Cabling Measurement questions

[Bareld Bruining]

Hello,

I am Bareld Bruining from STC Brielle in Holland, the certified FF training institute. I got a few questions regarding the cabling tests which are prescibed in the E&D guideline:
1.Why is the capacitance check on the wiring done? And what are the numbers telling you? I asume that according to the cable length you can calculate what the capacitance should be in your particular case. So the number don't tell you if your cabling is correct, capacitance wise?
2. measure resistance:
According to the guidelines you should disconnect the cabling at the host end and have it loose at the fieldside. Conductor to conductor should be bigger than 50kOhms. Obviously the resistance is an open loop if you disconnect the leads at both ends. Can anybody tell me what we can discover with the resistance measurent. How can you see when something is wrong?
3. If you use some kind of connectorboxes, should you measure resistance including the boxes or measuring al cables (trunk and spurs) seperately.
4. If you use the FBT-5 with the FBT-3 and check if the cabling is oke, this works for the LAS value, voltage level and noise. But if I am measuring with only 1 terminator on the segment I will read out a number of about 690 mV. According to the numbers in the FBT-5 manual you think that you have more terminators than 2? Does anybody withness the same thing and does anybody know how I should look at this?

[Hamad_1974]

Dear Bareld Bruining,

My name is Hamad Balhareth and I work for Saudi Aramco. We have just commissioned a new plant that uses this great technology, Foundation Fieldbus.

The tests you are talking about are indeed important when it comes to validating your installations and are integral part of your precommissioning phase. You do pose important questions that we came across during the precommissioning phase of our project. I will give it my best shot to answer your questions below as follows:

1- The capacitance check: The purpose of this check is to make sure that you do have two and only two terminators in a segment and also to check the quality of the ground signal. If you accidently installed three teminators, the capacitance value would increase to more than the recommended value, 1Micro-Farad. I have attached the installation guidelines from the FF website for your reference. I think it will answer most of your questions. Back to the capacitance issue, we did not measure the capacitance in our project for one simple reason and that is we could check the number of terminators using our good eyes. Normally, there would be one in the field with the mega-block and another one embedded in the power supply.

2- For the resistance issue, I think you misinterpreted the guidelines as they clearly spell out that you only disconnect the terminals at the host system coming in from the field. This is to make sure you don't have damaged section of the cable. Obviously, if you disconnect at both ends, then you're wasting your time. Again, this is spelled out in the guidelines.

3- Again you need to really know the intent of the test. The intent is to check that the cable is ok. Now, i suppose you can check every spur if you want, but we did not do that because the spur is normally a small section of the cable and not really that long. You normally would need to check the TRUNK because it is the longest cable of a segment and the possibility of damaging that cable while routing is more likely to occur.

4-For this question, I would suggest that you only stick with FBT-3 as it is more important. if you make sure that 1) and 2) above are complied with, then you're ok by default. The FF guidelines stipulate that a signal of more than 150mV to about 800mV is practically ok. Ideally, and theoretically you should see 1 Vp-p if you have two terminators (20mA * 50 Ohm). If you install three, then the signal is dampened to much lower value since your resistance is now smaller (20mA * 33.3 Ohm). If you install only 1, then your signal is now doubled that of two terminators (20mA * 100 Ohm).

I hope this helps and if you have more questions or need more clarifications, i would be more than happy to answer them or at least share with you my findings.

Regards,

Hamad Balhareth
Instrumentation Engineer
Saudi Aramco
Saudi Arabia

[jberge]

You check the capacitance because a too high wire capacitance will result in signal attenuation, possibly to the point where communicaiton fails.

You check the resistance to ensure that there are no shorts or leaks. If the cable is opened it would theoretically be infininate resistance, but there could be water filled housing or junction boxes, poorly inslulation, and other problems that result in less than infinate resistance.

[Mike ONeill]

All these measurements are made to establish whether or not there is cable damage, which can be broken cores or more likely, core to shield or core to core faults, which may themselves be intermittent and high resistance rather than dead shorts - ah, the wonderful life of the commissioning engineer!
Terminators are a real pain since most installation contractors simply throw them in everywhere and leave it to commissioning people to check them out. There are field device couplers with automatic segment termination which eliminates this problem.

[Hamad_1974]

Dear Mr. Mike,

We know of the field device couplers you're talking about, however, at the time when we installed those old terminators, there was no such thing as automatic segment termination. I guess technology is truly faster for anyone to keep up with.

Regards,

Hamad Balhareth
Saudi Aramco

[Bareld Bruining]

Dear Mike and Hamad,

Thanks for your reply. I know the function of the cabling measurements. However I feel that the numbers are a bit strange. especially the lead to lead of bigger than 50Ohm seems a bit strange to me. If I measure this I get an open loop every time. So why the 50Ohm? All other measurments are talking of an open loop?

[Hamad_1974]

Dear Bareld Bruining,

I absolutely do not know why 50K OHm, but i want to ask you a question. When you measure the resistance, how do you really perform the test? Can you give me a detailed explanation? I assume you disconnect the trunk at the FF power supply side and you measure the resistance directly. Do you disconnect the End Of LIne terminator? I was discussing this issue a couple of days ago, and it came to my attention that if you do not disconnect the EOL terminator, then you will notice that the resistance keeps climbing and you really can not get a steady answer. BUT back to your question, why 50K Ohm? I think this is a good question that I really do not know.

Can any of the FF gurus out there explain why?

Thank you for bringing up such a good question.

