Pipeline Water Boarding 101

Of course I remember George M. Nelson, as he maintained influence up and down the pipeline and Valdez, even when slip sliding away in a puddle because his badge went windblown from a Thompson Pass “blow”. Oh, and not to forget that Christmas bonus:

In the mid-1980s, the Trans-Alaska-Pipeline throughput was stampeding towards that ultimate goal, 2-miilion barrels a day was on the way. At the time, I was “Lead Instrument Technician” at the “Valdez Marine Terminal”. As far as the “Vapor Recovery” and the “Tanker Loading”, the control systems were functioning at near perfection to meet the demands, as we were aware this “wave” was coming home. So all our hard work in getting the control systems up to the Fluor design specifications was soon to pay off, as the “Marine Terminal” was ready and that was no easy accomplishment and took every minute of our undivided attention since “Oil In” to make it so.

And we went out of our way to enhance some of the control schemes, ones that were confronted with inferior motivation. Yes ready, as long as we could keep the “Training Wheels” greased for some of the “Pipeline Rejects” that had been sent to pasture in Valdez and thought they understood what the “Terminal” was all about. It was not a “pipeline”, but a dynamic system with many moving parts working together as one. The “Terminal” had over 200 control loops, the reason it was a 15-man crew of dedicated workers getting the job done safely. With the increased “throughput”, so would come more “water content” comingled with the “crude oil” and had to be accounted for - as part of the “Custody Transfer” criteria. Water and oil don’t mix? Think again. 

There is “comingled water” and “free radical water” and the conditioning that takes place at the Owner Company Prudhoe Bay “Processing Facilities” before it is accepted in TAPS can only remove a certain percentage. It takes time, so as more oil is sent south, more water was to be expected. To measure that water content, an instrumented process “slipstream” strips away a small “proportionate” stream of the mainstream, so that a representative “sample” can be “thieved” - oil speak. And Alyeska was required to approach “all things considered” with two criteria: BAT & BAM. Best Available Technology and Best Available Methods. With respect to the BAT, that would embrace the “slipstream” mechanics and the “Best of the Best” in present day technology to “thieve” a sample and the BAM, the best methods used in the “Oil Lab” to accurately disseminate the %water representative of the “mainstream” from that BAT sample. Simple concept, as it is not rocket science but still complicated. So a “Test” was performed in Valdez, under very controlled conditions to validate the sampling criteria that would eventually be used up and down the line at the various “Custody Transfer” points. Yes, we could afford to do our own “testing” and didn’t need no outside “interference”. That made the “Terminal” a bit different then the “pipeline”, as we had the expertise and the interest to accommodate that “Best of the Best for the Beast”. And this “testing” was paramount our dedication, it provides the documented proof should an “Owner Company” come sniffing, or one of those state regulators gets a little too nosey.

A “crude oil” storage tank was half-filled with oil and time-conditioned to become “dead”, allowing all the comingled water and free water to drop out and then “bottom drained” off as well as giving the tank rest time for the volatile hydrocarbons to boil off. This required time, to heat then re-heat the tank’s contents and then agitate, followed by rest until we felt it was “dead” oil. Ending up with a tank of “stable oil”, about 250000 barrels rested with a water content of “miniscule” and a vapor pressure of “zero” PSIA. Then the tank was “spiked” with a known amount of distilled water used in the steam boilers at the “Terminal”, so was also a known with respect to its chemistry. That was important due “mixology”. With the tank gauging and metered water additions, we had a very accurate assessment of the tank’s inventory. Then again, ferocious mixing until “thieved” samples were consistent with the hand sampled water content. It was the consistency we were after, before we could bargain for an “accuracy statement”. A “Test Slipstream” was set-up at the “West Metering” wherein Berth#5 and its “Custody Transfer” skid chosen as the “guinea pig”. The “Test Slipstream” was equipped with the BAT probes, all meeting the American Petroleum Institute criteria for “Custody Transfer” metering. Like “True Cut” and “Cliff Mock” isokinetic samplers and also the old trusted API probe - 45Degree cutter. The “test” was repeated, with varying amounts of “spiking” and in the end, very revealing results.

