Post-tensioning tendons and condensation at inspection openings

[ajk1]

In a post-tensioned above-grade open-air unheated garage in southern Ontario, I see that condensation can sometimes form on the slab and beam soffits and beam sides. Some years ago inspection openings in the beam soffits were chipped at low points of the tendon drapes to allow periodic inspection of the tendons for appearance and tightness testing by crowbar prying. The 16" ± square openings are closed up by placing some batt insulation under the tendon and then a steel closure plate, as shown in the attached schematic sketch. The plate is sealed around its perimeter with caulking (sealant).

Question:

Can the batt insulation be relied on to significantly changes the dew point of the tendon so that condensation is significantly less likely to occur? My opinion is that the batt insulation will have very little if any effect in this regard, because the steel tendon is in continuous contact with the mass of surrounding temperature and will quickly reach the same temperature as the surrounding concrete. But I am interested in your opinion. (If the plate perfectly seals the opening, perhaps the issue of condensation is a little less significant (I don't know) but it is hard to ensure that a perfect seal is always achieved).

 

[Canuck65]

I would agree as follows, the batt insulation will not result in any significant difference in temperature between tendon and surrounding concrete. It may result in a very, very small lag to the change in temperature of the tendon relative to the concrete, but in my opinion, an insignificant lag. For the insulation to work, there needs to be a source of heat (for example, ground insulation you have the heat coming from below frost depth).

As for dew point, that would be dependent on humidity and temperature, but should not vary between the surrounding concrete and exposed tendon significantly, for a given humidity level. For such small openings, it might be possible to add a dessicant to the cavity to minimize humidity.

 

[EngineeringEric]

Why not spray foam if condensation at this opening is of concern?

I would see that the batt and steel combination would act as a slight thermal short and will also allow exterior (humid air) into the opening... the sealant isn't perfect and the batt-concrete connection is less than air tight. But again, we are talking single degrees of temperature. I would believe that the air gaps is what could lead to condensation...

If the air can condensate on the surface of the concrete, then it would condensate directly to your tendon. Or at least that how my mental image works...

EDIT: I realized after submitting... spray foam means you can only inspect it once, maybe that defeats the purpose but at least it can be removed then reinstalled.

 

[KootK]

I also agree. For me, it's primarily about Canuck's point regarding the lack of a heat source. In this sceanario, that's only the thermal mass off the concrete which, as mentioned, would just create thermal lag at best. I'd actually prefer not to have the bat insulation. I'd be afraid that it would just end up being a wet sponge a lot of the time.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[Ingenuity]

ajk1,

I have little to add directly to the good info you have obtained thus far with respect to condensation.

I will say that when we conduct exploratory openings for unbonded PT systems (that will be open for a length of time, usually more than than 24 hours), we use a 2" wide petrolatum tape placed over (or around, if available) the exposed strand segments. The tape is easily removed and reused, is an excellent corrosion protection layer, a water repellent, and is very economic. We usually use a plywood cover and sealant to close-out the opening whilst under evaluation.

If you, or your company, are conducting an evaluation on the current condition of the PT system, some additional things to consider:

1. Get this document: Link: "Guide for Evaluation & Repair of Unbonded Post-Tensioned Concrete Structures"

2. Double check the strand diameter in the field: 5/8" dia strand is not very common in building structures, and especially so in UNbonded systems. It may well be 5/8" dia, but worth the double check, especially if 'do something/do nothing' type calcs are being used to justify a course of action.

 

[ajk1]

To Canuck65, EngineeringEric and Kootk:
Thank you for your comments. It is nice to get comments that deal directly with the question asked and do not go off on a tangent as can sometimes happen on this forum. It is also nice to have comments that agree precisely with my thinking.

To Canuck65:
I too have been wondering about putting in a desiccant, but I don't know how that would work over the long term (30 years +).

To Ingenuity:
The 2" wide petroleum tape sounds very interesting especially if it has a good bond. To-date Densotape has been used but I am doubtful it has a good bond ... preliminary exploratory examination where we have removed the metal cover plates show that the grease from the tendon sheath (they were grease injected several years ago) is leaking out in a number of locations. Do you have a product name for the petroleum tape?

The strand s definitely 5/8" dia. We have the shop drawings and also we have measured it on site several years ago. In my area, 1/2" dia unbonded is generally used in slabs, 5/8 dia in beams.

 

[ajk1]

I googled petroleum tape and it came up "Densotape". Hmmm.

 

[Ingenuity]

Yes, Densotape or Densyl Tape by Denso.

But you have to apply it correctly. When we excavate for exploratory openings to PT we try and chip around and along the exposed strand length and perimeter. Installation of the tape is then spirally wrapped around the exposed strand using short lengths of tape. Where the strand could not be fully encapsulated around the strand, it is pushed and molded into the profile of the strand so that the exposed strand is fully covered from the exterior environment.

We have used it for temporary corrosion protection to exposed strands for periods of months without breakdown of the material or fall-out of the wrap from the overhead tendon group.

