Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Does this design have Isolation problem considering ATEX ?

Status
Not open for further replies.

hm_fa_da

Full Member level 5
Joined
Sep 16, 2003
Messages
287
Helped
10
Reputation
20
Reaction score
4
Trophy points
1,298
Activity points
3,217
Hi,

I'm designing a device which connects to two Gas volume converter at same time to read out them by RS232 port.
Converter like Actaris Corus are completely Isolated by optocouplers and won't have problem.

But Converters like Elster EK220 RS232 Port is not electrically isolated from the main board ( the main board is connected to temp and pressure sensors ...)
Note all devices are battery powered separately. My device must supply 9V power for EK220 RS232 ports when connecting to it.
the schematic is like below:
EDA.png

Each device has it's own Battery as you see in schematic. but my device should power up conerters 9V input when connecting to them.

In Ek220, you see 10M resistor which is connected between 0V of main board and Aluminum box of it.

In my design, i doubt maybe there is a problem in design and some risks against ATEX considerations !

What's your suggestion ?
is this diagram okay ? or i have to use both power supply isolaters and optocouplers for TXD-RXD lines ?

I know in these devices sometimes barriers are used, but they are usually used when device is externally powered from i.e 220 to 9V or RS232 port is connecting to a device which has 220V inside.
 

You need an I.S. barrier on both power and RS-232 entering the hazardous area, for every device.

The Actaris Corus has no I.S. barrier for power or data ports, in the brochure https://www.accutherm.com.au/pdfs/Actaris%20-%20Corus%20-%20Brochure.pdf
They show you need an I.S. barrier on power and RS-232.

The EK220 also requires an I.S. barrier for power and serial data.

Your picture shows unsafe 9V power going into the hazardous area.
Your picture shows unsafe RS-232 (power) going into the hazardous area.

An opto-isolator does not prevent unsafe power from entering the device (volume converter).
So a possibility of spark exists on the wiring and enclosure of the device in the hazloc.
 
You need an I.S. barrier on both power and RS-232 entering the hazardous area, for every device.

The Actaris Corus has no I.S. barrier for power or data ports, in the brochure https://www.accutherm.com.au/pdfs/Actaris%20-%20Corus%20-%20Brochure.pdf
They show you need an I.S. barrier on power and RS-232.

The EK220 also requires an I.S. barrier for power and serial data.

Your picture shows unsafe 9V power going into the hazardous area.
Your picture shows unsafe RS-232 (power) going into the hazardous area.

An opto-isolator does not prevent unsafe power from entering the device (volume converter).
So a possibility of spark exists on the wiring and enclosure of the device in the hazloc.

Thanks for your reply, none of these devices are powered from high voltage power supplies (i.e 220->9V), all are 3.6V battery powered.
my device function is like Focus+ in this picture:
isolator.png

And I.S barrier is not used between Focus+ and Corus. i know Focus+ supplies power for Corus RS232 Part, because inside the Corus, the circuit is only externally powered and has no electrical connection to main board.

As i've seen I.S Barrier is used when External devices like RTUs are powered from 220/110V main powers ...
 

Apart from the first schematic i've sent which is not isolated at all and just talking about connecting like Focus+ and Corus, i don't see any difference between my device and Focus+ !

The 9V output of barrier is also connected to Corus, so if you cut and remove the left part in second diagram (RTU connection to Corus) like this:
isolator.png

The 9V power and RS232 Signals are going through wire to hazardous side.
Now if i use current limiting resistors + Zener diodes on RS232 signals and same thing for power line, i.e limiting current to 100mA, i think there would be no problem ?
The Voltage output in my device is not 9V in fact, the input range of correcter is 6-12V, i plan to use 6-7V.

in this PDF : www.omega.com/techref/pdf/z131-148.pdf

figure 4 shows the curves for voltage and current range for making ignition. my device working rate is much less than dangerous voltage and currents.

Considering Fig5,6,7 in the PDF shows that when 110V is used the barrier is needed. so when battery powered, there would be no high voltage fault .

For Data Isolation, you mean optocoupler can't be used ?! if so, what is used in barriers then ?!

Thanks & Regards.
 

There are more requirements for intrinsic safe equipment then considered so far. Besides being connected through voltage and current limiting barriers, you need to show that no component inside the device can heat up to the ignition temperature with the specified barriers. Also internal energy storage must not exceed a certain amount.
 
