Wednesday, August 20, 2014

New Paper Undermines Stove Industry Variability Study

A new paper written by Woodstock Soapstone, a New Hampshire wood stove manufacturer, calls a key stove industry study misleading and flawed.  The industry study says the inherent variability in wood stove testing suggests that the EPA cannot lower emissions standards below 4.5 grams an hour. 

The EPA posted the 7-page Woodstock Soapstone paper today as part of the official record that the EPA can use to determine the final rule due in February 2015.  It was written by the company’s CEO, Tom Morrissey, and says the industry variability study is “based on a data sample that is small, old and deeply flawed.” 

The late Paul Tiegs from OMNI
 test labs was a  prominent
critic of variability being caused
mainly by fuel.
Morrissey’s paper argues that variability has much more to do with whether a manufacturer is paying for emission certification than inherent variability in the combustion process. “When a manufacturer pays for certification testing, why are the results so much better than at any other time the same stove is tested?” Morrissey asks.  The Alliance for Green Heat believes, as the Morrissey paper also suggests, that test labs have a range of operating procedures allowed by the EPA that can result in variable emission results.  The late test lab icon Paul Tiegs, a founder of Omni test labs, became a champion of tightening testing protocols to achieve consistency in testing, and rejected the notion that variability was mainly caused by solid fuel, as the industry study argues.

If the EPA, states, and air agencies take the Woodstock Soapstone paper seriously, it could unhinge a major industry legal strategy in the fight against stricter air pollution limits.  Many officials in the EPA and state air agencies were already critical or at least skeptical of the industry variability study.  However, there has not been such a detailed critique from inside (or outside) the stove industry before this.

The variability study was produced and written by Rick Curkeet, a hearth products engineer at the Intertek testing lab and Robert Ferguson, a consultant who is now working for HPBA on the proposed EPA regulations.  It was released in October 2010 in anticipation of the proposed EPA regulations and is called the “EPA Wood Heater Test Method Variability Study: Analysis of Uncertainty, Repeatability and Reproducibility.” 

[Update: Rick Curkeet submitted a rebuttal to the Morrissey paper on Sept. 10 and it should be posted in the EPA docket soon. His 7-page rebuttal defends his original study and he concludes that he stands by his "analysis and conclusions."]  

The study showed that there is a very wide range of variability between proficiency testing and certification testing of wood stoves.  HPBA contends that “it is arbitrary for EPA to define ... a value that is lower than the precision range” of the test method.  HPBA says the minimum justifiable emission limit is 4.5 grams an hour, which Washington State adopted in 1995 and has since become a de facto national standard.  The variability study does not say who paid for it but HPBA confirmed that they provided partial funding and extensively vetted early drafts in late summer and early fall of 2010.

While Morrissey’s paper is by far the most direct critique of the variability study, most experts agree that the understanding of variability in wood stove testing can be assessed in many more ways than the data set used by this industry study.  For example, compliance test data could be used.  The EPA requires that all stoves be retested after they produce between 2,500 and 10,000 units.  This data, if it could be obtained from the EPA, would provide possibly a more important data set than the one the industry study chose to use.  Another data set will be from the “K list” changes.  Most stove manufacturers are re-certifying their stoves in advance of the new EPA rules so they will have 5 years before they have to test again.  HPBA has been encouraging its members to recertify stoves with “K list” changes as they are allowed to do by the EPA so that they won’t have to face higher testing costs that include cord wood testing and the uncertainty of a new test method that may be harder to pass.  HPBA had urged the EPA to grandfather all stoves under 4.5 grams an hour until 2020.  In any case, these retests will provide a new and better data set to assess variability.

Even if high variability could be established and confirmed using various approaches and data sets by independent experts, the poor relations between industry and state air agencies and other key players has undermined the ability of the two sides to agree on much.  At a November 2012 meeting in Minneapolis convened by NESCAUM, Greg Green of the EPA left the room and urged the two sides to talk more amongst themselves.  That strategy did not work and very little productive communication occurred for more than a year, a result that is likely not beneficial to the interests of HPBA industry members. 
Greg Green, Alison Simcox and Gil 
Wood of the EPA listening to testimony 
at the Boston hearing on the NSPS.

The hard line approach to critics is what prompted Morrissey to write his rebuttal of the variability study which begins as a defense of test reports of his own stoves that had been called into serious question by HPBA, Rick Curkeet and Roger Purinton at Jotul stoves in formal comments to the EPA.  Ironically, proof that catalytic or non-catallytic wood stoves can be consistently clean has become the biggest threat to the mainstream stove industry that HPBA represents.  This conflict between cleaner catalytic stoves and not-as-clean non-catalytic stoves became very heated and public in an EPA hearing in Boston on February 26, 2014.  But it was preceded by the release of study in 2013 by non-catalytic makers that dismissed the effectiveness of catalytic stoves to reduce wood smoke in real world settings.  If this public rift within HPBA had not happened, the stove industry would likely have made it through the EPA rule making with a much more unified voice. 

