Innovation in Reef Technology - Page 2

H20ENG · 03/11/2009, 06:16 PM

Luis,
I'd love to find a decent price on a bunch of CTs. Any leads??

All,
FWIW, you can already use the 0-10v output of the Profilux for way more than lights

There are 4 each 0-10V outputs built right in.

Spearo2953 · 03/11/2009, 07:37 PM

Here's a little info on varying the speed of motors:

AC variable frequency drives (VFD) vary the voltage and frequency to control the speed of a motor. The motor must be 3 phase AC. The VFD will convert your house single phase AC power to 3 phase AC power for the motor. You won't find an 3 phase motors in the retail trade,

DC variable speed drives vary the voltage to control the speed of the motor. I believe this is how Tunze, Koralia, etc. are doing their variable speed motors (24v dc).

I'd be happy to go into more detail, but i don't want to put anyone to sleep.

DaveMorris · 03/12/2009, 12:27 AM

That is why I think it is far more complicated to vary the motor speed within our application. Are you going to open up the sealed Baldor motor and set it up to run 3 phase? Yes it can be done, but not without a fair amount of work by someone with some real knowledge on it. I would guess that there are pumps out there thaat we could use, but aren't in the aquarium mainstream. I would love to do this with my closed loop.

BeanAnimal · 03/12/2009, 12:43 AM

Dave... it is fairly trivial to mount a 3-phase motor to a standard wet end. I.E a DART uses a 56 frame motor. You can get a cheap or surpluss 3-phase 56 frame motor anywhere. You can buy cheap VFDs from several online vendors (automation direct, etc) or high end VFDs very cheap on eBay. I prefer the Yaskawa drives myself.

BeanAnimal · 03/12/2009, 12:47 AM

Lets talk about current transformers....

Unless the controller has some fancy software designed to do RMS calcs, a CT is going to be pretty useless, as you are going to be reading instantaneous current/voltage values from the sinewave.. I.E the 0-10v signal is going to track the current/voltage sinewaves.

search for my name in rec.electonics and you will get the info needed to build the interface (it will be easiest with a micro).

buck50bmg · 03/12/2009, 05:43 AM

Bean

Im gearing up to start a SLC 500 PLC controller. I will be interfacing my RDII pump and use a flowmeter for GPH control. Also will be adding some 0-10 CT's. Just need to get my code in order.

I have to stop and think about how the industry is finally listening to the customers. Its great. Heck I would go with the Profilux but I have been bit by the PLC bug.

BeanAnimal · 03/12/2009, 07:27 AM

What flow meter did you choose? The profilux is certainly a nice toy/tool. I spoke with developer at our MACNA a few years ago... maybe some of my ideas rubbed off

buck50bmg · 03/12/2009, 07:50 AM

I have a GF Signet paddle wheel, (the one with the red top) I have the 1/2" housing but I need it in a 1". Too expensive right now for a hunk of PVC.

This PLC project is going to last me the whole summer. I have to try not to force it and let it be developed correctly. Im rusty with SLC 500 code as I have only done PLC-5, and that was 10 years ago.

luisagos · 03/12/2009, 05:53 PM

For those that work with real PLC can really appreciate what a real controller is and see how advance Profilux Controller really is.

The current sensors that I am looking at, do the RMS calculation within the circuit them self.

If you guys just go to their support site, your going to see how serious they are wanting to be the best controller on the market.

Just don't tell them that they already are, I want more functions.

