OPINION
MIRABAL, Justice.
This case involves (1) claims under the DTPA against an upstream supplier of com[306]*306ponent parts,1 and (2) claims under the theory of strict products liability based on alleged design defects. The products involved are a hot water heater and a broken drain valve. The jury found in favor of the plaintiff homeowner,2 and judgment was entered accordingly. We affirm in part, and reverse and render in part.
On February 16, 1991, Corbitt and his wife, Elizabeth, returned to their home in the FM 1960 area after spending the day at the rodeo. Corbitt testified that Elizabeth “went in the house [then] she came out crying that something was drastically wrong. When I went in it looked like a Brazilian rain forest with water pouring from every electrical fixture, from every air conditioning vent and the amount of water coming over the bannisters almost looked like Niagara Falls.... ” Approximately eight inches of water covered the Corbitts’ floor. The Corbitts had bought the house new 10 years before, on February 27,1981.
It is uncontested that Corbitt’s damage resulted from failure of a drain valve attached to one of the water heaters in Cor-bitt’s attic. State Industries manufactured the water heater in 1980, and Apcom manufactured the drain valve. Apcom is a wholly owned subsidiary of State Industries. The two companies were sued as separate corporate entities, and there is no claim of alter ego involved in this ease.
In points of error three and four, State Industries and Apcom assert the evidence is factually insufficient to support the jury’s answers to the following questions:
Question No. 1
Was there a design defect in the drain valve at the time it left the possession of Apcom, Inc. that was a producing cause of the property damage in question?
Answer: Yes
Question No. 2
Was there a design defect in the water heater at the time it left the possession of State Industries, Inc. that was a producing cause of the property damage in question? Answer. Yes
With regard to each question, the jury received the following instructions:
A “design defect” is a condition of the product that renders it unreasonably dangerous as designed, taking into consideration the utility of the product and the risk involved in its use.
Negligent failure to discover or guard against a product defect is not a defense.3 “Producing Cause” means an efficient, exciting, or contributing cause that, in a natural sequence, produced the occurrence. There may be more than one cause of an occurrence, but if an act or omission of any person not a party to the suit was the sole cause of the occurrence, then no act, omission, or product of any party to the suit could have been a cause of the occurrence.
When reviewing a claim that the evidence is factually insufficient to support a jury finding, the court of appeals must first examine all of the evidence. Lofton v. Texas Brine Corp., 720 S.W.2d 804, 805 (Tex.1986); Hollander v. Capon, 853 S.W.2d 723, 726 (Tex.App. — Houston [1st Dist.] 1993, writ denied). After considering and weighing all of the evidence, the court of appeals should set aside the verdict only if the evidence is so weak or the finding is so against the great weight and preponderance of the evidence that it is clearly wrong and unjust. Cain v. Bain, 709 S.W.2d 175, 176 (Tex.1986); Hollander, 853 S.W.2d at 726. The appellate court cannot substitute its opinion for that of the trier of fact and determine that it would [307]*307reach a different conclusion. Hollander, 853 S.W.2d at 726.
When reversing on factual insufficiency grounds, the court of appeals must detail in its opinion the evidence relevant to the issue and clearly state why the jury’s finding is so against the great weight and preponderance of the evidence as to be manifestly unjust, why it shocks the conscience, or why it clearly demonstrates bias. Jaffe Aircraft Corp. v. Carr, 867 S.W.2d 27, 28 (Tex.1993); Cropper v. Caterpillar Tractor Co., 754 S.W.2d 646, 652-53 (Tex.1988). Further, the court of appeals should state in what regard the contrary evidence greatly outweighs the evidence in support of the verdict. Cropper, 754 S.W.2d at 652-53; Pool v. Ford Motor Co., 715 S.W.2d 629, 635 (Tex.1986). In reviewing great weight points, the supreme court has admonished the courts of appeals to be mindful that the preponderance of the evidence did not convince the jury. Herbert v. Herbert, 754 S.W.2d 141, 144 (Tex.1988). Reversal is only warranted where the great weight of the evidence supports a different answer. Id.
