Steve,

I might be wrong in what I see in that picture, but you are using way too many of Arduino's digital pins for the flat connector which I assume is a keypad. If you go with the I2C protocol this will enable a serial transfer and you will free up almost all the digital pins. I built and run for the last two weeks one of this controllers (see the other post) and I can tell you that once you get fancy trying to control everything Duemillanove will not have enough pins !
Just my two cents.

cheers,
Marian

PS. btw excellent work ! Kudos.

You are right I'm using all those pins for the keypad. I'm using 12c for a lot of the project and had all those pins left over and decided to use them for the keypad. The main reason I chose this route for the keypad is the coding involved for a 4x5 matrix through I2c. After doing some quick math on the code space to use an I2c port expander for it I decided to save the room and put them right on the arduinio. However, under the monitor there are two port expanders with a total of 16 available IO's. It was a compromise I went with to save room for more code. Hope that explains it.




Hey Russ. Yes I want to have a design in place for SMD in case there is any desire to produce a prebuilt unit. I choose all DIP for the kit for ease of build by a DIY'r that might not have a lot of experience with a iron. I wonder what prebuilt SMD boards would cost at quantity 100? I should check it out.

Steve

Nice work. I can't wait to see the finished product!

Getting excited! I have been awaiting a controller with salinity monitoring for soooooo long that I could diy. Slow, continuous, automatic water changes anyone?

PH1, PH2, and ORP are wired and up and running on the prototype boards. This part took me two solid nights to wire. Schematics are good for those circuits as I built them straight from them. I'll be wiring up the Alarm buzzer, 4 switched inputs, Power and notification LEDs, and adding the I2c RJ11 jack this evening. The alarm buzzer I'm using is used in fire alarms and is crazy loud. Once these are done I'm taking a break for a vacation for the next couple of weeks. When I return I promiss to get EC sorted. Once EC is breadboarded and working I'll add the circuit to the schematics and build it into the prototypes I'm working on now. We should be on track to send off for PCB's by August 1st, which should keep us on schedule for Sept 1 finish date.

The new programmer is getting his prototype unit this weekend and has promissed to work on this project every night till it is done. All the coding that interfaces directly with the hardware is done so all he needs to do is build a menu system to make user configuration easy. I believe he has completed a computer simulator for all the code so things should move along quickly.

I also want to give a big thank you to Firewolf4 who is taking care of the PCB layout from my horrible schematics I gave him. This will buy me some time to work on finalizing EC and a few little details without cramming to get that done also.

Cant wait 'till its done!

Good job Steve 'O. Looking good. I hope you sell a million of them

OK I've run into a problem. After getting everything completed on the prototypes I installed everything into the tank and PH1, PH2, and ORP are reading funny. I recalibrated everything and it looked good, put it back in the tank and readings are off. If I remove the ORP probe and then both PH's read correctly. Insert the ORP probe and it lowers the PH by .2 or so. Put the orp probe in a separate container and the PH's read fine and the ORP reads fine. With everything in the same container then I have issues. HMMMMMMMM, not real sure what is going on. I neglected to test everything in one container at the same time. This problem can be adjusted out, however, it would mean calibrating PH with the ORP probe in the solution along with it. I've got to find a solution for this. I'm leaving for vacation tomorrow for a couple of weeks and will take a break from the project. Any ideas would be appreciated.

Steve

Sample each sensor at different times? I assume its not reading values constantly, or that that is needed.

Yea I tried this. Turns out sharing the ground line is the issue. I'd have to completely cut the connection with a mechanical relay to do it. Firewolf and I have been discussing using an optocoupler to isolate the lines. Keep the ideas coming.

I did notice Neptune Systems says this about thier new APEX controller. Perhaps the other controllers have the same issue. Wonder if anyone has pulled a orp probe from thier tank to see if it effects the PH reading.

With current setup, how hard would it be to add additional I2c devices?

Say if I want to use a DAC to control a servo, how would I go about programming?
http://www.geocities.com/bourbonstreet/3220/servobasics.html

Sorry for my I2c/arduino newbness.

i2c is no problem with this setup. There is a rj11 Jack to plug external i2c devices into it. You can check out reprap.org for more info on i2c port expanders and servo control.

