Re: Ruby the CM queen safety related issues

Post by **Tater** » Sun Sep 21, 2014 6:15 am

Maritimer wrote:One thing I've never covered is how to apply a flow switch to controlling your still's power. Using electricity, most people use a phase-controlled SCR. This is not what I personally use, so I can't give you my experience, but here is how I would do it.

The phase controller uses a potentiometer to control how much of the half-cycle is let pass.

microswitch, no relay for Ruby.jpg
The microswitch is single-pole double throw. In normal use when there is coolant deflecting the bucket, the switch's N.O. contact is closed. This directs the wiper of the pot to the SCR gate. If coolant stops flowing, the wiper is transferred to ground, cutting off power to the elements. The only way the elements can be powered is if there is coolant flowing. Just to get some thinking going, will somebody come up with a circuit to bypass the flow switch, which you might want to do, say, if you are boiling water for a ferment?

Relays are very powerful logic elements. In fact, the first digital computers were made with relays. They can count, add, make choices, etc, just like transistor logic gates.

The relay memory circuit is shown below:
relay memory for Ruby.jpg
The top of the diagram shows the normal way to draw relay contacts on schematics. Two parallel lines represent a normally open (N.O.) contact and two lines with a slash a normally closed (N.C.) contact. This allows you to have a relay's contacts anywhere in a schematic without having a mess of lines going to the kind of representation shown on the left.

At the bottom of the diagram is a schematic of a basic relay memory circuit. The N.O. contact is shown in series with the relay coil. When power is first applied to the circuit, the relay will stay un-energized. When the N.O. push-button is pressed, the circuit through the relay is completed and the relay contact closes. Now when the pushbutton is released, the relay stays energized through the now-closed contact. When the N.C. pushbutton is pressed, the relay will be de-energized and the contact will open. When the push-button is released the relay will stay de-energized because the relay contact is now open.

The circuit in the first diagram is probably OK for the flow switch, but if you have a flood, you have a serious problem and should turn off the power until you have fixed it. The cup that was filled by the flood and which activated the microswitch could have a little hole in it so that it will empty on its own. But that would re-start the power.

So here is a circuit that remembers the flood and needs a human to re-start the power.
contactor cct for Ruby.jpg
The microswitch is shown N.O., which is what a flow switch would be. A flood switch would be N.C., opening when the cup is filled. I've shown two 120 VAC elements. Anyone care to modify the schematic for one 220VAC element?

A propane version of the above circuits would need an electrically-controlled valve, of course.

M

That looks like whats called a lockout circuit. Nicely done could also add another relay that would sound off a alarm when it happens .Seen it done in refrigeration application.micro switches Ive dealt with handled contact voltage with no problems 120 - 480 with 208 volts around here being most common .lots coolers with defrost heaters add micro switch on contactor that are normally closed when contacter is open . that way cant run cooler and heaters at same time

Ruby · Post by **Ruby** » Sun Sep 21, 2014 6:51 am

Motor controls and control logic is my profession, I have a degree in industrial electricity and deal with industrial process control all of the time. I could help with logic diagrams and wiring diagrams if need be, let me know.

Maritimer · Post by **Maritimer** » Sun Sep 21, 2014 12:23 pm

Thanks Tater and Ruby. The controller I'm working on now uses a microcontroller. In the spirit of two technologies per problem, I have a contactor as well as zero-crossing SSRs. Most of this has been covered in my thread on artisan distillers: http://www.artisan-distiller.net/phpBB3 ... =30&t=8686" onclick="window.open(this.href);return false;" rel="nofollow but which only takes off here: http://www.artisan-distiller.net/phpBB3 ... 30#p121545" onclick="window.open(this.href);return false;" rel="nofollow . The circuit has advanced a little from that shown in the thread.