Defendants argue that there is insufficient evidence to support the jury’s findings that (1) the drain valve, and (2) the water heater, were “unreasonably dangerous as designed, taking into consideration the utility of the products and the risk involved in their use.” In Boatland of Houston, Inc. v. Bailey, the Texas Supreme Court discussed the strict liability standard of “defectiveness” as applied in design defect cases:
Whether a product was defectively designed requires a balancing by the jury of its utility against the likelihood of and gravity of injury from its use. The jury may consider many factors before deciding whether a product’s usefulness or desirability are outweighed by its risks. Their finding on defectiveness may be influenced by evidence of a safer design that would have prevented the injury.... Because defectiveness of the product in question is determined in relation to safer alternatives, the fact that its risks could be diminished easily or cheaply may greatly influence the outcome of the case. Whether a product was defectively designed must be judged against the technological context existing at the time of its manufacture.
609 S.W.2d 743, 745-46 (Tex.1980). With these guidelines in mind, we review the relevant evidence.
As relevant to the issue of design defect, there were four expert witnesses, two for each side. All of the experts agreed that all water heaters will eventually leak, usually from a hole caused by corrosion in the tank. Such a tank leak gradually gets bigger; it is not usually the type of leak that causes catastrophic flooding, like failure of a drain valve would. All the experts agreed that a drain valve should outlast the life of the water heater tank.
1. Corbitt’s expert witness, Roger Tate, a mechanical engineer at the University of Texas, testified that the drain valve was defectively designed because the plastic which was used (acetal copolymer) degrades in hot chlorinated water. He noted that chlorine is added to tap water in almost every part of the country. In Harris County, ordinary drinking water contains one to three parts of chlorine per million gallons of water. In his opinion, this drain valve was destroyed by a combination of hot water and chlorine.
According to Tate, testing done by Cela-nese4 around April 1978 (two years before the manufacture of the involved water heater), showed that acetal copolymer shrank 3.3% over a 12-month period in distilled water heated to 180 degrees Fahrenheit. Other Celanese tests showed that acetal copolymer shrank 3.3% over a six-month period in a 4-6% sodium hypoehloride solution (chlorine bleach) at room temperature. In October 1990, Stephen Sinker of the Engineering Plastics Division of Hoechst Celanese Corporation wrote an article for Modern Plastics Encyclopedia stating that chlorine concentrations aboye one part per million gallons of water will deteriorate acetal plastics.
[308]*308In Tate’s opinion, although glass fibers added to acetal copolymer by Apcom would make the plastic physically stronger, they would not protect the plastic from chemical degradation by chlorine.
Tate noted that the manual that came with the water heater says that “[ijnstallation of the water heater must be accomplished in such a manner that if the tank or any connections should leak, the flow of water will not cause damage to the area adjoining the water heater, or to lower floors of the structure. When such locations can’t be avoided, a suitable drain pan should be installed under the heater.” The manual says the drain pan should be one and one-half inches deep. In Tate’s opinion, such a drain pan would not afford any protection to a homeowner in the event the drain plug would fall off the water heater. In fact, the drain pan installed for Corbitt’s water heater exceeded the manual’s specifications — it was 10 inches deep — yet it did not provide any protection in this case when the drain valve failed. The drain pan is installed to catch normal leaks. In evaluating the performance of a product, one must consider the instruction manual as well as the product itself.
Tate further testified on cross-examination that Celanese promoted acetal copolymer for plumbing applications. A 1978 silicon acetal copolymer design manual “might have misled” State Industries or Apcom into believing that this was an appropriate material with which to manufacture drain valves. Tate noted that plastic is cheaper than brass.
