Mustang Prop Arrived

Homebuilt Aircraft & Kit Plane Forum

Help Support Homebuilt Aircraft & Kit Plane Forum:

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
16,833
Location
Memphis, TN
I enjoy Seguin’s videos. He is very conscious. Briefs the flight for himself as much as the owner. Has skills to understand the problem as much as fly.

The other one was a gear problem, trying to collapse. He temporarily saved it and was flying around waiting for the owner for advice, but cycling the gear overloaded the power system and killed the engine. There was foreshadowing on takeoff.

Both are sad to have happened, both had lucky human outcomes. Are both planes connected by the shop? Different jobs were done to both planes unrelated, but were they the same fingers of apathy?
 

Saville

Well-Known Member
Joined
Apr 28, 2014
Messages
472
Location
Boston Ma
You make it sound like this was the test pilot's fault.

For those not already knowing about this bird:
  • The plane had some hours with a smaller engine, Titan put in an LS, damaged it in ground runs, had recurring issues internal to the PSRU and engine (perhaps due to the ground run accident);
  • A new engine and new PSRU were on the bird. The owner was monitoring the PSRU closely. Plane was being test flown by test pilots;
  • The engine was running well on test flights by two different test pilots. No issues had popped with the new engine and PSRU;
  • On the test flight in question, the heretofore good landing gear started to fold one main on landing. Most of us would have had the subject accident then and there. Test pilot recognized the folding main and went around;
  • While making attempts to get the gear locked, a chunk of the electrical system went away, taking the EFII off line, presenting the engine failure and forced landing with one unlocked main gear leg;
  • Plane got on the runway with one main and tailwheel, bad main went back into the well;
  • Prop strike, wing tip hit, belly hit, off runway excursion and stop.
Issues identified after the fact should rightly lay at the feet of the system designer/builder. Issues found were:
  • Electrical system fault took the EFII off line. According to some really good folks, AeroElectric Connection's Bob Nuckolls among them, the only circuit elements between battery and each EFII box should be wires and one fuse or breaker, and probably should have a backup power circuit or entire backup ECU. This one seems to have had a circuit breaker protecting multiple circuits including EFII, and it opened at an inopportune time;
  • Gear is driven by a standard electro-hydraulic pump, but in addition to the normal pressure switches for keeping pressure on in either of up or down and locked positions, there seem to have been gear limit switches in the electric circuit for the pump. A small error in limit switch setting results in pump shut off without all three gear being down and locked. This system design, even with gear locked, prevents pump operation to hold the gear down and locked until the switch is deactivated. I would love to know the logic behind adding this switch and its failure modes;
If someone competent in system and wiring design had pored over the wiring scheme and done his own checks of the entire airplane, maybe these issues in circuit design, wiring and fuse/breaker sizing, pump cutoff switches would have been found and perhaps corrected. Would the owner even cotton to having a newcomer go through this already demonstrated airplane for faults in system design? Now that a rebuild is proceeding, I sure hope that the entire set of systems is reviewed with an eye to failure tolerant design.

One has to wonder if similar issues are present in other T-51's out there. If wiring system design is left to the builder - as it is in many homebuilts - then I suspect a failure analysis of as-built wiring and systems is a good idea when reviewing any homebuilt for flight by a new owner. Even if one won't make changes before further flight, the pilot should have schemes for managing failures when they occur.

A number of us recommend AeroElectric Connection. The book is excellent, and the list is ongoing support. All of the schemes will make for way less in the way of "dark and stormy night" tales. With electrically dependant engines, highly reliable and redundant power supply becomes essential.

Billski

Pretty sure you are referencing another T-51 and not TxFlyGuy's
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
8,906
Location
Saline Michigan
I was not even aware that they were different accidents.

Looked, Pantdino in CA, TXFlyGuy in Texas, now to edit posts and look into TXFlyGuy accident...
 