I'm using ATEX approved GSM module inside my device, i'm using 2x3.6V batteries (ATEX certified) in series that makes 7.2V for power supply, now i want to convert and regulate it down to 3.8-4V for GSM module.
currently i'm using Buck converter with 6.8uH inductor which the maximum energy stored in it for 2.4A current is about 19.5uJ. i saw in documents less than 20uJ is okay in ATEX certification. The module maximum current is 1.8A. so inductor energy would be about 11uJ. (less than 20uJ anyway)

1- Is this buck converter usage okay ?! or i have to use voltage drop regulator to avoid usage of inductor (so i'll lose Efficiency) ?
2- I'm using Ceramic caps in my board, i.e for the 3.8V line, if i use 2.2uF, energy stored in cap would be 15uJ. if i use 10uF, energy stored would be 72uJ which exceeds limitation. can i use 4 or 5 x 2.2uF separate caps in parallel to avoid more than 20uJ energy storing in each cap ?!!


Thanks & Regards.
 

Is your device located in the hazardous area or in the safe area? It's a bit confusing.

It doesn't make sense to put together some ATEX certified parts like GSM modem, batteries and expect the whole thing to meet ATEX, or connect to anything.
They will consider your buck-converter failing short-circuit and at 7.2V out. Where is this 9V power coming from?
 

Is your device located in the hazardous area or in the safe area? It's a bit confusing.

It doesn't make sense to put together some ATEX certified parts like GSM modem, batteries and expect the whole thing to meet ATEX, or connect to anything.
They will consider your buck-converter failing short-circuit and at 7.2V out. Where is this 9V power coming from?

I'm not going to get ATEX certificate at this step but trying to consider it's specs as much as i can. the reason to ask these questions is trying to design compatible device with ATEX not just put them together.

I'm using ATEX fuses exactly after battery for GSM Buck converter line, it's output is protected by two parallel 4.5V zener diodes, so if buck converter fails short circuit, the fuse will be open. the buck converter has short circuit and thermal protection features too. i don't see any difference between LDO and Buck converter at this matter (failing short circuit).
The device is not installed in zone 1 but i like to design it for Zone 1 too !
The Volume Converter input voltage range is 6-12V. i'm going to inject 6-7V to it not 9V.
The whole circuit is in IP67 box.
About the energy stored in components, is my last idea true ? (replacing 10uF cap by 5 x 2.2uF in 3.8V line i.e)
 

You need a good drawing of the system to figure out where ATEX will have a concern. Right now the design is all in your head and still hard to figure out what is located where.

For the stored (capacitive) energy calculation in 60079-11 it is the sum of all capacitors in the product+tolerances, assumed charged to the highest voltage available- unless you have an IS barrier on power to limit V.

Group IIC with 1.5 safety factor is 13.5uF maximum at 7.2V
This is what makes a SMPS hard to use in HazLoc - they need lots of capacitance.
 
Thanks for your guidance,

I think i can remove all SMPS circuits and use LDO instead and keep capacitance less than 13.5uF ... using no inductors too.
Assume my device is I.S (considering all other ATEX specs related to circuit design ...). i still need to connect to two separate Volume converters.
for Isolation, i didn't find I.S Isolated DC/DC converter.


First Solution and Question : If my Device be Designed I.S and can be placed in Zone1 and connect to two Volume Correcters i.e EK220 at same place without Electrical Isolation (Considering current and voltage limitations for voltage and RS232 data lines), this way breaks ATEX rules ?! all considerations about limiting current and voltage are done, but there is no electrical Isolation between two Correcters. is there any rule saying there should be Electrical Isolation ?!

Second Solution:
All devices and connections I.S, Electrical Isolation is done in this way:
AT-Check1.png

Notes:
1- Assume my device is designed I.S
2- The battery voltage 7.2V is connected to V input of Volume corrector after ATEX fuse and Resistor-Zener I.S barrier limiting current and voltage.
3- Relay used is ATEX certified.
4- Opto isolaters are ATEX certified.
5- The RS232 converter after opto isolator is I.S.
6- Relay on and data Exchange between my device and Correcter No1 and No2 is done separately. (Relay1=ON/Relay2=Off and vice versa)

Assume all devices used are ATEX certified, Now question is: does the way of connections used in this scheme match or break ATEX rules ?

Thanks & Best Regards.
 

I guess Galvanic Isolation is not mandatory in fact ... if data and power lines be I.S.
Forget my device,
Assume two EK220 are installed in Hazardous area, they both have RS485 port (common hardware pins with RS232). if we can connect to both of them through one common RS485 Line, then there shouldn't be need to isolate power line for them too (of course power coming from I.S power module) because RS485 pins in EK220 are already electrically connected to main board and there is no isolation there.
ِِDo you see any problem in this idea ?
 

No Comment ? :?:

I saw in IEC60079-11 Electrolyte and Tantalum capacitors can't be used in I.S circuit but i see tantalum caps both in Elster and Actaris Volume converters. why ?!
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top