According to interviews with non-catalytic stove manufacturers, they felt it was vital for the EPA to understand that very low emission numbers from catalytic stoves in testing labs did not accurately reflect emissions in peoples’ homes as catalysts often clog, are not replaced, and are not properly engaged and used by consumers.  This issue has proved to be key because the EPA proposed emission limits of 1.3 grams an hour in 2020, a number that only a few catalytic stoves can appear to meet.  The company that has produced tests showing it can meet it is Woodstock Soapstone.

The Woodstock Soapstone defense of its test results and critique of the variability study comes at an important time when the EPA is finalizing its new wood heater standards, known as the New Source Performance Standards (NSPS).  It is believed that as of September or October, the EPA staff will have made most of their key decisions to send to Washington for review and approval by senior EPA officials and EPA lawyers.  Attempts between HPBA and air agencies to reach any agreements behind the scenes could still be fruitful, but time is running out.  And even if any agreements could be reached, the EPA may not adopt them.

The regulation and emission limits for outdoor wood boilers are also hotly contested but the testing variability for boilers is not a big issue, nor is it an issue with pellet stoves.  One solution, supported in part by the Alliance for Green Heat, would be to set separate emissions standards for pellet stoves, catalytic stoves and non-catalytic stoves, based on how clean each technology has become.  The argument for separating pellet stoves from cat and non-cat wood stoves may be even stronger because they use a different and very uniform fuel and are burned in a much more controlled combustion setting.  The Catalytic Hearth Caucus, of which Woodstock Soapstone is a member, strongly opposed separate emission limits for cat and non-cat stoves as well.  Ultimately, HPBA did not recommend setting separate emission limits that Jotul and other non-catalytic manufacturers initially appeared to support.  At this point in the process, the EPA may have already decided to set a single emission standard for these very different technologies, as they had proposed.

If senior leadership at EPA sees reliable data that at least one stove can consistently be tested under 1.3 grams an hour, they now have a better legal foundation to stick to their proposed 1.3 gram an hour standard.  And this is the nightmare scenario that HPBA and non-catalytic stove manufacturers fear.

Stove experts like Tom Morrissey, Robert Ferguson and Rick Curkeet are not only savvy about how stoves are tested, they are also skilled number crunchers.  Interpreting stove test data is like any other data set: it can yield very different conclusions based on what data is used, how it is interpreted and what statistical methods are applied.  The EPA in turn has to assess the reliability of each study and they can reasonably expect that their assessments will be challenged in court.

Tom Morrissey, top, 2nd from right and
the team that designed and built the
Ideal Steel for a 2013 competition
The stove that Tom Morrissey says is reliably and consistently less than 1.3 grams an hour was specifically built to win the 2013 Wood Stove Design Challenge, and it did. The Wood Stove Decathlon judged stove on 5 categories: emissions, efficiency, affordability, consumer appeal and innovation.  The Ideal Steel Hybrid is certified by the EPA at 1.0 grams an hour and gets 82% efficiency.  A second place winner, the Cape Cod, was a similar hybrid stove by Travis Industries that is certified at even lower  0.5 grams an hour.  Travis however does not say that they can reliably or consistently test at such low emission numbers, particularly with cord wood.  Woodstock Soapstone says testing shows that their Ideal Steel Hybrid is as clean with cord wood as it is with crib wood.

The Alliance for Green Heat promotes cleaner and more efficient residential wood heating to reduce our reliance on fossil fuels and help families affordably heat their homes.  Founded in 2009, the Alliance is based in Takoma Park, Maryland and is registered as a non-profit, 501c3 educational organization.


  1. The EPA should butt out and let the industry grow. There is no good whatsoever that can come of this regulation. There is more pollution from forest fires and/or volcanoes than the hearth industry ever puts out. This is nothing more than a witch hunt and a power play by the feds. If a local area has a problem with too much smoke then let them deal with it locally - much more efficient and cost effective that way.

  2. At point to make. Ethanol for a vehicle fuel is always getting the "real cost" argument. Cultivating soil, planting, spraying, harvesting, trucking and such and then calculating the true expense of the use on the environment.

    I just don't see the small home owner, polluting with burning wood in their "newer good" stove. Burning in an outdoor stove that uses more wood than an indoor wood heater, yeah some (most) waste energy. How can burning natural gas, propane or fuel oil be good for the environment?

    Burn firewood and plant more trees. As with most decisions, the company (person) with the most to lose, will fight the hardest for the sake of losing money or to continue with the gains they are currently receiving.

    I have a good stove design. Yet, I have a home owner that wants me to fix a stove from another manufacturer. Should not the other company just simply build a better stove? If aloud to sell junk, companies will sell junk. Some regulation protects people and the environment.

