API 7: IF vs. REG vs. FH Threads 6

[cnuk]

Can someone enlighten me on why you would choose one of these rotary connections over the other? I have looked at API 7 and can see the differences are in the TPI, root radius, and thread height.

So, in practical terms, why would you choose one over the other?
1. TPI: it makes sense that the lower TPI are more robust, yet generally not as strong due to increased thread height.

2. Taper: doesn't more thread taper induce more hoop stress in the connection and make it weaker? What does more taper buy you other than ease of assembly?

3. Root Radius: obviously larger is better in terms of stress concentration.

Any insight and/or comments would be greatly appreciated.

Thank You

 

[DrillerNic]

Connection selection is part of drill string design. A good source of information on drill string design is the TH Hill DS-1 standard, or any number of courses are available.

Briefly, I think API NC connections are regarded as teh 'base case' for most drill strings. The API NC thread form (the V-038R thread) is superior to Reg, FH and IF in fatigue resistance. API NC replaced a lot of other API connections such as API IF (which doesn't officially exist any more). However out of habit, lots of people still talk about 4-1/2 IF and so on (note that the NC thread IS NOT the same as IF thread- if you want NC50 specify NC50 and not 4-1/2 IF; most people these days will give you NC50 if you ask for 4-1/2 IF, but some won't).

API Regular remains the thread of choice for a lot of BHA components, largely from habit I think, as API Reg is inferior to API NC. Bits often seem to come with API Reg, and I usually cross over to NC thread at the bit sub, if I can.

Most API FH connections were made obsolete by the API in 1968. I don't think I've ever used them. I think a couple of sizes are still popular due to gaps in the API NC range.

Then you get into the proprietary threads, some are modifications of API NC connections such as VAM EIS, Grant TF SST and Hi-Torque (I'm using VAM EIS on my wells at the moment), some are totally different such as Hydril, H90 etc. Due to their different IDs, ODs and strengths these connections can offer adantages to your Hydraulics and Torque & Drag if you're drilling long step out wells or have clearance probelms like TTRD wells or something.

 

[snubengr]

We still cut several more IF, Reg. and FH threads than we do NC in any given week, in a shop that primarily builds specialty wellhead equipment, wireline and downhole tools, crossovers and lately a fair number of PDC bit bodies (API 6A, 7, 16A and 16C licenses). I would say at least three or four to one. I would also say that if you are ordering your complete string from a tubular provider and you aks for 4 1/2 IF you may get NC50, but if you are ordering other equipment I think you are more likely to get 4 1/2 IF whether or not you order 4 1/2 IF or NC50. A number of the job shops/specialty shops here might have IF gauges, but few have NC gauges. I've only encountered one time when a consultant or engineer would not substitute a 4 1/2" IF for an NC50, especially after being told that he would have to supply gauges.

 

[cnuk]

Interesting. In API Spec 7, section 10.3 they list a table of equivalent connections. NC26= 2 3/8" IF, NC31= 2 7/8" IF, NC38= 3 1/2" IF, NC40=4 " FH, NC46=4" IF, NC50=4 1/2" IF. They don't seem to distinguish between the two after that, even listing NC###( ## IF) in the tables. IF are definitely rated higher than REG so that supports some of my thoughts on taper vs. strength. We mfg BHA components so that's why we see the API REG connections still coming through, as well as some IF...I assume now on the newer compoenents.

Thanks

 

[DrillerNic]

According to TH Hill, in 1968, the API took the most popular connections, gave all of them the better V-038R thread form which has smoother, more fatigue resistant root than the V-040 and V-050 seen in REG & FH or the V-065 seen in IF, XH and so on and called them numbered connections, NC38, NC40 & NC50 and so on, and made almost all the other connectionc obsolete.

So the old 4-1/2 IF is 'equivalent' to NC50 and will make up to it, but strictly speaking is worse as it has the sharper root form. However, people (especially drillers)are very traditional and continue to use the old names, so there's a fair amount of confusion, and from what Snubengr says, shops still cutting what is an obsolete thread, (according to my copy of DS-1).

But if I ask for NC50, and I get 4-1/2 IF, then that's not compliant! I don't work for a drilling contractor or a directional company so I'm not that involved in getting threads cut, but if I base a drillstring and BHA design on the specifications of NC50, and I get a fishing job or round trip due to fatigue failure in a drill collar 4-1/2 IF connection (let's face it, in the unlikely event of a collar failing, especially a large OD one, it will be by fatigue failure at the box connection), then I'm not going to be happy!!

However, as Snubengr says, if getting NC50 has a cost impact because I've got to supply gauges, then I'll probably accept 4-1/2 IF, but think twice about using that shop again in the future, as I'd expect somewhere with API 7 certificate to be able to cut an NC thread!

 

[cnuk]

Good information again DrillerNic. That makes more sense now why they call them "equivalent".

Thank You

 

[MrSticky]

NS-2 is available from Fearnley Procter Group in Houston and Aberdeen, 748600.

NS-2 covers more than DS-1 and is part of the Drill string and well equipment NS system NS-1, NS-2 and NS-14.

NS-1 covers the manufactuering of new equipment and is available on CD. (it was originally written by Shell in holland) but is now kept up to date by the above company.

Most UK oil companies require manufacturing to NS-1 - THis means you have to use API, as NS-1 is a top up sustem for API - it fills the hole in the API standards - at least some of them.

NS-1 is widely used around the world - lots of the smaller operating companies now use it too. It aligns with API in that IF connections are obsolete. So it is not just DS-1 that says this - it is API. I think even DS-1 new version references this manufactureing spec ( actually it references SQAIR - but NS-1 replaced SQAIR in 2002 so it effectively references NS-1)

Look in API RP 7G - you won't find any info for IF connections. So how are you going to calculate/look up MUT, yeild strength etc.,? Interesting huh?