2. Corbitt’s expert witness, Dr. Constantine Armeniades, a professor of Chemical Engineering at Rice University, testified that this drain valve failed catastrophically because it was made from acetal copolymer, which is subject to attack by chlorine or chlorine-producing agents in residential drinking water. Acetal copolymer is not an appropriate substance with which to manufacture drain valves, or to use in any plumbing applications. Apcom’s use of glass fibers to physically reinforce the drain valve may have actually increased the deterioration of the acetal copolymer by diffusing chlorine through the valve.
Dr. Armeniades knew as far back as the 1960s that acetal copolymer was subject to chemical attack by chlorine. He was involved in testing acetal copolymer in the 1960s, and the methods used were widely available to manufacturers of acetal copolymer, as well as to manufacturers of products made from acetal copolymer. Heat increases the rate of chlorine deterioration of acetal copolymer. In his opinion, this drain valve was defectively designed.
Traditionally, brass has been used for plumbing valves. Brass drain valves last more than 25 years, much longer than the life span of acetal copolymer valves. More recently, other plastics such as CPVC, chlorinated CPVC, and nylon are used for such applications as drain valves.
Further, according to Dr. Armeniades, the instruction manual gives the false impression that if a suitable location for installation of the water heater cannot be found, a drain pan will be adequate protection from water damage if the heater or plumbing fittings fail. The manual does not mention the danger of pressurized water escaping from the water heater if the heater or plumbing fittings fail. A catastrophic water loss due to the failure of a drain plug was foreseeable by Apcom and State Industries. Further, the installation of the water heater in the attic was foreseeable by Apcom and State Industries; builders install water heaters in attics routinely.
In evaluating the performance of a product, one should consider the manual as well as the product itself. A product is not defective today merely because technology has improved since the product’s manufacture. A manufacturer must consider the materials available at the time the product is designed and make the best possible choice.
Celanese recommended acetal copolymer for “plumbing applications” until the late 1980s. State Industries and Apcom were entitled to rely, at least in part, upon the representations by Celanese that acetal copo-lymer could be used in “plumbing applications.” Chemical and manufacturing data and literature in the late 1970s and early 1980s indicated that acetal copolymer was acceptable for “plumbing applications,” and [309]*309acetal copolymer was widely used during that period for “plumbing applications.” However, State Industries had available to it in 1978,1979 and 1980, a bulletin from Celanese regarding their acetal copolymer plastic. The bulletin showed that exposure of the product to chlorine (4% to 6% sodium hypo-chlorite) caused pitting in the plastic, and caused it to lose over three percent of its body weight. It also lost about three percent of its body weight when exposed to distilled water in ordinary room temperature. These test results should have put Apcom and State Industries on notice that they needed to do some specific testing to find out if their application was going to result in pitting, loss of weight due to heated water, or loss of weight due to exposure to chlorine. The tests would have needed to be very carefully controlled, and conducted over a period of years. Alternatively, the Celanese test results should have put Apcom and State Industries on notice that they should consider alternative materials.
Although he did not know whether Apcom boiled the acetal copolymer drain valves in tap water “for a number of years,” they should have. Manufacturers should check the failure rates of their products. Dr. Ar-meniades was aware of at least one other instance where a State Industries water heater drain plug had failed in the same way as Corbitt’s water heater, resulting in flooding of the home. This type of catastrophic drain plug failure is something that was foreseeable by Apcom and State Industries.
3. State Industries’ and Apcom’s expert witness, Robert Murphy, served as a senior project engineer for State Industries from 1968-1971, as chief engineer for Apcom from 1976-1992,5 and as a manufacturing engineer for Apcom at the time of trial. He testified that in 1970, the major brass manufacturer’s plant burned down and “just about shut all the water heater manufacturers down.” For about six years thereafter, water heater manufactures mainly used a combination steel drain valve. According to Murphy, it was during this time that Apcom experimented with acetal copolymer and field tested about 25,000 acetal copolymer drain valves. In Murphy’s opinion, drain valves made from acetal copolymer were better than brass valves because brass valves had to be machined after molding and tended to have imperfections that could not be seen with the naked eye; plastic valves, on the other hand, were ready to use after molding and were of more consistent quality. Brass drain valves tended to rust, causing the water heater’s anode rod to deteriorate more rapidly, which in turn meant less protection for the water heater tank.
