all high voltage stuff... not sure...
Yeah I got my test valve from that guy, and was going to pick up the other two, but I see you posted the eBay link and picked them up. They are a good deal for $80, a killer deal for $40. Just be careful, these aren't spring return open, so when the power cuts, they stay where they are. When you design for failure assume the valve is fixed closed.
That's a good price for the Johnson valve. Was it valve and actuator? I paid $100 for my Johnson valve and actuator and thought that was ok. L-) Also 4-20ma. I like Auber instruments. Got my RTD's there.
If you are going to go with the auber box, just get the one with double hole, you can use the second valve to control the product condenser flow in exactly the same manner. Might as well do them both at the same time. Or, just add another temp probe above the reflux condenser (vapor temp) and feed it to a standalone DIN temp meter, you'd at least have all the reflux temps and controls in one unit.
Just one more note, you need a pump that can handle high head with this valve, the orifice is small, it's not 100% through. A little pond pump is going to choke. Otherwise, city water will have enough pressure to work.
Easier to just use solenoid valves if you've already got a pwm controller that allows you to configure the duty cycle. It'll click, which is the downside, but the upside is that it's much faster - the 9104 takes a minute to swing fully open to fully closed. However, once the condensers start to get large, the thermal mass makes them slower anyhow. During a run it's only making slight changes to position.
I used to work for Johnson's in Adelaide, I still have all my manuals etc. Checked that valve and if its the one I think it is, it's designed for a system pressure of 250-400kPa, hence the little hole in the in it. Take @grims advice and use a pump that has a shit load of pressure more so than one that has a lot of flow. There should be a brass tag around the valve body with a full part number on it, send it to me and I can tell you the CV of the valve and work out a pressure range that it will work around. Also the valve should be piped on the outlet of the dephleg, not the inlet, the dephleg should be under system pressure.
@CothermanDistilling said:
If I wanted my BCS to control these valves, I think I could get one of these ~$100 dual channel converterd that can be DIN rail mounted...
It's a pity the BCS doesn't have analogue outputs. But yes the PWM to Analogue converter will do what you want. I buy lots of control gear from plccenter, good blokes.
@Mickiboi said:
Also the valve should be piped on the outlet of the dephleg, not the inlet, the dephleg should be under system pressure.
I'd never heard that, why is that?
As for the tag, mine's got - VG1241AD-9A4GGA
I suspect if you hunt around you can find a larger 1" valve to replace the 1/2" valve and be able to deal with a bit less head. Something like this guy here:
Is there some comparison page/info for all these different actuators? Seems there are some older ones and newer ones. What's the difference between VG1241AD-9A4GGA and VA9104-GGA-2S for example?
@grim said:
Also the valve should be piped on the outlet of the dephleg, not the inlet, the dephleg should be under system pressure.
I'd never heard that, why is that?
As for the tag, mine's got - VG1241AD-9A4GGA
I suspect if you hunt around you can find a larger 1" valve to replace the 1/2" valve and be able to deal with a bit less head. Something like this guy here:
The valve has a CV of 1.2 which means it is designed to work under low pressure/low flow with a low pressure drop across the heating/cooling coil (in our case the dephleg). The lower the CV the lower the flow through the valve. As a guide you lose a bit of control with a low CV/low flow setup. This is why 3 way diverting valves are a more popular choice as they tend to be less pressure dependant. The other reason is with a 3 way valve the pressure remains constant no matter what position the valve is in as the the flow in will always equal the flow out, therefore the only difference in pressure will be what the dephleg introduces into the system, which should be bugger all. With a 2 way valve, as the valve opens the pressure drop across it drops, as the pressure drops you get less control across the flow curve.
Have a look at this pdf page 4, 2 way piping package code 24.
The valve is designed to control under pressure after the coil. The pressure drop across coil helps to smooth out the flow curve, makes it more linear across the control range. It also stops trickle through/trickle down which can make the reaction/capacity of the coil erratic. You'll find it will just work better with the coil (dephleg) under pressure.
