I’ve spent many years and a great deal of my time and effort helping customers solve problems in the world of tablets and capsule manufacturing. The problems encountered in manufacturing often stem from a request to establish solutions for a defect, to troubleshoot, and to provide training. Solving the problems usually involves several skills and initiatives to establish the root cause(s) and help avoid repeating the same mistake over and over again. To solve problems and avoid repetition there is a methodology that I follow that I’d like to share so that everyone can more effectively overcome them on their own.

Common problems
Many problems in manufacturing stem from common fundamental mistakes and all of them are easy to solve. So please take a little advice, or at least hear me out. First, I want to define the most common problems with manufacturing and then I will add suggestions to help solving them.

To understand the problems is to understand the manufacturing cycle. The cycle in manufacturing starts with machine setup, operation, disassembly, cleaning, and back around to setup. Setup involves assembly of each of the machine components. If the parts don’t fit right or came from another machine, the parts are often forced into place, sometimes with a hammer. Suffice it to say most setups are lacking and the procedures that support the setups are usually so vague that nobody could follow them anyway.

Operation must start with a clear understanding of all machine settings and an understanding of what and how each adjustment will influence the dose. Many operators simply don’t know their own controls very well, how to correctly make changes and how to fine tune operations to make the best product they can.

After operation is over it is time to clear the product and rectify the batch. Now the machine is ready for disassembly. Correct disassembly involves inspection of each part for wear from the previous run, establishing a record and identifying items for repair, or replacement.

The final step is cleaning. Most companies have figured out what to clean with they just don’t clean thoroughly enough. On average complete disassembly is often not accomplished, therefore cleaning cannot be complete. When requested to inspect a machine, I ask for a clean machine—unfortunately most don’t know how. I have had operators panic because they have never seen their machine properly disassembled. 

Parts handling
The one thing more than any other that wakes me panicked at night is the sound of tablet press tooling—punches and dies—clanging together. People know this about me. I use the term “people” instead of “friends” because they intentionally used some old tooling to play a trick on me. They crashed the cart the tooling was on and all of the punches and dies fell to the floor on top of each other and then laughed really loudly as they ran in the other direction (not funny). But really it’s not far off.

Most parts aren’t damaged on the machine during operation. They are badly damaged during disassembly, cleaning, storage, and assembly procedures. If this was your grandmothers fine China dishes she would beat you with her broom if one little clink between dishes was detected. That’s how I want my machines and parts to be handled, just like they are made out of glass and your job is on the line if there is one little chip. I can count on my hands the number of companies in the world that are truly successful at this. Most are ex-baggage handlers from the airline industry. I recently was asked what I could possibly teach a team of tablet press technicians with an average of 22 years’ experience. One look at the press and I knew they weren’t professional enough to make the grade. They may have been press technicians for twenty two years but they wouldn’t have gotten past their first 60 day review as bad as the press and tooling was banged up.

Tablet presses and encapsulators are precision made machines. Step one in training is to explain that because it is an industrial machine that doesn’t mean it’s ok to treat it rough or abuse it. Small dings and dents impact machine performance. They can generate heat where it doesn’t belong. They can cause metal contamination, and reduce a production runtime significantly.

When to stop a machine?
The wrong answer is at the end of the batch. There are several factors that determine how long a machine can run. How clean it is before assembly, proper lubrication at startup, control of dust and fines in the final blend, and careful operation. Every batch is different. Just because you ran it successfully for more than a shift yesterday, doesn’t mean you can do it today. Fines (dust) is the enemy. These fine particles get airborne and land on the punches, absorb the lubricating oil and then cause them to become tight. This means it is time to stop running and start cleaning regardless of how close or far the end of the batch might be.

The Pit Crew mentality
Be ready to clean and repair. Get your tools out and treat downtime like a pit crew in car racing. Your goal is to get the machine back running as quickly as possible. If a tablet press or capsule filler isn’t running, you’re not making money. Normal downtime activities include Easter egg hunts for tools needed to complete the task. On average most companies don’t have the necessary tools to clean properly, so cleaning hard to reach places either takes much longer than necessary or it simply doesn’t get cleaned.

