George Lanoue, et al., Appellant,

This opinion will be unpublished and
may not be cited except as provided by
Minn. Stat. § 480A.08, subd. 3 (2016).

STATE OF MINNESOTA
IN COURT OF APPEALS
A17-1963

George Lanoue, et al.,
Appellant,

vs.

Better Water Industries, Inc.,
Respondent,

Thein Well Company,
Respondent.

Filed June 4, 2018
Reversed and remanded
Bratvold, Judge

Kandiyohi County District Court
File No. 34-CV-16-395

William D. Mahler, Will Mahler Law Firm, Rochester, Minnesota (for appellants)

Daniel A. Haws, Krista J. Pezewski, HKM, P.A., St. Paul, Minnesota (for respondent Better
Water Industries, Inc.)

Christopher A. Wills, Gordon H. Hansmeier, Rajkowski Hansmeier Ltd., St. Cloud,
Minnesota (for respondent Thein Well Company)

Considered and decided by Florey, Presiding Judge; Larkin, Judge; and Bratvold,
Judge.
2

U N P U B L I S H E D O P I N I O N
BRATVOLD, Judge
In an appeal from summary judgment entered in favor of respondents Better Water
Industries, Inc. and Thein Well Company in a negligence and products-liability action,
appellants George and Peggy Lanoue challenge the district court’s determination that one
aspect of their expert’s evidence lacked foundational reliability. Appellants do not dispute
that expert testimony is required to establish that damage to their dairy goat herd was
caused by exposure to excessive and harmful chlorine resulting from respondents’
negligence and the defective design of their water-treatment system. Because appellants’
expert opinion had sufficient foundational reliability f or his estimate of the amount of
chlorine in the goats’ drinking water, the district court erred in granting summary judgment.
Thus, we reverse and remand.
FACTS
Appellants George and Peggy Lanoue have farmed livestock for most of their lives.
In 2014, the Lanoues transitione d from raising hogs to goats. By the end of 2014, the
Lanoues had purchased approximately 400 dairy goats from several sellers. The Lanoues
became concerned about water quality. Their farm relied on a well and their water
contained methane and iron, which affected the water quality . The Lanoues sought to
improve water quality so the goats would drink more water and increase milk production.
Around August 2014, the Lanoues contacted respondent Thein about improving
water quality. Thein sold one water-treatment system to the Lanoues, but they were not
3

satisfied after installation because it did not remove enough methane and iron from the
water.
In February 2015, Thein submitted a sample of the Lanoues’ water to Chandler
Systems, Inc. (CSI) for analysis. CSI recommended a liquid chlorine-based water treatment
system. Preferred Pump, Inc., Thein’s distributor, purchased a Sentry I Open Air System
(the Sentry system) from respondent Better Water Industries (BWI), sold it to Thein, and
Thein then sold it to the Lanoues. On May 18, 2015, Thein installed the Sentry system at
the Lanoues’ farm. Before installation of the Sentry system, about fifteen goats had died
and the herd experienced an outbreak of coccidiosis, a protozoan parasite.
The Sentry system is designed to oxidize and sanitize the water, in part, by opening
the water lines to the air, which also releases and dissipates the methane. Because opening
the water lines to the air increases the risk of contamination, Thein recommended chlorine
to sanitize the water. Chlorine treatment also removes iron, manganese, algae bacteria, and
tannins from the water. In the Sentry system, chlorine “pellets” drop into the tank from a
“rotating gear.” The amount of chlorine is controlled by the gear setting, which ranges from
A-L. For example, setting G delivers a chlorine pellet every two minutes and six seconds.
The pellets drop through a tube into a basket ; water passes through the basket when the
water pump is running, dissolving the pellets.
4

Chlorine demand, as well as the desired chlorine residual, indicate the proper gear-
setting.1 The Sentry system manual describes how to calculate the chlorine demand and
recommends a chlorine residual of 1.5 to 2.5 parts per million (ppm). The manual also
recommends that, after chlorine treatment, water should be filtered through a carbon filter
or an Ag filter. A carbon filter will eliminate chlorine residual. BWI does not include a
filter with the Sentry system, and it does not make or sell them. Thein installed a carbon
filter on the Lanoues’ system.
The chlorine tank is depicted in the following diagram:

Thein installed a second tank next to the chlorine tank to allow a greater volume of
water to be treated. The water flowed from the first tank into the second tank, which

1 The chlorine demand is created by iron, manganese, algae bacteria, and tannins, and varies
with the amount of these elements present in the water. Any chlorine that is not “used up”
by these elements is chlorine residual.