If you want to use modulating valves then I suggest you get a 2 or 3 way diverting/mixing ball valve, these have a CV generally highr than 7. They are just about a straight through , 1/2" thread and 1/2" open ball internally, and would work much better on the dephleg, the dephleg has virtually zero pressure drop across it. You would have to put a lot of water through the dephleg to get a decent pressure drop in order to use a low CV valve. Its not really suited to a control valve application. Basically anything that has a medium CV will be suitable. So @grim this valve might be OK for what you want to do, pipe it up and see what happens, you will be suprised how much water comes through that little hole even with a pond pump of reasonable size.
edit. if you order a Johnson valve/actuator combination go for VG1241AG, VG1241AL or VG1241AN. These might be better suited if you have really low supply pressure the AN having the larger opening/higher CV.
Valve position vs flow (At 60psi municipal water pressure, full open) - 4 meters total hose (2 in, 2 out), 3/8th ID with push-fit fittings, zero head (with municipal mains it's probably irrelevant anyway).
At this higher PSI, the flow curve appears to be very nicely sloped at the lower end 0-50%, very good control of the flow rates, but flattens out above 50%. There is some variability there, since this was done by filling a big 2 liter graduated cylinder with one hand, and a stopwatch in another.
I've got an old mag drive pond pump around here that is rated for 2600 liters/hr, I'll plumb that up since that might be more typical of a small pumped system. I could always try the my big pump, a Taco 3 speed recirc pump that has a 20 foot head, this would be more typical in a large system.
Comments
Thanks guys. Lloyd you have hit on how I am going to do mine when I get it. The reason I am doing it this way is I have the valves, controllers, fittings etc already. And this is how condensers have always been piped in systems I have worked on. But I am not disagreeing with you. I have plenty of components, controllers, plc's, sensors and valves to muck around for ever and see what works best for me, the other reason I am doing it. I also have some stainless steel air bleed valves (I might mount on top of dephleg) and some non return valves so I can try both top and bottom fed, vapour and condenser water control and see how it goes. Thanks for your help guys. I have to get punkin to put sensor points everywhere for me.
Air bleed valve, that's it! Thanks @Mickiboi. Couldn't seem to grab it til you mentioned it.
Hate when something so simple can't be recalled....
Going to start writing my name in my underwear now, just in case. Llo... dammit, how do you spell that...
Ahhh, sorry when you said bypass I thought you were talking about a 3 way valve B-AB bypass like in my drawings. I have some stainless steel weld in air bleed valves, I might put one in anyway, it can't hurt.
I'm currently experimenting with a proportional valve and controller. The mixing valve approach is interesting, since it would keep flow rates in the reflux condenser high.
I'm using a Johnson Controls valve (VA9104-GGA-2S) and a matching Omega controller (CN4316-F1-R2).
A couple of findings - system works well in maintaining reflux condenser temp regardless of conditions, as coolant reservoir temp progressively increases, the valve opens wider to compensate.
Nice thing is on startup, when coolant temps are low, the flow rate needed to maintain temps is very slow - which might be an interesting option for those using public water supply and looking to reduce consumption. This is why I want to monitor the coolant in the reflux condenser, versus measuring temps at the outlet. When the supply water is very cold, the valve can actually shut down, resulting in a situation where there isn't sufficient flow to move the effluent past the sensor. What would happen is natural convection would move the water, up the the sensor, it would trip, valve open, but then overshoot. I suspect another easy fix would be to adjust the controller such that the valve never fully closed - but this would use more coolant.
I know a few others are using solenoids with pulse modulation - the clicking would drive me insane, but the impact is similar.
man... that stuff makes a danfoss valve look really simple ;-) but I do like it!
Also have a Danfoss, I couldn't help but pick one up to play with, but mounting that "sensor" is a PITA unless you build a separate housing (like you did). Or, you are a large German still manufacturer and you can engineer your condensers to accept it through a fitting.
I'm just using a tiny NPT thread stainless RTD sensor, I think the probe itself only extends out 3/4 past the threads. I don't have my hands on the big boy 8" condenser yet, but finding a 1/2 NPT that will thread into one of the top (or bottom) fittings shouldn't be any problem at all.
