Inaccurate tach = grounded?

[pilotmattk]

So a tach is minimum day vfr equipment per 91.205. To me that means an inop tach grounds the airplane. What about an inaccurate tach? Say one that is 3000-5000 rpm off and swings within that range? (bad cable). Is that considered inop? Should it ground the a/c?

Thanks,
Matt

 

[KennyFlys]

Sec. 91.205

Powered civil aircraft with standard category U.S. airworthiness certificates: Instrument and equipment requirements.

(a) General. Except as provided in paragraphs (c)(3) and (e) of this section, no person may operate a powered civil aircraft with a standard category U.S. airworthiness certificate in any operation described in paragraphs (b) through (f) of this section unless that aircraft contains the
instruments and equipment specified in those paragraphs (or FAA-approved equivalents) for that type of operation, and those instruments and items of equipment are in operable condition.

One could reasonably assume this means to be in "proper" operating condition. I certainly would expect as much and I don't see an inspector expecting less.

 

[Dan Thomas]

So a tach is minimum day vfr equipment per 91.205. To me that means an inop tach grounds the airplane. What about an inaccurate tach? Say one that is 3000-5000 rpm off and swings within that range? (bad cable). Is that considered inop? Should it ground the a/c?

Thanks,
Matt

I browsed the FARS but couldn't find the equivalent to our Canadian reg that says this:

http://www.tc.gc.ca/CivilAviation/Regserv/Affairs/cars/Part6/Standards/a625c.htm

8. Tachometers The accuracy of mechanical drag cup type tachometers, for fixed wing propeller driven aircraft, shall be checked on site annually, and be accurate to within the tolerances established by the aircraft manufacturer or, where no tolerance has been specified by the aircraft manufacturer, to within +\- 4% of engine RPM at mid-point of the cruise range.
(amended 2000/12/01; previous version

Dan

 

[AWACSEng]

If it's not idicating properly, might as well be inop.

 

[CJones]

I don't know that the accuracy of the instrument is the 'grounding' factor. I would assume that the CAUSE of the inaccuracy is what would cause a grounding. If it's a cable loose, what happens if it comes off completely? If it's more than the cable (on the engine side (sensor)), what happens if it completely fails? etc. etc. Most people can safely fly an airplane without looking at the tach, but what causes the tach to be unreliable would be my concern.

 

[Steve]

So a tach is minimum day vfr equipment per 91.205. To me that means an inop tach grounds the airplane. What about an inaccurate tach? Say one that is 3000-5000 rpm off and swings within that range? (bad cable). Is that considered inop? Should it ground the a/c?

Thanks,
Matt

WOW!!!

 

[Tom-D]

I just noticed it was fubar while I was landing.

Or,,

Oops, it does what?

It's the FAA burden of proof to show you used a unairworthy aircraft..

But that is pretty easy to do when the discrepancy was written up 5 times before.

 

[pilotmattk]

Woops that was a little bit of a brain fart coupled w/ iphone typing. The rpm was off by 300-500 rpm maybe 800 rpm at full power (well into the 3k range). That's hundreds, not thousands. I only noticed it b/c i was shooting approaches (VFR) and noticed that it was swinging weird and the numbers weren't matching up right. I went missed and noticed that level flight at 2400 was only giving me 80 or so ias. So i landed and squawked it.

This was a rental and what bothers me is that it was signed back into service (i was the first squawk in the book on this one). The problem was a cable. They ordered the part and fixed it a week or 2 later.

 

[pilotmattk]

Additionally, it seems a 2-3% inaccuracy is pretty common on a tach. I've seen this comparing an optical tach to a mechanical one. This is more like a 20% inaccuracy, however.

 

[PabloT]

By the book...I'd have to say the airplane should be grounded.

That said, most the inaccuracy in that tach is probably just fine in many airplanes assuming the pilot is familiar with the airplane.

Even though the odds of getting caught are rather small, I'd still rather opt to not fly it and not have to worry about explaining my actions to the FAA later.

 

[poadeleted20]

I agree with PabloT -- that plane as described ain't worthy for air.

