Shadin Miniflo-L malfuntion

[Peter T]

At startup, 2 flights ago, fuel flow started registering  0.0. Left it, tried it again the following day with the same results. I believe that the unit was factory installed in my 93 MSE. Will hopefully have a chance to check it out with my independant maintainance provider this weekend. Will hopefully find an obvious issue with the wiring from the transducer to the unit. Otherwise assume that the transducer needs replacing?

[donkaye]

22 minutes ago, Peter T said:

At startup, 2 flights ago, fuel flow started registering  0.0. Left it, tried it again the following day with the same results. I believe that the unit was factory installed in my 93 MSE. Will hopefully have a chance to check it out with my independant maintainance provider this weekend. Will hopefully find an obvious issue with the wiring from the transducer to the unit. Otherwise assume that the transducer needs replacing?

More than likely the transducer.  I have had that happen before and it was the transducer that needed replacement, then recalibration of the K Factor.

[Yetti]

I think the blue dye gunks up the impeller.   Should be able to clean it with carb cleaner.   or wiring issue.

[larryb]

Well I have both the miniflow and transducer removed for upgrade. Let me know if you need one or the other.

[Piloto]

Try soaking the transducer with carburetor cleaner. It works very well.

José

[Peter T]

 So I get to the airport today, and start it up. The miniflo comes to life and works perfectly.  Went flying and continued to work as always before. It is likely a connection issue and will look next time I have a reason to remove the cowl.

 Thanks for all the help.

[donkaye]

43 minutes ago, Peter T said:

 So I get to the airport today, and start it up. The miniflo comes to life and works perfectly.  Went flying and continued to work as always before. It is likely a connection issue and will look next time I have a reason to remove the cowl.

 Thanks for all the help.

Expect it to go bad again soon.  I say that from experience.

[David Lloyd]

For a year I had one that sometimes worked fine, sometimes zero.  Wasn't the wiring.  Replaced transducer and everything great.

[Vance Harral]

Concur with Don and David.  It will likely quit again, and it will likely be a failure of the transducer itself.  Ours exhibited this exact same intermittent failure pattern, and was ultimately fixed by replacing the transducer.  I did look carefully at the old one after removal, but I was never able to establish if the impeller was hanging up, or the connective wiring cracked inside the potted epoxy.  Solution is the same either way: replacement.

[M016576]

3 hours ago, donkaye said:

Expect it to go bad again soon.  I say that from experience.

Yep- same happened to me- intermittently failed indications, I thought it was electrical- cleaned the contacts, etc, but it was in fact the cube failing.

Edited October 6, 2018 by M016576

[PT20J]

The '94 J I purchased last summer is equipped with a factory installed Miniflo-L. The fuel flow would intermittently drop to zero when I bought it. The previous owner had sent the panel unit back to Shadin and it checked out OK and has the latest software version. I replaced the transducer with a new Floscan 201B from JPI (which now owns Floscan) and set the Miniflo-L K-factor to 28.9 to match the new transducer. It seems to read about 6% high, which surprised me as the transducer is rated at 0.5% at 16 gph with 1% linearity. 

I note that the Mooney factory installation varies from Floscan mounting requirements. Floscan says to mount with the wires facing up; Mooney mounts with the wires facing down. Floscan says no sharp bends within 6 inches; Mooney uses a 45-deg elbow at the inlet side.  

Of course, I can easily tweak the Miniflo-L K-factor setting to match actual fuel use. But, I'm suspicious of a digital system being so far off and I wonder if others have experienced this with the factory setup.

Skip

[Peter T]

On my 93 J the transducer is mounted with the wires up with a 45-deg elbow at the inlet. I would not consider 45 degrees to be a sharp bend. 

Also, I cleaned and back flushed the transducer with brake fluid, and it has not acted up since.

[PT20J]

16 minutes ago, Peter T said:

Also, I cleaned and back flushed the transducer with brake fluid, and it has not acted up since.

Soaked mine in Hoppes No. 9 and it worked better but still would sometimes hang up. FWIW, a tech at Shadin said that although Floscan claims 10,000 hours on the transducers, he's regularly seen them get flaky after about 20 years.

[carusoam]

Expecting linearity from an analog device can lead to a surprise...

liquid flow in pipes and tubes is very similar to airflow over the wing.  

Laminar flow leads to predictable results.

Things that get in the way of laminar flow... bends and elbows... even 45° bends... near the sensor...

who would think those little stall strips would do anything...? 

In plumbing, an L/D ratio of 10:1 is often used to have laminar flow at the sensor... but who’s got that much room under the cowl?

That also goes for the exit side of the sensor... believe it or not.

Around here... sensors are known to need cleaning after decades of use...

Other things of interest that get in the way of good data... include air bubbles entering the system...

So if you see some fuel stains up stream like at the selector valve, you are probably getting air entering the system, messing with the results...

K factors can be set pretty close from the factory, but a good calibration is still expected... using the bracketing technique... to narrow in on the final solution...

PP thoughts only, and stuff you can read about on MS...

Best regards,

-a-

[PT20J]

1 hour ago, Peter T said:

On my 93 J the transducer is mounted with the wires up with a 45-deg elbow at the inlet.

Interesting. Do you know if this is a factory installation? It doesn't match the IPC which shows wires down.

[carusoam]

The sensor is a paddle wheel with a magnet in the vanes...

The sensor is a simple pick up that recognizes the magnet passing...

The wires transmit the pulses back to the panel where the meter counts the pulses related to time.... the k-factor is used to convert pulses over time to gph...

Rightside up vs. upside down might have something to do with bubbles vs. gravity... But is more likely cleanliness of installation... wires down would keep moisture, oil, or fuel from collecting near the wires...