The AVTA would take the cake if the condensers had an appropriate fitting and the tubes were spaced in such a way that it could accept it horizontally.
Agree 100% when we get 12" condensers, we will make sure @Lloyd puts in ports for AVTA's ;-)
I just happened to look for the above mentioned VA9104-GGA-2S on eBay, and found 2 for $40 each with free shipping... comes with a 2-way valve, cannot beat the price until Lloyd sources them for us... ($200 on grainger with no valve)
Oh, @grim - I have to make another AVTA sensor housing or two in the next few weeks, if I have enough parts and material laying around, I will make an extra...
The 2 drawings I put up are diverting circuits, they divert water away from the condenser and would be better suited to controlling by vapour temp, when the controller is not calling for any cooling no coolant flows through the condenser. A mixed flow is below, constant flow, constant temperature in the circuit and better suited to condenser water control. Its a bit trickier to pipe up though as the pump must be in the circuit to maintain the constant flow. I am going to try both ways and see what happens. Also with a constant flow it does not really matter if you have flow to the top or the bottom since the amount of flow through the condenser is equal to pump flow, which is probably going to be heaps.
You can also use a 2 way modulating valve but the circuit would require some pressure relief at the pump. That's the good thing about 3 way modulating valves, you maintain a constant flow through the circuit regardless of whether you pipe it for mixing or diverting.
Done properly with a good PID controller I could just about guarantee you +/- 0.1degC with zero fluctuation.
I see the sestos D1S-CR-220 on eBay for $31, but probably worth the $69 of the Omron...
The D1S are great little controllers and very cheap at $31. They have the added ability of being able to control the valve manually. I have used them for years but I am going to use the ones @olddog sourced at work for a while so I have a few of these to play with, unfortunately no manual control on them.
OK, received the Johnson valves($40) and the sestos PIDs($29.99), looking at enclosures... thinking about just keeping the reflux condenser coolant operation separate from everything else in it's own box on the wall since it is 4-20mA, and runs on 24vAC where the product condenser is on the danfoss valve and the boiler power is all high voltage stuff... not sure...
Wall Mount Box for Single Controller @ Auber Instruments
Yeah I got my test valve from that guy, and was going to pick up the other two, but I see you posted the eBay link and picked them up. They are a good deal for $80, a killer deal for $40. Just be careful, these aren't spring return open, so when the power cuts, they stay where they are. When you design for failure assume the valve is fixed closed.
That's a good price for the Johnson valve. Was it valve and actuator? I paid $100 for my Johnson valve and actuator and thought that was ok. L-) Also 4-20ma. I like Auber instruments. Got my RTD's there.
Also, don't forget you need the 500 ohm resistor across the control lines using 0-20ma (black/gray).
On mine the control wires are gray and orange and the resistor is in series on the gray.
On the VA9104-GGA-2S I've got, the Orange is position feedback.
If you are going to go with the auber box, just get the one with double hole, you can use the second valve to control the product condenser flow in exactly the same manner. Might as well do them both at the same time. Or, just add another temp probe above the reflux condenser (vapor temp) and feed it to a standalone DIN temp meter, you'd at least have all the reflux temps and controls in one unit.
Just one more note, you need a pump that can handle high head with this valve, the orifice is small, it's not 100% through. A little pond pump is going to choke. Otherwise, city water will have enough pressure to work.
If I wanted my BCS to control these valves, I think I could get one of these ~$100 dual channel converterd that can be DIN rail mounted...
DUAL PWM TO 10V ANALOG CONVERTER @ PLCCenter
DUAL PWM TO 20MA ANALOG CONVERTER @ PLCCenter
any thoughts?
Easier to just use solenoid valves if you've already got a pwm controller that allows you to configure the duty cycle. It'll click, which is the downside, but the upside is that it's much faster - the 9104 takes a minute to swing fully open to fully closed. However, once the condensers start to get large, the thermal mass makes them slower anyhow. During a run it's only making slight changes to position.
Thanks!
