disfunktional
Member
awsum thanx man!! is there a thread that you could refer me to that uses this plc? or is the only one?
-
As of today ICMag has his own Discord server. In this Discord server you can chat, talk with eachother, listen to music, share stories and pictures...and much more. Join now and let's grow together! Join ICMag Discord here! More details in this thread here: here.
Application of Nano PLC In A Growroom
- Thread starter imnotcrazy
- Start date
I wire mine 1 DPDT contactor per ballast isolating the neutral as you spec'ed.
Mine runs magnetic ballasts fine since I first posted the thread (+14months), but as far as digital ballasts go, any idea why it would run a full 12/12 cycle through one set of contacts (Ballast channel) then, when the flip reapplied power after lamp change, the first ballast blew up. I wired in the spare and cycled the flip power. The program forced the DPDT contacts back the the same position the 1st ballast failed in. When the flop reapplied power to the replacement it also blew up.
Both had the same component failures but so did all the ones I looked at that had failed being used for regular ballast operation.
Do you use a different relay for digital ballasts? IMO, the DPDT changed the output circuit enough that it caused the filter capacitor and MOSFET to fail. Although I have several friends who blew up several of these using them as intended. I had the 2 that blew up from them dumping digital ballast all together and wanted to test. I haven't had a chance to get my hands on the parts to repair my digital ballasts or acquire some new ones to try some ideas out. You have any suggestions for us?
M
medi-useA
imnotcrazy-> Just got 'round to your thread and WHAM!
This thing's looking better and better as I read through.
And a tute with pics!
Do you mind If I post this link on my Automation LinkList?
muA
This thing's looking better and better as I read through.
And a tute with pics!
Do you mind If I post this link on my Automation LinkList?
muA
Hydrostatic
Member
imnotcrazy,
I am currently using White Rogers 94-389 type contactors in conjuction with C.A.P. Controls Nextgen 1000W ballasts without any problems since isolating each ballast to it's own individual contactor.
Contactor Specs: http://www.white-rodgers.com/wrdhom/pdfs/06_Cat_pages/Cat_06_pg0108.pdf
I am currently using White Rogers 94-389 type contactors in conjuction with C.A.P. Controls Nextgen 1000W ballasts without any problems since isolating each ballast to it's own individual contactor.
Contactor Specs: http://www.white-rodgers.com/wrdhom/pdfs/06_Cat_pages/Cat_06_pg0108.pdf
So one contactor 2 poles, except they don't have the capability of breaking the common to the bulb. Are you using them to power the ballast or to switch from one hood to another??
I am talking about switching to a 2nd hood every 12 hours: 2 rooms 12/12 1 ballast run 24/7
Again, I do not know if my setup would have an issue with a higher quality digital ballast. These were the Galaxy/HTG/MD Hydro/Digital Greenhouse 600W
Those are the same Definite Purpose relays I do use, Two Digital 600Ws still shit the bed.
Again several of the same ballasts failed in regular use, which was how I came across them.
Maybe just shitty ballasts...
Here's my box as of now:
Not sure what could be causing your digital ballasts to fail, they may be bad batch. I noticed that you didn't use terminal lugs to connect the wires to the screw terminals. Since it's stranded wire, as you tighten the screws, the pressure crushes the strands, this could be a problem. Best bet is to get a ratchet crimper and use insulated terminal lugs, the "round" ones. Nice box though, PLCs are great.
muttley
muttley
Hydrostatic
Member
Yes, one contactor 2 poles, just switching between 2 lamps per ballast. Common legs are indeed tied (2 lamps/1 ballast). I know that I don't need 2 poles, they were used only because of their smaller size and I thought I could double up on the space savings, not to mention, they were only $7.00 each. With only 1 ballast/contactor, they are getting the job done perfectly without any problems at all (so far).
Yeah, I had put them in my original box and was in a rush to expand so when I made the Flop bigger by 2 contactors and a new enclosure.
If those contactors handle digital ballasts, more than likely the one I attempted to use were lower quality. All the other ones that failed in normal operation failed in the same manner, Mosfet and filter capacitor shorted.
Is there a job title for someone you would call to make something like this and service it? This is something I will want to do in the future.