Excipient function and order of addition
On the formulation front I get a lot of calls about ingredients, specifically excipients. The first rule is an excipient must have a function. To improve flow, to control cohesivity, ejection, disintegration and dissolution are the places to focus first. The second issue is how to put the formula together. Ingredients are added in a specific order and are often part of a pre-blend of groups of ingredients. The objective is to achieve homogeneity—a complete blend. The secret about blending is that there is no such thing as a perfect blend. The traditional way to determine blend uniformity is through sampling and testing. Establish the “dead spots” within the blender, and test those areas compared to open mixing areas for blend uniformity. Ingredient addition should complement the characteristics of the paired ingredients. Compatibility is the key to success. 

Blend hold time
Get this one right and you can improve performance immediately. Don’t use a batch immediately following blending because the ingredients have been aerated. This can mean waiting 6-24 hours. The objective is to use the blended batch when it is most consistent. Use it too early and your yields are reduced. If the blend sits too long, it settles and segregates. The key here is that all formulas settle at different rates, which means that the useful period for peak performance is typically narrow. The rule is the more a product settles the narrower the window. The average product has a peak performance window of four days. Use it too soon and weight control and yields are all over the place; wait too long and compression issues like sticking, capping and delamination become more frequent.

Thou shall screen all ingredients
Probably one of the most common errors in manufacturing come from agglomerates. The rule is all ingredients must be sieved or screened before initial use. This eliminates agglomerates, improves uniformity in blends and levels the playing field. I know what you’re thinking: this one’s not for us! Well, listen up. It is the most common error in manufacturing. It influences everything throughout the operation. It’s too easy to accomplish and the downside is too risky. But I see it all the time, and you wouldn’t believe some of the products I see the problem with—I’m talking big and important.
Usually when I read deviations and CAPAs they have nothing to do with the real problem or the real solution. I have even seen procedures where it is up to the technician to determine if each particular batch needs it or not. Note: some products re-agglomerate on their own. A good example is magnesium stearate. Products like this are screened at use; prescreening doesn’t work. 

SOPs
Good SOPs and Batch Records are hard to find. Procedures are often drafted from machine manuals by non-operators. That is ok as long as they test for success. This is the best description of how to write a procedure: “Write what you do, and then do what you wrote.”

In other words, write a procedure as is being done, the way in which everyone and anyone can follow once they are trained. Then repeat the procedure to make certain it works. I do this all the time in my classes when we install tooling on a tablet press. The first two people create each step in the process. Then they follow each step exactly as they wrote it, amending the instruction to pick any details that were missed. Then we involve a third person to follow the instructions and test and modify, repeating the exercise until we have created a procedure that works.

The final key to the exercise is to get a buy in for the team. The buy in is an agreement that this is a procedure that must be performed following these steps; it’s not ok to do it their own way. They must believe in it and follow it. Once an operator gains confidence that a procedure is worthy they will follow it. The problem with many procedures that aren’t proven and tested for accuracy is that everyone comes up with their own version, and that’s not ok. We all know that a great many procedures aren’t robust and if they only did what the procedure required they would end up with success.   

Don’t change that adjustment
There are a number of myths in our business when it comes to machine operation. The average operator doesn’t understand each of the functions of a machine and how and when they should be made.

A good exercise in training is to identify each machine function, control and adjustment and define when and why they should use it. An example of this would be punch penetration on a tablet press. Most operators are told not to touch! That the adjustment is set at the factory and shouldn’t be changed. This is a misconception. The reason for an adjustment is so you can do just that. In the case of punch penetration, it is a tool to control air release during compression and the final setting has to do with a few key factors. Fine tuning the adjustment can make each batch perform better. If the procedure says to set the punch penetration at one setting rather than at a range that is based on slight changes in the tapped bulk density, it can mean the difference between one batch running considerably different than the next.