The API probe was a better sampler then the modern BAT devices, so it won the race. And that did not surprise us, as the main ingredient in the “Custody Transfer” recipe, it is the “Density” or as it is called for convenience and confusion’s sake, the “Specific Gravity”. And the only probe that was certified for grabbing a slipstream sample for volume calculations on a continuous basis in “real time”, it was from an API probe - ol’ Trusty. Look, oilmen have been doing this for a long time and the reason it takes so long for change, just ask anti-trust SOHIO! 

And special devices - Solartron Densitometers - would vibrate at known frequencies and then measure the phase shift in the slipstream sample and through a long winded Praunitz-Reid-Poling “multiphase hydrocarbon” algorithm, a density was evaluated and applied to the flow computers to adjust “flowing volumes” to “standard volumes”. It was a very complicated endeavor overall, but piecewise Simple Simon and the reason that when averaging any measurement of interest, simple time averaging was outsmarted by “Flow Weight Averaging”. And with the pipeline at the end of the day, right at 2400-hours the Net Volumes for the day would be broadcast and then the “24-hour composite” water measurement would be applied after it was verified by the “Lab Coats”. It was a reduction factor, that which lowered the anticipated “Custody Transfer” credits - so of course the “oil barons” wanted this deduction to be as accurate as possible.

With the API probe beating the competition by a factor of 2:1 in accuracy and actually seeing a representative composite sample of the “spiked water” with no handicap, so declared the winner, that saved a whole lot of money for initial costs if Alyeska was required to retrofit all the “Custody Transfer” points with new goods. The API probes were already in use up and down the pipeline, north to south and very easy to maintain - which adds to the overall reliability factor.  So this was good news, as if we had found a discrepancy that accounted for a “measurable loss”, that was justification for “making whole” the owner’s of the oil, including the State of Alaska. Yes, backtracking adjustments, that would have been a nightmare for the bean counters and we wondered if a “statute of limitation”…So yes, a time to leave well enough alone for the time being. Now before the test, we had evaluated and challenged the “Jistkoot” sampler for several months in Valdez, and it was a maintenance “headache”. Testing the “sampler” in parallel with an API probe, it was just too complicated a device and did not perform any better. Therefore this test was good, as it meant business as usual, we thought. Somebody in their infinite wisdom decided that the “Transfer Skids” would receive the “Cliff Mock” samplers, the device that was the poorest of performers? This is what the test results gave us, and we were very comfortable and confident that all “uncertainty” had been eliminated in the “Test” routine and procedures. And we could not get any answers as to why such a decision was made, except we understood the decision maker was not the person that would have to maintain the devices. And there was a safety risk involved, as these samplers have to be maintained by removal under pressure. And that was bothersome, especially when such was installed on the “Incoming Metering Skid” in Valdez, always active and under pipeline pressure. But it appeared to be a decision based on the fact the “least accurate” sample probe under test was consistent with a margin of error from the “Lab”? Yes indeed, the statisticians were involved! That “interference” we were starting to see infiltrate our domain here in Valdez. See, “lab coats” don’t always pay attention to details and the error with the “Jistkoot” was in favor of that “Human Error Factor”, consistent with the boring job of same thing different day mix A with B in test tube C shake well, so we got stuck with the inferior devices - and yes a full-time maintenance “headache”. That was the influence that Bragaw was having on the “Terminal”. See, we used to rely on an Owner Company Field Auditor from ARCO for advice and advise in the arena of “Custody Transfer” responsibility. Just like the advice and advise we received from EXXON Research & Engineering when Alyeska wanted to raise the Tanker Berths “loading arm rates” to faster fill a tanker - STOP! Need I say more.

Now we had the “proof in the pudding” from a very reliable set of consistent testing of which probe satisfied the BAT criteria, but went stuck with an inferior device because it was a better fit for the “Human Error” as part of the overall “fudge factor” criteria. What it meant, it was OK to be inaccurate as long as Peter and Paul were on the same detour dead-end dead-beat street. Now this was of no big deal in Valdez and soon the inferior probes were installed. And it would remain an out-of-sight out-of-mind thing of the past, and we had more important things to do in Valdez - like messing with DEC Dan Lawn’s head on the “Incinerator” air quality permits. Another story! All was good until I was asked to transfer up to Pump Station#1, because the “Custody Transfer” was in a state of “Human Error” extraordinaire and was cause for concern with the “Owner Companies” which could include the State of Alaska as “Royalty Owner”. At Pump Station#1, the accuracy and reliability and repeatability of the critical “Custody Transfer Cash Register” was all over the map, as instead of the station personnel maintaining the systems by the book, there was a concerted effort to “tweak” things and a “tail wagging the dog” which finally went completely out-of-control. That gets you to the point-of-no-return and the brink of failure. So I packed my bags and headed North and was not welcomed with a royal “red-carpet” greeting - as SCADA taking over a pump station priority that wasn’t a SCADA function to begin with coupled with a guy from Valdez called upon to fix things, that’s interference infringement and makes the station supervisor and personnel look shabby. Maybe it was payback, as in Valdez we were being inundated by a “pipeline mentality” so maybe this was a way of saying something in retrospect. Said again, the pipeline and “Terminal” were two distinct animals.