Murphy testified that Apcom ran tests on the acetal copolymer drain valves, consisting of placing the valves in two tanks of water for up to two and one-half years; one tank had boiling water, and the other had 160 degree water with 40 pounds of pressure. However, the only documents Apcom was able to produce showed the acetal copolymer drain valves had been tested in boiling water for 10-minute periods of time; Apcom did not have documents showing that it tested the valves in boiling water for one year or longer periods of time.
Murphy testified the life expectancy for acetal copolymer drain valves is 15 years, while the life expectancy for a water heater tank is seven to 10 years. However, Apcom did not have any documents to show that the acetal copolymer drain valve would last at least the life of the water heater.
Murphy further testified that, because all water heaters leak, proper installation is “critical” in order to minimize damage from water. A water heater should not be installed in an area of a home where it is likely to cause damage if a leak should occur.
According to Murphy, acetal copolymer is stronger and lasts longer than other materials Apcom had tested. Celanese’s tests were not performed on a drain valve, but rather on flat samples of acetal copolymer; therefore, “you would need to extrapolate” from and “temper” the Celanese information that showed the samples pitted and lost body weight when immersed in the chlorine solution and boiling water.
[310]*310Murphy farther testified that he is not aware of any other Apcom acetal copolymer drain valves on water heaters that have failed during normal operation. Apcom has made over 13 million valves of this type.
4. Another expert witness for State Industries and Apcom was Timothy Dunn, a chemical engineer with Dunn Laboratories. He testified that, in this case, the threads of the drain valve indicated a partial leak had existed before the final rupture of the drain valve. The cause of the drain valve breaking off was ordinary tap water heated to around 160 degrees, and its effects over a nine or 10-year period on the drain plug.
With regard to the Celanese test results, Dunn testified that 3% loss of body weight to a sample piece of acetal copolymer immersed in a chlorine bleach solution is not excessive considering that Apcom did not intend to use this plastic for storing bleach solution. Even so, in Dunn’s opinion, today acetal copolymer is not an appropriate material for water heater drain valves. In fact, Apcom started using an alternate material in 1984 or ’85.
According to Dunn, brass also has some negatives: brass drain valves diminish anode rods in the tank; are less consistently of good quality than acetal copolymer drain valves; and are likely to become cross-threaded, which will cause a water heater to leak.
As far as Dunn is aware, this is the first time an acetal copolymer drain valve has failed in this manner. He has been told that, according to Celanese, acetal copolymer could still be safely used for water heater drain valves today if (1) the valves are properly designed, and (2) the valves are properly tested.
After hearing all the evidence, the jury concluded the drain valve and the water heater were defectively designed. The jury is the sole judge of the credibility of the witnesses and the weight of their testimony; it may believe all, part, or none of controverted expert testimony. See Gray v. Floyd, 783 S.W.2d 214, 216 (Tex.App. — Houston [1st Dist.] 1990, no writ).
From the evidence, the jury could have concluded that in 1980, when the involved drain valve and water heater were manufactured, Apcom and State Industries were aware, or reasonably should have been aware, that acetal copolymer was adversely affected by a combination of chlorine and hot water. Therefore, they should have properly tested the acetal copolymer drain valves for an extended period of time, yet they had no documents showing they had in fact conducted such extended testing. If the jury believed that Apcom tested the acetal copolymer drain valves in boiling water for only 10-minute periods of time, as their supporting documents showed, the jury could have concluded the testing was woefully inadequate in light of the information available from the prior Celanese tests.