Ruby, that contactor-welding-itself problem, is that caused by arcing when releasing an inductive load? My Start and Stop switches also send pulses to the microcontroller, where they generate interrupts. That would allow a delay in turning on the SSRs at Start, and turning them off at the next zero-crossing at Stop. It's easy to do. Do you think the contactor will be slow enough on releasing? The load, of course, is purely resistive, and the contactor has varistor across it.

M

Ruby · Post by **Ruby** » Sun Sep 21, 2014 12:49 pm

The only thing that generates an arc is the current. Regardless of wether it is inductive, resistive or a reactive load. I'm not sure I understand your other question about the speed of release. As for overall speed, electricity works at the speed of light, it's pretty much instantaneous as far as contactor release. I do have a question about the microcontroller inputs, are they the same voltage as the control circuit, if not, I believe you will have a problem. Most controllers use a low voltage signal that is generated by the controller itself. That would make it hard to use the same button unless you add more than one sets of contacts to the button, which is easy to do. More info would help. Circuit control voltage has to match controller input voltage, but even if it does, you could still have problems if the controller input voltage is not derived from the controller power supply.

Maritimer · Post by **Maritimer** » Mon Sep 22, 2014 12:15 pm

Ruby, I'm well-qualified to design analogue and digital circuits at the signal level. When it comes to power, I use an SSR with an optically coupled control signal. As for electrical codes, I'm pretty well clueless.

When designing for safety and one hears that the contactor has a potential fault, then that potential failure needs to be taken into account. The contactor is the backup to an SSR failure, which is stuck-conducting. The probablility of them both failing at the same time is remote, but I'd like to make it even more remote.

Visible arcing is caused by the current during contact and release. If the contactor closes and opens with zero current, then that potential danger can be mitigated. Is arcing during the closed phase also possible? Aren't AC contacts sort of "self-cleaning" compared to DC contacts?

My question about the speed of the contactor concerns the arcing when opening. If the SSRs, which are in series with the contactor contacts, are non-conducting when the contactor contacts open, then that source of arcing is eliminated. I would think that a contactor would be designed to be somewhat tolerant to line noise, so it shouldn't respond to a missing half-cycle. If this is true, then there is enough time to turn off the SSRs before the contacts open.

M

Ruby · Post by **Ruby** » Mon Sep 22, 2014 5:34 pm

No current means no arc. Deenergizing the line side of the contactor will make it a little to no arc situation. Sequencing the contactors to the ssrs seems like over kill but is possible. You are correct that ac arcing differs from dc arcing. Dc arcing is considered a burner because there is no pause in the fire, so to speak. Ac changes direction and therefore pauses when it does. This makes it easier to quench the arc, resulting in less contact damage when opened under load. I wouldn't get to concerned with contact welding. It doesn't happen too often. Usually only under a fault condition. On a side note, I admire your dedication and elevated level of thinking on the safety side of things.

Post by **shadylane** » Mon Sep 22, 2014 6:25 pm

What do we want to do?
Automate a CM still, or add E-stop circuits to stop the process if something goes wrong?
There's a big difference.

Maritimer · Post by **Maritimer** » Tue Sep 23, 2014 6:16 am

Shadylane, my posts are about making the still safe, although I'm automating my own still. The difference between making a safe still and an automated one isn't all that great. Since a still must be safe before it can be automated, a lot of the sensors are already in place.

SoMo wrote:Two technologies can't replace 5 senses, no way no how.

That was from a while back, but on my walk this morning it popped into my head. Actually, your dog or cat, or even your pet frog, has all the senses to watch a still. But it takes understanding and the ability to react to what your senses are telling you, to do something when danger strikes.

The circuits I've shown can handle a single sensor and react to shut down the still. But to really simulate a person watching a still requires an array of sensors and some intelligence to decode their information. This can be done with a lot of relays and vacuum tubes (if you use 1940's technology) or with a microcontroller (using current technology).

Unfortunately, that takes making a safe still out of the hands of the average stiller.

So, what would comprise the minimum sensor array?

