Double block and bleed - how did this become standard? 3

LittleInch · Jun 14, 2018

A bit of along post, but hope you read through.

I'm in the process of critically looking at the isolation requirements for a Crude Oil pipeline system running in a class 600 and class 900 system. The pipework is relatively open and uncongested and fairly simple layout.

I remember when I first started in the pipeline industry that single valves ( either ball valves or gate valves) and possibly a blind flange, was all that you had between you and the pressurised fluid ( in my case refined products).

Nowadays in various project and Company isolation philosophies / standards you find requirements for DBB, generally two valves in series (not double seated valves), for all isolations where you have a potential for operating /working against pressurized fluids and where the pressure class where this starts can be as low as class 300 and definitely class 600.

My understanding is that this requirement crept onshore from offshore facilities, especially after a number of fires and explosions which affected those facilities in the past. However I would welcome any history lesson on how and where this practice became the "norm".

I can't find anything in particular in ASME 31.3, 31.4 or PCC-1, but I'm open to pointers on this subject ( or searching for a different word than "isolation") in any other published code or standard, not Company standards. Before you ask, the project I'm working on doesn't have any particular company or legislative requirements covering this issue (not in Europe or US)so it's up to us to create one and get it reviewed and approved. I have seen and am working through the rather good guide the UK HSE produce called HSG 253 safe isolation of plant and equipment.

What I would like to do / thinking about is to generally remove the second valve requirement and replace it with blank flange with a tapped bleed valve like the one shown below for isolations which end in a blank flange. This complies with Cat I positive isolation examples in HSG253 if I add a pressure monitor. The question for isolation where you remove the blank flange and or use a single valve is that it requires a risk assessment to show it is acceptable. However it is not clear what you need to consider to get to that point.

There are some good pointers in HSG 253 on what constitutes "short term" isolation - basically max one shift, monitor, don't leave unattended etc where you can then look at single valve isolations for doing short term work, but need to assess the risk.

So when coming to open the isolation, the bleed valve does exactly the same job as the bleed in a DBB system - i.e. check for leakage passed the primary valve. I've never been very clear on what happens in a DBB system if you do find leakage? Do you just drain away the hopefully small leakage, leave it open and then work against the second valve? Or what?

The risk here might increase a little, but for the fluid I have (not gaseous) I don't think this is sufficient to have to put a whole bunch of extra class 900 valves all over the place.

Most of the valves we're looking at would probably be operated only a few times so should be virtually new when exposed to the operation.

Any questions, comments, pointers, experience, things you've done recently gratefully accepted.

If I'm pushing the boundaries too far then please let me know your thoughts also - I recognise that times move on and what was acceptable 25 years ago may not be now.

Thanks

LI

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

Latexman · Jun 14, 2018

Did you read David Simpson's FAQ on DBB? It's not a history lesson, but it is another perspective.

Good luck,
Latexman

To a ChE, the glass is always full - 1/2 air and 1/2 water.

XL83NL · Jun 14, 2018

Cant be of any help to your question, but those bleed valves from your uploaded picture look rather interesting (i.e. short-build). What make/brand are they?

zdas04 · Jun 14, 2018

The FAQ that Latexman mentioned is faq378-1612. I also wrote a paper for engineering.com titled New Processes are Needlessly Reducing the Recovery from Onshore Gas Fields that covers some of the history (I wrote it from a gas perspective, but the nonsense imposed on us is not quite as onerous as what is being shoveled onto oil operations).

Funny thing about DBB--two floating ball valves with a bleed between them is always acceptable for DBB. A floating ball valve requires differential pressure to activate the seal. This means that the farside valve (relative to the work area) might be very well activated, but still leak. The nearside valve should have zero dP since the space between the valves is vented. In an oil operation this situation nearly always results in oil leaking into the isolated section around the floating ball if there is any leakage in the farside valve. This is "safe", but a single trunnion ball valve (with both seats activated by heavy springs) almost never puts oil in the isolated section but is "unsafe". This nonsense comes from a tendency to have a policy instead of a decision.

[bold]David Simpson, PE[/bold]
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist

LittleInch · Jun 14, 2018

XL83NL,

They are actually fro Swagelok, but others make them also. I've always really liked them but for some reason you don't see very many.

https://www.swagelok.com/en/product/Valves/Bleed-and-Purge#

zdas04,

I recall reading that article when it came our originally and re-reading it makes sense. I think you can drop the density bit from risk (density) as, for me risk is simply likelihood times consequence.