TXFlyGuy

Well-Known Member
Joined
Apr 25, 2012
Messages
2,314
Location
Republic of Texas
Both of these accidents were similar, in the electrical problem caused the CB to trip, killing the engine.

Lady Elaine, flown by Elliot, suffered very little damage. My plane, flown by William T. Koleno, suffered major damage.

He claims (boasts?) to have crashed no less than 12 T-51 Mustangs.
 

Appowner

Well-Known Member
Joined
Mar 30, 2021
Messages
84
Both of these accidents were similar, in the electrical problem caused the CB to trip, killing the engine.

Lady Elaine, flown by Elliot, suffered very little damage. My plane, flown by William T. Koleno, suffered major damage.

He claims (boasts?) to have crashed no less than 12 T-51 Mustangs.
Such a shame but certainly sounds like the test pilot is 100% at fault. But I'm curious, were you aware of the test pilots crash record with the T-51 before you gave him access? And if so, why in the world did you let him near it?

For some reason I hate to see a Mustang go in be it an RC model, home built or real deal. IMO there's a grace and beauty to that design that is unmatched by anything else in the air. Yep! I'm biased! :)
 

AdrianS

Well-Known Member
Joined
Jul 5, 2014
Messages
823
Location
Australia
Both of these accidents were similar, in the electrical problem caused the CB to trip, killing the engine.

Lady Elaine, flown by Elliot, suffered very little damage. My plane, flown by William T. Koleno, suffered major damage.

He claims (boasts?) to have crashed no less than 12 T-51 Mustangs.
That sounds like there's a serious design flaw in the layout of the electrical system.
The engine should still function no matter what else dies.
 

TXFlyGuy

Well-Known Member
Joined
Apr 25, 2012
Messages
2,314
Location
Republic of Texas
Such a shame but certainly sounds like the test pilot is 100% at fault. But I'm curious, were you aware of the test pilots crash record with the T-51 before you gave him access? And if so, why in the world did you let him near it?

For some reason I hate to see a Mustang go in be it an RC model, home built or real deal. IMO there's a grace and beauty to that design that is unmatched by anything else in the air. Yep! I'm biased! :)
This was the Factory Test Pilot. More hours in a T-51D Mustang than anyone on the planet.
No...I did not know about his number of crashes. But that is not the issue.
He chose to fly an airplane that was clearly not airworthy or in safe condition for flight.
And he did this against the orders of my mechanic.
 

Marc Zeitlin

Exalted Grand Poobah
Joined
Dec 11, 2015
Messages
1,119
Location
Tehachapi, CA
Sure, if you run on magnetos.
While I sympathize with your position and agree that in this case, the decision making (to go fly with known issues) was extremely poor and should never happened, the notion that an electrically dependent engine cannot be made substantially fail-safe from a power loss standpoint is specious at best.

Between the Aeroelectric Connection and just general FMEA practices, there are numerous known pathways to creating an extremely reliable and robust engine electrical support system.

I'll use my COZY MKIV as the first example. I have what is probably the most complex single piston engine electrical system there is, based on Bob Nuckolls' Aeroelectric Connection drawing Z-14 (IIRC), which is a dual bus, dual battery, dual alternator system with a cross-tie. My engine is electrically dependent, because I have two Lightspeed EI's. Other than the ground bus blowing up (an extremely low probability occurrence), there is no single point of failure, nor single component, which can cause the engine to stop working. Now, not everyone will want a system that complex, particularly if you don't fly IFR, but there are substantially simpler ways of achieving similar outcomes.

The Lightspeed EI instructions show a simple methodology to power one of the two EI's from a small backup battery (4 - 6 AHr), charged through a diode from the main bus. This battery does nothing but keep one EI running in the case that the main bus goes down or the master switch is turned off. It weighs about a pound, has a few wires and a switch, and will keep the engine running for close to 45 - 60 minutes - way long enough to get on the ground safely.

So, no - not just with magnetos. A properly designed electrical system would NOT have the failure mode that your plane experienced. Although the root cause of the crash was the idiocy of the test pilot, the crash would not have occurred with a proper electrical system architecture.