-Pressure sensor. This would probably cover the flooding problem. In order to eject a bolus of liquid though the vent, pressure must have built up to propel it.
-Flow switch for coolant.
-Temperature of the vent. This is the second technology to cover the flow switch. It would also react to liquid flooding.
-Temperature of the product exit for CM and VM stills. Again, this is looking for flooding.
-?

Analogue sensors such as temperature and pressure require trip points, so some kind of comparison needs to be done. This can be done with electronics, but at this point a microcontroller would be cheaper, more reliable, and more intelligent.

M

Ruby · Post by **Ruby** » Tue Sep 23, 2014 4:22 pm

Something to consider on the legalization side, if it looks like it takes a rocket scientist to make this hobby safe, then we really aren't helping the cause much. Most people are already lost by the conversation here.

Post by **T-Pee** » Tue Sep 23, 2014 5:38 pm

Most of what people think we do they learned by watching Moonshiners. Inconvenient truth, I know.

tp

Ruby · Post by **Ruby** » Wed Sep 24, 2014 2:25 am

I was trying to say that if a politician or agency thought it took an electrical engineering degree to make the hobby safe, we are going to have a hard time convincing someone that it should be legalized for the masses.

Post by **ShineonCrazyDiamond** » Wed Sep 24, 2014 4:04 am

Ruby wrote:I was trying to say that if a politician or agency thought it took an electrical engineering degree to make the hobby safe, we are going to have a hard time convincing someone that it should be legalized for the masses.

I hear ya there, and I don't mind being the first to admit that I stopped comprehending this thread a long time ago.
But think about beer. You can boil extract and hops and throw it in the corner for a couple Weeks, our you can take it to the next level, pull out your chemistry set, match ph and water build to match the city that the beer originated, use all sorts of high tech set ups, etc. there are different extremes in any hobby.

Personally, it was the diverse expertise of seemingly intelligent people in this forum that made it reputable. I would probably never given this hobby a second thought, or respect, if it seemed to be a bunch of idiots just trying to find a way to get wasted.

Maritimer · Post by **Maritimer** » Wed Sep 24, 2014 4:56 am

Ah, the morning walk! Gets the brain moving, too.

Ruby, I misinterpreted your post about welded contacts; you were talking about the push-button Start switch, which would power the elements before the contactor closed. Fortunately, the switch I'm using is rated at 15 amps and the maximum current is 16 amps, so it's still working. But by delaying the turn-on of the SSRs, this problem can be eliminated easily.

Yes, even wiring up a flow switch would be difficult for a lot of people, let alone a bunch of sensors.

So here is a possible solution. Wiring that requires only inserting the wire into a terminal block and screwing the contact onto the wire would be within almost anyone's capability. Now the ultimate thing that needs control is the SSR that is powering the element (or the electrically operated valve for propane). The SSR controls power to the element. Instead of using phase-control, use zero-crossing SSRs. This allows the beautiful Bresenham's algorithm to control power, assuming you are using a microcontroller: http://homedistiller.org/forum/viewtopi ... 0#p7087248 . It turns the AC line voltage on or off a half-cycle at a time, distributing the on and off cycles evenly over time. This also makes the potentiometer rotation proportional to the power.

The microcontroller I'm talking about is a type called One-Time-Programmable and costs anywhere from $1.50 to $10 depending on its number of pins. Since a microcontroller is necessary to implement Bresenham's algorithm, extra inputs on it can be used for sensors. This microcontroller would be on a circuit board with a screw-type terminal block to connect the pot, sensors, SSR control signals, and maybe even a small readout. The circuit board would cost, say $25. Anything from a basic flow switch to a more elaborate system with pressure and temperature sensors could be accommodated.

How would readers feel about such a solution?

M

Ruby · Post by **Ruby** » Wed Sep 24, 2014 1:46 pm

Actually the way you drew it the elements would not power without contactor closure. I was indeed talking about contactor welding issues not the push button. If you looked up some start stop ladder diagrams you would see what I'm talking about. Normally you use all auxiliary contacts and external relays. I was just trying to illustrate why it is not done the way you drew it. Contactor welding is fairly rare and usually because of fault issues incurred by failing loads, such as motors that need rewound. As for the algorithms, now you are sailing over my head. Wish I could be of more help, but it sounds like you are way more versed in the electronics portion of this than me. Plc's and relay logic is more my game.

