reef-pi :: An opensource reef tank controller based on Raspberry Pi.

robsworld78

Well-Known Member
View Badges
Joined
Feb 14, 2020
Messages
985
Reaction score
1,325
Location
Edmonton, Canada
Rating - 0%
0   0   0
@Ranjib
Error in a ph calibration?
maybe i have a problem in my ph probe, but, when i calibrate it, the program says "mid value" and "second value". There is any difference if "mid value" is > "second value" or "mid value" < "second value" because it seems that if the "second value" is less than the "mid value" the real value is not correct.
i made ax example, yesterday i calibrate my ph probe in my tank,
"mid value" = 7
"second value" = 4
now in my tank, the ph is 6.4! if i put the probe in the 7 calibration liquid and i read 7 (+- 0.1 decimals) the same in the 4 calibration liquid.
fortunately, with the liquid test, the ph is 8. (or everything in my tank are dead).
today i'll try with
"mid value" = 4
" second value" = 7

(i tried to read the program on github, but is very difficult without comment :( )
What circuit are you using?

As I understand calibrating Mid value should be 7.0 and second either 4.0 or 10.0. I think going other way around would mess things up. The value you expect to measure should fall between the two calibration values, as you expect 8 you should be using 7.0 and 10.0 for best results. Not sure if that could cause what you are experiencing though. Do you get close to 8 with no calibration? When viewing the calibration window you see the non-calibrated pH so you should be seeing something different when calibration is open vs closed.

EDIT: @Ranjib, not sure if you ever looked at presenting the slope of a pH probe so users can tell when a probe is bad or getting old.

 
Last edited:

Kurtzisa

Community Member
View Badges
Joined
Mar 21, 2020
Messages
79
Reaction score
70
Location
Parma, Italy
Rating - 0%
0   0   0
What circuit are you using?

As I understand calibrating Mid value should be 7.0 and second either 4.0 or 10.0. I think going other way around would mess things up. The value you expect to measure should fall between the two calibration values, as you expect 8 you should be using 7.0 and 10.0 for best results. Not sure if that could cause what you are experiencing though. Do you get close to 8 with no calibration? When viewing the calibration window you see the non-calibrated pH so you should be seeing something different when calibration is open vs closed.

EDIT: @Ranjib, not sure if you ever looked at presenting the slope of a pH probe so users can tell when a probe is bad or getting old.

I don't have at home the ph10 solution now, but my previous ph value was nearly correct from 7,9 to 8.2, I've deleted the ph probe to reset some strange value (for cable management I disconnect the probe but it was always in the water ), after re-adding it I've found this strange values.
I check the cable, i check the solutions (i have another probe in my calcium reactor), i'm using a self-made ph-board from Roberto Buti, so i read the raw value (like 13600 now. )

in the afternoon (im living in italy) i'll try to recablibrate the probe, middle value 4 and second value 7. i'm read both raw value and calculate the slope and the hypothetically correct value, then i will compare with the ph value of reefpi
 

Kurtzisa

Community Member
View Badges
Joined
Mar 21, 2020
Messages
79
Reaction score
70
Location
Parma, Italy
Rating - 0%
0   0   0
Recalibration done!
Mid value, ph 4 raw 32350
Second value, ph 7 raw 11600

Tank value, ph 7,83 raw 6106, ph test kit 7,8

Conclusion:
my ph probe has a lot of offset value, it's time to replace it
There is a difference between the two calibration
 

robsworld78

Well-Known Member
View Badges
Joined
Feb 14, 2020
Messages
985
Reaction score
1,325
Location
Edmonton, Canada
Rating - 0%
0   0   0
Recalibration done!
Mid value, ph 4 raw 32350
Second value, ph 7 raw 11600

Tank value, ph 7,83 raw 6106, ph test kit 7,8

Conclusion:
my ph probe has a lot of offset value, it's time to replace it
There is a difference between the two calibration
I think it's ok to use 7 and 4 to calibrate, just better to use 10, either way it should be close. Hopefully a new probe gets it sorted.
 