Steve

What exactly is i2c >

Looks like we lost a little bit of data during the database corruption. I'll just give a quick update as were we are today.

Menu System is coming along. We have run into some RAM issues when loading up the menu system. We are building a system now that only loads the parts of the menu that you are currently using and flushes the memory of the other stuff. The Menu System is very cool and easily expanded. Early tests show that our menu system "fix" should work fine.

I ran into issues with PH and ORP interferring with each other when installed in a tank together. This issue seems to be repeatable with some other controllers on the market. I think I have a good solution for the problem and the parts are ordered and should be here this week to correct that. It is possible that these parts could increase cost a little bit as that are not cheap. The parts will only take me about 15 minutes to install in our prototype controllers and will know immediately if it works or not.

I have ordered sample probes from a company out of China. These include PH, ORP, DO, and EC. Most importantly I need the EC to be able to get that final circuit up and running. At this time I'm not sure when to expect it as I'm waiting on a tracking number. I am building the EC circuit now and most of it can be tested without the probe using my O-Scope to verify the AC simulator.

My goal is to have the EC circuit and fix for the PH and ORP completed by the end of next week although without the probe I can't verify EC for final production.

The September 1st deadline is still possisble but can only happen if the EC probe gets here in time. We have a max of three weeks before the PCB design must be in the hands of the manufacture to make this date happen. My experience with anything coming from China is that it could take at least that long to get it. IF YOU HAVE AN EC PROBE THAT YOU CAN SPARE YOU COULD REALLY HELP US TO MEET THIS DEADLINE, PLEASE PM ME IF YOU CAN HELP.

I want to give Firewolf4 a big thank you for helping us AND being very patient with me while I sort out the last few issues so he can complete the PCB layout.

I also want to give an anonymous donor a big thank you for sending us a new PH probe for testing at the beginning of the project.

This project continues and will remain and open source project. The success of the final product is greatly influenced by the continued support of those of you in the reef community.

Steve

It's not an open source project until you open the source. Schematics and code. There are none yet, so you can't call it open source until you start to disclose details.

I stand corrected. There is always a lawyer in every group.

This project will be released as open source when complete. Armchair attorney's aside, the success of the final product is greatly influenced by the continued support of those of you in the reef community.

ok I got the first part of the EC circuit completed tonight. Getting the sinewave correct for the meter to work correctly is the most difficult part. The rest should be smooth sailing. I'm using a reference circuit from Belladona on the web. It's been modified from a dual polarity circuit to a single polarity 12v DC circuit using a 555 timer chip to get the negative voltage down to about -10.5 with ref. to ground. I could probably get the voltage a little closer to -11.5 with better diodes but I think this will work. Here is a pic.



I'm still waiting on the DC to DC converters for the PH and ORP fix I have planned, which should be here tomorrow. I'm also still waiting on my EC probe. Apparently, China doesn't get wire transfers very fast so they still haven't shipped yet.

Steve

Steve,

I got very good results for feeding dual power amps using a 7660a. With that and two 10uf capacitors you can get up to a (minus) 12V from a simple positive 12V.

cheers,
Marian

Thanks Marian, you just saved me a bunch of components and at least a couple of dollars. I'm going to order some in. Those look perfect.

EC circuit is up and running. Does it work? yes and no. I don't have a probe in so I built one from some spare parts. The circuit IS working and doing what it is supposed to do but the slope is jacked. Turns out the Belladona circuit I'm using is really designed for about 3ppt or less. I'm going to have to do the math on the resistors needed to get the slope to 0-50ppt with 5v 50ppt, 4v 40ppt, 3v 30ppt, 2v 20ppt, etc. If the EC probe is not here by next thursday I'm going to order another one so I can complete the circuit.

The parts to correct the PH and ORP issue are in but I didn't have time to get them installed this weekend. I'll find time to get them done this week.

Steve

I was going to make the same suggestion. An ICL7660 (cmos version of 555) and a couple of 10uf caps and you've got your negative voltage for your op amp. For an EC probe I like the AD 7705. It has 2 channels of 16-bit, differencial, sigma delta ADC (one channel for bridge and one for optional thermocouple). Biggest pain was finding a 1MHz resonator/crystal. Just not popular any more.