And I'd be happy to design that system for you for my standard hourly rate.
 

TXFlyGuy

Well-Known Member
Joined
Apr 25, 2012
Messages
2,314
Location
Republic of Texas
Any update on timelines etc. for the rebuild?
My mechanic and I are thinking 9 to 12 months.
But we are not simply rebuilding the aircraft. The center wing section is getting heavier ribs and outer skins. The fuselage side skins will be heavier also.
The plane will be better, and stronger than the original. A number of improvements are being made.

It will take at least 4 months to redesign and rebuild the center section.
 

TXFlyGuy

Well-Known Member
Joined
Apr 25, 2012
Messages
2,314
Location
Republic of Texas
While I sympathize with your position and agree that in this case, the decision making (to go fly with known issues) was extremely poor and should never happened, the notion that an electrically dependent engine cannot be made substantially fail-safe from a power loss standpoint is specious at best.

Between the Aeroelectric Connection and just general FMEA practices, there are numerous known pathways to creating an extremely reliable and robust engine electrical support system.

I'll use my COZY MKIV as the first example. I have what is probably the most complex single piston engine electrical system there is, based on Bob Nuckolls' Aeroelectric Connection drawing Z-14 (IIRC), which is a dual bus, dual battery, dual alternator system with a cross-tie. My engine is electrically dependent, because I have two Lightspeed EI's. Other than the ground bus blowing up (an extremely low probability occurrence), there is no single point of failure, nor single component, which can cause the engine to stop working. Now, not everyone will want a system that complex, particularly if you don't fly IFR, but there are substantially simpler ways of achieving similar outcomes.

The Lightspeed EI instructions show a simple methodology to power one of the two EI's from a small backup battery (4 - 6 AHr), charged through a diode from the main bus. This battery does nothing but keep one EI running in the case that the main bus goes down or the master switch is turned off. It weighs about a pound, has a few wires and a switch, and will keep the engine running for close to 45 - 60 minutes - way long enough to get on the ground safely.

So, no - not just with magnetos. A properly designed electrical system would NOT have the failure mode that your plane experienced. Although the root cause of the crash was the idiocy of the test pilot, the crash would not have occurred with a proper electrical system architecture.

And I'd be happy to design that system for you for my standard hourly rate.
As stated here, my system is a dual battery, dual bus system with the Efii Bus Manager.

The system itself is fine. Yes, there was an error in the wiring of the O2 sensors with the associated relay. The relay failed to do it's job.

Not to beat a dead horse, but my electrical system was rated at 40 amps max. On the crash flight, the amp meter was pegged at 50+. The ECU CB is 15 amps. It did not require any brains to predict what was going to happen.

So the Pilot-In-Command is either an idiot, or he did this on purpose.

We will look at a switchable battery backup, wired directly to the ECU.
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
8,906
Location
Saline Michigan
Both of these accidents were similar, in the electrical problem caused the CB to trip, killing the engine.
That is a sad correlation. On this thread: Phase One - Safety there are nine T-51 crashes documented. How many airframes are flying? I wonder how many of those are sudden stoppages with similar failure chains. Has anyone been counting the causes of them, looking for common threads, etc. If there is some hardware or wiring scheme or other issue routinely applied to T-51's that is resulting in sudden engine stoppage, it would behoove all of us to identify it, figure out fixes, notify the community of it, and stop wrecking these airplanes.

