Hot Insulation for Instrument Tubing 2

[CGonsalves]

All,

We procured some bare, 1/2" SS intstrument tubing for hookup to static and differential pressure transmitters (Steam/water) with the assumption that freeze protection would be handled down the road by our Project Engineers. After looking at the final schematics, this tubing turned out to be ~6" OD (!!), so I'm looking for other options for insulating this tubing. A contract spec refers to using a fiberglass tape as the insulation. This would be our easiest route, since it would handle unions and elbows without much hindrance. However, we haven't handled a situation like this before, so I wanted to see how others have approached a situation similar to this.

Any thoughts?

 

[danw2]

Tubing size is almost irrelevant with regard to the medium freezing. Impulse tubing to pressure transmitters is dead headed, so there's no flow through the tubing, so there's no continuous heat transfer from the process medium. So the medium in the tubing (and instrument manifold/valving and the instrument body) cools to ambient conditions because conductive heat transfer through SS impulse tubing/piping is very low.

Freeze protection involves BOTH insulation and heat trace (of all the parts - tubing, sensor body, valving) to maintain a medium temperature above its freeze point. Insulation-only of cooled-to-ambient medium still allows a temperature drop below freezing.

For electric heat trace, self regulating heat trace is the flexible plastic type sold in northern climates for residential water pipe and gutter/roof heating; Mineral insulated (MI) is a higher capacity/wattage heater element in a copper or stainless sheath.

There's steam heat trace that I've heard is more maintenance than electric.

Google will find you industrial process heat trace vendors.

 

[mk3223]

Agreed, the steam tracing system can be costly, if the steam is not ready available. The electrical tracing system can be controlled and monitored for the desired temperature setting.

 

[KevinNZ]

Are sure it is 6" NPS piping and not DN6 mm tubing?

6" piping would required tubing at the end to get connect to P and DP transmitters.

 

[CGonsalves]

Dan, thanks for the tip.

Kevin, I guess I wasn't exactly clear. It is .5" tubing, but the insulation/tracing came out to about 6" diameter, which is unnecessarily large.

Obviously, tubing bundles would be ideal, but since we have already purchased our tubing, I think we plan to just wrap the tubing and heat trace with fiberglass insulation.

Side note, and I'm not sure whether anyone is still paying attention to this, but I was wondering if there's a desired slope for the tubing? Horizontal runs cannot be completely horizontal, and our past projects have used a slope detail of dropping 1 inch for every foot of tubing. Just wondering if the only real desired slope is simply making sure that the tubing is always sloping down.

 

[danw2]

The definitive document on impuluse tubing installation is TUV NEL's Best Practice Guide Impulse Lines for Differential-Pressure Flowmeters
Link

The reason for sloping impulse lines is to prevent trapping vapor/gas in liquid impulse lines or trapping liquid in vapor/gas impulse lines. Slope provides a drain or vent path.

Being compelled to publish in SI measurement units, they state the politically correct appropriate slope as 8%, which is what 1 inch in 12 inches happens to be. See page 7.

 

[CGonsalves]

Hi Dan,

Thanks for your help and for the reference on best practice. I do have one more question about the tubing. On the same project, we noticed that some of our pressure transmitters were directed to be installed above their tap points. While I think we would be able to move them to lower platforms, I was wondering if the field installers could bend the tubing up to a certain point to allow for the change in height.

For example, one certain transmitter is 39 mm above the tapping point. Could I bend the tubing upwards to account for the change in height (and 8% slope of the line) as long as I'm making sure there's no local low point in that tubing line? Will there be issues with accuracy that result because of this?

 

[danw2]

Pressure transmitters for gas service are mounted above the pipe taps so liquid droplets drain back into the pipe by gravity. Liquid filling a gas service impulse line would eventually contribute an offset.

Pressure transmitters for liquid service are mounted below the pipe taps so gas bubbles rise upwards back into the pipe. Gas pockets in liquid service create anomaly pressure readings.

Pressure transmitters for steam are mounted so that condensation in the impulse tubing isolates the pressure transmitter from direct steam impingement to prevent killing/overheating the transmitter. The condensate is in vertical impulse tubing, but the top of the run, which can be condensate pot or a filling Tee, should slope back down towards the pipe tap, so that the vertical line remains full, but roughly horizontal feeder line lets condensate drain back into the pipe:

You can see each approach in the graphic below:

To your question - tubing is routinely 'bent'. The trick is to bend it without kinking it, where a kink could close off the passageway.