Maritimer · Post by **Maritimer** » Wed Sep 24, 2014 3:35 pm

Ruby, this is the path I was thinking about:

Before the contactor closes, the element is powered through the push-button.

M

Post by **T-Pee** » Wed Sep 24, 2014 3:42 pm

Nice ummm...diagram. Consider squaring it off.

tp

Ruby · Post by **Ruby** » Wed Sep 24, 2014 4:15 pm

Gotcha.....and also I like the diagram. Very proud.

badbird · Post by **badbird** » Wed Sep 24, 2014 6:36 pm

Maritimer, Is there any reason for using the dual contactor rather than a 3 phase contactor which others seem to be using with 2 elements?

Maritimer · Post by **Maritimer** » Thu Sep 25, 2014 6:07 am

Hey badbird, are you some kind of Right-Angle Conservative? Me, I'm a Skewed-Axis Liberal.

A three-contact contactor would have been preferable, I see now. It's not really a three-phase circuit, though. I figured I could switch off the hot lines, like you would with a single-pole switch on a light bulb.

I've been thinking signal-level instead of power-level.

Ruby, you're the expert. What would be the best way to do this, complete with the auxiliary contact?

M

badbird · Post by **badbird** » Thu Sep 25, 2014 6:45 am

Definitely Right-Angle Anarchist Conservative!

Three phase contactors are relatively cheap and easy to get. With mine I was planing on using the third leg like an auxiliary just for the lockout switching unless a better idea comes along.

Maritimer · Post by **Maritimer** » Thu Sep 25, 2014 9:02 am

Badbird, are you using electronics? Just curious.

M

badbird · Post by **badbird** » Sun Sep 28, 2014 5:38 am

Maritimer wrote:Badbird, are you using electronics? Just curious.

I did seriously consider designing something base on PIC microcontroler but I have enough buggy half finished projects hanging around at the moment

Maybe i will eventually go that way but for now I will use a couple of Sestos PID controllers that are lying around waiting for a job. One will control the elements for heat up (something like Swedes commercial units) and the second is the 4-20ma current loop version and will ''hopefully'' control a dc pump for reflux control.

The toy budget is a bit stretched a the moment so this is all happening slowly at present but it will get there.

badbird · Post by **badbird** » Fri Oct 03, 2014 3:36 pm

Hey Maritimer, you might find this ''off the shelf'' safety cutout controller interesting, ~$250
http://www.homedistiller.ru/hd-bezopasnost.htm you will need Google chrome to translate.

The controller is designed for the emergency shutdown of the AC of the equipment (heaters and water supply unit, other electrical equipment) in an emergency.
Emergencies:

increase in room temperature over 45C;
temperature rise in the tube connection with the atmosphere above the reflux condenser 75C;
overflow the receptacle;
spill liquid on the floor;
appearance alcohol vapor in the air;
occurrence of smoke in the room.

BB

Maritimer · Post by **Maritimer** » Sat Oct 04, 2014 6:17 am

Thanks, badbird.

But they don't include loss of coolant!

And I'm not including:
appearance alcohol vapor in the air;
occurrence of smoke in the room.

A smoke detector could be wired up as an input to the controller.

I looked for alcohol detectors and found this: https://www.sparkfun.com/products/8880" onclick="window.open(this.href);return false;" rel="nofollow

At $5, it's well worth adding.

M

badbird · Post by **badbird** » Sat Oct 04, 2014 4:49 pm

Maritimer wrote:Thanks, badbird.

But they don't include loss of coolant!

And I'm not including:
appearance alcohol vapor in the air;
occurrence of smoke in the room.

A smoke detector could be wired up as an input to the controller.