Can't wait to see your finished prototype.

Scott

Id deff buy 1!

Thanks Scott, the ICL7660's are on their way. Not sure at all what you are saying about the AD7705 but it sure sounds sexy. Like a Ferrari or something I downloaded the spec sheets on it tonight and I'll give it a look over and see if rethinking the circuit now is an option. Once I get the final product working we'll release all the schematics, code, files, etc. and open up changes from the public for the next release. The great thing is changing individual circuits in the final product should be easy and not require much if any code changing. Thanks for the input.

Steve



$899 for you

Steve - Dang, I wish I knew you needed the ICL7660s as I have about 15 extra! What else r u looking for?

Scott

Thanks Scott. I think I'm good now for everything. I ordered 5 of them Monday. I think it was like $10 shipped so no worries. EC was my last real circuit to complete. I've got to wire up one more IO expander for the switched inputs but that doesn't require any thought. If you have an EC probe I really could use that for testing. I built one, but without using a "real" probe I'm scared to ok the circuit for production. I'm going to redesign my PH and ORP Circuits to also accept the ICL7660's instead of the divided resistor setup I did also. I've got an EC probe coming from China but it might be a few weeks. Apparently, there was some language issues at play and they want more money from me, but I'm not about to send another wire to them as the wire alone is $40. I sent a nasty email today. We'll see if he ships anything. If he doesn't ship then I'm going to order one locally but that might be a couple of weeks out. Thanks for asking.

Steve

Looks like we have the probes on the way and have overcome any differences. I have some very good pricing on probes and should be able to sell PH and EC for about $25. Orp will be a bit more but I don't really understand why until I see them. They are all listed as Laboratory grade. They are also sending me a Dissolved Oxygen probe. I'm not sure if we will make a circuit for it but the probe is coming. Once I have the EC in my hands we should be able to wrap up the final PCB design and send it out for the first boards.

Steve

Steve,
How hard would it be to make each of the probe circuits interchangable sub assemblies? Lets say you want pH, ORP and say magnesium (ion specific probe). If it can be designed with internally changable sub assemblies that all utilize 3, 4 or 5 leads each then it would make the probe portion completely customizable and upgradable down the road. Estimate that a sub assy would contain the BNC, sub PCB, specific probe support components and a 5 pin header. The firmware changes would be akin to writing code saying "in PortA=pH, In Port B=ORP" etc.
and Yes, the past 2 weeks I've been slammed with work averaging 19 hrs a day so no urgency for the PCB updates

that is a crazy awesome idea. That way a person could have any configuration they wanted and it could support future probes out of the box. All we would need is a four pin header with +12v, -12v, GND, and voltage back to I/O. Then write the code to enable which function is on what port. We could design a basic pcb module dimension that all circuitry would have to fit on to be compatible with our main board. Lemme finish these two circuits and let's expand on that. The only additional cost would be the additional pcb. I'm guessing it would only be like 1 square inch each.

Steve

Soooo exciting!!!

You could even put a different value voltage divider on each board so when you plugged in a board the Arduino would recognise what type of board it is by the analog voltage it receives. All you would need to use is an additional analog pin. Thats what i was planning to do on a commercial version of mine before i gave that idea up and made it just for me!

VR

That would make things horribly simple.

This is a very in depth project. I dont know if you can make the deadline but you might want to enter it in this arduino project contest

Do you have a link?

oops here it is

No real update here. I haven't worked on this project in a couple of weeks as I had been preparing our companies layoffs. I HATE layoffs. Even if you are not getting the axe it is still very stressful. Godspeed for those who left us.

I am only waiting on my EC probes to finish this up. I have confirmation from China that they shipped last week so they should be here soon.

At this point I'm sure we will NOT make the September 1st deadline mainly cause we have been monkeying around with these probes. Once I have the probes in hand I'll be able to set a new deadline. Thanks for looking.

Steve

UPDATES..

