Return-Path: Received: from [24.25.9.100] (HELO ms-smtp-01-eri0.southeast.rr.com) by logan.com (CommuniGate Pro SMTP 4.2b1) with ESMTP id 3142861 for flyrotary@lancaironline.net; Fri, 02 Apr 2004 15:18:44 -0500 Received: from EDWARD (clt25-78-058.carolina.rr.com [24.25.78.58]) by ms-smtp-01-eri0.southeast.rr.com (8.12.10/8.12.7) with SMTP id i32KIASn016908 for ; Fri, 2 Apr 2004 15:18:12 -0500 (EST) Message-ID: <001701c418ef$a1836610$2402a8c0@EDWARD> From: "Ed Anderson" To: "Rotary motors in aircraft" References: Subject: Re: [FlyRotary] Re: Rusty's O2 / EM2 problems. Date: Fri, 2 Apr 2004 15:18:15 -0500 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0014_01C418C5.B865F4D0" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2800.1158 X-MIMEOLE: Produced By Microsoft MimeOLE V6.00.2800.1165 X-Virus-Scanned: Symantec AntiVirus Scan Engine This is a multi-part message in MIME format. ------=_NextPart_000_0014_01C418C5.B865F4D0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable MessageRusty, the O2 sensor in itself does not cycle, but only in = response to changes in the oxygen content of the exhaust. The reason = that "cycling" is seen in automotive applications is the EFI CPU of the = auto is "hunting" for the optimum ratio for emissions approx 14.7 = air/fuel ratio. So if it senses too high an oxygen contents (varies = from 0.45 volts or thereabouts to 0.0 volts) it increases the fuel = content which likely overshoots the desired ratio so then the computer = senses this overshoot through the o2 sensor and so quickly corrects the = mixture by decreasing the fuel content. This seesawing between over and = under the desired voltage (approx 0.45) causes the O2 sensor to send out = a "Sawtooth" wave form of voltage varying from under 0.45 volts to over = 0.45 volts and under certain conditions and probably approach 0.0 to 1.0 = volts in swing. =20 In the automobile use Lead will quickly decrease the sensitivity of the = O2 sensor to the point that it is relative useless for the emission = control purpose since it is trying to maintain an optimum by quickly = sensing and responding to an over or under oxygen voltage signal. In = our application we more or less use it as a trend indicator Rich, Lean, = Going Rich, Going Lean. Millisecond response time is not required for = our reaction time {:>). In our use, the O2 sensor simply sends out a voltage proportional to the = O2 content it sees and since there is no computer feed back it stays = relative stable (unless you change an engine parameter or mixture) or = changes slowly. Ed Anderson Ed Anderson RV-6A N494BW Rotary Powered Matthews, NC ----- Original Message -----=20 From: Dale Rogers=20 To: Rotary motors in aircraft=20 Sent: Friday, April 02, 2004 1:31 PM Subject: [FlyRotary] Re: Rusty's O2 / EM2 problems. Rusty, >>... A properly functioning oxygen sensor will show a >> rapidly fluctuating output voltage between approximately 0.1 and = 1.0 volts. >> The time taken for the voltage to change from 0.1 V to 1.0 V = (referred to as >> the lean to rich response time) should be about 300 milliseconds. A = similar >> time should be measured when the voltage changes from 1.0 V to 0.1 = V (rich >> to lean response time).=20 > =20 > Can anyone explain why the voltage is supposed to cycle between .1V = and 1.0V > ? Tracy said the EM-2 doesn't do any averaging, but just captures = the > voltage in a snapshot when it's time to measure that channel. If = that was > the case, it would seem that a "properly operating" sensor would = cause the > EM-2 to be erratic. =20 It *really* depends on the EM2 sampling rate. I would=20 hope that in 300 mS, the EM2 would have taken at least five=20 samples, but maybe not ... The O2 cycling is due to the processing delay of the=20 computer; the sensor gets read, and the computer sees=20 that the engine needs more (or less) fuel, so it adjusts=20 the injector timing; by the time that that