I looked for alcohol detectors and found this: https://www.sparkfun.com/products/8880" onclick="window.open(this.href);return false;" rel="nofollow

At $5, it's well worth adding.

Thought it was worth a mention just for the alcohol vapor and smoke detection sensors as they hadn't come up in the the other discussions on parameters worth monitoring, their implementation of the surface wetness "leak" sensor is also interesting, I have used similar with weather stations but hadn't considered for overflow.
I guess they think the 75C sensor above the LM condenser sufficient to capture loss of coolant events, maybe they are right as with the flow sensor i have any momentary drop in flow will trigger a shutdown.

Maritimer · Post by **Maritimer** » Sun Oct 05, 2014 5:21 am

badbird wrote:I guess they think the 75C sensor above the LM condenser sufficient to capture loss of coolant events, maybe they are right as with the flow sensor i have any momentary drop in flow will trigger a shutdown.

When designing for safety, I try to apply the two technologies rule to sensing and responding to every danger, so I'm using the vent temperature as a second sensor to lack of coolant flow.

Sounds like you have your flow switch in the holding path of a contactor. Is your flow switch SPST N.O.? It could be put in series with the DC supply to the potentiometer if you are using a phase angle controller.

If you need some delay in turning off, an auxiliary relay with a capacitor can be used. I'll design a circuit for you if you tell me how you are using the flow switch.

M

badbird · Post by **badbird** » Mon Oct 06, 2014 1:36 am

Hey Maritimer. You are probably right about the second sensor improving the reliability of the cutout function, though i don't intend running this thing unattended the safer it is the better.

The flow switch is SPSC NO (closed when water is flowing), the 60C bi-metal thermostat is normally closed. Ill run these in series through a 12v DIN relay controlling the contractor etc.

I have used a basic series diode - paralell capacitor arrangement across a relay coil in the past to add a bit of Hysteres to a battery LV disconnect system. It worked moderately well as I remember but am interested in better ways of doing this if you can suggest one.
BB

Maritimer · Post by **Maritimer** » Mon Oct 06, 2014 6:27 am

Hey badboid,

A relay with a capacitor is probably best for you if you want to avoid electronics. Looking at typical relays on DigiKey, a 12 VDC relay would require about 16 ma, which would make it about 750 ohms. If you wanted a time constant of, say 2 seconds, the product of the resistance and capacitance would be 2, which would make the capacitor about 2700 micro farads. If your flow switch were to suddenly have to charge such a big capacitor, there would be a very heavy current, so if we restrict the current to 250 ma, that would require a 48 ohm resistor.

Here is a possible circuit:

If you are using 12 VAC, you'll need a diode in series with the switches. But you'll still need a DC relay.

M

badbird · Post by **badbird** » Mon Oct 06, 2014 11:33 pm

Thanks Maritimer,
Thats pretty much the relay and contactor arrangement I am planing on and the 12v circuit is DC, I need it for PWM as well.
I don't have the relay yet so will check the coil resistance when it arrives but your delay circuit looks good so Ill go with those values providing the resistance is similar.
BB

Post by **HDNB** » Sun Oct 12, 2014 7:07 am

i skimmed this thread and may have missed but...
has anyone considered using a security panel for the sensors? pretty (extremly)reliable proven technology with trouble monitoring and all.
for example using a DSC "power" series controller like a PC1616 you could have 8 hardwired sensors (easily expandable wired or wireless) (security grade smoke, rate of rise, fluid, fluid level, temp, CO, LP, or anything else that has NO or NC contactors) with EOLR trouble monitoring (wiring faults) and use the alarm output (12vdc) for tripping your failsafe or the PGM to trip a failsafe and use the alarm for audible/visual warning. combined with an IT100 data bridge you can take all the status out digitally via rs232 to whatever automation controller you choose. existing apps will even send live time streaming data to your smart phone, and allow you to control it from there

the whole thing is battery backed up an monitors power and communications. best yet, it's already all invented and costs about 100 bucks.

Home Distiller