5

contained a pump. Next, the water passed through a carbon filter and a water softener, and
was delivered to the barn for the goats to drink. The Sentry system layout is depicted in the
following diagram:

After the Sentry system was installed it “flushed nonstop;” the Lanoues shut it off
and informed Thein of the problem . Thein decided to replace the “brains” of the system,
and the system was restarted in June 2015. Peggy testified in her deposition that she noticed
that some of the goats suffered “diarrhea and some bloating.” The Lanoues again shut off
the system on June 23, 2015, because they believed it was causing health problems in their
herd by providing too much chlorine. A total of 7 goats—3 kids and 4 does— died shortly
after the Lanoues started using the Sentry system.
The Sentry system was not used again until January 6, 2016, when it was restarted.
At that time, Mike Rivers, a Thein employee, replaced the carbon filter with an Ag filter,
which does not remove chlorine from water. Rivers tested the chlorine residual using “little
litmus paper test strips” on at least three occasions in January 2016, and testified in his
deposition that he found no chlorine. It is undisputed that Rivers did not “follow the testing

6

procedure” for using the testing strips. 2 On January 20, Rivers increased the amount of
chlorine being delivered to the tank by changing the gear setting.
After the amount of chlorine was increased, the Lanoues reported on January 22 that
the goats became progressively more ill based on the following symptoms: “[d]iarrhea,
bloating, definitely off feed, cutting back on milk.” On January 26, Rivers added a carbon
filter and tested the water. Rivers testified that he again tested the water with strips and
found little to no chlorine residual at this time. On January 27, Peggy tested the water using
the same litmus strips, and found that the chlorine residual level was “3 to 5 parts per
million total chlorine and 1 part per million free chlorine.” As a result, she decreased the
amount of chlorine by altering the gear setting.
On January 28, the goats were vomiting, foaming at the mouth, suffering from
diarrhea, and staggering, and a veterinarian was called to treat them. The Lanoues turned
off the Sentry system, and have not used it since that day. In the following weeks, the
Lanoues reported that their goat herd suffered from bloating, acidosis, diarrhea, vomiting,
pain during urination, loss of bladder control, polio, abortions, and thinning of hair. In
February 2016, Dr. Ryan Tollefsrud, a veterinarian, euthanized one goat, sent samples to

2 The district court stated in its “findings of fact” that Rivers did not use the correct type of
“test strips,” because the test strips that Rivers used were instead “designed for use in pools
and spas.” The district court also found that Rivers did not follow the manufacturer’s
testing procedure because “the strips were dipped in the water only once, rather than being
dipped twice as the instructions directed.” We note that on a summary judgment motion, it
is inappropriate for the district court to make factual determinations. See State ex rel. Hatch
v. Allina Health Sys., 679 N.W.2d 400, 406 (Minn. App. 2004). Because r espondents
conceded at oral argument that this fact was not disputed, we include it in our summary of
the record.
7

