DSS039: Does Size Matter - Why is the Standard Dust Explosion Testing Chamber 20-liters?

In this episode of the DustSafetyScience Podcast, we talk about minimum size requirements for dust explosion testing chambers. This is going to be part of a series where we discuss fundamental science concepts and how they apply to dust explosions.

At DustSafetyScience, two of our major goals are:

Connecting industry with service and equipment providers to make their facilities safe.
Connecting industry and scientific research being done by universities and other groups around the world.

Moving forward, we will continue to talk to industry experts, but we’re also going to talk about some fundamental science concepts and how they apply to dust explosions. The goal is to close the gap between the science itself and how we use it in the industry today.

In this episode, we answer the following questions:

What standards apply for dust explosion testing?
Why does vessel size matter in the first place?
Why do we use a 20-liter chamber?
How was vessel size originally determined?
What are some other dust explosion testing challenges?

What Standards Apply for Dust Explosion Testing?

The international standards for dust explosion testing are ASTM E1226 and E1515, which govern how to determine:

Pmax: The maximum pressure from an explosion in a closed vessel
KST: The size-normalized maximum rate of pressure rise under standard testing conditions.

These are the standards that are followed in North America through NFPA, but other guidelines are also used internationally. They include ISO/IEC 80079-20- 2, which is also an explosion testing for combustible dust, and VDI 2263-1.

These standards generally point to the 20-liter chamber as being a laboratory bench scale apparatus that can properly replicate industry scale conditions if used correctly.

Pmax is measured by increasing the dust concentration within an enclosed chamber and measuring the explosion pressure until it reaches its maximum (which is the greatest possible amount of pressure that it can be produced by that material). In other words, it determines how much damage a dust might be able to do to a closed container like a dust collector.

The more difficult part of combustible dust testing is determining how reactive the material is. This is usually specified as the volume normalized maximum rate of pressure rise. Under the specific testing conditions specified in the standards mentioned above, this can be used to determine the KSt of the dust. This parameter is used for engineering design including:

Designing suppression systems
Doing venting calculations
Ranking materials according to explosion outcome severity

This is a critical parameter, but unfortunately there is no underlying or unified scientific theory that can be used to determine the reactivity of a given dust sample. In most cases the physical and chemical steps for the underlying reactions are largely unknown. This leaves us with no way to estimate the speed of a flame that would propagate throughout a given dust cloud at turbulent conditions or under relevant industrial conditions. Instead we have to rely on experimental testing at the 20-L or 1-m cubed testing scales.

There has been a lot of discussion about how test results might differ from the reality in industrial scenarios. This is why facility managers need to bring in experts who can do proper testing, know the materials, and can provide appropriate safety recommendations based on the results from experimental testing.

Why Does Vessel Size Matter in the First Place?

Cubic-meter chambers have routinely been used for dust explosion testing, but their size is an issue. Smaller vessels like the 20-L chamber make testing easier. It has 50 times less volume than the cubic meter, so it needs 50 times less dust. In other words, if you need a bucket or a kilogram of dust for the 20-liter chamber, you need 50 for the one-cubic-meter chamber, and (hopefully) most facilities don’t have that much dust laying around.

The other issue is that the time required for testing is vastly longer for the cubic-meter chamber. Up to a week may be needed to do a whole KSt curve while in the 20-liter chamber, a KSt curve can be accomplished in a couple of hours. As a result, 20-liter chambers are more widely used: there are several hundred of them in the world and only about a dozen or so cubic meter chambers.

All of this begs the question: why 20-liters? Why not 5, or 15, 36, or 45 liters as the standard benchtop testing size?

How Was Vessel Size Determined?

It comes down to the fundamental properties of these testing chambers and understanding how they impact the rate of pressure rise.

A lot of this work is based on the findings of Swiss researcher Richard Siwek in the late 70s and early 80s. There are reports of testing done with 10000 individual explosions over 50 dusts to find the smallest volume capable of reproducing the scaled maximum rate of pressure rise in a cubic meter chamber.

One comprehensive report released by Richard Siwek is called Experimental Methods for Determination of Explosion Characteristics of Combustible Dust. Another good resource is a textbook called Explosions Course Prevention and Protection by Wolfgang Bartknecht.

From this work, two main parameters are defined that govern the impact of vessel volume or size on the maximum rate of pressure rise.:

The surface area to volume ratio
The maximum chamber volume

The surface area to volume ratio is important due to heat loss. If there is a large surface area on the outside of the chamber, the explosions are being cooled as they’re happening. If the chamber is at ambient room temperature, it’s actually cooling down the flame and the heat that’s building up inside as the explosion happens. There are some corrective measures that can be taken, like putting water jackets on to bring up the temperature of the chamber, but having a large surface area to volume ratio still negatively impacts the resulting rate of pressure rise.

In the report mentioned above, one of the plots shows several experiments comparing the maximum rate of pressure rise from different sized chambers compared to that of the one-cubic-meter chamber. The x axis is KSt from the one-cubic-meter chamber and the y axis the scaled maximum rate of pressure rise from the laboratory equipment. If they’re in perfect agreement, the one-cubic-meter chamber would give a 350 bar-m/s value and the smaller chamber would give also give 350 bar-m/s value. This would give a straight line running diagonally in the plot with a 1:1 slope.