Sure, a constant, predictable RPM error of known and small proportions (only a few percent) is a fact of life with an older mechanical tach due to normal wear. But this one is well beyond that level of error -- something's wobbling loose in there, and since part of that tach system is plugged into the engine, something that comes loose could end up loose in the gear box, which is a real bad place for something to be floating around loose.

At some point, it's pilot judgement, but given the wild inaccuracies of the tach described by the OP (even as modified by an order of magnitude), I'm saying this one is in the "it ain't fixed, so don't fly it" category.

 

[Tom-D]

I agree with PabloT -- that plane as described ain't worthy for air.

Sure, a constant, predictable RPM error of known and small proportions (only a few percent) is a fact of life with an older mechanical tach due to normal wear. But this one is well beyond that level of error -- something's wobbling loose in there, and since part of that tach system is plugged into the engine, something that comes loose could end up loose in the gear box, which is a real bad place for something to be floating around loose.

At some point, it's pilot judgement, but given the wild inaccuracies of the tach described by the OP (even as modified by an order of magnitude), I'm saying this one is in the "it ain't fixed, so don't fly it" category.

How does any pilot draw the conclusion that one error in a primary instrument is airworthy/safe to fly, and the other is not.

91.205 Powered civil aircraft with standard category U.S. airworthiness certificates: Instrument and equipment requirements.
(a) General. Except as provided in paragraphs (c)(3) and (e) of this section, no person may operate a powered civil aircraft with a standard category U.S. airworthiness certificate in any operation described in paragraphs (b) through (f) of this section unless that aircraft contains the instruments and equipment specified in those paragraphs (or FAA-approved equivalents) for that type of operation, and those instruments and items of equipment are in operable condition.
(b) Visual-flight rules (day). For VFR flight during the day, the following instruments and equipment are required:

(1) Airspeed indicator.

(2) Altimeter.

(3) Magnetic direction indicator.

(4) Tachometer for each engine.
(5) Oil pressure gauge for each engine using pressure system.

 

[Stache]

Just my two cents worth. If you read the airworthiness certificate it say to be airworthy to things have to be met. Meets the type design and safe condition. The safe condition in this case is the problem, which make the aircraft unairworthy.

During an annual inspection a good mechanic will use a hand held device to check the tack readings for error. Tach readings are considered limitations in the aircraft flight manual if they are off then the limitation can be exceeded. Pilots rely on limitations to operate the aircraft safety.

I would say take the aircraft to the local shop and have the mechanic check the tack calibration with a hand held tach. If it is not correct have it replaced or repaired. It’s not worth the risk as it is a required insturment.

Just one man’s opinion.

 

[Tom-D]

Just my two cents worth. If you read the airworthiness certificate it say to be airworthy to things have to be met. Meets the type design and safe condition. The safe condition in this case is the problem, which make the aircraft unairworthy.

During an annual inspection a good mechanic will use a hand held device to check the tack readings for error. Tach readings are considered limitations in the aircraft flight manual if they are off then the limitation can be exceeded. Pilots rely on limitations to operate the aircraft safety.

I would say take the aircraft to the local shop and have the mechanic check the tack calibration with a hand held tach. If it is not correct have it replaced or repaired. It’s not worth the risk as it is a required insturment.

Just one man’s opinion.

In the case of the Tach, ignorance is bliss, you as a plot do not know you have a discrepancy until you see something abnormal..

You have no reason to doubt the accuracy of the tach. other than to see it acting strange.

You have no requirement to do a full power run up to see if the engine will develop full RPM. You have no requirement to calibrate the tach in any inspection, or verify it as a pilot.

SO.....ignorance is bliss. you can't write a discrepancy with out seeing a abnormal occurance.

Wildly swinging needle, ? you have a discrepancy in a required instrument.

 

[Dan Thomas]

But this one is well beyond that level of error -- something's wobbling loose in there, and since part of that tach system is plugged into the engine, something that comes loose could end up loose in the gear box, which is a real bad place for something to be floating around loose.