the University of Minnesota diagnostics lab, and the lab reported that the goat suffered
from laminar neuronal necrosis, which can be caused by “ischemia, heavy metal toxicity
(e.g., lead) and by nutritional factors such as thiamine deficiency or excess sulfur
compounds in the diet.”
In July 2016, the Lanoues sued Thein and BWI and claimed that the Sentry system
added harmful levels of chlorine to their water, causing damage to their goats. The Lanoues
alleged that Thein was negligent in the sale, installation, and servicing of the Sentry system.
The Lanoues also alleged that BWI was negligent in the designing the Sentry system,
preparing the instruction manual, and in training Thein on how to install and service the
Sentry system. Finally, they alleged that BWI and Thein breached express and implied
warranties.
The Lanoues disclosed four expert witnesses: Peter Cartwright, a chemical engineer
and expert on water-treatment systems; Dr. Tollefsrud, the treating veterinarian; Dr. Sandra
Solaiman and Carol Delaney, each of whom opined on the goats’ health problems and the
Lanoues’ farm management.
BWI filed several motions, including a motion for summary judgment. BWI argued
that summary judgment was appropriate because the Lanoues’ expert opinion s lacked
foundational reliability, and without expert evidence, the Lanoues were unable to establish
that chlorine caused damage to their goat herd. In their response, the Lanoues disagreed,
but also withdrew their claim for breach of express warranty . To show that their expert
evidence was sufficient to survive summary judgment, the Lanoues submitted several
affidavits from their disclosed experts.
8

On October 31, 2017, the district court granted BWI’s motion for summary
judgment based on its review of the affidavit of Cartwright, the water-treatment expert.
The district court recognized that Cartwright offered five opinions, but concluded that the
Lanoues’ case “hinges” on proof that “an unhealthy chlorine residual was present in the
actual drinking water of [the Lanoues’] goats.” Because Cartwright’s expert affidavit
lacked foundational reliability , and expert testimony was required, the district court
concluded, “[t]here is no reliable evidence that a chlorine residual was present in [the
Lanoues’] goats’ water.” The court did not consider the Lanoues’ other expert evidence
offered to establish the impact of chlorine exposure on goats, nor did it decide BWI’s other
motions.
Finally, although Thein did not file its own motion for summary judgment, or timely
join BWI’s motion, the district court granted summary judgment to Thein because the
Lanoues had a meaningful opportunity to respond. 3 Thus, the district court granted
summary judgment in favor of BWI and Thein and dismissed the Lanoues’ claims with
prejudice. This appeal follows.
D E C I S I O N
A motion for summary judgment shall be granted “if the pleadings, depositions,
answers to interrogatories, and admissions on file, together with the affidavits, if any, show
that there is no genuine issue as to any material fact and that either party is entitled to a

3 At the hearing on BWI’s motions, Thein requested that “if BWI’s motion for summary
judgment is granted, we would also be granted judgment based on the Court’s own
motion.” On appeal, the Lanoues do not challenge Thein’s failure to timely join the motion
for summary judgment.
9

judgment as a matter of law.” Minn. R. Civ. P. 56.03. On appeal from summary judgment,
this court reviews de novo whether there is a genuine issue of material fact and whether
the district court erred when it applied the law. STAR Ctrs., Inc. v. Faegre & Benson, L.L.P.,
644 N.W.2d 72, 76-77 (Minn. 2002). We “view the evidence in the light most favorable to
the party against whom summary judgment was granted.” Id.
I. The district court erred in granting summary judgment based on its conclusion
that Cartwright’s expert opinion lacked foundational reliability.

Causation is an essential element of each of the Lanoues’ negligence and warranty
claims. Driscoll v. Standard Hardware, Inc., 785 N.W.2d 805, 816 (Minn. App. 2010)
(causation is an essential element of an implied warranty claim), review denied (Minn.
Sept. 29, 2010); Gilmore v. Walgreen Co., 759 N.W.2d 433, 435 (Minn. App. 2009) (same
regarding negligence), review denied (Minn. Mar. 31, 2009); J & W Enters., Inc. v. Econ.
Sales, Inc., 486 N.W.2d 179, 181 (Minn. App. 1992) (same regarding negligent design).
The Lanoues offer ed expert testimony to meet their burden in response to BWI’s
summary-judgment motion. Because what is a harmful chlorine level and what caused the
damage to the Lanoues’ herd is outside common knowledge, expert testimony is required.
Gross v. Victoria Station Farms, Inc., 578 N.W.2d 757, 762 (Minn. 1998) (“Expert opinion
is required to prove causation if the issue is outside the realm of common knowledge.”).
The Lanoues submitted Cartwright’s affidavit to prove the chlorine level or concentration
in the goats’ drinking water, among other issues. The district court rejected Cartwright’s
opinion on chlorine concentration as lacking foundational reliability. The Lanoues argue
that the district court erred because Cartwright’s estimates of the chlorine level in the goats’
10

