constantlylearning
Electrical
- Jan 5, 2006
- 39
Let me start this post by saying that I'm a protection and controls engineer with a good bit of automation experience and while I have a decent amount of industry experience with controls, I have little to no experience with intrinsic safety. As such, I am humbly asking for some clarification.
Background: For my home, I'd like to monitor the liquid level in my outdoor, above-ground, 250 gal propane tank. Initially, when looking at doing this, I thought it would be simple to just add a Rochester R3D sensor to a device such as an ESP32, Ardunio, etc. However, the more I dug into this, the more I realized that to "do this right", I really needed make sure that what I was doing was intrinsically safe.
After digging around at what it would take, I decided to just buy the TankUtility remote monitor. This worked great and allowed me to have access to that data through TankUtility's API. However, about a year after I bought this device, Generac bought TankUtility and shortly thereafter, my device "mysteriously" stopped working.
After this, I decided to revert back to "rolling my own" and have gone so far as to purchase an intriniscally safe diode barrier and enclosure from a company that I do believe is completely trustworthy, but would rather not mention. They've been very helpful, but I do not want to appear to be questioning their expertise in any way.
So, at this point, I'm about $800 dollars into this project and I am stalled.
More Background: The reason I'm "stalled" is that I don't fully understand a few things that the barrier supplier is telling me.
1.) I'm trying to understand the intrinsic safety calculations myself. I'd like to select my own cable. As far as I can tell, I need to do the following:
A.) For the allowable cable capacitance calculation, I "think" I need to use the one-way length between the barrier and the sensor to calculate total cable capacitance. I think the one way length is appropriate because capacitance is a "shunt" parameter in this context. To do this, I would add together the wire-to-wire capacitance and the wire-to-shield capacitance. <<--REALLY NOT SURE ABOUT THIS.
B.) For the allowable cable inductance calculation, I "think" I need to use the "two-way" length between the barrier and the sensor because I believe the total inductance is a series parameter and not a shunt parameter.
I go into a little more detail about this in my question 2.)
2.) The supplier that sold me the intrinsically safe barrier is telling me that I need to limit my cable run between the barrier and the sensor to less than 100 feet. However, I do not see any reason for this. Typical control cable capacitances and inductances are much less than 60 pF/ft and 20 uH/ft respectively. This means (I think) that I could potentially run multiple hundreds of feet before the capacitance and inductances of the control cable contribute enough to the total capacitance and inductance of the "allowable" totals as dictated by the barrier spec before it becomes a problem. To be conservative, I'm doubling the distance of the cable run in my calculations to account for the "out-and-back" distance of the sensor current and then I'm doubling it again to be conservative. I'm using the "Nom. Capacitance Cond-to-Other (Conds + Shield)" as my capacitance value for the cable. Am I wrong about the way I'm calculating total cable capacitance? I can't help but feel I'm overlooking something, but I can't seem to find what it would be. As I see it, the cable capacitance is small enough that it takes a large amount of cable (hundreds and hundreds of feet) before one starts to approach the limit as dictated by the barrier.
The propane sensor is a Rochester 9700-00001 with the following parameters:
Vmax = 14 V
Imax = 200 mA
Ci = 2.0 nF
Li = 4.8 uH
The barrier data is: Eaton MTL 7761ac
V = 12
Ohms = 90
I = 100 mA
Ca = 4.90 uH (still trying to verify this number)
La = 3.72 mH (still trying to verify this number)
3.) Also, the supplier is telling me that for an outdoor installation, it is acceptable to run a non-outdoor-rated cable in PVC conduit. This might be true, but I am not aware of any entity that would "give a thumbs-up" for this. Since this application requires that I run the cable underground (still through conduit), to ME this means the conductor should be rated for wet conditions or direct-burial. I'm willing to say that I might be wrong about this, but I'm concerned that a non-outdoor rated cable may allow water ingress, which I believe would greatly increase the capacitance of the cable.