Kurtzisa

Community Member
View Badges
Joined
Mar 21, 2020
Messages
79
Reaction score
70
Location
Parma, Italy
Rating - 0%
0   0   0
I think it's ok to use 7 and 4 to calibrate, just better to use 10, either way it should be close. Hopefully a new probe gets it sorted.
ph8 is near ph7, the point of calibration, so theoretically the value is correct

I also have a redox probe the ph increase but the redox goes down... problably I need another calibration. Probably the biggest raw value is the "mid value" the second one is the "second value"
 

robsworld78

Well-Known Member
View Badges
Joined
Feb 14, 2020
Messages
985
Reaction score
1,325
Location
Edmonton, Canada
Rating - 0%
0   0   0
ph8 is near ph7, the point of calibration, so theoretically the value is correct

I also have a redox probe the ph increase but the redox goes down... problably I need another calibration. Probably the biggest raw value is the "mid value" the second one is the "second value"
I found it, was looking for this yesterday. This explains nicely how pH is calculated and shows the formulas. It's all based on mV so I'm not sure how those analog values reef-pi gets from the circuit are translated. This is what my circuit is based on.

 

Kurtzisa

Community Member
View Badges
Joined
Mar 21, 2020
Messages
79
Reaction score
70
Location
Parma, Italy
Rating - 0%
0   0   0
I found it, was looking for this yesterday. This explains nicely how pH is calculated and shows the formulas. It's all based on mV so I'm not sure how those analog values reef-pi gets from the circuit are translated. This is what my circuit is based on.

the ph probe has a linear trend, so i can use the equation of a line passing through two-point to calibrate my probe and convert my mV to my ph. But, first of all, the signal (mV) needs to be amplified and elaborated by an ADC. So my raw signal is between +32767 -32768.
If i had a little slope but the signal in mV has always a linear trend and i can calibrate between my two point (4ph and 7 ph or 7 ph and 10 ph) the slope is irrilevant!
 

bishoptf

Valuable Member
View Badges
Joined
Jan 1, 2019
Messages
1,349
Reaction score
1,724
Location
Missouri
Rating - 0%
0   0   0
I am in the process of redoing one of my original reefpi since I want to add an additional light. I have 2 lights and one of them is what I thought just a straight forward 12v led light being controlled via pwm with a mosfet. The original reefpi is on a bread board and I have used one of these in the past - mosfet board On my new set up I have it working but the LED's are very dim, nothing like how it looks with the current reefpi that is in use. I am using an @Michael Lane goby hat and I have pwm and ground going to the mosfet board and have 12v input to the board and then the output from the mosfet going to the led's.

Anyone have any suggestions on why it appears to not be getting full power? It does change when I increase the value but even at 100% its more like 5%, just trying to think what I may have missed.

Thanks :)
 

robsworld78

Well-Known Member
View Badges
Joined
Feb 14, 2020
Messages
985
Reaction score
1,325
Location
Edmonton, Canada
Rating - 0%
0   0   0
I am in the process of redoing one of my original reefpi since I want to add an additional light. I have 2 lights and one of them is what I thought just a straight forward 12v led light being controlled via pwm with a mosfet. The original reefpi is on a bread board and I have used one of these in the past - mosfet board On my new set up I have it working but the LED's are very dim, nothing like how it looks with the current reefpi that is in use. I am using an @Michael Lane goby hat and I have pwm and ground going to the mosfet board and have 12v input to the board and then the output from the mosfet going to the led's.

Anyone have any suggestions on why it appears to not be getting full power? It does change when I increase the value but even at 100% its more like 5%, just trying to think what I may have missed.

Thanks :)
Do you have +12v coming out of the mosfet going into the light?
 

bishoptf

Valuable Member
View Badges
Joined
Jan 1, 2019
Messages
1,349
Reaction score
1,724
Location
Missouri
Rating - 0%
0   0   0
Do you have +12v coming out of the mosfet going into the light?
Yup, I think its the mosfet that is used in that board. I think it drops the voltage down to where the led's are not getting enough juice. I am going to pick up a IRLB8721pbf that is similar to what I have used in the past. I am using this in another circuit but its for moonlights and they may be dimmer but thats ok for moon lights.

I wish I had another one on had to try but thats my guess right now, it works but its really dim. :(
 

robsworld78

Well-Known Member
View Badges
Joined
Feb 14, 2020
Messages
985
Reaction score
1,325
Location
Edmonton, Canada
Rating - 0%
0   0   0
Yup, I think its the mosfet that is used in that board. I think it drops the voltage down to where the led's are not getting enough juice. I am going to pick up a IRLB8721pbf that is similar to what I have used in the past. I am using this in another circuit but its for moonlights and they may be dimmer but thats ok for moon lights.