OK, the EC probes finally made it from China, YAAAAY!! Last night I was busy and installed the ICL7660A's and they worked flawlessly. Very little voltage drop between the positive and negative with +12.2 on the positive side and -12.0 on the negative side. In fact power must have been an issue cause now the PH and ORP reading are rock stable with my first night of testing. I also wired a CAT5 cable to my test system and ran it out to a breadboard with my EC circuit. I haven't tested the EC probe yet as I was busy rewiring a bunch of stuff last night. Hopefully, if all goes well I should have everything buttoned up by next weekend so firewolf4 can finally get working on the PCB layout for me.

The plan is this. Once we have the PCB's back, built, and tested I'll put together the first kits to be available for purchase and we will release all the code and schematics as open source so you can build your own. I won't be updating the release date until firewolf4 has what he needs in his hands to finalize the PCB design. Sorry for the secrecy on the designs and code, but sometimes the easiest way to meet a deadline is to not get a bunch of input for the initial release but instead get public input and make changes for the next revision.

Steve

Steve,

glad to notice your progress. Can you please let me know what schematic you used for pH. I have tried several and I was not really happy. They are capable of reading between 1 and 14 with a rather reduced resolution. For as, the 7.5 - 8.5 range is important and a 0.1 error tends to be significant. We need to control the gain as well as the slope. Thank you.

Marian

ph ORP Isolation...I was thinking instead of using a voltage output from these cct's you could use a voltage-to-frequency IC. This would allow a 7,000 - 10,000hz output through an otpocoupler and straight into a DI pin of the processor. So having a DC-DC isolator for the power supply as you mentioned and a freq output would mean the ADC inputs of the arduino would not provide a common between the cct's.

PC interface:
If you enabled a serial output to a PC we could use zigbee to provide a wireless link between the PC and the controller. I have done some development on a web interface using dallas 1wire over the last few years. www.rjconway.homeip.net

Looking forward to the cct's when posted.

You should be able to adjust the slope and offset to run between those parameters without issue and get more resolution this way, however I'm not sure a 0.1 error would be to significant given the fact that most tritration tests could fall outside of this error easily depending on how it is administered and the perception of color. Temp compensation would be a more important factor I think in the end along with proper maintenance of the probe. What I found to be the most helpful is to read the probe 10 times and average the reading. This really smoothes out the probe and zeros in on a very accurate PH. The PH schematic I'm using is widely available on the internet and works very well. At first I made a bunch of changes to it then after being told about the ICL7660A's I've rewired everything to be identical to it and found it very good. The only changes I've made at this point is to add filter caps to the power inputs. The quality of the negative voltage is very important in my testing. Here is the link PH meter Hope this helps I'll be glad to help you any way that I can. One thing you want to keep in mind about the PH range is that some people will want to use the second PH to control a CO2 reactor and the PH readings need to be lower for this if the probe is inside the chamber.

Steve

bad inferno, thatnks for your input. I thought about this idea and using the Analog to Digital convertor to read this information. The only stumbling block for me is that it was outside of my scope of reference and I just don't know how to do it. I know the Dallas One Wire Temp probe that we use does this and the one wire library should be able to do everything we want but the learning curve seemed to be large for me. I've tried those DC-DC isolators and they don't work right for analog without a common ground between the controller and the op-amp (which nullifies the need for them LOL). I suspect the different ground potentials is screwing up the readings when using them. I think once the reference designs are released lets work on getting this Analog to Digital working correctly as this will be the solution for the Isolation we want. Since I have changed to the ICL7660A's the problem I was experiencing before has resolved itself. When you put the probe in it does make the PH move a bit but then it settles back down within seconds. It is something we want to fix but it isn't a deal breaker right now. I'll check out your link tonight and see what you are working on.

OK, I worked on the EC meter all night. The problem is this. The reference design we are all looking at is really designed for freshwater aquariums and reads about a max of 3ppt. For those of you who are thinking of using the circuit designed by isabella you are in for a treat. The first part of the opamp design that creates the AC is good, the second part that provides the gain for the probe is not right for our uses, the third part that converts to DC I believe is ok, and the fourth might need a little work that provides slope. The problem is with the 6.8k resistor between pins 6 and 7 in the gain loop. My plan is to replace that resistor with a pot so that I can adjust my probe ohm reading to 680 @ 40ppt. This should put the rest of the circuit back into alignment and correct the slope issues I am having. This is just a heads up for those of you working on this for your own projects. Anyway, I'll update more this weekend.