No, it won't do that. Lycs and Conts and others take the tach drive off the aft end of the camshaft, and the tach is calibrated to read engine RPM while turning at half that. There's a pretty stout shaft driving the tach output because it's also driving the oil pump. The tach cable is a skinny little bit of cable wrapped in a very fine spring and will break if the tach itself seizes up.

The swinging tach needle is due either to a really sticky tach cable inside its sheath, sticky because the oil or grease has dried out and formed gucky stuff that grabs the cable and makes it walk around the inside diameter of the sheath and can make it wind up and release repeatedy. Or the tach itself has dried-out bearings that are sticking and making the cable wind up and release when the bearings let go. Many tachs have a small oil hole atop the aft end of the case, right at the input. Drop some really light oil like Royco 363 in it. And some tach cables can be pulled out of their sheaths; solvent blown through the sheath will remove some of that crud and the rest is wiped off the cable. Lube it with a low-temp instrument grease like Lubriplate, or maybe some dry graphite.

Cessna calls that cable a "tachometer drive shaft."

Dan

 

[Tom-D]

I was once called and asked why the tach was leaking engine oil?

After questioning the owner said he had 5 hours on a brand new Lycoming, they had for gotten to instal the tach cable seal.

You as an A&P can't repair the tach. all you can do is repair/replace the cable or housing.

Can you check the tach calibration? how

 

[timwinters]

Can you check the tach calibration? how

With a prop tach.

My mechanical tach (overhauled 3 years ago) reads anywhere between 10 and 100 rpm low. The amount of variance depends on temperature, how long I've been flying, how the stars are aligned, and, apparently, many other factors unbeknownst to me.

I set up cruise power using the prop tach and pretty much ignore me mechanical tach.

I wouldn't recommend anyone work on a mechanical tach or cable anymore. All of my tachs have proven unreliable even when everything is working perfect. For me, it's time for an electronic tach the next time something something needs work. I've given up on mechanical tachs.

 

[bullwinkle]

Even if it were legal to fly with the tach that far off, why would you want to? Doesn't seem very safe. Other than to fly it to get repairs, of course. How could you do a proper mag check, for instance?

 

[jshawley]

Regarding mag checks (and that is all I'm addressing in this post), you're not really looking for exactly 1725rpm, +/- 4% or whatever.

What you're looking for is a drop of, say, 75 (or 50 or 100[--whatever your POH says]) rpm, in harmony with a slightly lower frequency in the sound of the exhaust. So, assume your tachometer is off by, say, +125rpm at 2500rpm (i.e. when the photo-tach is indicating 2,500rpm, the plane's mechanical tach is indicating only 2,375rpm--thus when you open your throttle to an indicated rpm of 2,500, the engine--and photo tach, if you were using it--is actually at about 2625rpm), and you check your mags at 1800rpm (indicated), then when you go to L or R, your drop will still be only, say, 75rpm. It's just that you actually tested your mags at a true engine speed of 1890rpm (assuming the error is linear and there is no offset error at 0rpm), and they dropped to 1811.25 (i.e. you'll see about [1710] or so on your tach ), rather than 1725rpm.

The Difference will still be [for all practical purposes] the same.

[I edited inside the brackets]

But please let me emphasize, I'm not addressing the airworthiness of a tach that is just a little out of spec.

I approach this from my perspective as a retired metrology technician; we used to see this quite a bit on mechanical bourdon-tube pressure gauges, as well as some electonic Measuring and Test Equipment (M&TE). Hope it helps!

Grace and peace to all,

 

[Tom-D]

With a prop tach..

Nope, You can check the prop for accuracy, but to calibrate you must have a test bench and the instrument repair station certificate. You must open the case and adjust the tention on the Hysteresis disk.

 

[poadeleted20]

Regarding mag checks (and that is all I'm addressing in this post), you're not really looking for exactly 1725rpm, +/- 4% or whatever. What you're looking for is a drop of, say, 75 (or 50 or 100) rpm, in harmony with a slightly lower frequency in the sound of the exhaust.

There is a reason for doing the checks at a particular RPM, primarily having to do with carburetor functionality. If you do the check at a significantly different RPM, the book drop limits are no longer valid. How different is significantly different? I don't know.