BeanAnimal · 03/12/2009, 06:25 PM

What current sensors are you looking at? Units with onboard logic can get kinda spendy. My application was to monitor RMS voltage and current on 24 outputs. I chose to use inexpensive current transformers at first, but found some very inexpensive 3pin to-220 packages (shunts with integrated drivers) that do a nice job of outputing a proportionate voltage and don't have the dangers associated with an open secondary CT. (I will find the model number of I can). I am using an atmega128 to do the calcs.

luisagos · 03/13/2009, 04:12 AM

Quote:

Originally posted by BeanAnimal
What current sensors are you looking at? Units with onboard logic can get kinda spendy. My application was to monitor RMS voltage and current on 24 outputs. I chose to use inexpensive current transformers at first, but found some very inexpensive 3pin to-220 packages (shunts with integrated drivers) that do a nice job of outputing a proportionate voltage and don't have the dangers associated with an open secondary CT. (I will find the model number of I can). I am using an atmega128 to do the calcs.

Here is some of them.

http://www.ampsolution.com/

http://www.cy-sensors.com/CE_Transducer_T.htm

http://www.phidgets.com/products.php...roduct_id=1119

BeanAnimal · 03/13/2009, 07:47 AM

ALL of those devices will output the 0-10v signal in the form of a sinewave if used to monitor 120VAC loads, useless for your purpose. They are just current transformers.

Remember 120VAC is delivered in the form of a 60Hz sine wave. At any given instant, both current and voltage will be at someplace along that sinewave. That is, the voltage will be somewhere between -170V and +170V and the instantaneous current will track accordingly.

Example. Your Current Transformer is sized to read 10V at 20A
5V at 10A
0V at 0A
assumung it is linear, connecting it to a 20A resistive load will produce an output voltage on the CT that swings between 0-10V at a freq. of 60Hz and tracks the zero crossing of the AC sinewave exactly. So unless the profilux monitors the 0-10V signal at a freq. greater than 60Hz and can perform true RMS calcs on those values... then the readings will be useless

luisagos · 03/13/2009, 11:49 AM

Why make things more complicated then it needs to be.

Take this transducer for an example.

http://www.cy-sensors.com/CYCS11-32H1.pdf

The measuring accuracy of this sensor is .005 thats 5ma of error at any given time.

No need for Profilux to do the math, this transducer is proportional to the average effective value (RMS) of the input AC current.

We are not using a VFD, so the freg will stay at 60 Hz.

Every 1 s we will get 2.5 samples, convert to a 0-10 dcv.
If the transducer wasn't doing the averaging, yes your right about the current samples being all over.

BeanAnimal · 03/13/2009, 06:17 PM

Quote:

Originally posted by luisagos
Why make things more complicated then it needs to be.

Take this transducer for an example.

http://www.cy-sensors.com/CYCS11-32H1.pdf

The measuring accuracy of this sensor is .005 thats 5ma of error at any given time.

No need for Profilux to do the math, this transducer is proportional to the average effective value (RMS) of the input AC current.

We are not using a VFD, so the freg will stay at 60 Hz.

Every 1 s we will get 2.5 samples, convert to a 0-10 dcv.
If the transducer wasn't doing the averaging, yes your right about the current samples being all over.

I am not making things more complicated than they need be. I am pointing out something that most people do not understand.... and that not all of the devices you linked to are suitable

My comments were in regard to straight CTs, Hall Effect units and/or shunts that do not do the averaging. I just glanced at the links you posted (and am very familiar with the phidgets unit). The phidgets unit does not do RMS averaging and I did not look at all of the offerings from ampsolutions or CY.

As I mentioned, units that do have internal averaging circuits are pretty spendy. $75-$150 is not a problem if you only need to monitor 1 or 2 devices. In my case, the goal was to monitor 24 devices

$2-$5 was my price range per device.

luisagos · 03/13/2009, 07:21 PM

I will admit the phidget amp circuit is not suited for the application i am looking for, but I have them on my list because I was going to contact them to see if they are willing to take them to the next level.

If you read my thread, you will see I am very careful with what I say.
My favorite word is reference or this is what I am looking at.

I am even looking at the killawatt meter, to see if it can be modify to suit our needs, they are cheap and do a fairly well job with a neat package.

I haven't commited on the sensors yet, because I am still waiting to see what Profilux is going to do with the new power bar they are designing.