I used to work for Johnson's in Adelaide, I still have all my manuals etc. Checked that valve and if its the one I think it is, it's designed for a system pressure of 250-400kPa, hence the little hole in the in it. Take @grims advice and use a pump that has a shit load of pressure more so than one that has a lot of flow. There should be a brass tag around the valve body with a full part number on it, send it to me and I can tell you the CV of the valve and work out a pressure range that it will work around. Also the valve should be piped on the outlet of the dephleg, not the inlet, the dephleg should be under system pressure.
It's a pity the BCS doesn't have analogue outputs. But yes the PWM to Analogue converter will do what you want. I buy lots of control gear from plccenter, good blokes.
I'd never heard that, why is that?
As for the tag, mine's got - VG1241AD-9A4GGA
I suspect if you hunt around you can find a larger 1" valve to replace the 1/2" valve and be able to deal with a bit less head. Something like this guy here:
Johnson Controls VG1241CN, 1" ball valve, NEW @ eBay
Is there some comparison page/info for all these different actuators? Seems there are some older ones and newer ones. What's the difference between VG1241AD-9A4GGA and VA9104-GGA-2S for example?
VG1000 Series Two-Way, Plated Brass Trim, NPT End Connections Ball Valves with Non-Spring-Return Electric Actuators @ Johnson Controls (PDF)
Believe the VA9104xxx is the product code of the actuator alone, and the second code VG1241xxx is the combination of actuator and valve.
But sounds like Micki's got the scoop and can school us on what the ideal part number would be (there are hundreds).
Is this your valve and actuator?
1/2" NPT 2-Way Brass Valve with VA9104 Proportional Actuator @ SupplyHouse
The valve has a CV of 1.2 which means it is designed to work under low pressure/low flow with a low pressure drop across the heating/cooling coil (in our case the dephleg). The lower the CV the lower the flow through the valve. As a guide you lose a bit of control with a low CV/low flow setup. This is why 3 way diverting valves are a more popular choice as they tend to be less pressure dependant. The other reason is with a 3 way valve the pressure remains constant no matter what position the valve is in as the the flow in will always equal the flow out, therefore the only difference in pressure will be what the dephleg introduces into the system, which should be bugger all. With a 2 way valve, as the valve opens the pressure drop across it drops, as the pressure drops you get less control across the flow curve.
Have a look at this pdf page 4, 2 way piping package code 24.
Johnson Controls Piping Packages (PDF)
The valve is designed to control under pressure after the coil. The pressure drop across coil helps to smooth out the flow curve, makes it more linear across the control range. It also stops trickle through/trickle down which can make the reaction/capacity of the coil erratic. You'll find it will just work better with the coil (dephleg) under pressure.
If you want to use modulating valves then I suggest you get a 2 or 3 way diverting/mixing ball valve, these have a CV generally highr than 7. They are just about a straight through , 1/2" thread and 1/2" open ball internally, and would work much better on the dephleg, the dephleg has virtually zero pressure drop across it. You would have to put a lot of water through the dephleg to get a decent pressure drop in order to use a low CV valve. Its not really suited to a control valve application. Basically anything that has a medium CV will be suitable. So @grim this valve might be OK for what you want to do, pipe it up and see what happens, you will be suprised how much water comes through that little hole even with a pond pump of reasonable size.
edit. if you order a Johnson valve/actuator combination go for VG1241AG, VG1241AL or VG1241AN. These might be better suited if you have really low supply pressure the AN having the larger opening/higher CV.
Yes, that is it.
Pipe it up and give it a whirl.
Oh it whirls, no problem there.
Valve position vs flow (At 60psi municipal water pressure, full open) - 4 meters total hose (2 in, 2 out), 3/8th ID with push-fit fittings, zero head (with municipal mains it's probably irrelevant anyway).
At this higher PSI, the flow curve appears to be very nicely sloped at the lower end 0-50%, very good control of the flow rates, but flattens out above 50%. There is some variability there, since this was done by filling a big 2 liter graduated cylinder with one hand, and a stopwatch in another.
I've got an old mag drive pond pump around here that is rated for 2600 liters/hr, I'll plumb that up since that might be more typical of a small pumped system. I could always try the my big pump, a Taco 3 speed recirc pump that has a 20 foot head, this would be more typical in a large system.