 

[jshawley]

There is a reason for doing the checks at a particular RPM, primarily having to do with carburetor functionality. If you do the check at a significantly different RPM, the book drop limits are no longer valid. How different is significantly different? I don't know.

Thanks for pointing that out, Cap'n Ron, as I neglected to do so (and I also saw a typo error or two in my previous).

What's important is, say your mechanical tach has index marks for every 100rpm, with, of course, the 1000rpm index marks silk-screened a little differently, with the numbers duly printed, and the needle is indicating about 1/4 between 1800 and 1900rpm. By convention in the metrology discipline, all I can reasonably say is my rpm is 1850. I know it's 1825 by gut feeling, and even perhaps by a history of "reverse calibration"; however, I could not write 1825 down on a Test Report--only 1850.

This was why we tended to perform "reverse calibrations" on analog M&TE--we could adjust the Calibration Standard's output to center the needle on a cardinal point, then read off the Calibration Standard's output, which was at least one digit more precise, and was (usually) at least 4 times more accurate.

 

[john smith]

With a prop tach.

Sure, if you want to spend the money right away.
The cheap and easy test is to find a mercury/sodium vapor or fluorescent light on the airport.
At night, park the airplane with the tail towards the light.
Set the parking brake.
Start up.
Slowly increase the the throttle until the the prop appears stationary. Note the indicated RPM on the tach.
Slowly increase the throttle until the prop stops again, record the tach.
Repeat until you run out of throttle.

 

[N801BH]

Sure, if you want to spend the money right away.
The cheap and easy test is to find a mercury/sodium vapor or fluorescent light on the airport.
At night, park the airplane with the tail towards the light.
Set the parking brake.
Start up.
Slowly increase the the throttle until the the prop appears stationary. Note the indicated RPM on the tach.
Slowly increase the throttle until the prop stops again, record the tach.
Repeat until you run out of throttle.

You need to add that the light is a great way to confirm the tach is working ok. In America the powergrid is set to 60 cycles so every 60 rpm should stop the prop, ie, the prop should look stopped at 1200,1800, 2400 rpm and all the other multiples of 60. Very accurate.. Another thing to consider is some engine/prop/airframe combos have a rpm range that is prone to harmonics so if your tach is off 200 rpms and it is labeled not to run in that range and you cruise in the bad range you would never know the true rpm or if it is really trying to break something.. Bottom line is if it looks suspect treat is as such..YMMV

 

[Tom-D]

To calibrate is to adjust

Calibration of instruments
Calibration can be called for:

with a new instrument
when a specified time period is elapsed
when a specified usage (operating hours) has elapsed
when an instrument has had a shock or vibration which potentially may have put it out of calibration
whenever observations appear questionable
In non-specialized use, calibration is often regarded as including the process of adjusting the output or indication on a measurement instrument to agree with value of the applied standard, within a specified accuracy. For example, a thermometer could be calibrated so the error of indication or the correction is determined, and adjusted (e.g. via calibration constants) so that it shows the true temperature in Celsius at specific points on the scale.

You can only check accuracy not "calibrate".

 

[Tom-D]

Next question,

When does the FARs require your tach to be correct?

 

[jshawley]

"You can only check accuracy not 'calibrate'."

True, in the sense spelled out in your post. However, if we found the Unit Under Test close to being out of tolerance, we would, after having taken the "As Found" or "As Received" data, adjust it closer to nominal then take "As Left" data prior to returning it to service; if the "As Found" data were out of tolerance, we would also write up an "Out of Tolerance Report" to be sent to the end user for his evaluation and analysis/response and, of course, adjust the unit back into tolerance (and take "As Left" data).

 

[sba55]

There is a reason for doing the checks at a particular RPM, primarily having to do with carburetor functionality. If you do the check at a significantly different RPM, the book drop limits are no longer valid. How different is significantly different? I don't know.

In what way do the book limits matter in this particular instance? What does this have to do with the carburetor?