I emailed them recently about a flip unit muttley, appreciate the referral. I will be looking for someone close by so that they can custom tailor the system to my needs onsite, and also service 24/7. This will be a year down the road for me right now, need to get the operation up and running first, and I will cadillac it out later. So not just any electrician lol, a qualified one. I want someone who is a plc expert, I will find them 
Hah love it! - great minds think alike
Hah love it! - great minds think alike
First of all congratz on the excellent forum; quick intro, electrical professional with 30 years experience in industrial, machine and building control services installation, commissioning fault finding. Long time grow enthusiast. And I thought I was the only one doing this! Tip of the hat to Imnotcrazy and the rest of you guys for spotting the growroom potential of microPLCs/Smart-Relays and their derivatives. I've been working with these devices in a specifically growroom context since early 2003 and have designed/produced/supplied/installed systems from simple 2lgt controllers up to full on 4 x 24 light master & slave controllers that also ran all the irrigation duties, the environmentals and monitored the genset at the same time. The sky is the limit with these devices and I heartily commend this forum for spotting them and recognising their potential ad benefits. Back in 03 I was asked to do a electrical safety general survey of a certain major hydro retailer's display areas and the of the various bits of kit they had plugged into the mains that were accessible to the public. What I saw frankly amazed me, not so much the retailer's electrical installation (in good nick) but some of the items for sale were some of the ropiest looking sacks of cr@p I'd ever seen offered for sale and was amazed that they'd got their CE safety marks okaying them for retail sale to the public. I'll also state that I couldn't believe how generally "low tech - no tech" the state of the art was (with a very few exceptions) I mean THIS was their 'flagship' controller!!
I mean what can you say, I was screwing stuff like this to factory walls 25 YEARS AGO!!! and they were charging top dollar for a lash up like this? They were also expecting complete novices to bend and terminate 25mm/sq cables into it! It was a mickey mouse segment timer, barely adequate contactors a standard domestic consumer distribution unit and bunch of 'pound-shop' twin sockets mounted onto the reverse of a rough-arsed offcut of kitchen worktop (fire ratings anyone? - who cares?) Laugh? I thought they'd never dry! "What a rip off and what a total crock of sh1t" thought Carter "is this truly the best a man can get?" and immediately repaired to his lair, cracked a six of Carlsberg and fired up AutoCAD for an all night session to right the wrongdoings of the hydroponic industry. Well just the initial thinking through of the realities and technicalities of the growing process took around 3 months to nail down into concrete requirements for an 'all-in-one' grow-room controller. The few that I'd seen such as the H******master looked way too flimsy, like something a home electronics buff might knock together and thenI saw that they were charging £95 for a "contactor" which was in fact nothing but a monkey ass cube relay!! and that really confirmed that growers were (are?) being royally bent over and given it roughly with no lube! not £25 or even at a pinch £35 but £95!The cheeky bar-stewards! I don't like devices that rely on one long single printed circuit board because they flex and dry solder joints open up, circuit tracks fracture, if the unit gets a Poot light stashed on top of it then it immediately becomes £800 worth of landfill. Bare phono connectors for signal inputs?? In a high humidity grow room? Ffs do me a favour; f’kin amateurs! No environmental sealing on the casings and so on and so on, and unbelievably, no earth/ground connections in some cases!! But yet it still gets licenced for sale to Joe public? Then I turned to computer/software control and embedded processor driven devices but you have to ask yourself the question “how often does Windows drop he ball and wander off into a corner?” Answer: far too often to entrust MY PRECIOUS grow to the vagaries of Mr gates’ wobbly, crash prone operating system. Same went for remote control via internet, no way would I trust ‘Internet Exploder’ to have the final say on the state of the grow! Plus who really wants to be sending streaming vid of their installation across the net? Not me for one. Eventually I decided to go with PLC as it’s what I know and trust from my industrial installation and machine tool experience. I did so because the ultimate requirement for ANY grow room controller has to be ABSOLUTE RELIABILITY. If you’re spending all your time wet-nursing iffy software and temperamental device interfaces then what the hell is the point? You may as well revert to mechanical segment timers and have a (relatively) stress free life. Industrial process controllers just keep going remorselessly day after day, month after month, year after year, no fuss, no drama. The only trouble was that the devices I was used to were full on, £10,000 a rack items and way out of reach in terms of cost and complexity for the layman. Then I came across Seimens “Logo” range of micro PLCs and from the spec sheets I could see immediately that this was the perfect platform. I contacted one of my local wholesalers and he ran through its capabilities re analogue sensing and I ordered a pair of them along with an expander and he threw in the download cable and software for free. I ordered a few casings in varying sizes from another firm I know and spent the next three months playing around with features 'n options, component layouts and program writing and the first prototype device was built into a floor standing 'pedestal' type unit with a sloping front for the plc that looked like something off a Star Trek set! (I'll try and dig out the CAD images for it.) I then got a bit more conventional and put this fellah together which ran 2 x 1mtr/sq flood & drain tables, the 2 lights and the intake/exhaust fans.