That sounds like there's a serious design flaw in the layout of the electrical system. The engine should still function no matter what else dies.
First I am worried over the fleet of T-51's. I support finding the source or sources or these shut downs, fixing them, and letting the community know how to greatly reduce the odds of having these failures. The devil is in the details. Detail needed here is data on the planes out there. Why did the crashes occur to them? Problem Solving has been around a long time, and bunch of recognized tools exist that all need DATA. Without data, we are guessing. Let's gather some data before we go leaping to conclusions:
  • Is the rest of the fleet built with the same system details and/or schemes and thus equally vulnerable? This is Search for Causes in Common;
  • Is any particular hardware or wiring schemes or other functional issues common to the crashes and not to the uncrashed? This is an Is-Is Not;
  • Was there a cascade of issues leading to engine failures? This is Failure Mode Analysis;
  • There are more and more will keep being invented...
Yes, single thread power to EFII is vulnerable, and redundant power sources, properly configured for high reliability, and automatically cascaded so it comes up without a lot of troubleshooting is a good thing. Is that the source of the other accidents? We do not know. So, anybody have any real data on other crashes or for that matter on configuration of any T-51's out there along with any unplanned engine shutdowns?

Going from worrying over the fleet of T-51's to worrying over TXFlyGuy's T-51...

As stated here, my system is a dual battery, dual bus system with the Efii Bus Manager.

The system itself is fine.
Given the data we have here, I must contend otherwise:
  • The engine cutout on takeoff;
  • Something was drawing a lot of power and no one has said yet what that something was or that it has been fixed;
  • Something caused a 15 amp CB on the ECU circuit to pop, and the source of that high amperage in that circuit has yet to be identified and fixed;
  • The only item reportedly changed is HEGO deletion, but TXFlyGuy says the HEGOs are self limiting on current, so the HEGOs either are not the high draw OR the HEGOs can draw big power. This would require some testing to check before believing you have already got the root cause of the high power draw;
Without the source definitively identified and definitively fixed, this failure is likely still out there waiting to cause the same trouble again. In addition, if one problem was built in, you may have more. Seriously, go looking for more, and then test extensively on the ground before lifting wheels again;

[/QUOTE]Yes, there was an error in the wiring of the O2 sensors with the associated relay. The relay failed to do it's job.[/QUOTE]

IIRC we were told that the Bosch HEGO's shut themselves off if current exceeds some nominal amount. That would preclude the HEGOs from drawing enough power to be our problem, and deleting the HEGOs does not seem to constitute fixing the source of the shutdown. That or the HEGOs in question have been checked and do draw big power under flight conditions, in which case you have closed the loop on HEGOs sucking the batteries dry. But both can not be true simultaneously...

Not to beat a dead horse, but my electrical system was rated at 40 amps max. On the crash flight, the amp meter was pegged at 50+. The ECU CB is 15 amps. It did not require any brains to predict what was going to happen.
I am concerned that you have way more confidence in the electrical system than it has earned... This is your airplane, and your butt will be riding in it. Please be skeptical of the electrical and EFII stuff.

Let's do a "for instance" check on the charging system concern. Let's say we had two 24 Ah batteries, that is most likely a little less than 48 Amp-hours of actual stored electricity. For argument's sake, let's also say 80 amps were being drawn while the alternator was putting out all it could. The alternator is fine with this and so are the batteries, except that they will eventually run out of power. With reasonable batteries, it would still take a half hour (48 Amp-hour/80 Amp is 0.6 hours) of running to draw the battery down and voltage to sag away towards shutdown of electronics. I suspect it took a lot less time than that to get airborne, but maybe the battery capacity is a lot lower than we are postulating. We might still have bus voltage data stored on the EFIS or engine monitor of bus voltage, and may be able to check out this path. I strongly suspect that ECU shutdown did not happen due to a power deficit in the airframe, but data will tell. Maybe the batteries have gone soft and the failure was drawn down batteries not an opened CB.

A circuit breaker opening has been cited as the circuit cutout to the ECU in the accident. Do you think this is what happened or do you know it from solid evidence? ECU CB open at the crash site with no others open would qualify nicely, but was it open? CB's respond to amperes passing through them in the very recent past. Modest overloads take a long time to trip, larger overloads take less time. The CB protecting the wiring to the ECU won't open because battery voltage is sagging or spiking or even melting down and leaving the airplane. If the CB was open at the crash site, replacement is a good idea, but more than that, why did it open? Usually the answer is current through it went high. Until you know why current through the ECU circuit went high and fix it, you are just waiting for a repeat of this event.