  3. “This leaves the only logical conclusion. Variability
    in wood heater emission testing results for any given
    appliance is most likely a function of the random na‐
    ture of burning wood, no matter how tightly you try to
    control the process."...

    I agree with Tom Morrissey that the above statement from HPBA is misleading.
    Paul Tiegs stated at a B-415 meeting that I was at, that in in-house testing at OMNI, they were able to get 6% repeatability in a series of 6 repeat runs with cribs, by tightening the moisture spec. Unfortunately, he never published this data.

    In our own testing on masonry heaters, we got 10% repeatability with cribs constructed from run-of-the-mill Home Depot SPF 2x4's that were, however, carefully matched for moisture.

    Similar to Morrissey, we got cleaner results with cordwood than with cribs.
    If we want repeatibility, then the current fad towards cordwood testing is misdirected, in my opinion.

    Paul Tiegs was a big advocate of crib testing. Cribs can be specified and repeated exactly, lab to lab. How do you do that with cordwood? - that's the $64.00 question. We are putting the cart before the horse.

    For instance, an informal analysis of the results from the Wood Stove Decathlon yielded a variability of 40% (cordwood), which is in line with what HPBA is claiming.

  4. Interesting. A "paper" undermines a "study" - if that's the standard, we should all get busy writing!

    Of course, the great unsaid is that consumers burn ONLY cord word and not of an even moisture content. Also, having one stove built and tested to the standards is not really "proof of best technology" any more than the Honda Insight was in 1999. Did the EPA then adjust the standard so that only the Insight fit into it? Of course not. On the same token, more power to those doing the research and improving the real world burns.

    I don't think there is a single example of a standard being set as high as one product demonstrated when tested in-house by the manufacturer - it would be a shame if that were allowed to stand.

    Tightening up is a good thing - but it should represent the real difference in cord word burns in consumer houses. Perhaps the proponents here would be willing to donate the million or so needed for a basic study?

    Let me make a prediction here. If the tight standards are allowed to stand, there will be a few basic designs on the market which will be copied (by China and others) within a year or so....if that's the goal, then so be it. But if innovation within a relatively wide goad (>2 or 2.5 g, for example) is desired, then the standard should be set higher to allow for all the differences in moisture, chimneys, operators, stove sizes, etc.

  5. Sometimes we need to be dragged to a better way, kicking and screaming...remember with automobiles in the 70' makers said emission standards would put them out of business...and there was growing pains but in the end no reasonable person can say it was a bad move...The same holds true with woodstove i'll take my epa certified stove over a smoke dragon any day.

  6. Sadly I really want regulation to be the last straw. I would prefer to see standards for testing so that everyone is on the same playing field as far as results. I personally love the energy star ratings labels on products and I do in fact go looking for the most efficient and cost effective options when I buy other products. I would like to see this same sort of data put on products so that consumers know what they are buying, how it is rated, and how it stacks up to other products on the market. To point to the fuel and say it is a fuel quality issue does have some merit, but I think these testing standards have to assume some degree of averages or even sample worst case. These companies that make these products then can use this as a marketing tool to show how advanced, or efficient their products are. Honestly many people burn wood because of availability of the wood in their locations, and because it is a cost effective. Since the consumer is already focused on cost they will pay attention to those products that get them the biggest bang for their buck both in up front costs and in the lifetime costs of using the products. Market forces can go a long way molding the products that are available if the data is there so that the customer can decide. While we hear the sky is falling all the time when it comes to new regulation and I am strongly opposed to the one size fits all regulation standards when you could just take the time to inform the customers of these products of the differences and they will move the market in a direction that makes sense. I like the Woodstock Ideal Steel product, but it is still out if reach for many people in terms of cost(and yes I know it is already on the low end of that product category cost). But for this sort of tech to reach average people the cost needs to come down. I also have reservations about the catalytic combusters long term reliability and the possible variablity due to problems that come up with the catalytic combusters. This is an ongoing maint issue, and an ongoing cost that adds complications to the product that quite likely will degrade product performance in the long run. Again costs and simplicity need to be in the center of focus. Now does that mean I wouldn't buy a catalytic stove? No it doesn't, but it does still give me reservations that make me look at the cleanest non-catalytic options first.

    If however the industry cannot see its way to developing a standard for testing and labelling then maybe we are forced to some degree of regulation in this regard so that customers are informed and the data is available to everyone. I hate to see it, but setting a testing and labelling standard may be required. I don't think you need to go as far as to set a standard on emmisions. You put the data out there for the customer to see and explain what it means and watch the market change. People are not stupid, if you give them the information and options they will make different choices than they do today. I know I am looking for the most efficient and reliable stove I can find right now which is how I found your blog post. Thanks for sharing this info as it helps me to understand some of the issues with this industry and why it is so hard to get good information.