particular=20 expended charge gets to the O2 sensor, the computer has=20 taken more readings, inducing a small amount of "overshoot". =20 Then, seeing that, it compensates in the other direction. =20 As long as it stays in the range that produces more than .1v=20 and less than 1.1v, the engine judged to be running=20 efficiently. Ideally, at a stable RPM, the oscillations=20 would "settle" to somewhere near the middle of the range,=20 but processing delay assures that this is all but=20 impossible. That's my understanding ... Dale R. COZY MkIV #1254 -------------------------------------------------------------------------= ----- Considering that my O2 sensor exhibits similar behavior as = yours and I also measure downstream EGT temps (after turbo) right beside = the O2 sensor and only see temps in the 700F range, I can't help but = think that maybe the low temps are possibly the cause of the problem. =20 My EGT probes are about 10 inches from the port, and the O2 sensor is = about 40 inches farther downstream. I'm confident that I can get the = temps to 800F at the O2 sensor, but I may not have the day I installed = the sensor. I was seeing temps in the 1000 to 1100 range at low rpms, = but the other day, when tied down, I was reaching 1600's at full = throttle. That was with the old sensor though. I'll probably be doing = some more running tomorrow, so we'll see what I get. I also have the = old mixture gauge installed as well as the EM-2. =20 Here is a site that explains O2 sensors- = http://www.ngkspark.com.au/tech_info_oxygen_sensors_index.htm Here is an excerpt from that site: How To Test Oxygen Sensors? Using an appropriate connecting device, connect the sensor output to = your oscilloscope; do not disconnect the sensor from the ECU. Run the = engine at approximately 2000 rpm. A properly functioning oxygen sensor = will show a rapidly fluctuating output voltage between approximately 0.1 = and 1.0 volts. The time taken for the voltage to change from 0.1 V to = 1.0 V (referred to as the lean to rich response time) should be about = 300 milliseconds. A similar time should be measured when the voltage = changes from 1.0 V to 0.1 V (rich to lean response time).=20 If the sensor output is constant or the response time is too slow the = sensor should be changed.=20 Can anyone explain why the voltage is supposed to cycle between .1V = and 1.0V ? Tracy said the EM-2 doesn't do any averaging, but just = captures the voltage in a snapshot when it's time to measure that = channel. If that was the case, it would seem that a "properly = operating" sensor would cause the EM-2 to be erratic. =20 Hopefully, tomorrow's testing will reveal some more useful info. =20 Rusty (looking to fly again this weekend) -------------------------------------------------------------------------= ----- >> Homepage: http://www.flyrotary.com/ >> Archive: http://lancaironline.net/lists/flyrotary/List.html ------=_NextPart_000_0014_01C418C5.B865F4D0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Message
Rusty,  the O2 sensor = in itself does not=20 cycle, but only in response to changes in the oxygen content of the=20 exhaust.  The reason that "cycling" is seen in automotive = applications is=20 the EFI CPU of the auto is "hunting" for the optimum ratio for emissions = approx=20 14.7 air/fuel ratio.  So if it senses too high an oxygen contents = (varies=20 from 0.45 volts or thereabouts to 0.0 volts) it increases the fuel = content which=20 likely overshoots the desired ratio so then the computer senses this = overshoot=20 through the o2 sensor and so quickly corrects the mixture by decreasing = the fuel=20 content.  This seesawing between over and under the desired voltage = (approx=20 0.45) causes the O2 sensor to send out a "Sawtooth" wave form of voltage = varying=20 from under 0.45 volts to over 0.45 volts and under certain conditions = and=20 probably approach 0.0 to 1.0 volts in swing. 
 