I wish I had another one on had to try but thats my guess right now, it works but its really dim. :(
Is the PWM signal only +3.3v @ 100%? Check if you see +3.3v on signal wire going into module at 100%, if so and the output from module is only say 8v instead of 12v then it would seem like mosfet needs 5v signal to turn on fully. Maybe when you turn the PWM pin to 100% it only goes to +2v instead of +3.3v which would point to the pca9685 or Pi pin.
 

bishoptf

Valuable Member
View Badges
Joined
Jan 1, 2019
Messages
1,349
Reaction score
1,724
Location
Missouri
Rating - 0%
0   0   0
Is the PWM signal only +3.3v @ 100%? Check if you see +3.3v on signal wire going into module at 100%, if so and the output from module is only say 8v instead of 12v then it would seem like mosfet needs 5v signal to turn on fully. Maybe when you turn the PWM pin to 100% it only goes to +2v instead of +3.3v which would point to the pca9685 or Pi pin.
I'm pretty sure the pwm coming from the hat is only 3.3v and that is causing the issues. I took some measurements and it appears that when no load is on the output it scales from 10.5v to 12v at 100%. But when the light is added to the mix it drops it down to 8v so I think I need a mosfet with a lower gate voltage which is in fact what I used in my earlier build. I've ordered some but will take a day or so to arrive but I think that will solve my issue. Thanks for helping think through this, it's been awhile since I've had to troubleshoot any mosfet/pwm based stuff.

:)
 

Fish_Sticks

Active Member
View Badges
Joined
Jan 22, 2019
Messages
446
Reaction score
970
Rating - 0%
0   0   0
Thought I'd share my quick experience setting up a raspberry pi / reef pi for the first time.

Fairly tech savvy, so I didn't have any issues getting it going; however, I found the raspberry PI OS installation (and youtube/web guides) a little... diverting and outdated, which may be trouble for some of our less savvy folks.

The guides had me looking for 'raspian' (lol) and downloading third party imagers - seems clear these are made useless by raspberry pi's official imager, and that raspian is now called raspberry pi os. The software name change was clear after a quick google search, but I wasted about 10 mins of time wondering what exactly I was looking for or using to 'install' the main raspberry operating system to the SD card.

Basically all you gotta do is download the raspberry pi imager (basically an installer in laymans terms, but ya) from:
Insert your micro sd to you pc. run the raspberry pi os imager, and click two buttons to 'install' rasberry pi os to your microsd via raspberry pi's imager, as it contains the OS.

After that, insert your sd to the pi, insert your monitors hdmi, keyboard usb, mouse usb, insert the pi power plug to turn it on, and you can use the pi just like an actual computer now with a display and everything to connect to the internet, and download the reefpi installer from directly inside the raspberry pi. Im sure I missed a few things, but the process is way simpler than it seems.

After that, it took ~5 mins to reboot, and be in the reefpi dashboard. Another 10 mins to configure some settings using the reef-pi interface. And around 20 mins to setup email alerts in the reef-pi - primarily because there were a few extra steps in gmail's settings to get smtp to work, but Rob has a great guide in this thread on how to do that.

Hope some people find this helpful - the whole thing can be done in 5-10 minutes tbh.
Lmk if I'm just being a noob and if I missed some important steps/features :D but it turns out to be incredibly straight forward, and I'm pretty confident someone with far less tech experience could get it going. I think the biggest hurdle for most people would surprisingly be figuring out what their PI's local IP is to get to the dashboard.

You can figure that out by logging in to your internet router (the black internet box) and looking at your connected device list. In my case it was 192.168.1.111 - but I had this knowledge from configuring private runescape servers back in the day lol - as well as a little javascript, setting up websites via cpanel, etc - although I am definitely no super genius and technically dumb to most people in this thread, there are a lot of people who may find that the most difficult part.
 
Last edited:

bishoptf

Valuable Member
View Badges
Joined
Jan 1, 2019
Messages
1,349
Reaction score
1,724
Location
Missouri
Rating - 0%
0   0   0
Anyone using the Nicrew Hyperreef LED, the one that can be controlled via 0-10v pwm? I have my reefpi working and taking measurement it is putting out on both channel accordingly but when I plug it in to the Hyperreef it does not do anything, lights stay at the same intensity. These are being fed by a pca9685 with frequency set to 1000, does anyone have this working with these lights and if so what else is needed in order to get them to work, supposed to be like Kessil 0-10v analog but it no workie for me yet. :(

Thanks :)
 