But over time, things came together for myself and a new found friend up north and at the same time Alyeska was replacing the outdated “Custody Transfer” flow computers with modern digital devices over outdated analog devices - as part of that BAT criteria. That would eliminate the “Human Error Tweaking” that allowed the “Cash Register” to “TILT”, as being digital there are no temptations to massage the messenger. At the time, the ARCO Kuparuk system had already been updated, when that field came online. So the first system to retrofit under my watch, it would be the “Lisburne” transfer skid, due to come online in a few months.

I was sent to Texas to act as an “Auditor” for Alyeska, specifically for SCADA, during the commissioning and testing of the “Custody Transfer” skid being developed by Daniels. An opportunity to get heavily involved in what was to become the Alyeska standard up and down the pipeline when measurement for deliveries and receivership was of concern with respect to “ownership”. Right off the bat, there came disagreements, between Alyeska’s “Measurement Specialist” and the Daniels’ Engineers, an entity well known in the industry to build “Custody Transfer” stuff and that company also advises the API when standards are revised. These Daniels guys knew their stuff, no second guessing that fact of the matter. When it was explained to me the bone-of-contention, of course Daniel’s was right, and then a thought came to me. Since the same equipment is already in use up at PS#1, has it been modified to be in error? Finally, the Daniels’ engineers refused to work with the Alyeska “guru” and he went frustrated and ran back to Alaska. In the following days, I worked directly with the Daniels “flow measurement” engineers and provided “oversight” to the reprogramming efforts to meet the current API standards for “Net Volume” accountability, instead of using an Alyeska supplied “PROM” that had been modified to calculate the density in error? This was major, especially if the devices up north were embedded with the same revision programming that had been supplied by Alyeska for this acceptance testing - it don’t work right! Unless for a “weird” reason Alyeska was using this “error” outside the norm. This delayed the acceptance test a few days, which delayed the skid moving north. In a memo from Alyeska’s management, it is noted: “your dedication was visible during the initial testing of the Lisburne Metering Module. It is quite evident that several questionable situations developed during the testing and your recognition brought about several days of detailed testing. In the final analysis it is your technical expertise and dedication has ensured that we are ready to support oil in. Thank you for a job well done”. So I am not making this up and when I got back to Alaska, when back up north that was the first thing I checked. Sure enough, some of the “flow computers” were using the wrong “density reduction” algorithm. How this was allowed, oh well that “Human Error” carries on even for a digital device that is supposed to eliminate “tweaking - where there’s a will there’s a way. And what a fight, as I did not have the equipment to burn new “proms” to right a wrong and the “Measurement Specialist” was adamant that the reduction was correct, it wasn’t. Now at this time in Uncle Al’s history, the “Measurement Department” had very little say in what to do and could only offer up recommendations take it or leave it as the station ruled, which meant at PS#1 it was SCADA that ruled as it was an executive decision that SCADA was to maintain responsibility for the “Custody Transfer”.

And this “Measurement Specialist” could provide no information that fostered a trust he knew what he was doing. Now I had worked with this guy in Valdez, and the local management finally got rid of him because even with an electrical BS degree, he fumbled with the relationship between “parallel and series” resistances and ask him about “impedance”, please don’t electrocute me! So he was sentenced to the Bragaw Street incompetency “boneyard” because Alyeska didn’t believe in “Your Fired”, maybe even a second chance opportunity. Low and behold, he resurrected himself! This reminds me, of the “boneyard”. For some reason, that “boneyard” of candidates not likely to succeed due merits in accomplishments but kiss-an-ass because incompetency rules, it morphed into the “Measurement Department”, a gang that thought they knew what was best for well…errors rule! So I understood how to approach this “mess” that stunk of that “boneyard” barking dog. Over time, one finds a ways and means to not surrender, but call the bluff of interference.