BTW. we probably lost 99% of our reader with the last posts, hehe.
But I will promise once the smoke clears (bad pun, lol) it will be plug & play devices.

BeanAnimal · 03/13/2009, 09:29 PM

Good luck with your project

When you get a price and a small qty source for the CY CT part, let us know.

03/11/2009, 07:37 PM	#27
Spearo2953 Registered Member Join Date: Feb 2009 Posts: 17	VFD primer Here's a little info on varying the speed of motors: AC variable frequency drives (VFD) vary the voltage and frequency to control the speed of a motor. The motor must be 3 phase AC. The VFD will convert your house single phase AC power to 3 phase AC power for the motor. You won't find an 3 phase motors in the retail trade, DC variable speed drives vary the voltage to control the speed of the motor. I believe this is how Tunze, Koralia, etc. are doing their variable speed motors (24v dc). I'd be happy to go into more detail, but i don't want to put anyone to sleep.

03/12/2009, 12:27 AM	#28
DaveMorris VP, SDMAS Join Date: Nov 2006 Location: Sunny San Diego, CA Posts: 1,309	That is why I think it is far more complicated to vary the motor speed within our application. Are you going to open up the sealed Baldor motor and set it up to run 3 phase? Yes it can be done, but not without a fair amount of work by someone with some real knowledge on it. I would guess that there are pumps out there thaat we could use, but aren't in the aquarium mainstream. I would love to do this with my closed loop. __________________ Dave My 130g tank: http://www.reefcentral.com/forums/showthread.php?t=2387989 Current Tank Info: 130g build in progress

03/13/2009, 11:49 AM	#38
luisagos Registered Member Join Date: Nov 2008 Location: Florida Posts: 1,453	Why make things more complicated then it needs to be. Take this transducer for an example. http://www.cy-sensors.com/CYCS11-32H1.pdf The measuring accuracy of this sensor is .005 thats 5ma of error at any given time. No need for Profilux to do the math, this transducer is proportional to the average effective value (RMS) of the input AC current. We are not using a VFD, so the freg will stay at 60 Hz. Every 1 s we will get 2.5 samples, convert to a 0-10 dcv. If the transducer wasn't doing the averaging, yes your right about the current samples being all over. Last edited by luisagos; 03/13/2009 at 12:27 PM.

03/11/2009, 06:16 PM	#26
H20ENG Pro builder/aquarist hack Join Date: Jul 2002 Location: NORCAL (Vacaville, CA) Posts: 5,125	Luis, I'd love to find a decent price on a bunch of CTs. Any leads?? All, FWIW, you can already use the 0-10v output of the Profilux for way more than lights There are 4 each 0-10V outputs built right in.

03/12/2009, 12:43 AM	#29
BeanAnimal Registered Member Join Date: Feb 2003 Location: Pittsburgh Posts: 20,772	Dave... it is fairly trivial to mount a 3-phase motor to a standard wet end. I.E a DART uses a 56 frame motor. You can get a cheap or surpluss 3-phase 56 frame motor anywhere. You can buy cheap VFDs from several online vendors (automation direct, etc) or high end VFDs very cheap on eBay. I prefer the Yaskawa drives myself.

03/12/2009, 12:47 AM	#30
BeanAnimal Registered Member Join Date: Feb 2003 Location: Pittsburgh Posts: 20,772	Lets talk about current transformers.... Unless the controller has some fancy software designed to do RMS calcs, a CT is going to be pretty useless, as you are going to be reading instantaneous current/voltage values from the sinewave.. I.E the 0-10v signal is going to track the current/voltage sinewaves. search for my name in rec.electonics and you will get the info needed to build the interface (it will be easiest with a micro).

03/12/2009, 05:43 AM	#31
buck50bmg Moved On Join Date: Jul 2007 Location: Pittsburgh, Pa Posts: 792	Bean Im gearing up to start a SLC 500 PLC controller. I will be interfacing my RDII pump and use a flowmeter for GPH control. Also will be adding some 0-10 CT's. Just need to get my code in order. I have to stop and think about how the industry is finally listening to the customers. Its great. Heck I would go with the Profilux but I have been bit by the PLC bug.

