Maxi doser and profilux

[Dave-T]

Unfortunately, it is never as simple as we might think. Variables and resources in the code are calculated and defined just right, it is not a simple arbitrary limit and we will probably have to change the variable type and review different parts of the code to change this.

For the volume issue when using the desired flow, yes we are already aware of that. Sorry. Will be fixed ASAP.

Gaël

Yes, well I’m a professional SW developer, I do have a sense of how difficult these things are. It sounds like maybe your code uses short integers instead of long ints or floats?

But anyway, for the volume issue, where does this leave me? Can I trust it to be accurate if I set the desired rate equal to the max rate I measured during calibration?

And Is it safe to use max speed (3), or would it be more accurate or reliable at 1 or 2?

 

[Dave-T]

I see there will be a fix coming for this. I saw the same thing. I thought I did something wrong so I did a second set of my 5 calibration method. Doing it again and entering the new number fixed the issue. Might be a work around until the fix comes out.

Ken - I don’t follow. What fixed the issue?

But even if you had to set the pumps at different desired speeds because they calibrated differently, you could still use the Ken technique with timers - you’d just need two timers instead of one.

 

[KenO]

Ken - I don’t follow. What fixed the issue?

But even if you had to set the pumps at different desired speeds because they calibrated differently, you could still use the Ken technique with timers - you’d just need two timers instead of one.

Well I thought I had messed up the calibration. After the first calibration I did a dose of 125ml and got a different number. So I did another calibration. After the 2nd calibration (remember I do 5 calibrations in a row and measure the total amount and divide by 5) I did the same 125ml dose and got 125ml measured. So I thought it was just an error in my first round of calibrations. Why it worked I have no idea.

 

[Dave-T]

Well I thought I had messed up the calibration. After the first calibration I did a dose of 125ml and got a different number. So I did another calibration. After the 2nd calibration (remember I do 5 calibrations in a row and measure the total amount and divide by 5) I did the same 125ml dose and got 125ml measured. So I thought it was just an error in my first round of calibrations. Why it worked I have no idea.

You’re saying that you were able to enter a different rate for the “desired dose” field than the max rate, and it would dose at that rate? Because that shouldn’t be possible, with the current bug Gael confirmed.

 

[KenO]

You’re saying that you were able to enter a different rate for the “desired dose” field than the max rate, and it would dose at that rate? Because that shouldn’t be possible, with the current bug Gael confirmed.

Maybe a misunderstood the issue. I thought the issue was not dosing the correct amount after calibrating and setting a desired dose. So maybe my issue was just a miscalculation.

 

[Dave-T]

No, the issue is that the "desired flow" field doesn't do anything. It will continue to run at the fastest rate, no matter what you put there.

So I'm working a little more on this now. I'm really concerned about trusting these pumps to do AWC. If one pump stops working, starts working intermittently, or the relative rate between the pumps diverges - I will be screwed. Overfilling or draining the sump, or throwing salinity off.

I'd still really like to know if it's advisable to run the pumps at max speed. That is my preference, but if that increases the likelihood of one of the issues above, then I will run them slower.

But stuff happens, so I need to plan for the worst. I'd like to stop the pumps from running if my salinity or sump level goes into alarm. I can set up a programmable logic to evaluate that state. Unfortunately, the pumps don't let you tell them to stop based on an alarm condition or programmable logic. But switch channels do! So if I have a PL that includes those alarm checks, combined with a timer, I can use the Ken technique to run the pumps and keep them from running during an alarm state.

Does anyone see anything wrong with that?

 

[KenO]

No, the issue is that the "desired flow" field doesn't do anything. It will continue to run at the fastest rate, no matter what you put there.

So I'm working a little more on this now. I'm really concerned about trusting these pumps to do AWC. If one pump stops working, starts working intermittently, or the relative rate between the pumps diverges - I will be screwed. Overfilling or draining the sump, or throwing salinity off.

I'd still really like to know if it's advisable to run the pumps at max speed. That is my preference, but if that increases the likelihood of one of the issues above, then I will run them slower.