Steve

I don't recall if the AC control method was ironed out, but I ran across this setup for another arduino-based reef controller.

Will it be possible (via the usb port or something) to upload the readings to a computer? It's been talked about, but I couldn't find a definitive answer.

-Margrave

Yes I have AC sorted with all solidstate circuits. It provides zero crossover switching and has a snub circuit. It isn't fully tested yet but is working as we speak. The snub circuit should improve the typical solid state design when working with inductive devices like motors. These tend to hang up solid state relays and keep them open. This design should be much improved over mechanical switches in that it should work much longer. Once I get the main controller done I'll focus attention on it. That link you provided I believe is jener8tionx's controller. Very nice setup and I believe he has made some more progress on his deal since I last saw it. On one of the threads I noticed a Make project link that I believe is an update of his box. Maybe he can chime in. Do a search for "automated reef controller build" in NR and I believe you will find it there. There is some very good information on that thread and these guys have done a lot of research to get where they are. I hope I didn't upset him earlier with my "armchair attorney" comment as I didn't mean to offend him and really admire the work he has done. To answer your second question, yes we will have readings sent to the USB and TX/RX ports on the arduino. So addition of ethernet cards that transmitt serial data and capturing data from USB is possible.

Steve

I couldn't wait to get home to tests the EC circuit with my ideas, so I left work early and headed to Chili's for a Margarita for brain nurishment and then home to make my changes. The result is....

IT WORKS

I made four samples of test solutions. a 40ppt, 30ppt, 20ppt, 0ppt. 40ppt reads 2.0 volts, 30ppt reads 1.5 volts, 20ppt reads 1 volt, and 0ppt is 0 volts. Multiply volts by 2 and again by 10 and you have ppt. My 30ppt kept reading 1.6 volts which would have translated to 32ppt and I couldn't figure it out. I went back and tested my sample with my refractometer and it was 32ppt. Although the circuit is working fine I would like to modify it so that it reads 40ppt to 4 volts, 30ppt to 3 volts, etc. I'll play with it some more, but for now it looks like the last major hurdle is overcome.

Some more good news today. I talked with my friend that is completing the menu coding and he has solved the RAM issue. Right now our code with existing menu is only 14k. With menu loaded we never get over 800 bytes at this point. The code is very elegant and we are very excited to release it when complete.

So all that is left is to try and modify the voltage output on the EC to get it to the volts I like, test switched outputs again, and button it up. We should be on schedule to get the final schematics to firewolf4 for PCB work by the end of next week.

Thanks for everyone's support. We are getting very close now.

Steve

Wow it's been at least a month since I've checked up on this thread. Sounds like things are coming along nicely. In my quest to build an arduino-based controller I am still trying to source some cheap solid state relays...

Good luck and I'm excited for the release!

If you only looking for a couple of amps look here:

http://www.alliedelec.com/Search/ProductDe...G3MB-202PL+DC12


They are 2A which will be good anough for alot of the pumps used in our industry!!!

Hope this helps,

VR

OK I cleaned up my breadboard and connected it to the arduino and wrote the code and now I have a good value to the screen for PPT. I'm going to try and get a video together of the home screen with all the info and a sneak peek of the menu next week after I get the schematics updated. Hopefully my kids won't pull all the parts off the breadboard, LOL.

I'm waiting with baited breathe for the updated files . I will be out of Town (in Chicago for work seminars) from the 18th through 23rd of this month. If my laptop could handle either of the CAD programs I'd work on them at the hotel.

You have email

One big step closer guys!!

Do you have an updated ETA on when this will first be available to a first generation purchase? I'd love to use something like this, but I'll need to start saving my pennies! And dig out my soldering iron

Great project and I cannot wait to see the final product. I'm new to nanos and controllers (have been running 75g reef tank manually for many years) and was just starting to look into an Arduino based system.

Are you selling the plans, or are you going to sell finished product? Too hard to find that in the 10 pages of posts.

Thanks!