The only thing you're going to accomplish with this "runup" is to check that both mags are working and that the plugs are connected, working, and not fouled (although the latter never happens if you lean correctly). I do this at some RPM that also allows me to check the prop governor once. This is usually some RPM above idle somewhere, and it's certainly not as high as the book recommends (I like my prop).

-Felix

 

[poadeleted20]

In what way do the book limits matter in this particular instance? What does this have to do with the carburetor?

Bill Scott of Precision Engine explained this to me once a long time ago, but I fear I have forgotten the details. Perhaps Charlie Melot or someone else familiar with the precise aspects will chime in. What I do remember is that if you don't do the run-up at the right RPM, a problem could be masked within the "OK" drop range, or you could get a drop outside the "OK" range without having any problem at all. Ditto the carb heat check (if you have a carburetor, and Felix doesn't).

 

[Tom-D]

Bill Scott of Precision Engine explained this to me once a long time ago, but I fear I have forgotten the details. Perhaps Charlie Melot or someone else familiar with the precise aspects will chime in. What I do remember is that if you don't do the run-up at the right RPM, a problem could be masked within the "OK" drop range, or you could get a drop outside the "OK" range without having any problem at all. Ditto the carb heat check (if you have a carburetor, and Felix doesn't).

THe mag and plugs may fire good at idle, but not under a greater load.

 

[poadeleted20]

When does the FARs require your tach to be correct?

The regs require your tach to be operating properly at takeoff (91.213(a)) and during flight (91.7(a)). Further, if it fails in flight, you are required to "discontinue the flight" (91.7(b)). Lest we get into a "correct" versus "inoperative" discussion, the FAA maintains that malfunctioning is the same as "inoperative" (Administrator v. Bass, Administrator v. Bernstein).

The only thing I can't tell you is how far off it has to be before it reaches a legally unairworthy condition. I've flown planes with tachs that were known to be off 100-150 RPM after checking them with a strobe tach so I knew what I had based on what the instrument read -- kind of like a compass correction card. I'm not entirely sure the FAA would say that's within the letter of the law, but the errors were consistent and repeatable, so I felt comfortable with the safety factor.

 

[gismo]

You need to add that the light is a great way to confirm the tach is working ok. In America the powergrid is set to 60 cycles so every 60 rpm should stop the prop, ie, the prop should look stopped at 1200,1800, 2400 rpm and all the other multiples of 60. Very accurate.. Another thing to consider is some engine/prop/airframe combos have a rpm range that is prone to harmonics so if your tach is off 200 rpms and it is labeled not to run in that range and you cruise in the bad range you would never know the true rpm or if it is really trying to break something.. Bottom line is if it looks suspect treat is as such..YMMV

Just a note for clarity: 60Hz AC power causes 120Hz modulation of most light sources (electronic ballasts used in modern fluorescent lamps run at approximately 20KHz). A two blade prop illuminated by 120Hz modulated source will appear to stop with just two stationary blades showing at mulltiples of 3600 RPM all of which are too fast for most aircraft. At lower RPMs you get:

RPM 2bladeImages
900 4
1200 3
1800 2
2400 3

RPM 3blade images
600 4
800 3
1200 2
1800 4
1600 3
2400 1

 

[sba55]

Bill Scott of Precision Engine explained this to me once a long time ago, but I fear I have forgotten the details. Perhaps Charlie Melot or someone else familiar with the precise aspects will chime in. What I do remember is that if you don't do the run-up at the right RPM, a problem could be masked within the "OK" drop range, or you could get a drop outside the "OK" range without having any problem at all. Ditto the carb heat check (if you have a carburetor, and Felix doesn't).

True, I don't, but I'd love to learn more about this.

Does anybody know the details? How does the carburetor come into play here?

-Felix

 

[gismo]

True, I don't, but I'd love to learn more about this.

Does anybody know the details? How does the carburetor come into play here?

-Felix

I'm not buying Ron's position yet. First of all with a smooth running 4 or 6 cylinder HO engine, pretty much the only thing indicated by the RPM drop on one mag is the timing for that mag, if the drop is small the timing is likely advanced (bad for the engine) if it is large the timing is likely retarded (produces less power). But I don't think that a consistent 100 RPM tach error will affect this enough to matter. And the folks at GAMI and APS have convinced me that a mag check with a very lean mixture is far more likely to uncover a problem with the ignition system than anything else and the "normal" drops in that case are typically well beyond POH limits.

