George

Thank you, for your questions following your “Plastic Bag” experiment with my Low-Level Monitor and your Nighthawk C O Detector.

I am REALLY Happy to respond to these Questions because, hopefully, you can pass this INFO on in your Future Presentations to help “Explain” a number of Issues that UL, ULC, CSA, and the C O Alarm Manufacturers need to expend a WHOLE LOT MORE EFFORT in trying to “Clear-Up, and “Clean-Up”, for the average Consumer.

First on the “Elapsed Times” of Activations of my Monitor and the Nighthawk Detector, are dependent upon two factors that I don’t presently have available to me. These are: from the “Start” of your “Test”, and the “TIMING” thereof, ... how long, [how many Seconds] from the Time you opened the Valve on the 100 ppm Calibration Gas, was it until the C O Concentration inside the Plastic bag reached the Monitors “Set Points” of 10, 25, and 50 PPM, and then did you EVER achieve 100 ppm C O Concentration inside the plastic bag ?

My answer to your question #2 will “Explain” the “Main reason for the difference in the “Elapsed Time” Readings and Activations.

1. My Monitor was Intentionally Designed to be a “Special” Low-Level “Health” type Monitor, ..... not to meet the “Commonly Considered” ,”SERIOUS” Levels of C O Exposure in the Area of 10% COHb, or more. I suggest that you read the Very First Paragraph of my Owner’s Manual.

The UL / ULC and CSA Standards are written in a fashion to permit the use of a Wider Variety of Less Accurate Sensors and Microcontrollers, enabling the Production of inexpensive, “Listed” products for the “Retail” Marketplace. Because this “lack” of Strict Quality Control resulted in “HUGE” False / Nuisance Alarm Problems, and VERY inaccurate Digital Display Readings, [which is a TRUE reflection of the Poor Accuracy of the Sensor / Control functions], ... Tremendous Pressure was brought to bear on UL / ULC / CSA to REDO Calibration and Display Features. As a Result, NONE of the above Standards will PERMIT ..... ANY Alarm to Display less than 30 PPM on the Digital Display. [In fact, there was MUCH discussion about Eliminating the Digital Display Feature, ... TOTALLY.] Also, the Current Versions of UL / ULC / CSA, .....

REQUIRE that the Alarms IGNORE 70 PPM, for AT LEAST 60 Minutes, ... and “Could” Ignore 70 PPM for 239 Minutes, ... and STILL BE LISTED.

While these Levels of CO ...”MAY” be OKAY for “HEALTHY” Individuals, ..... I CHOOSE ..... to Protect EVERYONE ! ! !

[ I do want to mention to ALL, that UL HAS NOW OFFERED to “LIST” my Monitor as a “Special USE” product, under UL-2075, ..... which I Intend to do as soon as I complete two NEW Additional Models.]

Question #2. There are Several Factors involved in the “Difference” in the C O PPM Displayed on the Units in your Plastic Bag “Test”. First, I need to mention that your photograph is Clear Enough to see that on the FRONT of the Nighthawk Unit, ... it CLEARLY says: “DETECTOR”. This tells me that this Unit was Manufactured before Oct. 01. 1998, [as of that date, UL, ULC, & CSA REQUIRED that ALL C O Units be called “Alarms”, ... Not “Detectors”.] and therefore, that it contains a MOS, [ Metal Oxide, [usually Tin,] Semi-Conductor, Solid State Sensor, [most likely a Figaro Model 203, or it could be a FIS C O Sensor], because After Oct. 01, 1998, Nighthawk did NOT use MOS Sensors.

Therefore, this also tells me that this is a Product that takes a Sample of the “AIR”, and CO Present, only ONCE, every 2 ½ Minutes, ... and because of the “Actual Requirement” that FINALLY started being “ENFORCED” to employ “Time Weighted Averaging” by this 1995 UL-2034 Requirement, [ as was not true under the 1992 Version of UL-2034], this Model of Nighthawk C O Detector, as per UL Requirement, would have HAD to DELAY Alarming until at LEAST 16 Minutes had Elapsed before sounding an Alarm at 100 PPM, and it “Could” have DELAYED Alarming for up to 89 Minutes. The MAIN Reason that the Digital Display was SO LATE in displaying a C O Reading, [compared to my Monitor] was that the C O 100 PPM Calibration Gas was “FILLING” the plastic bag Very Quickly, ... and since my Monitor takes a NEW C O Reading EVERY TEN SECONDS, it could have actually UPDATED the Digital Display 15 TIMES, ... by the “Time” that the Nighthawk unit took its SECOND C O Reading.

I probably should also mention that the Nighthawk’s Owner’s Manual of that Model C O Detector [those made before Oct. 01,1998, although MANY, MANY, MANY were actually SOLD AFTER that date] stated:

The digital reading tolerances are listed below:

Ambient: 80 F [+/- 10 F], atmospheric pressure +/- 10%, 40% +/- 3% relative humidity.