drinking water relied on the Sentry system’s specifications (i.e., the tank volume),
independent tests of water composition to determine chlorine demand (e.g., iron
concentration), and the chlorine content of the BWI pellets, as described in deposition
testimony and in the manufacturer’s information.
We review the district court’s evidentiary rulings, including rulings on foundational
reliability, for an abuse of discretion. State v. Loving, 775 N.W.2d 872, 877 (Minn. 2009);
Jacobson v. $55,900 in U.S. Currency, 728 N.W.2d 510, 525 (Minn. 2007) (stating that a
district court’s determinations regarding foundation for expert witness testimony will not
be reversed absent abuse of discretion). Expert testimony is admissible under Minnesota
Rule of Evidence 702 if, (1) the witness qualifies as an expert; (2) the expert’s opinion has
foundational reliability; and (3) the expert testimony is helpful to the trier of fact. Doe v.
Archdiocese of St. Paul, 817 N.W.2d 150, 164 (Minn. 2012). If the testimon y involves a
“novel scientific theory,” it also must satisfy the Frye-Mack standard. Id.
Here, the district court limited its analysis to the second element, foundational
reliability, which required the court to: “(1) analyze the proffered testimony in light of the
purpose for which it is being offered, (2) consider the underlying reliability, consistency,
and accuracy of the subject about which the expert is testifying, and (3) determine if the
proffered evidence is reliable.” Pfeiffer ex rel. Pfeiffer v. Allina Health Sys., 851 N.W.2d
626, 638 (Minn. App. 2014), review denied (Minn. Oct. 14, 2014).
In his affidavit and report, Cartwright, a chemical engineer with 43 years of
experience in water treatment, estimated the chlorine concentration in the goat drinking
water by analyzing the volume of water in the Sentry system tanks, the chemical
11

composition of the water to determine chlorine demand, and the amount of chlorine added
to the water based on the number of chlorine pellets dropped into the water, depending on
several different gear settings. According to his affidavit, Cartwright estimated that when
the gear was at a G setting, the chlorine concentration in the goat s’ drinking water would
have been between 7.2 and 7.9 ppm; at the K setting, the chlorine concentration would
have been between 21.9 and 25.6 ppm; at the intermediate setting between K and L, the
chlorine concentration would have been between 35.0 and 38.7 ppm. Each of the settings
that Cartwright considered came from deposition testimony stating the setting was used in
the Lanoues’ Sentry system.
Cartwright’s affidavit provided several “[o]pinions . . . based on a reasonable degree
of scientific certainty.” First, he opined that there was “no need for any chlorine whatsoever
to be in the” goats’ water. Second, he opined that the Sentry system pellet feeder was set
incorrectly resulting in “excessive and harmful level of chlorine in the water fed to the
goats.”4 Cartwright concluded that the Sentry system treatment of the goat water on the
Lanoues’ farm resulted in chlorine concentrations of “26.5 ppm or over six times higher
than the EPA limit for human consumption,” and far higher than BWI’s recommended
residual, which is between 1.5 and 2.5 ppm, as stated in the Sentry system manual.
The district court determined that Cartwright’s opinion on chlorine concentration
lacked foundational reliability because: Cartwright’s estimates were not reliable to