03/12/2009, 07:27 AM	#32
BeanAnimal Registered Member Join Date: Feb 2003 Location: Pittsburgh Posts: 20,772	What flow meter did you choose? The profilux is certainly a nice toy/tool. I spoke with developer at our MACNA a few years ago... maybe some of my ideas rubbed off

03/12/2009, 07:50 AM	#33
buck50bmg Moved On Join Date: Jul 2007 Location: Pittsburgh, Pa Posts: 792	I have a GF Signet paddle wheel, (the one with the red top) I have the 1/2" housing but I need it in a 1". Too expensive right now for a hunk of PVC. This PLC project is going to last me the whole summer. I have to try not to force it and let it be developed correctly. Im rusty with SLC 500 code as I have only done PLC-5, and that was 10 years ago.

03/12/2009, 05:53 PM	#34
luisagos Registered Member Join Date: Nov 2008 Location: Florida Posts: 1,453	For those that work with real PLC can really appreciate what a real controller is and see how advance Profilux Controller really is. The current sensors that I am looking at, do the RMS calculation within the circuit them self. If you guys just go to their support site, your going to see how serious they are wanting to be the best controller on the market. Just don't tell them that they already are, I want more functions.

03/12/2009, 06:25 PM	#35
BeanAnimal Registered Member Join Date: Feb 2003 Location: Pittsburgh Posts: 20,772	What current sensors are you looking at? Units with onboard logic can get kinda spendy. My application was to monitor RMS voltage and current on 24 outputs. I chose to use inexpensive current transformers at first, but found some very inexpensive 3pin to-220 packages (shunts with integrated drivers) that do a nice job of outputing a proportionate voltage and don't have the dangers associated with an open secondary CT. (I will find the model number of I can). I am using an atmega128 to do the calcs.

03/13/2009, 07:47 AM	#37
BeanAnimal Registered Member Join Date: Feb 2003 Location: Pittsburgh Posts: 20,772	ALL of those devices will output the 0-10v signal in the form of a sinewave if used to monitor 120VAC loads, useless for your purpose. They are just current transformers. Remember 120VAC is delivered in the form of a 60Hz sine wave. At any given instant, both current and voltage will be at someplace along that sinewave. That is, the voltage will be somewhere between -170V and +170V and the instantaneous current will track accordingly. Example. Your Current Transformer is sized to read 10V at 20A 5V at 10A 0V at 0A assumung it is linear, connecting it to a 20A resistive load will produce an output voltage on the CT that swings between 0-10V at a freq. of 60Hz and tracks the zero crossing of the AC sinewave exactly. So unless the profilux monitors the 0-10V signal at a freq. greater than 60Hz and can perform true RMS calcs on those values... then the readings will be useless

03/13/2009, 07:21 PM	#40
luisagos Registered Member Join Date: Nov 2008 Location: Florida Posts: 1,453	I will admit the phidget amp circuit is not suited for the application i am looking for, but I have them on my list because I was going to contact them to see if they are willing to take them to the next level. If you read my thread, you will see I am very careful with what I say. My favorite word is reference or this is what I am looking at. I am even looking at the killawatt meter, to see if it can be modify to suit our needs, they are cheap and do a fairly well job with a neat package. I haven't commited on the sensors yet, because I am still waiting to see what Profilux is going to do with the new power bar they are designing. BTW. we probably lost 99% of our reader with the last posts, hehe. But I will promise once the smoke clears (bad pun, lol) it will be plug & play devices.

03/13/2009, 09:29 PM	#41
BeanAnimal Registered Member Join Date: Feb 2003 Location: Pittsburgh Posts: 20,772	Good luck with your project When you get a price and a small qty source for the CY CT part, let us know.