Schreiber

Leviathan
View Badges
Joined
Jun 28, 2016
Messages
468
Reaction score
593
Location
Knoxville
Rating - 0%
0   0   0
Anyone using the Nicrew Hyperreef LED, the one that can be controlled via 0-10v pwm? I have my reefpi working and taking measurement it is putting out on both channel accordingly but when I plug it in to the Hyperreef it does not do anything, lights stay at the same intensity. These are being fed by a pca9685 with frequency set to 1000, does anyone have this working with these lights and if so what else is needed in order to get them to work, supposed to be like Kessil 0-10v analog but it no workie for me yet. :(

Thanks :)

Never used it. How do you have your 3.5mm jacks wired? I found this chart on their product page that may be helpful.

1618709446418.png
 

bishoptf

Valuable Member
View Badges
Joined
Jan 1, 2019
Messages
1,349
Reaction score
1,724
Location
Missouri
Rating - 0%
0   0   0
Never used it. How do you have your 3.5mm jacks wired? I found this chart on their product page that may be helpful.

1618709446418.png
Yeah thats how I have it wired up, I have verified with a MM that tip and ring have voltage and it goes up and down based on what I have the channel set to but the light just stays doesn't budge from its current setting. Maybe that function is not working, never tested it and have no way to test it but right now I am getting nothing.
 

elysics

Valuable Member
View Badges
Joined
Jan 15, 2020
Messages
1,591
Reaction score
1,590
Rating - 0%
0   0   0
Never used it. How do you have your 3.5mm jacks wired? I found this chart on their product page that may be helpful.

1618709446418.png
Are you sure it's 10v pwm control and not 0..10v analog? And your multimeter probably can't tell the difference, while a lamp might.

Do you have a breadboard and a potentiometer or two? Then I'd put those between a 12v power supply and the lamp channels to test if they work at all and then you can measure with your multimeter without it lying
 

dmolavi

Well-Known Member
View Badges
Joined
Jan 3, 2015
Messages
509
Reaction score
646
Location
United States
Rating - 0%
0   0   0
Anyone using the Nicrew Hyperreef LED, the one that can be controlled via 0-10v pwm? I have my reefpi working and taking measurement it is putting out on both channel accordingly but when I plug it in to the Hyperreef it does not do anything, lights stay at the same intensity. These are being fed by a pca9685 with frequency set to 1000, does anyone have this working with these lights and if so what else is needed in order to get them to work, supposed to be like Kessil 0-10v analog but it no workie for me yet. :(

Thanks :)
Is it 0-10V analog or 10V PWM?
 

bishoptf

Valuable Member
View Badges
Joined
Jan 1, 2019
Messages
1,349
Reaction score
1,724
Location
Missouri
Rating - 0%
0   0   0
Are you sure it's 10v pwm control and not 0..10v analog? And your multimeter probably can't tell the difference, while a lamp might.

Do you have a breadboard and a potentiometer or two? Then I'd put those between a 12v power supply and the lamp channels to test if they work at all and then you can measure with your multimeter without it lying

Is it 0-10V analog or 10V PWM?
All I know is these lights Nicrew Hyperreef led say they can be controlled with a Kessil controller which is a 0-10v input, which I believe it should be analog input like this - hyperreef link

However I am not using a robotank, I am using an @Michael Lane goby hat and have one of his old mars aqua boards - Mars aqua board and here is how I have it wired up, I do see voltage output based on the pwm setting.

mars-aqua-pic.png

But so far the light does not change intensity when hooked up, nada. :(

Anyone has any thoughts let me know.
 

elysics

Valuable Member
View Badges
Joined
Jan 15, 2020
Messages
1,591
Reaction score
1,590
Rating - 0%
0   0   0
All I know is these lights Nicrew Hyperreef led say they can be controlled with a Kessil controller which is a 0-10v input, which I believe it should be analog input like this - hyperreef link

However I am not using a robotank, I am using an @Michael Lane goby hat and have one of his old mars aqua boards - Mars aqua board and here is how I have it wired up, I do see voltage output based on the pwm setting.

mars-aqua-pic.png

But so far the light does not change intensity when hooked up, nada. :(

Anyone has any thoughts let me know.
what does "does not change" mean? Stays at 100%? Stays off?

Different approach: do the voltages also show up at the end of the cable, or did you just measure the output at the board?

Also, what happens if instead of connecting the cable to the board, you connect the ground line to ground, and the two other lines to a resistor that also goes to ground?
 
Back
Top