4 Cartwright also opined that Thein’s failure to test the water “after the initial installation,”
to determine whether the goat water “contained excessive levels of chlorine,” was negligent
and that BWI’s decision to “increase the level of chlorine” in January 2016, without
reliably testing the chlorine concentration in the goats’ drinking water, was also negligent.
12

establish the “actual level” of chlorine concentration in the goats’ water; Cartwright failed
to reliably establish the dissolution rate of the chlorine pellets, which is affected by several
variables including chlorine demand; and Cartwright failed to calculate the “amount of
chlorine” that a carbon filter “might have removed from the water.” We disagree with the
district court’s view of Cartwright’s affidavit. Because Cartwright’s affidavit established
foundational reliability for the data he used to reach his estimate of the chlorine level in the
goat herd’s drinking water, the district court erred in requiring evidence of the “actual level
of chlorine” in the goats’ drinking water.
An estimate by an expert has foundational reliability , as long as it is based on data
that has sufficient evidentiary support. See State v. Myers, 359 N.W.2d 604, 610-11 (Minn.
1984) (stating that the “cause of many physical and emotional ailments” that “cannot be
demonstrated to an absolute certainty” are “nevertheless, the subject of expert testimony”).
Expert testimony that offers estimates may not be excluded as conjectural simply because
it is not absolutely certain. Ingram v. Syverson, 674 N.W.2d 233, 236 (Minn. App. 2004)
(stating that an expert witness “is not required to speak with such confidence as to exclude
all doubts in his mind, but may render his testimony in the form of an estimate of opinion,
couched in expressions that fall short of absolute conviction of accuracy ” (quotation
omitted)), review denied (Minn. Apr. 20, 2004). Additionally, expert opinion about
estimates based on a reliable method may not be excluded as speculation. See Loving, 775
N.W.2d at 878 (rejecting defendant’s foundational reliability challenge where the expert
adequately “followed approved procedures” and both parties agreed that method was a
reliable means for determining gunshot residue). The weight and credibility to be given to
13

expert opinions should be “decided by a finder-of -fact and not by the district court.”
Pfeiffer, 851 N.W.2d at 639 (“The reliability of appellant’s expert opinion testimony with
regard to causation goes to the relevant weight of that testimony rather than to its
admissibility.” (quotation omitted)).
To estimate the chlorine concentration of the goats’ drinking water, Cartwrigh t
followed the method espoused in BWI’s manual for the Sentry system: that chlorine
residual may be calculated by subtracting the chlorine demand from the amount of chlorine
added to the water. Cartwright determined chlorine demand with reference to tests of the
Lanoues’ water that were performed at Thein’s request in 2015 and 2016 by National
Testing Laboratories, Ltd., Dairyland Labs, and CSI. After considering the known volume
of the water in the two Sentry tanks (based on BWI’s specifications) and the amount of
chlorine dropped in the water, Cartwright estimated the chlorine concentration of the goats’
drinking water.
The district court was understandably troubled because Cartwright stated in his
affidavit that the “actual concentrations in the goats’ drinking water will never be known,
and can only be estimated.” But Cartwright’s estimates of the chlorine concentrate in the
goats’ water were based on known data, the foundation for which respondents did not
challenge. Because Cartwright provided sufficient foundation for the factual data he used
to formulate his estimates, the district court erred when it concluded that the Lanoues
offered “no reliable evidence that a chlorine residual was present” in the goats’ water.
To be sure, BWI may test Cartwright’s opinions through cross-examination and the
presentation of their own expert witnesses. But these matters go to the weight and
14

credibility of the expert opinion and not its admissibility. In fact, each of the district court’s
other criticisms of Cartwright’s opinions identifies likely topics for attack by the
respondents.
For example, the district court also determined that Cartwright’s estimates of
chlorine demand are not reliable because Cartwright did not consider elements other than
iron before reaching his demand calculations. Also, the district court found that Cartwright
admitted that the amount of iron varied greatly, and any attempt to calculate the chlorine
demand would “basically be a guess.” Cartwright nonetheless estimated chlorine demand
based on data from water tests performed by National Testing Laboratories, Ltd., Dairyland
Labs, and CSI at Thein’s request. Finally, Cartwright’s affidavit opined that, other than
iron, “[a]ll other compounds reflected in the water sample would have no effect on chlorine
demand.”
The district court also criticized Cartwright for failing to reliably establish the
chlorine concentration of the water after it exited the Sentry system. But any difference
between the chlorine concentration of the water in the tanks and in the goats’ trough goes
to the weight or credibility of Cartwright’s opinions, not to their admissibility. Ultimately,
the weight and credibility to be given to Cartwright’s opinion should be decided by a
finder-of-fact and not by the district court. Pfeiffer, 851 N.W.2d at 639.
Additionally, the district court determined, and BWI now argues, that Cartwright
failed to consider the amount of chlorine that a carbon filter might have removed from the
water. Several factual allegations regarding use of a carbon filter are contested. For
example, the record shows that a carbon filter was used while the Sentry system operated
15