There was evidence that Apcom and State Industries should have foreseen that their hot water heaters would be installed in attics, and thus there was the risk of substantial flooding if the drain valve would fail. Even if the jury believed the water heater should not have been installed in the attic, this would not have relieved State Industries and Ap-com from responsibility in light of the jury instruction on “producing cause.”
The credibility of Apcom’s and State Industries’ witnesses may have been negatively affected in the eyes of the jury by their repeated testimony that this is the only case involving such a failure of their acetal copo-lymer drain valve. Dr. Armeniades, the professor of Chemical Engineering at Rice University, had himself reviewed another case involving a State Industries water heater with an acetal copolymer drain valve that had failed in the same way as Corbitt’s. The jury may not have believed much of the testimony of the witness for Apcom and State Industries.
The only evidence to the effect that brass was in short supply related to the early 1970s; there was no evidence brass was in short supply at the time Corbitt’s water heater was manufactured in 1980. There was ample evidence from which the jury could have concluded that brass was a safer, well-tested and better alternative at the time of manufacture. Brass, though more expensive, had been used for many years in the manu[311]*311facture of drain valves. Further, there was evidence that a combination steel drain valve was commonly used between 1970 and 1976, and there is no indication in the record that such a valve could not have been safely used on Corbitt’s water heater in 1980. Apcom itself stopped using acetal copolymer for water heater drain valves in about 1984. Ap-com’s and State Industries’ own witness admitted that today acetal copolymer is not an appropriate material for water heater drain valves.
Based on a review of all the evidence, we cannot say the jury’s answers to jury questions numbers one and two are so against the great weight and preponderance of the evidence as to be manifestly unjust. As noted in the Boatland case, “[bjecause defectiveness of the product in question is determined in relation to safer alternatives, the fact that its risks could be diminished easily or cheaply may greatly influence the outcome of the case.” 609 S.W.2d at 746. There was clear evidence that acetal copolymer was not an appropriate material for water heater drain valves; that State Industries and Apcom had notice before 1980 that acetal copolymer may not be appropriate for such use; and that there existed safer, available, and cost-effective alternatives at the time of the manufacture of the drain valve. The jury could have concluded from the evidence that Apcom did not conduct the types of tests reasonably necessary to insure the safety of its drain valves, opting to use the less expensive material for its drain valves irrespective of the unnecessary risks involved.
We overrule points of error three and four.
In point of error one, appellants assert the evidence is legally and factually insufficient to support the jury’s finding that Apcom had violated the Deceptive Trade Practices Act6 through unconscionable action that was a producing cause of damages to Corbitt. In light of the recent Texas Supreme Court opinion in Amstadt v. United States Brass Corp., 919 S.W.2d 644 (Tex.1996), we agree.
It is uncontroverted that Apcom was an “upstream supplier of a component part” of the water heater. Apcom manufactured the drain valve that was supplied to State Industries, and State Industries manufactured the water heater using the drain valve as a component part. The water heater was installed in the home that Corbitt later bought. Because Apcom’s alleged DTPA violation did not occur “in connection with [Corbitt’s] purchase of his home,” Corbitt cannot recover from Apcom under the DTPA, although Cor-bitt may recover under other theories. Amstadt, 919 S.W.2d at 646, 648.
Accordingly, we sustain point of error one.
In point of error two, appellants assert the trial court abused its discretion when it allowed Corbitt to file a trial amendment relevant to his DTPA cause of action. In light of our ruling on point of error one, this point of error is moot. Therefore, we do not reach its merits.
In a cross-point, Corbitt complains that the trial court erred in overruling Corbitt’s objections to the testimony of appellants’ expert witnesses. In light of our rulings on points of error one, three and four, the cross-point has become moot, and we, therefore, do not consider its merits.
We reverse the portion of the judgment against Apcom that is based on the DTPA claim, and render judgment that Corbitt and Standard Fire Insurance Co. are not entitled to recover from Apcom the “additional damages” and attorneys’ fees set out in the judgment. In all other respects, we affirm the trial court’s judgment.