Sensor: Thoroughly conditioned [minimum 7-day warm-up period] stabilized, and non-contaminated.

Tolerance

Reading [of displayed reading]

_____________________________________________________________________

0-999 ppm + / - 40% + 15 ppm

You asked if I agreed that the “0” C O Reading is “Mis-Leading”, ... I would prefer to point out that ANY C O Detector, or Alarm, that was manufactured, or PURCHASED, more than TWO YEARS AGO, .................. should be REPLACED AT ONCE.

Frankly, other than my own Monitor, I would not RECOMMEND, to my 4 Children, my 9 Grandchildren, or my 6 Great-Grandchildren... ANY C O ALARM, ... NOT LISTED to CSA-6.19 - 01, ................... UNTIL, ..... UL starts to require “Time of Manufacture”. And “Lifetime” Reliability Testing.

ALL Sensors, ... of ALL Technologies, ... are subject to “Cross Sensitivity” from other gases, ...”IF”, ... they are NOT FILTERED to PREVENT IT, ... and Even Then, ... it is NEVER PERFECT ! ! !

Some Technologies, ... more so than others, ... and the “Percentages” of “Filtering Success”, also varies GREATLY. This “Variation” is an even MORE SERIOUS PROBLEM when you are dealing with a Sensor , and / or , “Controller Chips”, that do NOT Properly Compensate for changes in Temperature and Humidity.

My Personal experience has Proven to me, over and over again, that ALL of the SENSOR Technologies “CAN” be made into more accurate, reliable, and repeatable Detectors / Alarms ... “IF”, ... the Manufacturer would only spend the “Time and “Money” to make it so.

ALL “Un-Filtered Sensors”, ... “Could be” ... effected by the Your ENTIRE LIST, , ... over “Time”. Once again, I would refer you to GRI “Testing”, and Cross Sensitivity “Testing” offered by Sensor Manufacturers of the Unfiltered, versus, Filtered Sensors.

”My” Sensor, for example, has several gases to which it would “Normally” have a Cross-Sensitivity ratio of about 16%, that, after “Proper Filtering” to Compensate, ... is “Reduced” to .005 %. In an Electrochemical Sensor, the “Filtering” Process is done via a “Series” of layers of membranes, [referred to as the “Stacking” Process], it is the Number, and “Type” of filter membranes used in this “Stacking” process that determines the “Degree” of Cross-sensitivity Elimination that is achieved.

None of your “List”, would Cause my Monitor to go into a “False Alarm”, in “Normally Expected” Volumes in the “Residential “ environment.

At “Higher” than “Expected” Levels, the greatest effects would be from Hydrogen and Toluene, [which is really the potential problem substance found in “Super Glue”, etc.].

If these, or ANY OTHER GAS / SUBSTANCE, is in the AIR in your home in sufficient Volume to “Cause an Alarm”, .... it would be enough to indicate a potential SERIOUS “HEALTH” Concern, .... that NEEDS to be CORRECTED IMMEDIATELY ! ! !

Once again, I refer you to the List of “Additional Gases” ... “Proposed” to be included in the “Cross Sensitivity” Testing by GRI, and Paul Clifford, of Mosaic Industries, Inc., which was VOTED DOWN by ALL of the C O Alarm Manufacturers that sit on the UL-2034 ..... and ALL, except one, that sat on the CSA-6.19 C O Technical Committee.

The “Testing” to these “Additional Gases” would have resulted in a 100% Failure rate of the SF Detection C O Alarm, as well as several others. [See GRI Report of Aug.,’00]

4. I also was curious about a GRI report from two years ago on the performance and reliability of the UL/CSA approved home alarms. In this report they stated that one brand (they don't specifically mention the name) had a 38% failure rate - brand new, right 'out of the box' and about 1/2 had a 100% failure rate alarming when a person exposed to CO would have a 2.5 - 10% COHb level as UL/CSA requires they do. I just don't understand how such a dismal report can seemingly be ignored by UL and CSA. I also am concerned that the GRI report does not specifically mention the brand names of the nine different alarms they studied. To me that is like the National Highway Traffic Safety Administration (NHTSA) crash testing automobiles but not releasing the names of the cars that failed the tests. Wouldn't that information be critical for consumers purchasing home CO alarms? The report also stated that the 'self check' feature on the 'approved' alarms only tested the electronics, not the sensor and that this led to a situation where a consumer would believe the alarm was working while the sensor would not provide any protection. How does your monitor check for proper operation of the sensor?

You MUST always remember that “Standard” Requirements, ... are NOT WRITTEN to “FORCE” the production of the VERY BEST Products Possible, .... they are Written to “Accommodate” the “EXISTING” products that are “presently” available in the Marketplace, .......... at the “TIME” ..... the Standards are Written.