4.) Am I correct in assuming that only the cable length between the barrier and the sensor is important as far as "allowable capacitance" and "allowable inductance" are concerned? My understanding is that any cable capacitance/inductance "upstream" of the barrier is not a concern because because the barrier limits the energy delivery downstream of its location.
Background: For my home, I'd like to monitor the liquid level in my outdoor, above-ground, 250 gal propane tank. Initially, when looking at doing this, I thought it would be simple to just add a Rochester R3D sensor to a device such as an ESP32, Ardunio, etc. However, the more I dug into this, the more I realized that to "do this right", I really needed make sure that what I was doing was intrinsically safe.
After digging around at what it would take, I decided to just buy the TankUtility remote monitor. This worked great and allowed me to have access to that data through TankUtility's API. However, about a year after I bought this device, Generac bought TankUtility and shortly thereafter, my device "mysteriously" stopped working.
After this, I decided to revert back to "rolling my own" and have gone so far as to purchase an intriniscally safe diode barrier and enclosure from a company that I do believe is completely trustworthy, but would rather not mention. They've been very helpful, but I do not want to appear to be questioning their expertise in any way.
So, at this point, I'm about $800 dollars into this project and I am stalled.
More Background: The reason I'm "stalled" is that I don't fully understand a few things that the barrier supplier is telling me.
1.) I'm trying to understand the intrinsic safety calculations myself. I'd like to select my own cable. As far as I can tell, I need to do the following:
A.) For the allowable cable capacitance calculation, I "think" I need to use the one-way length between the barrier and the sensor to calculate total cable capacitance. I think the one way length is appropriate because capacitance is a "shunt" parameter in this context. To do this, I would add together the wire-to-wire capacitance and the wire-to-shield capacitance. <<--REALLY NOT SURE ABOUT THIS.
B.) For the allowable cable inductance calculation, I "think" I need to use the "two-way" length between the barrier and the sensor because I believe the total inductance is a series parameter and not a shunt parameter.
I go into a little more detail about this in my question 2.)
2.) The supplier that sold me the intrinsically safe barrier is telling me that I need to limit my cable run between the barrier and the sensor to less than 100 feet. However, I do not see any reason for this. Typical control cable capacitances and inductances are much less than 60 pF/ft and 20 uH/ft respectively. This means (I think) that I could potentially run multiple hundreds of feet before the capacitance and inductances of the control cable contribute enough to the total capacitance and inductance of the "allowable" totals as dictated by the barrier spec before it becomes a problem. To be conservative, I'm doubling the distance of the cable run in my calculations to account for the "out-and-back" distance of the sensor current and then I'm doubling it again to be conservative. I'm using the "Nom. Capacitance Cond-to-Other (Conds + Shield)" as my capacitance value for the cable. Am I wrong about the way I'm calculating total cable capacitance? I can't help but feel I'm overlooking something, but I can't seem to find what it would be. As I see it, the cable capacitance is small enough that it takes a large amount of cable (hundreds and hundreds of feet) before one starts to approach the limit as dictated by the barrier.
The propane sensor is a Rochester 9700-00001 with the following parameters:
Vmax = 14 V
Imax = 200 mA
Ci = 2.0 nF
Li = 4.8 uH
The barrier data is: Eaton MTL 7761ac
V = 12
Ohms = 90
I = 100 mA
Ca = 4.90 uH (still trying to verify this number)
La = 3.72 mH (still trying to verify this number)
3.) Also, the supplier is telling me that for an outdoor installation, it is acceptable to run a non-outdoor-rated cable in PVC conduit. This might be true, but I am not aware of any entity that would "give a thumbs-up" for this. Since this application requires that I run the cable underground (still through conduit), to ME this means the conductor should be rated for wet conditions or direct-burial. I'm willing to say that I might be wrong about this, but I'm concerned that a non-outdoor rated cable may allow water ingress, which I believe would greatly increase the capacitance of the cable.
4.) Am I correct in assuming that only the cable length between the barrier and the sensor is important as far as "allowable capacitance" and "allowable inductance" are concerned? My understanding is that any cable capacitance/inductance "upstream" of the barrier is not a concern because because the barrier limits the energy delivery downstream of its location.