In the automobile use Lead will quickly = decrease=20 the sensitivity of the O2 sensor to the point that it is relative = useless for=20 the emission control purpose since it is trying to maintain an optimum = by=20 quickly sensing and responding to an over or under oxygen voltage = signal. =20 In our application we more or less use it as a trend indicator Rich, = Lean, Going=20 Rich, Going Lean.  Millisecond response time is not required for = our=20 reaction time {:>).
 
In our use, the O2 sensor simply sends = out a=20 voltage proportional to the O2 content it sees and since there is no = computer=20 feed back it stays relative stable (unless you change an engine = parameter or=20 mixture) or changes slowly.
 
Ed Anderson
 
Ed Anderson
RV-6A N494BW Rotary Powered
Matthews, NC
----- Original Message -----
From:=20 Dale = Rogers=20
Sent: Friday, April 02, 2004 = 1:31=20 PM
Subject: [FlyRotary] Re: = Rusty's O2 / EM2=20 problems.

Rusty,

>>... A properly functioning oxygen = sensor=20 will show a
>> rapidly fluctuating output voltage between=20 approximately 0.1 and 1.0 volts.
>> The time taken for the = voltage to=20 change from 0.1 V to 1.0 V (referred to as
>> the lean to = rich=20 response time) should be about 300 milliseconds. A similar
>> = time=20 should be measured when the voltage changes from 1.0 V to 0.1 V=20 (rich
>> to lean response time).
> 
> Can = anyone=20 explain why the voltage is supposed to cycle between .1V and = 1.0V
>=20 ?  Tracy said the EM-2 doesn't do any averaging, but just = captures=20 the
> voltage in a snapshot when it's time to measure that=20 channel.  If that was
> the case, it would seem that a = "properly=20 operating" sensor would cause the
> EM-2 to be erratic. =20

   It *really* depends on the EM2 sampling = rate.  I=20 would
hope that in 300 mS, the EM2 would have taken at least five=20
samples, but maybe not ...

   The O2 cycling is = due to=20 the processing delay of the
computer; the sensor gets read, and = the=20 computer sees
that the engine needs more (or less) fuel, so it = adjusts=20
the injector timing; by the time that that particular
expended = charge=20 gets to the O2 sensor, the computer has
taken more readings, = inducing a=20 small amount of "overshoot". 
Then, seeing that, it = compensates in=20 the other direction. 
As long as it stays in the range that = produces=20 more than .1v
and less than 1.1v, the engine judged to be running=20
efficiently.  Ideally, at a stable RPM, the oscillations =
would=20 "settle" to somewhere near the middle of the range,
but processing = delay=20 assures that this is all but
impossible.

That's my = understanding=20 ...

Dale R.
COZY MkIV #1254

        Considering that my O2 sensor exhibits similar behavior = as yours=20 and I also measure downstream EGT temps (after turbo) right beside the = O2=20 sensor and only see temps in the 700F range, I can't help but think = that maybe=20 the low temps are possibly the cause of the problem.  
 
My EGT probes are about 10 inches from the = port, and the=20 O2 sensor is about 40 inches farther downstream.  I'm = confident=20 that I can get the temps to 800F at the O2 sensor, but I may not = have the=20 day I installed the sensor.  I was seeing temps in the 1000 to = 1100 range=20 at low rpms, but the other day, when tied down, I was reaching = 1600's at=20 full throttle.  That was with the old sensor though.  I'll = probably=20 be doing some more running tomorrow, so we'll see what I get.  I = also=20 have the old mixture gauge installed as well as the=20 EM-2.  
 
 
Here is a site that explains O2 = sensors-  ht= tp://www.ngkspark.com.au/tech_info_oxygen_sensors_index.htm
 
Here is an excerpt from that=20 site:
 
How To Test Oxygen=20 Sensors?

Using an appropriate connecting device, = connect the=20 sensor output to your oscilloscope; do not disconnect the sensor from = the ECU.=20 Run the engine at approximately 2000 rpm. A properly functioning = oxygen sensor=20 will show a rapidly fluctuating output voltage between approximately = 0.1 and=20 1.0 volts. The time taken for the voltage to change from 0.1 V to 1.0 = V=20 (referred to as the lean to rich response time) should be about 300=20 milliseconds. A similar time should be measured when the voltage = changes from=20 1.0 V to 0.1 V (rich to lean response time).

If the sensor = output is=20 constant or the response time is too slow the sensor should be = changed.=20
 
 
Can anyone explain why the voltage is = supposed to=20 cycle between .1V and 1.0V ?  Tracy said the EM-2 doesn't do any=20 averaging, but just captures the voltage in a snapshot when it's time = to=20 measure that channel.  If that was the case, it would seem = that a=20 "properly operating" sensor would cause the EM-2 to be erratic. =20
 
Hopefully, tomorrow's testing will reveal = some more=20 useful info. 
 
Rusty (looking to fly again this=20 weekend)

>>  Homepage: =20 http://www.flyrotary.com/
>>  Archive:  =20 = http://lancaironline.net/lists/flyrotary/List.html
------=_NextPart_000_0014_01C418C5.B865F4D0--