As I have mentioned to you in a Previous Response, as long as UL will Continue to “List” a C O Alarm that States in their Owner’s Manual that the “Accuracy” of their Digital Display is: “PLUS 100%, ........ MINUS 40%, ..... you can always EXPECT that the Actual Field Testing, ..... WILL indicate FAR WORSE results. Unfortunately, MOST Manufacturers will build their products, “Just Good Enough’ to pass “Whatever”, the “REQUIRED” Standard DEMANDS.

FYI, ... GTI, who is now in control of all previous GRI “Test” Results, has THE NAMES of the “Specific” Brand name Products “TESTED”, has indicated that they “May”, ..... name, “Names”, .... as Steve Weismer, their Former Director of “Health & Safety”, had Suggested in the Spring of 2000, ...... prior to his retirement.

I Personally DO HAVE, the NAMES & RESULTS; however, while I WAS given “Verbal” Approval to “RE-Produce”, ..... in “Printed” form, this INFO, ... I have NOT yet received Permission to post it on my Website.

If you “Review”, .... ANY, of my ”COMPLAINTS” about the UL, ULC, CSA, ... C O “Standard Requirements”, ... you are VERY MUCH AWARE, that MY “BASIC” Complaint is:..... SENSORS are NOT Required to be “Monitored”, ..... and that UL does NOT REQUIRE, . “Time of Manufacture” Testing, ..... or “LIFETIME” Reliability Testing.

My Sensor is “Monitored” by checking the relative “Current Flow” between the “Reference” Electrode and the “Counter Electrode”, Minus the Sensing Electrode “Data - Imput”, to Establish a “Sensor Activity Baseline”. This reading is “Verified” every 10 Seconds, as the Required “Baseline”, ... and a “Drift” in Excess of 35%, or more, will result in an “ERR” Warning.

Remember that a GOOD Electrochemical Sensor, [Acid-Based], is much like an “Alka-seltzer tablet”, that “Does Nothing”, ... until you drop it into a Glass of water, ... then a “Chemical” Reaction Occurs..... Likewise, with an Electrochemical Sensor, ... it is just sitting there, doing NOTHING, ... UNTIL a Molecule of C O enters the Sensor, ... then the Chemical reaction Occurs, ... and EVERY Molecule of C O , ... Creates so many “Nana-Amps” of Electricity, ..... Producing an EXACT Reading as to how many PPM of C O is Present in the Sensor. Presently, this is the MOST ACCURATE C O Detection Technology Available in the Consumer Marketplace.

The “Reason” that my Monitor is NOT “APPROVED” as a “SMOKE DETECTOR” is:

b. I “Presently” use a “Graduated” Series of Alarm Signals at 10, 25, & 50 PPM of C O Present, ..... this would HAVE to be CHANGED to an Instant “FULL” Alarm, .... using the “REQUIRED” SMOKE “AUDIBLE ALARM SIGNAL”, ... NOT the C O “Signal”.

No, “Presently”, ..... NO other Residential C O Products Manufacturer Provides the Extensive “Data Recall’ that my Monitor provides for the First Responder / HVAC Contractor.

Once Again, ..... this is NOT “Rocket Science”, ..... it is just a matter of “Spending” a little Extra Money to use “Better” Components, ..... like the 12 Bit Microcontroller that I “CHOOSE” to use, ..... not ONLY for the Extra “Space” that it gives me for the “Recall” Features, “ERR” Warnings etc., but because it also provides FAR Better Sensor “Resolution”, for Very Accurate Sensor Calibration and Control.

Although, I DO NOT “Utilize it”, ... the “Resolution” provided “Could” permit “Accurate Calibration” down to 1/8 of ONE, PPM.

If the unit was Manufactured, [ NOT "SOLD" ] AFTER OCT. 01, 1998, ... it will say that it is an "ALARM", ... and that it will "ACTIVATE" at 70 ppm between 60 to 240 Minutes, 150 ppm between 10 & 50 Minutes, or 400 ppm between 4 & 15 minutes.400ppm [between 4 & 15 Minutes has remained the same in ALL of the Versions of UL-2034]. A few of the "BETTER Alarms may say 70 ppm between 60 &189 minutes. [Call me & I will EXPLAIN this some time][Some models that I have seen will say ONLY@ 400 ppm, within 15 Minutes, on the Alarm. [1st Alert Model FCD4, manufactured in March, 2002.].

DETECTORS Manufactured Prior to Oct. 01, 1998, "Could" Display ANY ppm that they WANTED, [ NONE Were Truly ACCURATE], the Owner's Manual for the 110 V, MOS Nighthawk of that Vintage, said that the Digital Display was Accurate PLUS, or MINUS 40%, PLUS 15 ppm.] the 110 Volt Model DID start showing a C O Reading at 7 ppm. [Although it was RARELY Correct at ANY Level.]

I Truly Appreciate your Questions; however, I Appreciate your Interest in Carbon Monoxide Poisoning, and your Dedication to C O Safety and Education, EVEN MORE !