What I was interested in was whether you had any success in reversing the creep of plant / offshore / high risk attitudes back to something more fit for purpose?

Or your view over whether DBB is required for many pipeline applications / locations.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

zdas04 · Jun 14, 2018

My two articles (the other one requires an SPE membership to see) have been accessed/downloaded about 25k times. Presentations I've done based on them have been well attended/well received, but that is a couple of hundred more people. Upwards of 500 people have been in my classes and gotten the whole spiel. In other words, only a tiny fraction of the impacted community have ever seen this discussion. If there is any movement in a positive direction it is too small to detect. A couple of engineers have contacted me with success stories involving moving a few topics from "policy" to "engineering judgement", but they are few and far between.

DBB is never needed (based on regulatory/code compliance) for "normal operations", and it is only one of the available options for some classes of "maintenance activity". Some companies have policies to pre-position DBB valve sets for future undisclosed maintenance activities, but they can sometimes be forced to see reason by the excessive cost of these complex valve sets.

[bold]David Simpson, PE[/bold]
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist

davefitz · Jun 14, 2018

Valves leak and also their stems can sieze up- yet the chemical plant or power plant still needs to isolate and work on equipment without killing the maintenance workers. Some engineers only view the plant layout and specs fromt the viewpoint of a designer trying to simplify a design and reduce costs, but there is a real need to look at it from the perspective of the maintenance worker and the plant manager who wishes to minimize the number of fatalities he needs to apologize for.

"...when logic, and proportion, have fallen, sloppy dead..." Grace Slick

zdas04 · Jun 14, 2018

davefitz,
I hope nothing in what I've said implies that I'm OK with killing maintenance workers, I was one for too long to have that opinion. I've been a facilities engineer for 38 years, done too many projects to count, operated several gas and water gathering systems for many years and never gotten an operator or maintenance worker injured let alone killed. My point in everything that I write is in line with Mike Rowe's "Put Safety Third" campaign. In other words the first line of defense against injuries is an individual worker committed to his or her own safety. The second line of defense is a sensible design. Corporate safety initiatives are a distant third.

Competent people with proper training are the single best method to improve safety statistics. If a competent worker finds that a valve won't move or that a required isolation valve is leaking through, then that competent worker will reassess the conditions and implement a plan "B" that keeps him out of the infirmary. Passing responsibility for individual safety to company safety initiatives is the single best way to create complacency that is the root cause of more injuries than any other possible cause.

For example, in my last job as an employee I managed a fleet of wellhead compressors of various sizes. To optimize this fleet under changing conditions, we made over 350 compressor moves in 7 years. No injuries, no spills, no safety or environmental incidents, not even a first aid event. I did not buy into the safety culture and required that the contractor provide competent workers and consistent leadership. Average cost of a compressor move was $6k. My replacement bought into the Supply Chain and PSM concepts and used the low bidder for each of the 3 compressor moves she did in the next 10 years (average cost $100k/move) and untrained, complacent workers had 5 OSHA reportable injuries on those compressor moves. The difference? I treated the workers as competent, valued individuals and required them to think. Her process treated them as interchangeable cogs that were under the protection of safety policies and substituted safety forms for actual safety.

I didn't mean for this to become this long, but you pressed one of my (many) hot buttons.

[bold]David Simpson, PE[/bold]
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist

georgeverghese · Jun 14, 2018

Probably cannot add much value to this discussion, but @LI doesnt say if this DBB is meant for positive isolation also, meant for maintenance activities requiring personnel access. If it is, then a spec blind and a blanked vent should be interposed in the DBB. Presumably not, since this is a crude oil pipeline system? Where the fluid is toxic or has some HSE concerns, the LP vent should be routed to a closed remote vent system.
In the few years I've led Hazops, I've had to cut short plant operator discussions on many occasions on this same topic( many of these seasoned sunburnt operations folks had more than 10years field operations experience) that threatened to derail the progress of the Hazop. So I wouldnt be suprised to see this thread being active for a long time.

KevinNZ · Jun 14, 2018

I often look at DBB valves on each side a control valves and then have guess which of the 5 (7 counting the bleeds) valves in a row will be first to need isolation be fix it. Only the valve in the middle actually has DBB.