Denso also make a moldable mastic, but that is more suited to volume-filling of voids or intricate profiles.

 

[ajk1]

Ingenuity:

Thanks for your comment. Very interesting and appreciated.

1) Doesn't that allow the grease to gradually seep out of the sheath where the spiral wrapping starts and ends? Maybe you find that your "moulding" of the tape stops that from happening...but have you checked a number of them say a couple of years after installing, to see if there has been any seepage out of the grease?
2) In the structure I am looking at, they did not chip above the strand (bottom of beam) and they did not spirally wrap. They applied the Densotape over the strands and a few inches each side of the strands. Any comment?
3) If there are multiple (say 12) adjacent closely spaced strands in the beam, doesn't that make it difficult to chip above the strands (bottom of beam), and to spirally wrap the Densotape around all the strands?

 

[Ingenuity]

ajk1,

You are welcome.

1]: We usually place a big bunch on extra Densotape at both ends of the exposed strands where the sheath was cut. For our past projects we seldom have issues with 'leaking/seeping' grease - usually the opposite, where the grease has dried up to a paste consistency. However, with antiquated PT systems there was no specification for grease at the time, so manufacturers used 'what ever they had', so I am not too surprised if you have seeping grease in your tendons.

2] and 3]: I agree that chipping behind tendons can be difficult, especially in beams. We use small Hilti TE-6 chipping hammers - real slow and tedious between tendons - but if the exposed length of strand is 24" or so, and there are no stirrups in the way, there is some flex in the strand to wedge adjacent strands apart and remove some concrete. Usually if there is 10 or 15 tendons in a beam, they are grouped in a maximum of 5, with clear spaces between each grouping, but if the beam width is reduced I can understand if they get clumped together. We have sometimes drilled small diameter Kwik-Cons into the adjacent concrete and tie-wire supported Densotape to the strands. Just really trying to protect the exposed surfaces from the exterior environment, and it is only temporary, albeit easy to remove and check strand condition every so often.

A bit off topic: I will add that we often take grease samples from tendons and have them tested for water content (ASTM D6304 or spectrochemical analysis) and compare them to current PTI specifications for maximum water % (currently 1000 ppm). We have projects from as recent as 2008 come back with water contents of more than 5 times the current specification, due to water ingress through non-sealed live-ends.

High water content grease is usually easy to determine visually due to emulsification:

Good grease:

Bad grease:

 

[ajk1]

To Ingenuity:

Excellent photographs. I appreciate what you say about the older greases and overall protection. I have been periodically investigating problems with p.t. corrosion since about 1986. The problems I have seen have all been with the "push-thru" type sheath. I have not to-date seen problems with the extruded type sheath, which did not come into general use in my area (southern Ontario) until roughly about 1992. I believe that the extruded type sheath came into use in the U.S. quite a bit before it was adopted in Ontario. Even the extruded type sheath can admit water thru the ends or thru the inevitable cuts in the sheath that are made on site by dragging rebar over the sheaths, etc. I insist on no void system with watertight end encapsulations. When we tested a claimed water tight end encapsulation we found it was not quite watertight--the manufacturer then went thru several cycles of tweeking the encapsulation and re-testing for water tightness, and eventually got it right. "Trust but verify".

Unfortunately, the post-tensioning industry was not very open about the corrosion problem, in my opinion, which may account for the relatively late adoption of extruded sheath system in my area. However my current situation is directed to tendon sheaths that were grease injected a few years ago in an effort to increase the tendon durability (not my idea).

My opinion is that if you can "flex" a 5/8" dia tendon in a 16" long opening, to insert the Densotape, the the tendon must have 1 or more broken wires. I know that you say that your hole is 24" long, and perhaps you are referring to a 1/2" diameter strand, so perhaps that might have some flex...I don't know.

Follow-up questions:

- Are you concerned that every time that the Densotape is removed to inspect the strand, that fresh supply of humid air and oxygen is admitted, which may re-initiate or sustain the corrosion process?

- Assuming adequate concrete cover to the strand, the concrete provides resistance against the breakout of the strand from the beam, if a strand fails (all 7 wires fail usually after the 4th wire breaks, the 5th, 6th and 7th go at once). After that concrete is removed, how do you satisfy yourself that the strand won't break out of the beam soffit and cause damage or injury? Is the closure plate design with fastening to resist that?

- Would you happen to have a photo illustrating the tendons with Densotape etc. installed?

- Do you use batt insulation to provide fire protection where the concrete cover has been removed? Do regulatory authorities accept that? Is there UL rating for it?

Thanks again for your comments. They are right on and very useful.

 

[ajk1]

I see that no one so far has responded to my last series of questions. I can perhaps understand that because I suspect that some of these questions have unhappy answers (I hate dealing with corroded tendons. Never a really elegant answer to them, and the actual situation is impossible to reliably determine). But if there is anyone out there who has dealt with these questions and has come up with answers that would satisfy what a reasonably prudent and knowledgeable engineer would do, I would be grateful, even if your answers are hedged.