According to the results presented in this paper, researchers found that the surface area to volume ratio impacts the agreement in a specific way. They found the larger the surface area to volume ratio, the lower the maximum rate of pressure rise predicted in laboratory equipment (the slope was less than 1:1).

With the Hartmann Tube the slope is much less than perfect agreement. With five liters, it’s a little bit better but not quite there. Same with 10 liters. As the vessel volume increases you start to see better and better agreement between the laboratory scale chamber and the one-cubic-meter vessel.

The maximum chamber volume is the second important parameter and this is related to the radius of the chamber. So on the same plot that was referenced earlier, perfect agreement will be diagonal. Researchers found that this line instead goes horizontal at some maximum value which depends on the total vessel size. A horizontal line indicates that the smaller equipment is unable to produce a scaled maximum rate of pressure rise above that value. In other words, a chamber of 5 liters may only match the one-cubic meter chamber up to say 100 bar-m/s (for example, the actual number depends on the material). After that value a higher maximum rate of pressure rise could not be produced.

For the Hartmann Tube, they found the maximum to be quite low. For the five-liter chamber, they found it to be around 350 bar-m/s. For the 10-liter chamber, it was higher, and so on. They realized that if the size is too small, you reach a plateau where higher rates of pressure rise can’t be predicted even though they are captured in the larger cubic meter chamber.

There are a lot of factors that could cause this, and it’s still an open scientific challenge; However, in Experimental Methods for Determination of Explosion Characteristics of Combustible Dust it is stated that the volume has to be large enough for the combustible dust to fully develop its reactivity.

On thing that could cause this is the thickness of the dust flame. For example, if the dust flame has a radius on a similar magnitude of the vessel radius it will not be able to fully develop during the experimental test. This will have a big impact on the recorded maximum rate of pressure rise and could result in it being measured as lower than it should be. This might be one explanation for the impact of vessel size on this maximum rate of pressure rise that can be recorded.

So, how did we end up with 20 liters? In the paper, researchers took the results from the laboratory testing equipment for the Hartmann Tube (1.3 liters) the five-liter sphere, 10 liter-sphere and 20-liter sphere, divided them by the results in the one-cubic-meter chamber, and found a linear relationship with vessel volume. This linear relationship produced better agreement for scaled maximum rate of pressure rise as they increased the vessel size.

When researchers extrapolated the linear relationship to perfect agreement, they found 16.1-liters as the minimum vessel volume for laboratory testing that could reproduce results in the one-cubic meter chamber. They then went with a slightly larger volume on 20-liters.

If this all sounds complicated, that’s because it is! 10,000 individual tests were done just to get to that data point and to complicate things further, we don’t yet have certain fundamental science knowledge available to understand how dust flames propagate.

What Are Other Challenges With Dust Explosion Testing?

There are several other challenges. We looked at three of them in this episode.

Maintaining Uniform Dispersion and Turbulence

It is difficult to get a uniformly dispersed dust cloud inside one of these chambers. The test used a uniquely designed nozzle to get uniform dispersion, however it is still difficult to do.

That is why it is very important to follow a standard procedure like ASTM E1226, ISO IEC 80079, or VDI 2263. Otherwise the results will be different from the standardized body of knowledge that it has taken 50 years to develop. Maximum rates of pressure rise determined using alternative dispersion methods should not be labeled as KSt, as they did not follow these standards.

Ignition Strength and Marginally Explosible Dust

The second challenge is ignition strength and marginally explosible dust. Maybe low-reactivity dust is a better term because it is still dangerous and causes loss in industry.

ASTM standards put some recommendations in their testing procedure for a go/no go test and how that applies for low-reactivity samples. Sometimes the recommendation is to go to a cubic-meter chamber. The details can be complicated, which is why you need to be dealing with experienced parties who can advise you on what kind of testing should be used.

High-Reactivity Dust and Radiation

Particular metals have a radiation effect that impacts rates of pressure rise as the cloud gets larger and the vessel volume gets larger. Radiation is released from the particles, and its strength is proportional to temperature to the power of four.

Some metals, like aluminum, have very high flame temperatures. As radiation is emitted from the flame front, it can actually heat the particles upstream from the cloud before the flame gets there. As the flame gets stronger and the downstream particles are preheated even more, the flame grows and accelerates due to radiation.

We don’t have a great scientific understanding of what the exact degree of increase in the flame speed is, which is why some standards multiply the rates of pressure rise by a factor of two or by another safety factor. That’s because we’re still trying to figure out the impact of radiation on for metal dust and this is an open challenge for the research community.

Conclusion

While all of these scientific details may not be as useful to lumber mill workers, if you’re a consultant or expert in the field, perhaps the information in this episode will provide you with some value.

If you have questions about the contents of this or any other podcast episode, you can go to our ‘Questions from the Community’ page and submit a text message or video recording. We will then bring someone on to answer these questions in a future episode