Now I had an engineer friend that worked over at Kuparuk and he introduced me to the ARCO “Measurement Specialist” who recognized me by namesake right off the bat, because of my association with the ARCO field auditor when I was working in Valdez. With that, I was able to “kopp” a set of “proms” since ARCO used the same metering equipment as Alyeska, under the same API criteria as outlined in the “Pipeline Tariff Agreements” and was made aware that ARCO had for a long time thought there was a problem in Alyeska’s measurements, as the balance was problematic. Low and behold, a day after I installed the correct “density reduction” algorithm proms, I get a happy call from the ARCO measurement guy, and it was the first time in months that the “leak detection” system was not in alarm - as that system looks at an indifference in the “Outgoing to Incoming” Metering skids, between ARCO Kuparuk and Pump Station#1 which means an easy balance. OK, easy if everything is up-and-up, but it something was “tweaked” to satisfy a fibonacci fantasy, forget it! With the wrong calculations, of course there was a mismatch. That was good for ARCO, that now a stubborn problem had been resolved and once again a working relationship established between the parties as it seemed some friction had ruled. But then the Pump Station supervisor is telling me that the Measurement Department through OCC is saying something is wrong. See, I never let on that I changed the “proms” and what the assholes were doing, using the error supplied by “miracle barrel proms” to offset the inaccurate “water samplers” and I just happened to have a very good understanding what they were up to, and even though it calculated out a perceived “error” it also entertained another error. This is not how we were supposed to do business - just to look good as that is how a Joe Hazelwood can get behind the wheel of a tanker and reck havoc.

I never let on what had been changed, as I understood what I was up against with the “Bozo Bus” mentality. A few days later the “Measurement Specialist” shows up with a new set of “poms” as these guys were experts in cheating, make things balance out so everybody is happy and when a bullet hole pokes a hole in the mainline, just change the “proms” and no leaks go back to sleep. Now when the “proms” were changed, then the Kuparuk pipeline was soon showing an imbalance and a leak detection alarm cause for concern, not for Alyeska but for ARCO. I get a call from ARCO and explain what was going on. Don’t mess with an “Owner Company” as the very next day the “proms” that reduce North Slope crude were re-installed. The big problem, this is a “Custody Transfer” point. A legal recognition of the commodity ownership, so what was going on was, well just more tweaking for lack of understanding what it all meant in the “big picture” show. But it was not going to stop, and that would be assured with the Endicott production soon to come online and when the ARCO and SOHIO flow computers were changed out. Even though there was less and less room for comforting that “error” through “tweaking” and by this time there would be no more “proms” for the misguided, not on my watch, it was only a matter a time things would get screwed up again. What was rather interesting, when I found some free time and went back in the Station Logs, this “prom” changing had become a routine - try this, now this better this one. So I held my ground, and for awhile it worked. In fact, it re-focused attention back on the “water samplers”, the inaccurate ones shoved down our throats that had been proven to be inferior during dedicated testing in Valdez.

One day I was sitting in a meeting with the powers-to-be from the station and a few of the “boneyard”, I mean “Measurement Specialist”, and low and behold they start talking about some of the “Lab” work and how it is performed, for testing the “water cut”. And right off the bat, suspicion on how Howey does his job, as they go on to say that his work results are too consistent? Yes, like maybe he was faking the results. See, Howey performed like a robot, same time same dedication day in day out. He was the epitome of a stable testing regime scientist, whereas the others would initiate the same testing but in a random fashion, as break-room coffee time sometimes takes a priority. And of course I spoke up and silenced the ridicule. So it was sad, that this is what…it doesn’t come from a professional attitude but this had become par-for-the-course for the “boneyard” mentality. It was becoming a “game”!

And because I still worked for the SCADA, I was allowed to purchase “Standard Lab” test equipment, so if ever there came a concern from the “Measurement Department” a device or inputting device questionable, we could quickly and easily dispel any concerns. They were trying everything, to make a system of record without error, through falsification and I witnessed it. OK, it was for their own good and did it hurt anyone - as it was more in tune to an ego problem. So we in SCADA had the wagons covered, and this was oozing out the weakness in the system of merits, that there was never a reason to believe in a balance acceptable if under false pretenses, just too complicated with way too many variables. But that is OK. Live with the knowns and take appropriate action instead of living with the “unknowns” which is accomplished by “tweaking” things inappropriately and then flying blind. It is best to know the state of affairs then to have to engage in a guessing game.

