CT controller was upgraded from MkV to MkVIe recently. The fuel gas meter is a Rosemount 3095MV multi-variable transmitter. Prior to the MkVIe upgrade I never had any comm issues using a Hart 275, 375 or 475 communicating with this device. While performing a calibration last week, the Hart signal kept dropping out using a 475. I checked connections and performed a Hart diagnosis (23.5 V), but had to continuously cycle power on the 475. Initially I assumed a faulty cable. Changing cables did not resolve the problem. I inserted a 250 Ohm resistor and it helped some, but comm still dropped out but would return much quicker. Has anyone seen this problem? I have never needed to add a resistor before. The Hart 475 in question does operate flawlessly otherwise.

 

Clearly, the difficulty you're seeing is associated with the upgrade. Something changed. Problem is, there's no response 3 days later because you've isolated most of the forum readership, who, like me, don't have a clue what is involved in a MkV to MkVIe upgrade.

I can see a peaker plant that probably uses gas turbine generators across a corn field on the north side of I-88 when I drive to the Rock Island, but that's as close as I've come to one of those beasts.

But I've done hundreds of HART hook-ups.

The purpose of the loop resistance is to generate a voltage signal strong enough for the HART modem/chip to read the signal successfully. So, if adding resistance helped, there's probably some issue with the HART FSK signal. That could be noise drowning the signal, insufficient signal, or the ill effects of multipoint grounding.

Insufficient signal can happen when a large capacitance eats up the signal.

- Was the cabling changed so the wiring run is now twice the distance?

- was the cable changed from one type to another?
HART foundation has a spec on cable capacitance (pf/ft) based on total cable length, but I don't have that available.

Was the DC power supply feeding this particular transmitter changed? Ripple from a power supply will interfere with HART. One time I ran into a HART problem that was caused by a breadboard power supply. When I substituted a little 2.5W commercial switcher, HART worked fine. I didn't have a scope to look at the power line, but there's a reason the HART Foundation has a noise spec for power supplies (maximum ripple 47 to 125 Hz is 0.2Vp-p; max noise 500 to 10KHz is 1.2mV rms)

- Was parallel conductor cable substituted for twisted pair (it's just low voltage, bell wire ought to work)

Noise: Was the wiring relaid so that the transmitter wiring now parallels a newly installed VFD or other electrical hash generator?

Was a loop powered device, like an indicator added? Sometimes the device adds noise to the loop in a frequency range that gets stripped off by the analog input, but which affects the HART signal.

 

David_2,

I think you'd make a great gas turbine field <b>engineer</b>, with heavy emphasis on the 'engineer' (as opposed to the field service people, technicians really, being tasked with trying to install, commission and troubleshoot GE-design heavy duty gas turbines these days). A field engineer is someone who can analyze problems and use a deep knowledge of how inputs and outputs work and how they should be wired and configured, and can "read between the lines" when necessary, and can use logical thought processes to troubleshoot and make processes work properly. It's also someone who doesn't think their entire job is to park their arse in front of an HMI monitor and doesn't need to have any knowledge of where field devices are, how they work, how they need to be wired, and how the control system is built to accommodate the inputs and outputs. If it's not on the HMI in ToolboxST or on a CIMPLICITY display, well, then it's not important. And, it's someone else's problem. You understand the entirety of what control system is, including inputs, outputs, and P&IDs (something OEM field service people aren't trained to read or use and quite frequency ignore at their own peril).

A Mark* Speedtronic turbine control is a version of Speedtronic turbine control systems, Mark Vs being produced from approximately 1991/1992 to until about 2001/2002. It was--and still is for those using them--a very capable turbine control system, but had some initial issues around weak printed circuit cards (which was eventually resolved). But, the thing that really caused people to dislike the Mark V (and to still dislike the Mark V) is the number of files which had to be modified or opened and read to understand the control system and how it was wired and configured. And, although the initial documentation was adequate it wasn't very well received and therefore ignored. (It was king of like an old DOS manual; it didn't tell you how to move a file from one drive/directory (before the MOVE command came along!); the reader had to know that the destination directory had to exist, and then use the COPY command to "move" the file to the destination directly, then use the DELete command to remove the file from the source directory. If you didn't have a clue about turbine control and ladder logic and how various types of I/O worked, well, you weren't going to get any of that from the Mark V Manuals.)

The Mark VIe is just the latest and greatest Speedtronic turbine control system. And, as such, there have been lots of changes and modifications to the type of hardware (oh, yeah, a LOT of people really disliked the form factor of the Mark V--myself included) used. I haven't actually seen all of the I/O terminal boards, but I've heard that some mA input circuits use 100 ohm precision resistors for dropping resistors and I would bet this is one of the problems. That can be remedied by putting additional resistance in the circuit, but that would also require changing the scaling of the input signal to match--not difficult to do.

I'm told that one of the reasons 100 ohm dropping resistors were used was so that some signals could be "shared" with other inputs/circuits, outside the Mark VIe (such as DCS panels and CEMS (Continuous Emissions Monitor Systems) panels, which quite frequency use either 250 ohm dropping resistors or 500 ohm dropping resistors. Although "hockey pucks" are inexpensive these days, most of the field service personnel don't know much about them, and the OEM doesn't usually provide them and buying components these days in the field is damned difficult (unless one gets a P-Card, and those aren't easy to get, which is what happens when people abuse purchasing- and expense account privileges--but I digress, again).

Anyway, while the Mark VIe is a functional equivalent of a Mark V, it does so with different hardware and configurations. One has to RTFM (read the fine manual) to find out about things like dropping resistor values, and many people just take these kinds of things for granted. Dropping resistor values aren't listed in ToolboxST, and since there generally aren't wiring drawings provided with many Mark VIe upgrades these days (and the ones that are seem to be devoid of this type of information) so one has to refer to the manual, GEH-6721, The Mark VIe System Guide, for such details. And, the authors didn't make it very easy to find things--though a quick perusal of the manual when there isn't a pressing need for solving an immediate problem will yield amazing results.

And, all the other suggestions you made were spot on, as well. Usually, the wiring isn't changed--but sometimes it is. And, lately, I've seen--and we've all read about--people using the "wrong" wire for purposes other than what it was intended, just because, well, wire is wire, right?

 

No changes other than the MkV - Mk VIe upgrade. I have never seen this problem prior to this occurrence. I did not have time to locate my old 375 to see if that needs a pull down resistor for comm to this transmitter. Due to limited generator downtime I don't get too much play time.

 

Thank you, CSA!

I will be sure to check GEH-6721. Due to a lack of manpower and a mountainous volume of never ending modifications to the plant where I work, training is often done in the form of 'troubleshoot and OTJ". The Mk VIe upgrade was required to accommodate the additional IO for the installation of a new liquid fuel recirc system. We were provided all the pertinent GE docs for the upgrades but time is not a luxury like it once was in this industry. Maybe a couple of forced outages will finally open the eyes of my facility managers and they will have no alternative other than to ship us out and have us properly trained.

Again, thank you!

 

To troubleshoot HART and other field communications refer to the
troubleshooting guide found here:
http://www.eddl.org/SiteCollectionDocuments/TechPapers/ag_Field Communication Troubleshooting.pdf