in June 2015, even though the goat herd became sick. Cartwright’s affidavit alleges that it
is unclear whether this filter was “properly sized;” and Cartwright opines that “it is
unknown if the carbon filtration would have been sufficient” to filter chlorine from the
water. The record also shows that Thein removed the carbon filter from the Sentry system
when it was operated in January 2016. From this, the Lanoues infer that the amount of
chlorine that a carbon filter might have removed is irrelevant. Because the evidence
regarding whether use of a carbon filter effectively eliminated chlorine allows “reasonable
persons to draw different conclusions,” we conclude that there is a genuine issue of material
fact on this issue. DLH, Inc. v. Russ, 566 N.W.2d 60, 71 (Minn. 1997).
Finally, the district court criticized other aspects of Cartwright’s expert opinion,
such as the demonstrations that were described in Cartwright’s affidavit: (1) the restart of
the Sentry to “simulate normal usage of the system;” (2) the “bucket” test; and (3) a yellow
ball demonstration.5 Because the district court criticized these demonstrations, but did not
rely on its criticisms to reach its summary -judgment decision, and Cartwright did not rely

5 First, in April 2017, a Cartwright representative restarted the Sentry system to “simulate
normal usage of the system.” For this test, Cartwright used a K gear setting, filled the first
tank with water in the evening, then removed 400 gallons the next morning, and removed
100 gallons in the middle of the day. Cartwright then measured the chlorine concentration
at two depths of the tank; he found between 182 and 200 ppm within one inch of the bottom
of the first tank and between 0.16 and 0.21 ppm within the top one inch of the top of the
tank. In this simulation, Cartwright did not measure the amount of chlorine in water
emptied from the tank. Next, the Lanoues dissolved a chlorine pellet in a bucket under a
garden hose, and measured the resulting chlorine concentration. The amount of the chlorine
residual increased with time, and, when measured 80 minutes after the bucket was filled
“[t]he Ecolab drop test showed 16 ppm and the [litmus] strip test showed 3-5 ppm.” Finally,
the Lanoues completed a “demonstration” in which they dropped a ball down the Sentry
tube, and the ball fell to the bottom of the tank.
16

on these demonstrations to reach his estimates, we do not consider them. We merely note
that experiments are admissible if they are conducted in “circu mstances substantially
similar to those existing in the case at issue.” See State v. Ritt, 599 N.W.2d 802, 812 (Minn.
1999) (quotation omitted). But substantial similarity is not required when an expert
employs an approved method that the parties agree is a reliable and relevant demonstration.
See Loving, 775 N.W.2d at 878. The admissibility of these demonstrations does not affect
our decision that Cartwright’s chlorine-concentration estimates have foundational
reliability.
We conclude that Cartwright’s af fidavit demonstrates sufficient foundational
reliability for his expert opinion that estimated the chlorine concentration of the goats’
drinking water. Thus, the district court erred in granting summary judgment.
While the Lanoues also offered expert veterinarian evidence regarding what caused
the damage to their herd, we do not consider that evidence because it was not considered
by the district court to reach its summary -judgment decision. Because the respondents
raised separate challenges regarding the veterinarian evidence and these challenges were
not decided by the district court, we remand for further proceedings consistent with this
opinion.
Reversed and remanded.