Thursday, September 30, 2010

Hmmm...

I must apologize for somewhat neglecting this blog.  I've been so busy with my MBA and work.  I looked at some of the analytics for this blog and noticed that the most popular post seems to those around troubleshooting.  If there's anything in particular you are having trouble send me an email and I'll make sure to reply directly or present a post on the topic.  I also noticed that the Cobb test troubleshooting post is the most popular post.  Perhaps I should write a little bit more on this topic.

Either way I will try to pay more attention to this blog and post more often.  Thanks for reading.

Monday, July 19, 2010

Troubleshooting Tips: Worms

Worms are a formation defect resulting from strings of improperly dispersed fiber.

They can be caused by an incorrect slice setting, excessive water from the breast roll shower, and poorly set deckle boards.  They can also be caused by an improperly set forming board and too high of a total head.

Once the source of the worm issue is identified the proper adjustments should eliminate the worming.  If the source is not known, carefully adjust probable cause areas to identify the source.  Of course, you should go through the adjustments that would cause the least possibility of a break before the ones that are more likely to cause a break.

This is a fairly straight forward process.  The backtender or quality person should be the ones who detect the worms.  The wetend should be notified to find the cause and rectify the situation.

Tuesday, June 8, 2010

Back to Paper

So I've been keeping up with the news and the oil spill is still a huge problem.  I apologize for not posting as much lately.  There are a couple of reasons for this, I wanted to leave my oil spill solution suggestion up at the top of the blog for a while and I've been a little busy of late. 

Don't worry though.  I'll be posting some new stuff here in the next couple weeks.  I think I will talk a little about some flex roll technology being installed for some press rebuilds.  I will also try to add more troubleshooting tips as these seem to be the most popular pages.  If there is anything in particular you'd like to hear about just shoot me an email.

Thank you for being a part of this blog, and I hope you've found some of the information to be helpful.

Friday, May 14, 2010

Off Topic: Oil Spill Solution

This post has nothing to do with making paper.  It has to do with something much more important.  The current oil spill in the Gulf of Mexico is a huge environmental issue.  I have thought about the problem over the past couple of days and this is what I have come up with.  My hope is that it gets in front of the right people and may provide a solution or a piece to a solution to this environmental tragedy.

I am very concerned with the current oil spill in the Gulf of Mexico. Due to this concern, I would like to offer up a possible solution. This is a rough solution that can be refined for implementation.

The current issues with siphoning the oil is the formation of natural gas (methane) hydrate. In order for this hydrate to form water needs to be present in a cold and/or high pressure environment. The phase diagram below illustrates the methane hydrate relationship between depth and temperature.


Figure 1. Methane Hydrate Phases

As depth increases, pressure increases. A higher pressure allows the hydrate to form at higher temperatures. For example, at 1400m (~4600ft) the temperature needs to be at or below 60F for the hydrate to form.


Figure 2. Gulf of Mexico Water Depth Temperature

Figure 2 shows the water temperature at various depths for the Gulf of Mexico. At a depth of 1400m, the temperature is around 5C or 41F. At this depth and temperature, with the presence of water, hydrate will form.

The proposed solution, which is illustrated on the attached page, addresses the hydrate issue. In order to reduce or eliminate the formation of hydrate, the oil should be siphoned at a point much further away from the source of the leak. The oil must be directed so that it can be concentrated and siphoned efficiently. To direct the oil, a cylinder should be placed around the source of the leak. This cylinder can be made of concrete or steel. It needs to be able to withstand the turbulence from the outflow of oil. It also needs to be heavy enough so that it can anchor a reinforced spiral ventilation duct similar to the one shown below.

Figure 3. Ventilation Duct

The big question mark is how long the duct needs to be. It could be anywhere from a few hundred feet to a few thousand feet. However, it does not need to come all the way up to the surface. Its purpose is to direct the flow of the oil up to a depth where hydrate formation is unlikely if not impossible so that another line which will come in from the top of the duct can efficiently siphon off the oil. The duct will act as an artificial column of oil.

The attached design of the solution shows two pumps, one for siphoning the oil and the other one near the base of the cylinder around the oil leak source. The purpose of this pump is to pump out water from the column. The specific gravity of oil is lower than water so it floats. In implementing the proposed solution, some water may be trapped in the column. While some water in the column does not present a problem, a significant amount of water will reduce the efficiency of oil recovery and also increase the possibility of hydrate formation. By pumping water out of the column near the base, this will create a column of mostly oil.

A pipe or hose should be placed inside the column with a positioning ring. This ring will keep the siphoning hose or pipe in the middle of the column. The siphoning pump should be set at a rate identical to the rate of oil leaking. Once the flows are matched up the column will achieve plug flow. Under ideal plug flow conditions, 100% of the oil coming from the leak would be recovered. However, this will be difficult to achieve without automatic controls. I believe by placing cameras near the top of the column and at various points along the column an operator could monitor and control the flow and recover over 90% of the oil. If the siphoned flow is too high water will flow in from the top of the column. If the flow is too low oil may escape through the top of the column.