Common unit operation mistakes
In the weighing department common mistakes are properly labeling clean, to be cleaned, and in use hand implements like scoops. If it’s not labeled it must be considered not clean. Another common mistake is to have more than one ingredient open in the weigh room dispensing area, which can and will result in cross contamination. Open only one ingredient at a time. Weigh, record, and document and close each ingredient container and make sure they are properly separated from other ingredients.

Granulating is defined as controlled particle agglomeration. Forming granules reduces “fines,” or dusty particles that exceed 10-20% of the total volume. In tableting exceeding 20% fines results in decreased performance of flow, compression, ejection, friability, disintegration and dissolution performance. We know that the higher the percentage of fines causes tablet weigh and hardness variations. The higher the percentage of fines the greater the likelihood of capping, delamination, black spots, and more frequent cleaning stages resulting in increased downtime. The answer is to make sure that the granules and the particle size distribution is on target. This is why we must have a range for tablet press speed. If there are more fines we are most likely going to need to sacrifice speed and performance. 

Milling and size reduction to control oversized particles and granules is important too. The higher the percentage of larger particle and granules results in weight and tablet hardness variation. The general rule is to establish that no particles can exceed 20 mesh (850 microns). Some go a little larger, maybe as large as 16 mesh. Our studies and findings show that the higher the percentage of oversized particles and granules the more likely there will be weight and hardness variations, which can have a direct impact on controls sticking and delamination.

When it comes to high shear mills that use sharpened blades, most companies don’t have a standard to follow for blade wear. When a blade loses its sharpness the result will be more fines. The other common issue is tracking screens. Make sure that each screen has its own logbook so you can track which screen was used. Remember, all screens are not created equal. A thicker screen of the same aperture will result in a smaller particle size because other trajectory of the particle moving thru the aperture is on an angle. The higher the speed and the thicker the screen the smaller the particle becomes. 

Blending is a science all to its own. Blend volume, and how long the ingredients are blended is important to understand. A blend of ingredients goes from an unblended state to a blended homogeneous state. As blending continues the ingredients will unblend. Always screen or sieve ingredients prior to blending, all ingredients. Always follow order of addition, as stated previously. I’ve asked operators many times if they follow the order of addition or do they ever just dump it in. You don’t want to know the answer because it’s not good. It is not ok to deviate from a procedure, this is why a procedure must be proven, practiced and enforced. This is why we have double checks.

In tablet coating one of the most common errors is making certain there are undissolved lumps in the solution prep tank. Undissolved coating materials clog guns and cause bearding—buildup of dried coting materials on the gun tip. Once bearding occurs cleaning the gun inside of the pan is the second most common mistake. Allowing these dried materials to fall onto the tables will result in tablet defect.

I’ll end with a note for my supervisors. A common mistake is to tell the operator to hurry up during the disassembly, repair and cleaning cycles. First find out why it is taking longer. Most times there is a good explanation. Make certain when you say hurry up that the request isn’t misconstrued to take a short cut from the procedure. If it is taking longer to clean there usually is a reason. It could be because there were more fines in the last batch and the machine is dirtier and it will take longer to clean and changeover. Again, most operators don’t have the best tools to do their job, or don’t have good training on how to properly and effectively disassemble, and clean. A supervisor needs to identify w¬hen this occurs and rather than push, offer more support and better tools so the operator can be more effective.
 
There’s more, lots more. But over the years I have learned that these are very common problems and events that can easily be solved. I’m wide open to criticism, experiences and continuous improvement. Each of these topics could be a complete and full article in and of itself. Give us the feedback and let us know how we can help improve the manufacturing process. 

---

Mike Tousey is the CEO and Technical Director of Techceuticals LLC, Cleveland, OH, USA. Techceuticals is a training, troubleshooting and optimization company that helps solid dose companies perform better. He is often called into provide Risk and Gap assessments to determine areas of improvement. His forte is solving tablet and capsule defects, working thru the process from end to end.