Yet this incompetency was like a virus, and for some reason these assholes thought they were god’s gift to Isaac Newton. At the time, with everything working accordingly, SCADA would soon be on the out, as the station management and the pipeline management never accepted the fact that department had to bail the station out in the “Custody Transfer” obligations.

Now one of the critical issues confronting the pipeline was the “vapor pressure” as that was limited by the “Operating Agreement” with the “Producers”. And with more NGL blending with the crude oil, this “vapor” was of concern. Now there were analyzers involved in trying to measure this “pressure” and once again, the “Lab” could perform a Reid Vapor Test for verification. But this was a science that was flawed, as even though the “RVP” is critical in the refinery process and is well documents and can be measured, for “crude oil” it is still in its infancy as very few worry about it. With the TAPS, it was a big concern. Once again, the on-line analyzers were suspect of not delivering a reasonable assessment of the “Vapor Pressure” in the crude, and that evident in the increased vulnerability in the “slack line” regions along the pipeline. So SCADA set up a test device, wherein we could test the “nozzles” used to create a “vapor pressure” to certify the device when placed into service for daily continuous “Vapor pressure” measurements. OK, we did this once again to be proactive in efforts to “trump” the “Measurement Departments” incompetency, as we knew that would soon be the case - blame the analyzers. It backfired on them once again. But after SCADA dedicated a lot of time and effort and forever calling the “bluff” of the “Measurements Department’ crying wolf, it was time to transfer the “Custody transfer” responsibility back to the Pimp Station proper. And for real, the SCADA was accused of harassing the measurement guys and giving Ranger Rick a migraine! And even though we had stood our ground and honest-to-god doing the best we could and refused to “tweak or fudge” results, we knew that as soon as we gave the keys back, that incompetency would reign again.

On my last meeting with the station personnel, as I had been tasked to build up a crew of station technicians that would be responsible for the “Custody Transfer” and felt comfortable those obligations had been met, in that meeting where the “Measurement Specialists” were present, just before break a note was handed over from the “boneyard” to the Station personnel. Like in a secret fashion, out-of-sight out-of-mind. It was a “tweak” factor that would do basically the same thing as the “proms”, make everything look good!

So what does it all mean? Tanker wrecks and “bullet holes”! Alyeska was becoming a breeding ground for this incompetency - how to look good for the 5^th Floor at Bragaw. After leaving Uncle Al, I became a contractor and was asked to write a program for the Alyeska engineering department, using a Daniels 2500 flow computer to measure “naptha” soon to be used at PS#8 to run new generators. I provided the program and never once was there any hesitation to use it, as Alyeska had a fine engineering staff. Soon after, due my knowledge of measurements and familiar with the Daniels 2500, which was the BAT at the time, I was asked to perform a similar job to measure the “flare gas” vented at a Crude Oil Topping Unit. Same thing, as the 2500 was designed for such critical measurements and the library of available program code made it an easy job. Basically a follow-the-leader, as somebody had already done all the leg work. And with this project, it would follow the API recommendations for “flare gas” measurement. Soon after delivery, the engineer wants the program revised, basically gutting the intent of the API standards to accurately measure such a gas. It was a stripped down “flow equation” that had no credibility, not in the “measurement” world. It was a waste of effort. Which bothered me, so I called the engineer to find out what had changed, why such a drastic change that would provide no semblance of a reliable measurement? He said, the “boneyard” was to blame, as this measurement fell under the purview of the “Measurement Department”, as part of the over-all pipeline balance that is part of the “Custody Transfer” INCOMPETENCY. Was I surprised, of course not and the reason I could no longer stick with Alyeska gainfully employed, it wasn’t worth it. See, when I grew up my father taught me how to behave and I will leave it at that - thank you dad! Let’s hope that “Pipeline Leak Detection” works, as if you thought the EXXON Valdez wreck recked havoc, an undetected leak along the 800-mile pipeline because someone wanted to look good and “tweaked” things, it would make Joe Hazelwood the lessor of evils.