and a rendered CAD screenshot of my first "electric gardener"...
It worked like a charm even without the full temperature sensing capability. I was using a simple two state room-stat device; ON/OFF whilst researching viable (=cheap & accurate) temperature sensing devices. Gentlemen; I have your solution to el cheapo temp sensor construction and interfacing for these devices.(of which more anon)
After another few months trawling device data sheets I came upon the 'holy grail' of temperature sensing and put together the above temp sensor that could be made accurate to tenths of a degree for under £10! and incorporated that discovery into the Mk2 version which would now switch states through a simple triac based speed controller (yuk) to more accurately track the environmentals.
and in real life it looked like this, very neat and tidy I thought...
This was getting more like it. I'd also got rid of the dreadful domestic socket outlets that infest commercial grow room products, as far as I'm concerned (speaking as an electrical designer/engineer/site foreman/maintenance man with 30yrs experience behind me) and as far as the 17th Edition of BS7671 Requirements for Electrical Installations (the regs) are concerned a growroom should be treated as an "area requiring special considerations". I don't think standard sockets have any place in wet environments with elevated shock risks from HV ballasts, pumped water systems etc. I scrapped them in favour of compact twist locking connectors rated at 20 amps! They use these puppies on outside broadcast equipment and such, a bit pricey but absolutely bombproof. I will never use another IEC (kettle plug type) connector ever again. As well as the enhanced electrical integrity and safety it allowed me to cram the whole lot into the above casing which measures:7 width=200mm height=240mm depth=105mm ...amazing! Of course I then got a few requests for variations on the theme and th beauty of the PLC heart of the system is the ability to cut 'n paste common elements from various different control programs and build custom units. One thing I've learned is that even though ALL grow rooms work on and must adhere to the same biological, horticultural and electrical principles NO TWO ROOMS OR GROWS ARE EVER THE SAME. Every one, I've found, is different in terms of what people want, the scale, the degree of control req'd, additional features and the "can you do this...?" type of off the wall ideas that need to be worked out and made to happen. It's fascinating to ork it all through and immensely satisfying when you hit >Program start >YES? >OK and it all clicks into life with reassuringly hefty clunks from pukka 'man sized' contactors. Of course it was only a matter of time before one craves more output capabilities. (I've never yet run short of inputs) and they just started getting bigger...
gotta love that sexy CAD rendering software! and for real...
Controlled 18 lights plus the fans plus all the pump timings
and then someone wanted a three phase 30 way unit based on this design...
...which was a bit of a backward step in design terms as I had to revert to using standard (bulky) modular DIN rail circuit breakers instead of the tiny 16mm sq panel mounted units I'd managed to source. Adding a few expander units to the side of the base unit PLC made it possible to put together really quite sophisticated control panels such as these bad boyz...
and finally... although it looks complex it's really just repetition of basic elements...