The only place I'vee seen real IF connections was india - all other marked 4.5IF have in actual fact been NC-50 but refered to as IF as that was was was requested on the order. The machine shop manager said - yes we know they really mean NC-50 so that what we cut - we just put 4.5IF on the documentation so as not to confuse them. And that was for a Jar company!

 

[DrillerNic]

Cheers MrSticky

But it's interesting to get confirmaiton of what I've read about API numbered connections vs IF connections.....what happens to your BHA design when you get real 4-1/2 IF connections and not NC50 connections in your BHA?

 

[MrSticky]

Nothing happens to your BHA design, most likely as your BHA will not usually have NC-50 - they will be provided by Schlum or Breakerfuse and they in my experience use NC-50 or more usually XX Reg of some size.

Let's say you did get the IF profile in a BHA component, since BHA connection fatigue is the most frequent failure mechanism around the world you would get a fatigue failure in less rotating hrs than an NC would have (probably about 60% of the time at a guess and without doing FEA!)

If on DP - then the risk of fatigue is less as DP usually fails in the tube (also by fatigue). Connection fatigue is much less likely in a DP connection (perhaps becuase we use NC-50???!!!) - however, there would be a decrease in the connections resistance to fatigue if it was 4.5IF.

But more likely you will see 4.5"IF on a crossover or similar kit that has been lying around for a while. It should be picked up on inspection or by your QC chaps.

Crossovers are what I call "Cheap and Critical"!!

Same goes for these though - fatigue resistance is reduced. You risk getting more fatigue cracks in your equipment, higher chance of a fishing job. You can put a cost/risk against it. Typical (Average) twist off costs £? depends whaere you are in the world - but in India not so much, offshore deep water, £2000000.

Despite looking I have not seen a real 4.5"IF for a long time. Although, I have seen lots of NC-50's disguised as 4.5"IF, you can use die-stamps or protectors to do this.

Smiles... MR S

PS bet you don't know where the code 4.5IF comes from...

 

[DS1Guru]

Cnuk:

Generally you will not find anyone cutting actual IF connections any longer. The IF threadform was replaced by API in 1968 with the NC threadform. The real IF threadform has a a V065 0.015 inch thread radius and was very prone to fatigue failures. The NC connections have a V038R thread form and a 0.038 inch root radius, which is much more fatigue resistant. Out of the non-proprietary connections on the market today, the NC connections are considered superior to FH and REG. Both FH and REG have the V050 thread form (REG also has V040 on smaller sizes), which is a 0.025 inch thread root radius (worse for fatigue than NC). When you look specifically at 8 inch collars, your choice is typically NC56 vs 6 5/8 REG. Studies have shown the NC56 to be far superior in terms of fatigue performance compared with the 6 5/8 REG for two reasons. First, the thread root radius is larger which lowers the stress concentration and second, the 6 5/8 REG has a 2 inch per foot taper rather than a 3 inch per foot taper. This is critical because the vast majority of failures on larger collars occur in the box. The 2 inch taper on the 6 5/8 REG connection means much less cross-sectional area in the critical area at the back of the box (where most fatigue failures in boxes originate). When taken together, this can mean a 3 - 10 fold increase in fatigue life when choosing the NC56 over the 6 5/8 REG. You can read “Use NC56 Connections on 8? Drill Collars and Cut 1?or 3/4? Pin Stress Relief Grooves on Rotated BHA Connections NC38 and Larger”, IADC/SPE 87191, 2004 for more information. In terms of torsional capacity and tensile capacity, I think the different connection types you mention are comparable size for size. Down in the lower BHA where most connections are run in compression (may or may not be bockled depending on hole inclination), pin neck tensile capacity is generally not an issue. Torque is also the lowest in this region, so torsional capacity is generally not needed in excess of the API recommended MUT (60% of the torstional strength of the connection) unless you are underreaming with a bullnose and rotating at a high RPM. On the whole (not counting any proprietary connections like XT, HT, DSTJ, WT, TM2, etc.) NC connections will have superior fatigue performance and relatively equivalent tension and torsional performance. As far as IF goes, a lot of orders are still placed for "IF" and "IF" is still stamped on connections, but 99.9% of the time, I would bet the connection is actually the NC equivalent. One other point of fact: you can actually screw a true IF into a NC of the same size. Example: NC38 will screw together with 3 1/2 IF.

The DS1(TM) Standard (sponsored by more than 40 companies around the world - operators, manufacturers, drilling contractors, and inspection companies) does not reference any of the NS manuals in any way. The Third Edition of DS-1(TM) is three separate volumes. Volume 1 covers manufacturing and inspection of new equipment (Drilling Tubular Product Specification), Volume 2 covers Drill Stem Design and Operation, and Volume 3 covers Drill Stem Inspection (used equipment). If you want a true comparison of DS-1(TM), NS-X, and API / ISO, decide for yourself by trying to read and use each of them. Which is easiest to read, understand, implement, and find the information you are looking for? Which has the most technological advances consistently and is the most up to date?

Mr. Sticky:

The connection name 4 1/2 IF derived from being welded to 4 1/2 inch externally upset (EU) drill pipe tubes that were internally flush (IF).

P.S. What actually weighs 19.50 pounds per foot on 5 inch 19.5 ppf drill pipe?

 

[rbgos]

Just come across this thread - very useful, thanks!

With all this talk of IF threads having worse fatigue resistance than NC, does anyone have the actual data in terms of SCF and what fatigue curve to use? I'm trying to do a fatigue analysis on some drill-pipe which has either 4.5"IF or NC50 - probably NC50 from what most people have stated above.

Thanks a lot, Richard