An alternate theory is that ship's power was doing strange stuff in response to the overload pulling on the batteries. Hmm. What scenario would make the CB on the ECU circuit pop? CB's and batteries are very resistant to spikes. Batteries will smooth out the spikes and CB's listen to current through them over time, usually by warming up a resister. The black box bus maintainer is another story. Something else likely had to cause the CB to open.

Another theory can exist that the Bus Manager has issues - it is a Black Box, we do not know what is inside it or what faults or undesirable interactions it might have.

Main battery contactors require amps to hold it on, but if the battery was drawn down enough to open the contactor, the ECU was probably out of the game too. Bob Nuckols is telling us that none of us should be powering our ECU solely from the switched side of the battery contactor. Got that? That is what always-hot battery buses and diode bank switching is for. Maybe that function is carried by the bus manager, but you would have to check that path too.

We will look at a switchable battery backup, wired directly to the ECU.
Maybe your troubleshooting brain can manage the airplane and get all the way to engaging backup ECU power aft the engine sags away at 200 feet, but most of us will already be on the ground before we can get that far. I much prefer power to run the engine automatically cascade and throw an alarm I can deal with once we get some safe altitude. Lots of good options exist to do this. AeroElectric Connection is a worthwhile read. There are multiple ways presented to making sure that the ECU is powered despite individual system faults.

Billski
 
Last edited:

Marc Zeitlin

Exalted Grand Poobah
Joined
Dec 11, 2015
Messages
1,119
Location
Tehachapi, CA
... Another theory can exist that the Bus Manager has issues - it is a Black Box, we do not know what is inside it or what faults or undesirable interactions it might have...
Semi-minor nit here - the FlyEFII Bus Manager (of which I've installed one at the owner's insistence, and worked on another) is not totally a black box - the internals and wiring diagrams are available on the FlyEFII web pages at:


Now, I'm not a fan of this thing - it gives SLIGHTLY more automated fallover to a backup battery without the full redundancy of a Nuckolls system. It's clear (mostly from direct inspection of the functionality, as well as discussions with the designer, Bob Paisley at FlyEFII) that no real failure analysis was done on the system (or any part of the FlyEFII system, for that matter).

But by analyzing the BM and it's functionality, as well as using all of the logic you already posited, it should be possible to determine why the engine failed, and that wasn't because the pilot took off - it was because of a problem with the electrical system that was NOT fail-safe.
 

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
16,833
Location
Memphis, TN
Essentially something like this needs parallel electrical systems. Options to tie them together for emergencies, but the engine needs its own 100% and have everything else run on the other concurrently. This plane is big enough to handle this level of redundancy. I wouldn’t even run the engine instruments on the engine buss unless they needed tied. If we were talking a VW powered plane, it’s too much, but this or bigger. Two equal of everything minimum. The engine shouldn’t know anything else is going on.
 

Vigilant1

Well-Known Member
Joined
Jan 24, 2011
Messages
7,630
Location
US
AeroElectric Connection is a worthwhile read. There are multiple ways presented to making sure that the ECU is powered despite individual system faults.

Billski
But by analyzing the BM and it's functionality, as well as using all of the logic you already posited, it should be possible to determine why the engine failed, and that wasn't because the pilot took off - it was because of a problem with the electrical system that was NOT fail-safe.
One problem is that many of the suggested improvements, as good and warranted as they are, would require modification of the wiring harness and some fundamental questioning of the circuit design itself. There seems to be a lot of resistance to this, as TXFlyGuy paid a company that he trusts a lot (about $10k-$15k) for that harness (and EFI programming). I'm sure for that price it is a thing of beauty, though the wires might not be going to the most suitable places.
Bold added:
The dealer made a big emphasis on the quality of their wiring, and wiring harnesses. Yes, he used the term Mil-Spec, as if it was a big deal. This is one big reason that I want to go with Fischer Motorsports.
And their experience in aviation is impressive.
The MoTeC wiring harness dictated that the sensors be connected here. This came straight from Hutter Performance when asked this question. There was no other option without an entire new harness design.
The Bosch O2 sensors have self protection built in. If there is a short, or an over amperage problem, or any problem with an individual sensor, they shut down and take themselves offline.