So there y' go folks, hope I've chipped a few possibilities into the ring and if anyone's got a question or needs advice I'm moe than happy to help a bod out with an answer. particularly when it comes to keeping folks in an un-electrocuted condition! TTFN CARTER
Hah love it! - great minds think alike
First of all congratz on the excellent forum; quick intro, electrical professional with 30 years experience in industrial, machine and building control services installation, commissioning fault finding. Long time grow enthusiast. And I thought I was the only one doing this! Tip of the hat to Imnotcrazy and the rest of you guys for spotting the growroom potential of microPLCs/Smart-Relays and their derivatives. I've been working with these devices in a specifically growroom context since early 2003 and have designed/produced/supplied/installed systems from simple 2lgt controllers up to full on 4 x 24 light master & slave controllers that also ran all the irrigation duties, the environmentals and monitored the genset at the same time. The sky is the limit with these devices and I heartily commend this forum for spotting them and recognising their potential ad benefits. Back in 03 I was asked to do a electrical safety general survey of a certain major hydro retailer's display areas and the of the various bits of kit they had plugged into the mains that were accessible to the public. What I saw frankly amazed me, not so much the retailer's electrical installation (in good nick) but some of the items for sale were some of the ropiest looking sacks of cr@p I'd ever seen offered for sale and was amazed that they'd got their CE safety marks okaying them for retail sale to the public. I'll also state that I couldn't believe how generally "low tech - no tech" the state of the art was (with a very few exceptions) I mean THIS was their 'flagship' controller!!
Hydrostatic
Member
Carter,
Could you please provide us with more detailed information with regards to the components necessary for the "el cheapo temperature sensor" I am very interested.
Also, thank you for sharing those photos of your work, it is a great example of what standards most of us in the community should demand, or in the case of DIY projects, standards that we should be striving to achieve.
Could you please provide us with more detailed information with regards to the components necessary for the "el cheapo temperature sensor" I am very interested.
Also, thank you for sharing those photos of your work, it is a great example of what standards most of us in the community should demand, or in the case of DIY projects, standards that we should be striving to achieve.
From scanning through the threads/posts it seems like this is the missing link, i.e. cheap, simple and reliable real time temperature monitoring as opposed to 'off board' discrete switching devices like domestic room stats and the like. This little fellah will allow you to do things like differing day/night setpoints, over-temp shutdown of specific lights/banks of lights. In my programs I will usually incorporate a message text in the stack giving a permanent readout of current temperature. The device you need to look at is the National Semiconductor LM35DZ: in a transistor sized ('TO92') three leadout plastic package. It looks like this...
Here's the datasheet for it...
http://pdf1.alldatasheet.com/datasheet-pdf/view/8866/NSC/LM35.html
Basically you stick 12 volts up its arse and in return it gives you back via the centre leadout a very linear voltage output that varies precisely 10 millivolts per degC with varying ambient temperature. Moreover this output voltage is referenced in such a way that it is directly readable by either of the two AI (analogue inputs) available on the micro-plc. (inputs 7 or 8)
I at first envisaged having to incorporate all sorts of stabilisation, buffering and signal conditioning to it in order to get a usable working signal out of it but the spec sheets don't lie and from that it seemed that it should be a simple plug 'n go exercise. Seeing as they are offered in anti-static packaging I took the precaution of ordering 10 of them anticipating a few prototyping fry ups and started at the very basic end, volts in = signal out and see what happens.
A quick lash up on a 'bread-board' showed that it did indeed provide an instant and strongly varying output signal in reponse to just touching the casing - looking good to go so I transferred it to a sliver of 'Veroboard' (15holes x 9holes) as below; it really doesn't get any simpler... here's the track side view (CAD model)
and from above...
The only things left to do were to sort out a DECENT 3 pole locking plug and connector (not some fucking boy-scout phono plug! lol) and set the gain and threshold settings in the ANALOGUE THRESHOLD TRIGGER 'properties' dialogue box as below...
Actually I initially set the gain at unity (1) set the threshold temps and hit GO and bugger me backwards up popped the current temperature on screen and the relays clicked in perfect response - fantastic! The values you see above are the result of me fine tuning the gain and offset in order to get it spot on throughout the anticipated operating range. Frankly it was just me being a retentive smart arse but if you're gonna do something, at least do it as best you can. Essentially this gave me a working accuracy of with+/- 0.2 degree C which for our purposes is more than good enough.
The next thing was to see how far the signal could be sent down copper. The 'bread-board' prototype worked fine with 1mtr of screened 4core signal cable, the Veroboard Rev1 model had 5 mtrs of screened so I thought I'd try it with the 15mtrs I had left on a part coil. Same again; no problem at all! As a final "can I get this signal to degrade beyond use?" I hooked the sensor up to a 50mtr coil of crappy UNSHIELDED telecoms type solid conductor 'string' and fook me it STILL gave back a good signal! (albeit with a little adjustment to the offset/gain to trim out the ludicrous volt drop/copper losses.) As a final test I held the whole lot (still wound on the coil!) next to a triac controller which was working hard trolling a fan motor at the slowest speed it could handle then razzed it with an ancient unsuppressed drill that freaks out every TV in a 400yd radius. No spurious voltages, spikes or false triggering could be induced that I could detect without a scope.