But stuff happens, so I need to plan for the worst. I'd like to stop the pumps from running if my salinity or sump level goes into alarm. I can set up a programmable logic to evaluate that state. Unfortunately, the pumps don't let you tell them to stop based on an alarm condition or programmable logic. But switch channels do! So if I have a PL that includes those alarm checks, combined with a timer, I can use the Ken technique to run the pumps and keep them from running during an alarm state.

Does anyone see anything wrong with that?

I would have to disagree with you on the desired flow. I have mine set to 30ml/min and my calibration number is 60+. My pumps are running at 30ml/min. I've measured it. I've also done specific dose amounts and used a stopwatch and a container to confirm the 30ml/min. So not sure what you are seeing with your pumps and the desired flow setting. I'm running rev 1.45 on my pumps.

 

[KenO]

I'm not running at the higher speeds like 2 or 3. Is the issue with the higher speed settings?

 

[KenO]

When I was testing my theory with using the switch channels and the desired speed this is what I did. Remember the pumps are set to a desired level of 30ml/min. I created a timer to run for 5 min. I tied it to a virtual switch. I let the timer turn on/off. I had put containers to collect the water in the output of the pumps. I got 149mls out of the one pump and 150mls out of the other. If the pumps ran at the calibration setting I would have had 2x+ that amount of water collected.

 

[KenO]

If you are worried about the conditions you stated above, why not use a set of optical sensors in your sump to measure the low and high point during the water change. You could either use them as just an alarm or use them as part of the AWC. If you use them as part of the AWC. You can set a max time a pump can run during the process if the pump runs longer than expected the p4 will shut it down. I've tested the GHL dual sensor AWC and it works. You could even use a different set of pumps to make the AWC go faster. They would be plugged into normal switch channels. If you have an auto top off and put the power for it into a switch outlet you could use a delay to turn off that ATO outlet for the length of time of the AWC.

 

[KenO]

For doing big water changes I think the continuous mode is better if you want to minimize salinity changes. It's hard to keep pumps in sync with each other. You can set an alarm using the conductivity probe and stop the AWC if the level goes too high or too low, but it would be a slow change vs removing and adding 43L at one time. Just my 2 cents.

 

[Dave-T]

I would have to disagree with you on the desired flow. I have mine set to 30ml/min and my calibration number is 60+. My pumps are running at 30ml/min. I've measured it. I've also done specific dose amounts and used a stopwatch and a container to confirm the 30ml/min. So not sure what you are seeing with your pumps and the desired flow setting. I'm running rev 1.45 on my pumps.

I can't explain that. I believe you, but mine are different. When I lowered the desired flow and measured the output, they were still running at max flow. Gael acknowledged that what I described is a known bug in the latest rev. Maybe I'm on a newer rev than you? I'll check what rev I'm on.

 

[Dave-T]

If you are worried about the conditions you stated above, why not use a set of optical sensors in your sump to measure the low and high point during the water change. You could either use them as just an alarm or use them as part of the AWC. If you use them as part of the AWC. You can set a max time a pump can run during the process if the pump runs longer than expected the p4 will shut it down. I've tested the GHL dual sensor AWC and it works. You could even use a different set of pumps to make the AWC go faster. They would be plugged into normal switch channels. If you have an auto top off and put the power for it into a switch outlet you could use a delay to turn off that ATO outlet for the length of time of the AWC.

That would work if one of the pumps totally wigged out, but it wouldn't help if a difference in pump output caused my salinity to drift over time. Level sensors aren't accurate enough, plus I have ATO running to maintain the sump level anyway. I want to err on the side of caution. It's not the end of the world if the AWC stops running. But it would be nice to trust it enough to keep it running while I'm away from home traveling or something, and have it shut off if anything looks awry.

 

[Dave-T]

For doing big water changes I think the continuous mode is better if you want to minimize salinity changes. It's hard to keep pumps in sync with each other. You can set an alarm using the conductivity probe and stop the AWC if the level goes too high or too low, but it would be a slow change vs removing and adding 43L at one time. Just my 2 cents.

43 liters is only about 3% of my system volume.

 

[KenO]

43 liters is only about 3% of my system volume.

I would look into the dual sensor AWC method. Depending on the pumps you pick, it would be a much faster process. You don't have to worry about calibrating pumps. You can set alarm conditions to stop the process if needed. Again just my 2 cents.