IME the typical 1700 RPM full rich mag check is only good for finding a dead magneto or a lead fouled plug, and either of those would show up on a fast idle mag check which is far less likely to damage the exhaust system. A carbon fouled plug will also show up with either but those are virtually always cleared during a full power takeoff roll long before they affect the takeoff in any measurable way. One might argue that the plugs are stressed more at the higher cylinder pressure associated with runnup RPM than at idle but the difference is rather small and not very meaningful IMO.

 

[Tom-D]

The regs require your tach to be operating properly at takeoff (91.213(a)) and during flight (91.7(a)). Further, if it fails in flight, you are required to "discontinue the flight" (91.7(b)). Lest we get into a "correct" versus "inoperative" discussion, the FAA maintains that malfunctioning is the same as "inoperative" (Administrator v. Bass, Administrator v. Bernstein).

The only thing I can't tell you is how far off it has to be before it reaches a legally unairworthy condition. I've flown planes with tachs that were known to be off 100-150 RPM after checking them with a strobe tach so I knew what I had based on what the instrument read -- kind of like a compass correction card. I'm not entirely sure the FAA would say that's within the letter of the law, but the errors were consistent and repeatable, so I felt comfortable with the safety factor.

How about at annual time,, 43-D Item 4

(4) Instruments—for poor condition, mounting, marking, and (where practicable) improper operation.


Could your A&P-IA legally sing the annual as airworthy with the tach as you say: "malfunctioning"

"Lest we get into a "correct" versus "inoperative" discussion, the FAA maintains that malfunctioning is the same as "inoperative"

But I also agree with your refferences.

 

[sba55]

I'm not buying Ron's position yet. First of all with a smooth running 4 or 6 cylinder HO engine, pretty much the only thing indicated by the RPM drop on one mag is the timing for that mag, if the drop is small the timing is likely advanced (bad for the engine) if it is large the timing is likely retarded (produces less power). But I don't think that a consistent 100 RPM tach error will affect this enough to matter. And the folks at GAMI and APS have convinced me that a mag check with a very lean mixture is far more likely to uncover a problem with the ignition system than anything else and the "normal" drops in that case are typically well beyond POH limits.

IME the typical 1700 RPM full rich mag check is only good for finding a dead magneto or a lead fouled plug, and either of those would show up on a fast idle mag check which is far less likely to damage the exhaust system. A carbon fouled plug will also show up with either but those are virtually always cleared during a full power takeoff roll long before they affect the takeoff in any measurable way. One might argue that the plugs are stressed more at the higher cylinder pressure associated with runnup RPM than at idle but the difference is rather small and not very meaningful IMO.

Lance, I agree 100%. Which is why I don't bother with "runups", other than switching to both mags sometime during taxi to see the EGT rise. In-Flight LOP checks reveal much more.

-Felix

 

[jesse]

Lance, I agree 100%. Which is why I don't bother with "runups", other than switching to both mags sometime during taxi to see the EGT rise. In-Flight LOP checks reveal much more.

-Felix

I don't think Lance was advising a complete lack of any run-up--as much as he was saying it isn't necessary to do the mag check at high power settings.

 

[sba55]

I don't think Lance was advising a complete lack of any run-up--as much as he was saying it isn't necessary to do the mag check at high power settings.

No he wasn't. I was just saying that the extension of what he says is the reason why a lot of folks, as well as APS, don't do a runup.

 

[poadeleted20]

Lance, I agree 100%. Which is why I don't bother with "runups", other than switching to both mags sometime during taxi to see the EGT rise. In-Flight LOP checks reveal much more.

Your method can mask a bad plug/plug wire/segment of the mag.

 

[gismo]

Your method can mask a bad plug/plug wire/segment of the mag.

Please elaborate on both what you believe to be Felix's "method" and how it would mask (leave undetected?) an ignition problem.