Try it chaps! Then send wads of cash, complementary crates of ale, proposals of marriage etc. to yer man Carter.
As for casings; well there's only one component so use your imagination. I've managed to squeeze one of these into a plastic 'project' box that measured about 20mm x 15mm x 10mm deep. The thing to bear in mind is that how you enclose it will affect how quickly it reacts to temperature changes. If you simply leave the top of the device poking through the casing in free air then its reaction time is pretty instantaneous and if you just fed this signal through the Trigger Block then its output would rattle the relay to pieces in minutes; so quickly does it react when at the switching setpoints.
There's two ways to mitigate this. First, you can clamp the LM35 to a small chunk of aluminium with a blob of epoxy and this will significantly damp out the speed of temperature change seen by the device's internals. You can even totally encapsulate it and the 3 sleeved solder connections in a 10mm tube and chuck it in your nute reservoir. The second component of the fix is to introduce an OFF DELAY block after the ANALOGUE THRESHOLD block so that once the Trigger switches 'high' it sets the OFF DELAY for a fixed period no matter what happens with the swinging of the raw analogue signal. Technically it's called "DE-BOUNCING" for reasons which should be self explanatory.
You're very welcome, I've made a bit out of this little circuit configuration over the years so now I'm donating it to a worthy home i.e. peeps who will appreciate it for what it is. Hope you guys get good service from it.
PLCs? - Child's play...
CARTER
p.s. The LM35 series is the metric Celsius calibrated device, for reading in degrees Farenhiet it'll be the LM34 you're after.
Great thread , +rep to you guys!
No doubt , this is the most stable way to automate a grow.
Great find on the temp sensor , Carter. You don't mind sharing your discrete fan controller circuit too?
What we need now , is a list of the different PLC's. (inputs , outputs , price , expansions, etc)
I 've looked at the following brands:
Siemens LOGO , Omron ZEN , Mitsubishi Alpha , Crouzet Millenium , Telemecanique Zelio..
What it all boils down to is wich input/output combination to buy...
It seems like they all comes in basically three versions:
1: 230/110V - relay outputs , 230/110v inputs , no analog I/O.
2: 12/24V -relay outputs ,12/24v digital inputs, some has analog inputs.
3: 12/24V -transistor outputs , 12/24v digital and analog inputs.
I would like analog inputs for temps , CO2 , and eventually ph.
I guess I would need a trasistor output for the fan controller.
Does anybody know if I can use a 12v transistor output or analog input module on the 230v version of the Siemens LOGO?
Anyhow , it seems to me that the Omron ZEN 20 CPU or the Crouzet Millenium SA 20R are better choices , with 12 inputs , and 8 outputs.
Does anybody have a soulution for a cheap analog RH%-sensor?
No doubt , this is the most stable way to automate a grow.
Great find on the temp sensor , Carter. You don't mind sharing your discrete fan controller circuit too?
What we need now , is a list of the different PLC's. (inputs , outputs , price , expansions, etc)
I 've looked at the following brands:
Siemens LOGO , Omron ZEN , Mitsubishi Alpha , Crouzet Millenium , Telemecanique Zelio..
What it all boils down to is wich input/output combination to buy...
It seems like they all comes in basically three versions:
1: 230/110V - relay outputs , 230/110v inputs , no analog I/O.
2: 12/24V -relay outputs ,12/24v digital inputs, some has analog inputs.
3: 12/24V -transistor outputs , 12/24v digital and analog inputs.
I would like analog inputs for temps , CO2 , and eventually ph.
I guess I would need a trasistor output for the fan controller.
Does anybody know if I can use a 12v transistor output or analog input module on the 230v version of the Siemens LOGO?
Anyhow , it seems to me that the Omron ZEN 20 CPU or the Crouzet Millenium SA 20R are better choices , with 12 inputs , and 8 outputs.