KernOily · Jun 15, 2018

Great question. As far as I know, DB&B is not a 'standard' per se in the way ASME or API defines standards, but DB&B has become a 'best practice'. The operators I work with have defined their DB&B requirements based on engineered risk analyses. The DB&B method varies by the process: sometimes two valves with a bleeder, sometimes one valve with a bleed and a spec/slip blind; sometimes an air gap/drop-out spool. Valves used for isolation are always mechanical positive seal, e.g. gate, and not the type of valve that requires an upstream pressure to effect the seal, as David said.

Whichever double isolation method is selected, it needs to be fully vetted via formal risk assessment using the established industry risk-evaluation methods. If there is an incident resulting from the use of any proposed double isolation method, with subsequent legal action, the pipeline operator (in this particular case, but it could be the facility or asset owner) will need to be able to show that his double isolation method meets the standard of 'what another operator of equivalent experience and competency would do when exercising a reasonable degree of care and custody'. That means: documented risk analysis, and, recall that a tapped blind will need a B31 area replacement calc to determine reinforcement needed, if any, at the nipple/bleed valve connection to satisfy the applicable Code requirement. One can't just cut a hole in a BF, fillet-weld on a TOExPOE nipple, and screw on a small ball valve. Not saying you would, but I see it all the time. A lot of times a weldolet groove-welded to the BF will satisfy the area replacement requirement but one needs to run the Code calcs to justify and then file them away where they can be found in the future. Then there is the issue of SOPs for use of the DB&B, training to the SOP, and documentation of the training.

The combination of a full engineered risk assessment, Code calcs to back up the connection design, rigorous published SOP for use of the DB&B, and documented training of proposed personnel would put this design and your operation in the top tier. Good luck and let us know what you finally decide. Pete

LittleInch · Jun 16, 2018

Thanks for responses.

Dave F

I think there are some interesting issue sin your response.

First whilst I would admit that DBB has become the current practice, it's not easy to define "best" practice. The petrochem industry tends to go in a bit of a herd like manner on some things and I think this is one, hence my post. "Deined ...risk based analyses". Hmmm. Sounds a lot like people with no responsibility for cost and efficiency sit around taking worst possible cases and creating "simple" rules for design engineers to follow. Where are the studies showing that DBB actually reduces risk?

My whole point is whether double isolation in fact decreases risk in line with the additional cost. Nothing is risk free when it comes to pressurized hydrocarbons, but there is an accepted principle that there has to be a practical limit and consideration of cost when it comes down to very small reductions in risk.

One aspect no one has responded to so far is what difference any of this makes if the upstream / pressure side valve actually has a leak? Do you then with a DBB continue using the second valve? If not (I assume not) then why not just have some sort of monitoring port / bleed / valve to check good isolation on one valve? If it's ok then fine, drain off / vent off and break containment. Anything taking more than one shift or with highly flammable material then blind off. If there's a leak in the U/S valve then fix the valve. To use the bleed line as some sort of vent or drain in a continuous manner doesn't sound like a good plan to me. If it is the plan then why can't I use the body vent / bleed in a double sealing valve? I've seen this not permitted in lots of cases.

The point also is how to classify the fluid. Most hydrocarbons just get lumped together as "hazardous" regardless of whether it's 2 phase, gas or in my case fairly sticky single phase liquid. I would argue there is a risk difference here which the "simple" rules invented to cover all cases with one "rule".

There is big money involved here and that's the main issue - gradually smaller decreases in risk, if in fact they can be calculated, cost more and more for the same reduction. At what point is it impractical??

I don't know what I'm going to do yet, but it's been interesting airing this here and other responses / contributions welcome.

[Edit] I missed out that for a centre tapped say 1" screwed connection this is permissible in B 16.5 for all sizes above about 2" without any calculation or re-inforcement. Clearly you can't hang a huge amount off it and I've seen some pretty dodgy things in my time on the end of BF's, but screwed tapped connections for sure are permitted, especially if it's with the small bleed valve like the photo I posted above or even a screwed 1" valve. So yes you can just drill a hole in a BF if you follow the rules in B 16.5 and not do anything more. There are a number of previous posts here on the subject. It's off centre or more than one hole that you ruin into trouble over and need to do ASME VIII calcs.

LI

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

Double block and bleed - how did this become standard? 3

LittleInch

Petroleum

Recommended for you

Latexman

Chemical

XL83NL

Mechanical

zdas04

Mechanical

LittleInch

Petroleum

zdas04

Mechanical

davefitz

Mechanical

zdas04

Mechanical

georgeverghese

Chemical

KevinNZ

Mechanical

KernOily

Petroleum

LittleInch

Petroleum

Similar threads

Part and Inventory Search

Sponsor