Regards,

Hamad Balhareth

[tibor]

Hamad,

I agree to your replies excluding one. I prefer oscilloscope measurement instead of FBT-3 because the shapes, levels and values can be seen not just digital values therefore I use that during all the projects... Any comment?

Tibor FARKAS
P+F agent
Hungary

[rezabejd]

Gents,

Has anyone ever "caught" anything by doing all these tests, that wouldn't have been revealed upon connection to the host?

I'd like to hear someone say, "yeah buddy are we ever glad we did all those tests". Maybe if you were concerned about a marginally competent and / or unscrupulous installer leaving you with a cobbled-up mess?

My instinct / impulse would be to let the host find problems and do the tests on an exception basis. Is this a recipe for disaster? Am I being naive?

I do agree with "at least" a spot check of termination "quality" (e.g., tightness) as this did cause us some nuisance issues.

Regards & HNY
John Rezabek
BP

[tibor]

rezabejd,

1. Yes, there are above some measurements what I do'nt do usually.
2. Good idea to let the host evaluate the signal shapes based on a knowledge base but it is not enough to make evaluation only at the host end. I met several segments where there were "looks like good" shapes at the host side but at the field devices (in the field) were terrible, therefore the device could not be on the live list continuousely. Now there are available FF devices on the market which has diagnostics regarding to quality of communication packages, similar to in the host. We have to measure in the field, this is my quiet strong opinion.
3. Based on my experiences there are need a little bit more activity than to check the termination quality. If you do not any oscilloscope measurement at the host AND in the field you can not know whether your segment is running just a little bit over the threshold of stabiltity or running strongly stable. The port and device statistics are not enough to get reliable information.

Regards,

Tibor FARKAS
D.E.A.K. Process Control Ltd.
P+F; Metso agent
Hungary
Tel.: +36 25 507 815
Mobile: +36 30 389 87 44
tfarkas@deak.hu
www.deak.hu (http://www.deak.hu)

[Bareld Bruining]

Dear Hamad,

I have witnessed the same as you did. The resistance is climbing if you have the trunk disconnected on the PS side and the EOL terminator still in. By the way this means that you measure with one termintor because you disconnected the terminator at the power conditioner.
I think the climbing of the resistance has to do with the fact that the terminator is not really an resistance (only), but there is a capacitor in it too. This will result in the climbing resistance measurement. This would mean that we have to do the measurements without the terminators?
Still the 50 Ohm question stays. Does anybody has any comments on this?


Kind regards,

Bareld Bruining

[tibor]

Dear Bareld,

Yes, the resistance is climbing up when you measure the loop resistance of cable including serial connected 100 Ohm resistor and also serial connected 1micro F capacitance. As usually all the capacitance generates huge resistance (more MOhm) when you measure it by DC method, it is normal. This is the reason why we talk about impedance at a frequency instead of resistance, in this case at 31.25kHz (now it is AC point of view). So, the result should be a huge resistance in this case.

The reason why usually it is less than some MOhm that usually there are 8-12 parallel connected slaves on the segment and those parallel connected internal resistance can be measured already. Approximate calculation: I=12V/50Ohm --> I=240mA

If you need any further feel free to contact me.

Many rergards,

Tibor FARKAS

[Hamad_1974]

Dear Bareld and Tibor,

Here are few comments regarding what you guys have written:

1- Yes Tibor The scope meter is better than just an FBT-3 if you have the luxury of time. Normally, on large base projects you really do not have enough time to experiment. So, we really want the shortest time to tell whether a certain segment has a problem or not. But, definitely I think the scope is much better.

2- Yes Bareld, if you disconnect the terminals at the FF power supply you are now only using one terminator, but you will notice the resistance climbing. However, if you disconnect the EOL terminator which is the one in the field, now you do not have any terminators and you could easily determine the cable resistance. If you notice in the FF guidelines, they only ask you to remove the terminals at the FF power supply and then measure the resistance. Maybe we need to discuss why.

3- I see you guys mention 50 Ohm and NOT 50K Ohm. The FF rules call for 50K Ohm and not 50 Ohm.

4- Tibor, I do not know where you got that formula from I = 12V/50 Ohm? and why 12 Volts? Could please explain?

5- I still do not know why the 50K Ohm as it has been previously addressed by Bareld. I am guessing that it maybe the value at which the FF signal would be marginally degraded because of cabling problems. For example, degraded cable quality, loose terminals, etc. I guess someone did this calculation and came up with the 50K Ohm. Any comments?

Thank you

Hamad Balhareth

[tibor]

Dear Hamad, Bareld,

The 12V/50Ohm is an approximate calculation as I wrote above. If you measure the at host side disconnected segment cable with one terminator (100Ohm serial connected 1 microF) by DC method I've no other idea how can you measure less than some MOhm just if there are connected parallel the internal resistance of field devices.

I could wrote e.g. 10.6V (FISCO)/50Ohm also which means 212mA. However the 200-230mA is a realistic current consumption of 8-12 slaves. That is all I've tried to explain. Of course their resistance strongly depends on are there any short circuit protection applied or simply junction boxes are installed in the field.

The Guideline tells ">50kOhm (increasing)" and the increasing tendency showes the both leads and resistor (100Ohm) with capacitance (1micro) are OK because the capacitance able to charge, so there are no short circuit neither lead breakage (Note1, page 83 of Engineering Guidelines).

The lead-lead capacitance can show how many terminators are on the segment (Note2, page 83 of Engineering Guidelines).

And the asymmetric capacitances are described in Note3, page 83 of Engineering Guidelines.

--------------------------------------

Regards,
Tibor FARKAS