Does anybody have a soulution for a cheap analog RH%-sensor?
Mulletsoda
Member
http://www.sparkfun.com/datasheets/Sensors/Weather/SEN-09569-HIH-4030-datasheet.pdf
There are cheaper places than SparkFun (they are out of stock anyway), that was just the first datasheet google gave me.
Holy crap, though... I've been looking at getting more involved in some MCU action, but I've never even heard of the brands you mentioned. I landed on AVR based chips, super cheap, fairly easy to use, great user base. Atmega328 (yes, the arduino) and ATtiny2313... Maybe they're a different category, though, because being able to sink/source 200mA of 5V is considered high current from the datasheets I've read.
My first temp sensor was the LM35, it works pretty well. I've since decided on Maxis 1-wire sensors, as I can stick a dozen on the same digital pin, instead of one per analog input like one of those linear voltage chips. You feed them voltage and they only need one wire they use as a data line. They even have a parasitic mode, where they need a single wire (the data pin) and ground, and they're powered from the data line. www.maxim-ic.com/reliability/dallas/DS18S20.pdf
Worrying about the output current capabilities (like the voltage) kinda seems... not the point to me. Whatever it is, you can make it into what you need. A 5V 20mA signal from my MCU hits a 'darlington array' (ULN2003, $.50) which allows a 12V 500mA current to pass through, more than enough to trip a mini power relay, which could trip... you get the point.
My biggest criteria in selecting a control unit has been number and types of inputs and outputs. Does it have PWM, what resolution is the ADC (those analog inputs) in it? Does it have any onboard goodies, like UART (for USB), or hardware based RS-232?
There are cheaper places than SparkFun (they are out of stock anyway), that was just the first datasheet google gave me.
Holy crap, though... I've been looking at getting more involved in some MCU action, but I've never even heard of the brands you mentioned. I landed on AVR based chips, super cheap, fairly easy to use, great user base. Atmega328 (yes, the arduino) and ATtiny2313... Maybe they're a different category, though, because being able to sink/source 200mA of 5V is considered high current from the datasheets I've read.
My first temp sensor was the LM35, it works pretty well. I've since decided on Maxis 1-wire sensors, as I can stick a dozen on the same digital pin, instead of one per analog input like one of those linear voltage chips. You feed them voltage and they only need one wire they use as a data line. They even have a parasitic mode, where they need a single wire (the data pin) and ground, and they're powered from the data line. www.maxim-ic.com/reliability/dallas/DS18S20.pdf
Worrying about the output current capabilities (like the voltage) kinda seems... not the point to me. Whatever it is, you can make it into what you need. A 5V 20mA signal from my MCU hits a 'darlington array' (ULN2003, $.50) which allows a 12V 500mA current to pass through, more than enough to trip a mini power relay, which could trip... you get the point.
My biggest criteria in selecting a control unit has been number and types of inputs and outputs. Does it have PWM, what resolution is the ADC (those analog inputs) in it? Does it have any onboard goodies, like UART (for USB), or hardware based RS-232?
Great! a 0-10V RH% sensor would be better , but that one will do.
The LM35 seems like a no-brainer , so I'll stick with it for now.
My goal is to make this as cheap and simple as possible.
Only thing I'm missing now is a working fan-control..
Next options would be CO2 and Ph sensors... any ideas?
edit:
I think I've found my PLC..Millenium 3
It has 16 inputs , including 6 analog 10bit. You can allso use a potensiometer on the inputs , to make dials for things wich needs to be adjusted , like how much water , humidity target , etc.
It also has 10 outputs, choice between relays and transistors. Transistor version has PWM on four outputs.
And it still retails for under 200$
.manitu
The LM35 seems like a no-brainer , so I'll stick with it for now.
My goal is to make this as cheap and simple as possible.
Only thing I'm missing now is a working fan-control..
Next options would be CO2 and Ph sensors... any ideas?
edit:
I think I've found my PLC..Millenium 3
It has 16 inputs , including 6 analog 10bit. You can allso use a potensiometer on the inputs , to make dials for things wich needs to be adjusted , like how much water , humidity target , etc.
It also has 10 outputs, choice between relays and transistors. Transistor version has PWM on four outputs.
And it still retails for under 200$
.manitu