I have a slave maxi 2.2. When I was on the rev 1.44 level I noticed a GHL doser showing up on my home network. After the 1.45 rev it's gone. Makes me think that the logic board inside has the stand alone wifi chips on it.

 

[Dave-T]

I would look into the dual sensor AWC method. Depending on the pumps you pick, it would be a much faster process. You don't have to worry about calibrating pumps. You can set alarm conditions to stop the process if needed. Again just my 2 cents.

I have a slave maxi 2.2. When I was on the rev 1.44 level I noticed a GHL doser showing up on my home network. After the 1.45 rev it's gone. Makes me think that the logic board inside has the stand alone wifi chips on it.

That's a viable alternative, but I still like using the doser better. If I did relatively large water changes, then that would definitely affect the salinity since I'd be removing all the water at once and then replacing it all at once. And I could only control the amount of the water change via manually moving the level sensors.

But I'm finding a small wrinkle in your method using timers. GHL Timers don't allow second level resolution. So if the input and output pumps differ in their rates, then you can't really remove/add the same amounts. This is solved by setting the pump speed if it wasn't for that bug that is affecting me but not you.

 

[KenO]

That's a viable alternative, but I still like using the doser better. If I did relatively large water changes, then that would definitely affect the salinity since I'd be removing all the water at once and then replacing it all at once. And I could only control the amount of the water change via manually moving the level sensors.

But I'm finding a small wrinkle in your method using timers. GHL Timers don't allow second level resolution. So if the input and output pumps differ in their rates, then you can't really remove/add the same amounts. This is solved by setting the pump speed if it wasn't for that bug that is affecting me but not you.

Why would you need to manually move the sensors? One sensor would be set to the low point when the 43L is removed. The other sensor would be the normal level of the sump.

I did do AWC previously using 2 timers and 2 pump heads calibrated at different rates. It's not perfect and like you said you only have minutes you can set. My setup was around 650 gallons, so overtime I would get some level of drifting. But you also get changes in salinity with your skimmer if you do a wet or dry skim. I did more of a wet skim. You just have to calculate the time for each pump's calibration level. The desired flow makes the process much easier. If my salinity dropped, I would have my ATO pull premixed SW instead of RO/DI water. That would slowly raise the salinity back up. I'm in AZ so I would get between 3.5 to 5 gallons of evaporation per day depending on the time of year.

I see your in Boston, years back I use to work for HP Medical in Andover and lived in Tyngsboro. That was back in the early 90's.

 

[Dave-T]

Maybe we're talking about two different things. I thought you were talking about starting with the level at the level of the high sensor, removing water until you reach the low level, and then refilling until it's back to the high level. So the amount of water changed is dependent on how low the low sensor is.

Meanwhile, I worked at HP Medical in Andover in the early 90's too! I will PM you to continue this discussion...

 

[KenO]

Maybe we're talking about two different things. I thought you were talking about starting with the level at the level of the high sensor, removing water until you reach the low level, and then refilling until it's back to the high level. So the amount of water changed is dependent on how low the low sensor is.

Meanwhile, I worked at HP Medical in Andover in the early 90's too! I will PM you to continue this discussion...

GHL has a video on how to do the 2 sensor water change. I followed the instructions and it works.

 

[Dave-T]

A fly in the ointment, for using the Maxi Doser for AWC. I just realized that the dosing pumps will continue to run, even after they know that their container is empty. This is bad. If your salt water container runs dry, both the fill and drain pumps will continue to run on whatever schedule you set up for AWC. I think they should stop running. I could see why it wouldn't be a problem if a pump continued to run if you're using it for dosing or something, but it is a problem in this case.

Can anyone think of a way around this? I thought I'd use programmable logic to control the pump and it would shut off if the pump was in an alarm state due to an empty container, but it looks like PL can't access dosing pump alarm states.

I think I could add a level sensor to my mixing tank and work off that, but that's one more possible point of failure and adds complexity. I don't get why you can't have the pump shut off if it knows the container it draw from is empty (or the container it's filling is full).

I guess I'll enter a support ticket on this.