X-Virus-Scanned: clean according to Sophos on Logan.com Return-Path: Sender: To: lml@lancaironline.net Date: Thu, 04 Oct 2012 13:59:28 -0400 Message-ID: X-Original-Return-Path: Received: from mail-pa0-f52.google.com ([209.85.220.52] verified) by logan.com (CommuniGate Pro SMTP 6.0c1) with ESMTPS id 5793441 for lml@lancaironline.net; Thu, 04 Oct 2012 12:00:15 -0400 Received-SPF: pass receiver=logan.com; client-ip=209.85.220.52; envelope-from=pjdmiller@gmail.com Received: by mail-pa0-f52.google.com with SMTP id hz10so601870pad.25 for ; Thu, 04 Oct 2012 08:59:40 -0700 (PDT) Received: by 10.68.141.46 with SMTP id rl14mr23189060pbb.2.1349366380703; Thu, 04 Oct 2012 08:59:40 -0700 (PDT) X-Original-Return-Path: Received: from [192.168.1.105] (S010620aa4b008706.cg.shawcable.net. [174.0.117.10]) by mx.google.com with ESMTPS id b10sm4454582pav.17.2012.10.04.08.59.38 (version=TLSv1/SSLv3 cipher=OTHER); Thu, 04 Oct 2012 08:59:39 -0700 (PDT) From: Paul Miller Content-Type: multipart/alternative; boundary="Apple-Mail=_2290A791-F895-4488-A9D6-85E89DD8F4C5" X-Original-Message-Id: <79EFA6B6-9608-4804-B170-FDE916D62DE9@gmail.com> Mime-Version: 1.0 (Mac OS X Mail 6.1 \(1498\)) Subject: Re: [LML] Re: Probes gone wild X-Original-Date: Thu, 4 Oct 2012 09:59:36 -0600 References: X-Original-To: "Lancair Mailing List" In-Reply-To: X-Mailer: Apple Mail (2.1498) --Apple-Mail=_2290A791-F895-4488-A9D6-85E89DD8F4C5 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=windows-1252 I like the freq generator. Easy to build one with parts as I have all = the old Radio Shack circuit books. There indeed may be a problem = somewhere between the tank and panel, I rang out all those wires but the = results are still so sporadic that its worth the try. Thanks and I will = update when I get somewhere. BTW, my right side kept going to empty = too. I traded up for the 10-point converters and now have different = problems. Paul. On 2012-10-04, at 9:17 AM, Colyn Case wrote: > Paul, >=20 > not sure what is going on there. I also don't know much about the = geometry of Legacy tanks. If they were perfectly rectilinear, you = would expect a linear change in frequency as you add fuel. Obviously = tanks aren't linear or they wouldn't bother with a 10-point system. = However, I would record the frequencies when you do your next = calibration. If it doesn't look reasonable (especially if freq is = jumping around erratically ) then you probably have a problem with the = probe. >=20 > You might consider, though, plugging a frequency generator into your = system instead of the probe just to make sure the problem isn't = somewhere else. I had problems with mine going to "empty" when it = wasn't and tracked it down to grounding problems in that path. You = would have to talk to EI or Princeton about how to plug in a freq = generator. In the extreme case you could order a new probe but hook it = up before you put it in the airplane and see if you can bench test it = like Adam did. >=20 > On Oct 4, 2012, at 8:43 AM, Paul Miller wrote: >=20 > Adam, Colyn: This thread has been helpful for me. I have a = problematic right probe in the Legacy that I'm dreading pulling the wing = to replace. If I adjust the wires I can get a good signal but it is not = consistent. We fixed the wiring in place with some RTV to prevent = movement and I then swapped the old Princetons for 10-point models and = tried to re-calibrate. Oh, it was very painful. I would fill one tank = while draining the other, I would miss a setpoint and have to start = over. Finally, got it done but the results are not good on either tank. = The right stays full now full time and left won't come off full for at = least 5 gallons. >=20 > BTW, I did use the frequency meter on the Princetons and saw the 2khz = and 4khz ranges but never understood those results until Adam's testing. = I am going to re-calibrate at least one more time next month so do you = have any thoughts on how best to do this with the 10-point units? I do = put the wings and tail on jacks so there is no movement. I drained = about 3.3g each time (or filled) and I waited about 2 minutes between = fills for the units to settle down. I think it might need more time to = settle down though. I will put the portable scope on the wires to make = sure I'm getting the square wave too. Thoughts appreciated. >=20 > Paul > Calgary > On 2012-09-05, at 10:28 AM, Adam Molny = wrote: >=20 >> Hi Colyn =96 I=92m measuring the signal before it goes into the = Princeton converter. The square wave looks fairly clean and is the = correct amplitude (5v). When I originally bench tested the probe, I = taped over the holes near the head, stood it on end, and filled it with = fuel. Several aluminum chips came out when I poured out the fuel after = the test, presumably left over from the manufacturing process. I=92m = wondering if any contamination could be causing a complete or partial = short between the center conductor and the outer tube. The only other = possibility would appear to be some sort of failure of the electronics = within the probe head. Short of pulling the wing and replacing the = probe, I=92m not sure what else I can do at this point. >> =20 >> -Adam Molny >> From: Colyn Case [mailto:colyncase@earthlink.net]=20 >> Sent: Tuesday, September 04, 2012 11:17 PM >> To: lml@lancaironline.net >> Subject: Re: [LML] Re: Probes gone wild >> =20 >> Adam, are you measuring that right at the probe or is there something = in the middle? >> I have not had anything like this occur with my EI probes. I did = however have a lot of problems with the previous (old) princeton = converters. If memory serves, these convert from freq to 0-5V. The = new 10-point Princeton converters have been great. >> =20 >> =20 >> On Sep 4, 2012, at 4:18 PM, = wrote: >> =20 >>=20 >> Posted for Adam Molny : >>=20 >> My EI fuel probes output a square wave whose frequency is = proportional to >> fuel level. Before installation, bench testing showed that the probes = put >> out 4kHz when dry and 2kHz when full. I calibrated my Dynon SkyView >> accordingly. Now, at 70 hours the left probe has become unreliable. A = few >> weeks ago the frequencies changed to 3kHz empty and 2kHz full. I = calibrated >> the Dynon, but now the probe is putting out 3kHz when full. >>=20 >> Has anyone had similar problems? I'll have to remove the left wing to = get >> the probe out, so I'd like to have a plan of action first.=20 >>=20 >> Thanks, >> Adam Molny >> Legacy N181AM=20 >> 72 hours hobbs time >>=20 >> =20 >> [We have an EI capacitive probe on our test stand and noticed that it = is subject to large swings relative to temperature. Not the probe = itself, but rather the oscillator module whose frequency is driven by = the probe. The easiest way for me to fill our 280 gallon tank is to = pull the truck out into the Florida sunshine and expose the oscillator = to full sunlight... soon as it heats up the tank tops itself right = off... damnedest thing you ever saw. Only problem is it won't stay = full. ] >> =20 >> =20 >> -- >> =20 >> For archives and unsub = http://mail.lancaironline.net:81/lists/lml/List.html >> =20 >=20 >=20 --Apple-Mail=_2290A791-F895-4488-A9D6-85E89DD8F4C5 Content-Transfer-Encoding: quoted-printable Content-Type: text/html; charset=windows-1252 I = like the freq generator.   Easy to build one with parts as I have = all the old Radio Shack circuit books.   There indeed may be a = problem somewhere between the tank and panel, I rang out all those wires = but the results are still so sporadic that its worth the try. Thanks and = I will update when I get somewhere.  BTW, my right side kept going = to empty too.   I traded up for the 10-point converters and now = have different problems.  Paul.
On 2012-10-04, at 9:17 = AM, Colyn Case <colyncase@earthlink.net> = wrote:


On Oct 4, 2012, at 8:43 AM, Paul Miller = wrote:

Adam, Colyn: This thread has = been helpful for me.  I have a problematic right probe in the = Legacy that I'm dreading pulling the wing to replace.  If I adjust = the wires I can get a good signal but it is not consistent.  We = fixed the wiring in place with some RTV to prevent movement and I then = swapped the old Princetons for 10-point models and tried to = re-calibrate.  Oh, it was very painful.  I would fill one tank = while draining the other, I would miss a setpoint and have to start = over.  Finally, got it done but the results are not good on either = tank.  The right stays full now full time and left won't come off = full for at least 5 gallons.

BTW, I did use the = frequency meter on the Princetons and saw the 2khz and 4khz ranges but = never understood those results until Adam's testing.   I am going = to re-calibrate at least one more time next month so do you have any = thoughts on how best to do this with the 10-point units?  I do put = the wings and tail on jacks so there is no movement.  I drained = about 3.3g each time (or filled) and I waited about 2 minutes between = fills for the units to settle down.   I think it might need more = time to settle down though.   I will put the portable scope on the = wires to make sure I'm getting the square wave too. Thoughts = appreciated.

Paul
Calgary
= On 2012-09-05, at 10:28 AM, Adam Molny <Adam@ValidationPartners.com> wrote:

=

Hi Colyn =96 I=92m measuring the signal before it goes into the Princeton converter. The square wave looks fairly clean and is the correct = amplitude (5v). When I originally bench tested the probe, I taped over the holes = near the head, stood it on end, and filled it with fuel. Several aluminum chips = came out when I poured out the fuel after the test, presumably left over from the manufacturing process. I=92m wondering if any contamination could be causing a complete or partial short between the center conductor and the = outer tube. The only other possibility would appear to be some sort of failure = of the electronics within the probe head. Short of pulling the wing and = replacing the probe, I=92m not sure what else I can do at this = point.

 

-Adam Molny

From: Colyn Case [mailto:colyncase@earthlink.net] =
Sent: Tuesday, September = 04, 2012 11:17 PM
To: lml@lancaironline.net
Subject: Re: [LML] Re: = Probes gone wild

 

Adam, are you measuring that right at the probe or is there = something in the middle?

I have not had anything like this occur with my EI probes. =   I did however have a lot of problems with the previous (old) princeton = converters.   If memory serves, these convert from freq to 0-5V.   The new 10-point Princeton converters have = been great.

 

 

On Sep 4, 2012, at 4:18 PM, <marv@lancair.net> <marv@lancair.net> = wrote:

 

 

--
 
For =
archives and unsub http://mail.=
lancaironline.net:81/lists/lml/List.html

 

=



= --Apple-Mail=_2290A791-F895-4488-A9D6-85E89DD8F4C5--