Bending the tubing will not affect the accuracy of the pressure transmission through the line, but accuracy of DP flow does depend on the equal head of condensate on both legs, or zeroing out a fixed difference in elevation.

Zeroing the transmitter by closing off the impluse lines (and opening the equalizing valve) at the 3 or 5 valve manifold does not correct for head offset in the filled impulse lines.

 

[CGonsalves]

Dan,

Thanks for your input. My department has been trying pretty hard lately to get our issues straightened out. There were three projects that had utilized incorrect information. Frankly, our domestic projects don't require us to draw up impulse lines, nor do we quantify any lengths, so these past three projects (which BTW are overseas) have been a pretty difficult situation now that we have to go back to our drawings from two years ago. We had no idea what rules needed to be followed nor how important this was.

Are the elevation requirements any different for static pressure transmitters? I noticed the 'Best Practice' document only addressed flow meters and differential pressure transmitters, but I'm wondering about PTs that measure the gauge pressure of the process fluid.

There are actually quite a few instances where placing a transmitter below the tap is just not possible. For instance, a water-service process tap is 0.5 m from Grade, and the customer requires that transmitters be at least 0.6 m above their respective platforms. To further add on, there are spacing restrictions (valve handwheel, drain piping) that don't allow for us to place the transmitter right next to the tap.
Is there an alternative that would allow me to place the transmitter above the tapping point?

EDIT: Dan, I'm re-reading your response again. Is it acceptable for steam-service PT's to be above the tap location, as long as it still maintains self-draining? I guess my confusion all along was why the transmitters needed to be located where they should be.

 

[danw2]

Your observation mirrors mine - the national and corporate standards/best practices focus on differential pressure where the DP range can be very low, like 3-4 in w.c. on an averaging pitot tube, so installation errors or air/gas pockets that affect the reading even slightly can create a large error. Yes, the pressure is supposed to be the same everywhere in closed vessel, but experience shows that liquid build-up in gas impulse lines and air pockets in liquid impulse lines create accuracy issues at low pressures. The transmitter elevation/location is spec'd to avoid those issues.

For gauge pressure measurements, typically at higher pressures, those relatively tiny DP errors diminish to the level of where the error is small enough to be in 'noise' range of the accuracy spec.

My approach for gauge pressure is that if the pressure is not a low pressure that would use a pressure gauge with a bellows or a diaphragm, (0-60 inches) then mounting a transmitter is similar to mounting a pressure gauge, with an eye to not letting radiant heat kill the transmitter electronics (gauges have no electronics to worry about) and enough impulse tubing length to drop conducted heat from the process.

My observation is that the commonly accepted "temperature drops 100 Deg F for each foot of deadheaded impulse tubing" is a myth, it's way too conservative. Temperature drop through dead headed SS tubing is more than that; stainless steel heat conduction is very poor. I've touched my hand to a 1/2" SS dead headed impulse tube coming out of a 1,950 Deg F furnace at 8 inches from the furnace wall and I did not burn my hand (I spray painted the tube black first and checked it with IR).

There are two styles of gauge pressure transmitters - in-line and single head. In-line has a process stem with the process connection vertical out the bottom of the stem. Single head looks like a DP but only one process head that has the process port connection; the other process head is closed/blind (atmospheric reference only).

The single process head style can have a vent valve on the backside. If someone opens that backside vent valve in a conventional wet leg, steam service impulse tube installation, then only ambient temperature condensate comes out.

With a transmitter mounted above the steam line with no wet leg, opening the vent valve allows steam flow at steam temperature. That's bad for transmitter and is likely to cook it. (not good for people, either)

I've seen a number of in-line gauge pressure transmitters mounted just like a steam-service pressure gauge, with a pigtail syphon pre-filled with water (condensate keeps the pigtail filled filled over time).

Frequently, the transmitter gauge is piped off to side slightly to avoid close proximity to pipe, but with a pigtail syphon.

I'd avoid gauge pressure transmitters for steam service that have an integral vent valve.