The big hurdle for this solution is the design of the reinforced spiral duct. The materials necessary for a viable duct are available. However, the problem may be in the production of a duct that is long enough. While the general design of this solution seems sound, the details still need to be ironed out. Note that this is not a permanent solution, but it would alleviate the oil leak until the relief wells are operational.
If you have any questions, concerns, or comments please feel free to contact me.  My email address is nhiemcao@gmail.com.

Tuesday, May 11, 2010

Troubleshooting Tip: Curl

Curl is an issue that can creep up on you from time to time.  There are three types of curl to be concerned with; machine direction (MD), cross direction (CD), and diagonal.

For MD and CD curl adjustments to the jet-to-wire ratio or rush/drag can be made to alleviate the issue.  Generally if you are having a MD or CD curl issue, moving the jet-to-wire ratio closer to one or moving the rush/drag closer to zero will help.  This adjusts the fiber orientation to alleviate the curl issue.  Reducing refining and reducing broke content will also help.

Other things to look at when dealing with MD and CD curl are the size press and coater.  Make sure that the sheet is picking up the same amount at the size press.  The same should be done for the coater.  Also check to make sure that the mixtures in the run tanks are consistent with each other.  Unless you are running out of one run tank, this can be a source of trouble.

Diagonal curl is a different animal altogether.  It can come seemingly out of nowhere.  The first thing to look at when this issue pops up is to see what has changed.  For example, was there an outage prior to the issue and if so what was changed out during the outage that could have caused the problem.  The lexan sheets in the headbox can be a source of trouble.  Make sure that the new sheets are the same as the old sheets and that the manufacturer hasn't changed anything in their production process.  Other causes could be uneven wear of metering blades or rods at the size press.  This will cause an uneven coating that could cause diagonal curl.  Uneven press loading may also cause diagonal curl.

Out of the three curl types, diagonal is the most difficult to troubleshoot.  Remember, curl like any other issue is caused by a change in the process.  Sometimes the change is something that can be controlled, and sometimes it is due to the raw materials you have to work with.  Either way make the necessary adjustment and make a note for future reference.

Sunday, April 25, 2010

Brighter Brights and Whiter Whites

I was thinking about something the other day, and I thought I'd share it with you.  It seems like there's always a competition between paper companies to see who can make the brightest/whitest sheet.  The conventional method is by adding optical brightening agents (OBAs) or optical whitening agents (OWAs) at the wetend of the machine.  This seems to work fine but you can only brighten or whiten so much before you lose the effectiveness of the OBA or OWA.

So what else can you do?  Well you can innovate and make a better OBA or OWA.  This would require some collaboration with a chemical company and could take years to develop.  So I started thinking about this and here's what I came up with.

Since the OBAs and OWAs are added at the wetend they have a limited dwell time.  I wonder if increasing this dwell time would allow the OBAs and OWAs to achieve greater brightness and greater whiteness.  If nothing else it could improve OBA/OWA efficiency, get more brightness or whiteness with less chemicals.  So now the question is where can you add it?  How far back can you go? 

What about the pulp mill?  Why can't it be added back there?  It's definitely worth a trial right?  I am not a pulp mill guy though so you should talk with your pulp mill manager to gain insight into where some good addition points might be.  Also remember that if you are going from a grade the requires OBAs or OWAs to a grade that doesn't require OBAs or OWAs that it might take a little while to get all the OBAs or OWAs out of the system.  Ideally you would want to try this at a mill that makes OBA/OWA required paper most or all of the time.

So the concept is to base-load the chemicals in the pulp mill and add as necessary in the wetend to reach your target brightness/whiteness.

Friday, April 23, 2010

Start-up Checklists

I know most people hate checklists, especially operators.  They have a checklist they go through when they start-up, but usually it's in their head.  Implementing formal checklists for each operator is a good way to avoid costly mistakes. 

Start-ups after an outage are very hectic.  Everyone has been working for a long time.  They're tired and ready to go home.  Having a checklist helps the operators go through the necessary steps, it also keeps the operators accountable. 

So how do you go about implementing these checklist?  The first step is to develop the checklist.  The operators should be involved in this process so that they take some ownership of the checklist.  Once the checklist is finalized, it must be determined who will actually carryout the checklist.  For example a supervisor can go through the checklist with the operator to make sure that the proper steps have been taken.  Simply giving the checklist for the operator to go through by themselves is not a great way to implement this tedious task.  Operators would soon discard the use of the checklist and go back to their old ways.  Having senior operators be responsible for the checklist would be a good way to provide some accountability.

The checklist should be turned in and kept for at least the next two outages.  They should also be revised from time-to-time as new equipment is installed or modifications are made to current piping and equipment.

This is a fairly simple way to improve start-up efficiencies.  Remember it's all in the way you implement the checklist.  Once it is executed poorly, it will be difficult for operators to buy into the program.