[Ken Osment]

Specific to our situation where we have fine hard particulates in our feed we use a plan 32 primary flush and a plan 62/65 secondary steam quench.

[Ken Osment]

Regardless of whether it was a progressive failure or a spontaneous failure thrust is a matter of pressure balance. Generally speaking thust bearing like steady state condition, they are hydrodynamically designed for a certain trust condition and do not tolerate rapid or excessive thrust loading. We have experienced both spontaneous and progressive thrust issues on our cutting pumps and the two predominant causes are progressive wearout due to erosive degradation, or rapid changes in flow to the process which is either due tothe valving switching suddenly or the cutting bit becomes plugged. To really nail the cause for you specific situation further information is required as trust bearing failure is almost always a symptomatic cause. 

[Ken Osment]

We have had to redesign some components to make our rotary joints more reliable. Of key importance make sure that the washpipe does not deflect off center when pressured resulting in loss of seal energization and eventual leakage from the lower seal. The second area to pay attention to is the surface finish and coatings used on the lower seal. You need to minimize the friction generated when the seal is energized or running dry.

[Ken Osment]

We have had some success using the NOZ-CHEK model of check valve in place of the swing check valves. We also have had success modifying the existing swing checks after first understanding the root cause(s) of the failures.

[Ken Osment]

Our two newest cokers have accelerometers mounted in three locations which indicate when the drum has been cleaned to that level. Bently-Nevada also has a new system which involves usinf a temporary accelerometer mounted on the vapour line or top head which can detect when the bit is cutting coke and when it is simply hitting the drum. We are currently looking at adding the latter system to our existing cokers to aide the operators in this matter.

[Ken Osment]

We utilize an API plan 32 with plans 62/65 on the containment seal.

[Ken Osment]

The Flowserve fall arrest system utilizes two modified Block Stop (Suppliers name) fall arrestors which work quite well when new. Care must be taken to ensure the linkages are inspected and adjusted to grip their respective cables at the same reaction height as they do wear or premature failure will occur. ( We have seen a couple malfunction due to this issue) The Ruhrpumpen mechanism is a far more robust system as mentioned, and is my overall preference, but both work as designed if they are properly maintained. We did have to make some modifications so that the Flowserve system components so that they could adequately be adjusted and maintained.

[Ken Osment]

Care must be taken when looking at these recomendations as the DSTHF is especially prone to erosive degrdation th the the areas of the interstage bushing and second stage hub. We have taken extraordinary measures to try and design this suseptability out of this design, but have decided a three stage or greater pump with a properly designed piston assembly is the way to go. The DSTHF or HVX are good choices if erosion is not an issue.

[Ken Osment]

We have two different winch systems one circa 1964 and one circa 2000.
Both systems are broken down  and stocked as serviceable assemblies namely drive, gearbox and drum where each can be serviced jointly or independently depending on the available servicing window. Also where the servicing window is very short we stock a complete winch assembly and simply change over the cable. With our 1964 winches this is not possible as they utilize a single drive system which is shifted to the either of the two winch drums, as a result the servicing intervals are much longer.
 
We have designed a replacement winch system for these older winches that take into account our knowledge and experience of our decoking system operation as well as the recommendations of a detailed Reliability Centered Maintenance Study.
Generally both systems are quite reliable if properly maintained, although we have run into a reliability issue with the Flowserve   I-R winches if the band brakes are not properly maintained, in that the drive shaft as it enters the planetary gearbox is prone to fracture if only the motor disk brake is functional.
 
Because our cokers are on the front end of our operation and supply all of the feed for the operation we tend to stock what seems an excessive amount of spares.

[Ken Osment]

Your safety limits are typically physically mounted on the structure and some of the newer systems are also double redundant. As for where you are with the drum, a proximity counter or encoder mounted to either the crown sheave or idler sheave are quite accurate, these locations are not that good for the absolute safety limits as cable growth and or miscounts can effect the position. As an indicator they work fine. Eight of my cokers have encoders mounted on the winch drums. There is some error with this method as the payout rate differs with the number of wraps on the winch drum.

[Ken Osment]

We operate two separate upgraders, each utilize radial split pumps, one service is single stage the other two stage. The first upgrader is 2 x 100% the second upgrader is growing from 3 x 50% to 5 x 25%. From experience your optimal arrangement is the 3 x 50% arrangement as you do not lose total production in the event of a trip, and you can reduce capacity by shutting down one pump such as in the case of a furnace pigging or partial outage. Reliability takes an exponential decline as additional units are run in parallel, as pump condition is never matched and weaker units are backed out by stronger units. This can be mitigated by a proper control regime consisting of individual recycle loops, but this is rarely ever done due to the added complexity and initial cost. We are also adding a third upgrader and it looks like it will be using a three stage barrel volute design in either 3 x 50% or 4 x 33% operation. I am still hoping for individual recycle lines. 

[Ken Osment]

We have seven IDECO (Dresser) 15ton units, and seven Ingersol-Dressor P-15-4AX units, which are essentially the same but of different vintages. These are now handled by Flowserve Inc. There are other companies which service these, and we service our own if no significant restorations are necessary.  I can e-mail you the technical contact within Flowserve if you want, or ask away and I may be able to help. 

[Ken Osment]

We have had our share of issues with threaded drill stems, we also have a coker set which still uses hexagonal single piece drill stems which have thier own unique issues. Largely the threaded stems are trouble free if you pay attention to them and perform periodic inspection of the joint transitions. We also utilize a number of interlocks which reduce the shock loading on the assemblies, and are currently looking at the autoswitching tools which has been successful in elliminating strain on the stem due to coke falls. We also have to continually slow turn our rotary joints as they freeze up when the temperatures approach -40C. I’d be most willing to share any insites you have on the topic.

[Ken Osment]

Our new coker set uses a Delta-Valve top head system. As such we are looking at retrofitting our existing 12 drums. Real Estate is our obvious hurdle however there are many other systems which accomplish much the same thing allbeit not quite as elegantly. The bottom line is operator safety, and anything we can do elliminate the necessity of being close to coke drums during deheading and cutting the better.

[Ken Osment]

We are looking at implementing a trial of the auto-shifting cutting tool. I’m most interested in the specifics of what you needed to change/modify to move over to the new tool? One concern of ours is if the auto switching device will tolerate temperatures as low as -45C.

[Author]

Posts