Yes, there are other options. But this would have involved a complete new wiring harness design. At the time, we did not feel that to be necessary.
Yes, there was an error in the wiring of the O2 sensors with the associated relay. The relay failed to do it's job.
 
Last edited:

PiperCruisin

Well-Known Member
Joined
Jan 17, 2017
Messages
393
Location
Idaho
I'm not great with electronics, but I recently had a high amp load on my aircraft. The source was a shorted ammeter field wire (rubbed against the engine mount).
 

Saville

Well-Known Member
Joined
Apr 28, 2014
Messages
472
Location
Boston Ma
One problem is that many of the suggested improvements, as good and warranted as they are, would require modification of the wiring harness. There seems to be a lot of resistance to this, as TXFlyGuy paid a company that he trusts a lot (about $10k-$15k) for that harness (and EFI programming). I'm sure for that price it is a thing of beauty, though the wires might not be going to the most suitable places.
Bold added:
TxFlyGuy

Many many posts in Phase One - Safety and now here are suggesting to you that putting anything else on the ECU power bus is a mistake.

As shown above, you've said that it would require a whole new harness.

My answer to that is - so what? A whole new harness would have prevented the crash - wouldn't it? The flap issue - while dumb to go flying with it in my opinion - did not cause the crash.

I think Post # 170 by 12Notes really says it all:

"Reset. There are people on here who have identified a problem with your, or Avionics 1st's wiring decision. They are trying to help. You have told us how it is wired..........

What this decision has done is guarantee that one type of emergency will absolutely cause an immediate much and more serious second emergency. This is easily fixed by moving just the power wires of one item to a different breaker. There is very, very little logical reason to keep it the way it is, it is a simple fix, and does not cause any further undue risk. ..............The same event, if the wiring is kept the way you have it, would result in the need for an emergency landing with a relatively high probability of damage to the plane, and a not insignificant possibility of injury or death.

In the risk/reward analysis, there is substantial risk increase for the reward of not moving two wires a few inches over, and possibly having to install a new breaker. It's your plane, you can tell us that you just don't want to do it this way, but don't defend it as a good or even remotely rational decision."

For the longest time you argued that the G3X is "essential".

It is not.

You are talking about a secondary system - well maybe that's ok but putting the ECS system on it's own bus and breaker is simple
and, in my opinion, necessary.

I wrote, in that thread that I, too, have faced major changes to a design mistake and was reluctant to do it. I fought and wrestled with myself and rationalized , but in the end did the work. So I know how you feel about that.

In post #52 above you are still talking about a switchable backup:

"We will look at a switchable battery backup, wired directly to the ECU."

You don't have time to think of the switch. S'ok to have it, as I say, but why not do everything you can to assure you won't need to think of it in the 3 seconds you have between engine stop and ground crunch?

Don't accept a band aid.

We are not arguing this because we think you are stupid. We want you to completely succeed. We're trying to help.
 

TXFlyGuy

Well-Known Member
Joined
Apr 25, 2012
Messages
2,314
Location
Republic of Texas
Thanks for the input. Everything is on the table at this point.

There were no major problems with the electrical system concerning current draw until after the 8 O2 sensors / heaters were installed.

Until then, normal amp readings were 19 to 24. On all engine runs, taxi tests, and the first few test flights.

After O2 sensor installation, amps were as high as 50.

The aircraft was fully powered on the final flight. All buses powered, essential and main. Even after the crash.

8 O2 sensors with heaters can pull 16 amps. They were wired into the circuit for the ECU. That is a 15 amp breaker.
 
Last edited:
Top