Later installations moved the FF sensor away from the engine... avoiding the vibration that comes with that... the FF doesn’t change anywhere along the line... unless there is a leak...

PP thoughts only..

Best regards,

-a-

[PT20J]

Here’s a link where someone took one apart. Forgot one more thing: Instructions that came with transducer said not to mount it to the engine. I might try removing it from the engine mounting and rotating it 180 degrees. Still interested if others have run into this problem with the Mooney design. 

https://www.euroga.org/forums/maintenance-avionics/3929-shadin-floscan-201-fuel-flow-transducer-internals

[carusoam]

Photo cell and cleaning make a lot of sense... so much for the magnetic theory... 

And... that pinpoint being down on the bearing surface makes sense for gravity... there aren’t two needles, one on each side?

Best regards,

-a-

[jcovington]

6 hours ago, PT20J said:

Still interested if others have run into this problem with the Mooney design. 

My 1996 J model had the fuel transducer hard mounted to the engine with the wires facing down (original from the factory). In the first few years of ownership I replaced a couple of fuel transducers and put them back on the hard mount with the wires facing down. Joey Cole convinced me on the third iteration that the transducer was mounted incorrectly and I should change to the JPI recommend mounting. He removed the transducer from the hard mount and let it free hang per the JPI installation manual. That has been about a 1000 hours/8 years without a failure. I believe this is a case where Mooney got the design wrong and you should follow the JPI requirements.

Jim

[thinwing]

I wouldn’t use an aggressive solvent like hoppes or especially carb cleaner because of damage to the plastic roter

[PT20J]

A lot of folks on BeechTalk recommend it with no ill effects, so I thought I'd give it a try. Since the transducer wasn't working reliably anyway,  I didn't have much to lose. As one of the BeechTalk guys pointed out, Hoppes comes in a plastic bottle .

[Yetti]

18 hours ago, PT20J said:

A lot of folks on BeechTalk recommend it with no ill effects, so I thought I'd give it a try. Since the transducer wasn't working reliably anyway,  I didn't have much to lose. As one of the BeechTalk guys pointed out, Hoppes comes in a plastic bottle .

The new stuff does.  The old stuff came in a glass bottle.

[PT20J]

I've been doing some research and I've decided that the Mooney transducer mounting is all wrong. It's difficult to get any good technical support now since JPI bought Floscan. I can't find anyone at JPI that knows anything about the transducers, and the Floscan phone number just rings a salesman. Shadin tech support just defers to Floscan. There is actually some pretty good info on Floscan's website for the marine versions (most of their business). The marine version is essentially the same design but it's cast from zinc instead of aluminum and they don't come with calibration data according to the Floscan salesman. So, combining information from the marine installation instructions and the aviation installation instructions and some installation instructions from Van's, here's where I think Mooney got it wrong:

Per Floscan marine installation instructions: "Flow Sensors must be placed in a horizontal section of fuel line at a low point in the fuel system. Fuel should travel, “Up-hill” when exiting the sensor. Its outlet should be at least 1 or 2 inches lower than the fuel pump inlet." This is, I believe, why it's supposed to be mounted with the wires coming out the top. The outlet is slightly higher than the inlet when mounted this way. Mooney mounts it upside down. The Mooney location also puts the transducer outlet at about the same height as the injector inlet. The instructions that come with the aviation version say this is OK, but I think lower would probably be better.

Other less severe mounting issues with the Mooney design are that it is mounted to the engine whereas the instructions that come with the transducer suggest not mounting it to the engine to reduce vibration. Also, it's a good idea to mount it in a straight section of fuel line - Van's suggests at least 6" of straight line on either side - and Mooney puts a 45 degree fitting on the input. Speaking of fittings, Mooney IPC calls out aluminum fittings. Bad idea to put aluminum NPT fittings into an aluminum casting as it can gall. Steel fittings are specified by Floscan and should be used. All Floscan installation instructions (marine and aviation) say not to use Teflon tape which certainly is appropriate since you don't want strands of tape clogging the transducer and injector inlet screen. However, the marine instructions say to use a fuel proof sealant on the NPT threads whereas the aviation instructions say not to use any sealant or you will void the warranty. This makes no sense. Pipe threads aren't designed to seal without some sort of sealer as there's a spiral channel between the mating threads. I'll take a voided warranty over a fuel leak under the cowling any day.

Byron did a very nice installation that seems to meet all the Floscan requirements. Look toward the end of this thread:

Skip

[carusoam]

1) For the fuel to go up hill at the exit... a proper mechanical arrangement will be needed...

 

2) The reason for this angle is to most likely, aid in the elimination of any air bubbles away from the sensor....

• a down hill arrangement will have bubbles trying to come back to the sensor...
• an uphill arrangement... the outlet tube should be higher... than the inlet tube...
• if there are no bubbles... the sensor won’t notice if it is going uphill or down...

3) If going with Teflon tape... it is possible to use extra caution during the application to avoid loose teflon strings entering the fuel system... with the conic shape of the NPT threads, the seal occurs along the length of the mating threads, not just the first few...

4) according to the pics posted, The sensor seems to have a single pointed ‘needle’ bearing(?) on only one side of the impeller. This would indicate that gravity may have an effect on the impeller design... less noticeable with fuel in the line. Blowing air through the line (for testing) it might be more noticeable....

5) The 10:1 entry and exit tube lengths are a ‘typical’ need for this type of instrument... it is possible to demonstrate that fuel flowing through a 45° fitting has a minimal effect... once this is demonstrated, everyone can copy the layout and get the same results....

6) The challenge is to be the first person doing a new instrument layout.  Lots of risk of it not working as planned... and difficult to figure out why... or what to do next...

PP thoughts only, not a mechanic.

Best regards,

-a-