discussion- J230D potential electrical fire - firewall penetration

[Thruster88]

This from RAAus accident and incident. Perhaps the fix should become a fleet wide thing, a solenoid to isolate the battery, yes please.  Sorry to sound like a broken record but every GA aircraft has to have that.   

 

26/4/2018	Gawler Airfield	SA	Jabiru	J170D	Jabiru	2200B	The aircraft was destroyed in a hangar fire and another aircraft that was parked next to this aircra...
The aircraft was destroyed in a hangar fire and another aircraft that was parked next to this aircraft was heat damaged and written off. There was also smoke damage to the hangar. Investigation showed fire started in the engine bay. It is suspected that the voltage regulator failed. All Jabiru aircraft owned by the Club have now been fitted with solenoids, to isolate the battery from the electrical system.

[Kenlsa]

T88,

the fire was quite exciting with 170, FK and Grob 102 along with the hangar damaged or destroyed.

All our fleet are trickle charged when not in use. All the holes lined up that afternoon; the smart charger turned out to be not so smart, the battery cooked leading to a small fire and the fuel tap was left on so the heat destroyed the fuel line and the fire was fed by the gravity fed tanks. All over in a few minutes.

 

So solenoids fitted and taps off at the end of the flight.

 

only problem now is I get zapped on my watch band if I’m not careful when checking the air filter!

 

Ken

[facthunter]

Monitoring battery temp is a good start. . Not difficult either. .Lead acid batteries left discharged will sulphate up which for all practical purposes is the end of it's useful life. ALL batteries contain energy. There's enough in a charged one to make a lot of heat and it can be from internal shorting.. Nev

[RFguy]

So, what does have to be dealt with with a total battery isolator is that the alternator-charging system must not be allowed to push the voltage up and damage the avionics

 

- even 1 millisecond  would be enough to do damage.

 

There are two ways around this

1)  have a battery isolation contactor and a charging system contactor. The charging system MUST be disconnected FIRST before the battery, otherwise the avionics will likely be damaged, The difficulty is that the alternator-charging contactor MUST let go first. must must must. Enforcing that could be done with the right combination  of contactors from a single switch.

 

HOWEVER- the alternator charging system CANNOT be unloaded without the possibility of damage and or overload in the charging system, so it needs a "load diversion bank". This can be in circuit all the time and would eliminate spike based avionics damage.. A diode and a bank of ceramic resistors will do it.  This would mean an alternator contactor does not have to be precisely timed.

 

In some aircraft the  alternator can be switched off, this is usually with field exciter systems, not permanent magnet systems like the Jab that must be loaded up. 

 

BTW - I believe this oversight of an unloaded permanent magnetic alternator to be a very common oversight of even experienced avionics people.

 

Edited December 26, 2020 by RFguy

[M61A1]

Why not disconnect the alternator phases before the regulator?

[old man emu]

Lots of talk about causes and cures, but has anyone sent in a report to CASA and RAAus to make it official?

[RFguy]

21 minutes ago, M61A1 said:

Why not disconnect the alternator phases before the regulator?

Because the alternator windings is a constant current source and that is what pushes up the voltage.

 

If you open circuit the windings, the constant current behaviour of the inductive component will attempt to keep the current going and ramp up the voltage until something gives/arcs.

 

Ever had a relay coil you had your fingers on give you a bite or a kick ? Well - multiply that by 100 for energy available here.

 

The winding type, magnet type, rate of change (magnet speed past the coils etc) will in practice limit the maximum voltage reached, but it is likely to be maybe 50 to 100V peak. This will depends a bit on RPM etc. The alternator produces energy, you have to do something with with it. If the windings are disconnected this also may saturate the cores (due to high volts per turn) and thus bias up the cores with an  unwanted  result.

 

Edited December 26, 2020 by RFguy

[skippydiesel]

17 hours ago, RFguy said:

OldK and Skippy

I will expand on your questions etc on this tomorrow. Star Trek wifey time.

  

Waiting????

[Blueadventures]

43 minutes ago, skippydiesel said:

Waiting????

Bit pushy, he has until midnight and still have answered ‘tomorrow’.

[RFguy]

ha ha. EMily's birthday today. have my mum here looking after the kids.

 

Rotax : Well, from what I read, similar issue with voltages going up with unloaded. The regulator is a bit slow. The 25,000uF capacitor slows everything down to some degree, and most certainly smooths out the high frequency components. but that cap is a BAD design fix  if the regulator to battery circuit is switched, because if it is switched, when the cap  gets connected across the battery, it will (almost) weld the contacts of that switch, be it a tooggle switch or a contactor/relay

Moat certainly, use a 105deg C rated capacitor. Not a 85 deg C one.

 

Again, the bus needs a fast dissipative voltage clamp like the jab (using the existing regulator)  

 

Oh and that cap that was recommended in another forum a few years ago was both under voltage(25V)  and under temperature- only rated at 85 deg C... might last 250 hours at 85 deg C and the high ripple current. Needs to be a 40V cap if there is no dissipative voltage clamp.

 

Edited December 26, 2020 by RFguy

[skippydiesel]

On 25/12/2020 at 5:33 PM, RFguy said:

well not that you know, or have checked. I would think it equally applies to all.

Anyway, this is a JABIRU forum. no talk of rotax here, thanks. please start another topic or PM

🙂 🙂

RF - I was not trying to subvert this elitist Jab thread 😁 just letting readers know that, seven years after purchasing my SSB RB16CL-B, the battery appears to be still taking a full charge (14.2 V) and delivering nice crisp engine staring. I dont know why this would be - can only assume its something to do with superior Rotax😏 technology.

 

(Star Trek starting to be a bit repetitive - would not dare make a comment on" wifey🤣")

[RFguy]

Skippy, you are not entirely correct about the rotax setup and wiring. It is wrong also.

 

While I am sure the big capacitor on the rotax would help power bus noise down, if  it reduces noise in radios that indicates to be a poor regulator system, and also that the wiring is in error.

 

And if that 22000uF capacitor is down stream of a switch that is wrong wrong wrong also.... numerous gross basic electrical errors.

and.....

alternator + and - must go to the battery terminals.. direct wire (via switch or contactor or fuses etc) . NOT via some intermediate panel.

 

If a common firewall negative is used this must be connected to the neg battery terminal with something like 0 or 2 AWG short cable, and have a large copper bus bar

NOT a stack of ring crimp terminals stacked on top of each other on a single bolt  !! ROFL.

 

I will fix my Jabiru, and show how it is done right, with the minimal of work.  Same goes for ROTAX airplanes I suspect. wrong wrong wrong.

 

And the big stainless firewall is not a great ground, either, stainless steel may have considerable resistance depending on its composition.

 

Forgive me for my flippance. I am amazed just how poor LSA manufacturers understand electrical wiring techniques.  It's attrocious.

 

No wonder electrics in planes has numerous idiosyncracies and poor reliability.

mutter mutter !!!!

Edited December 26, 2020 by RFguy

[Old Koreelah]

15 hours ago, RFguy said:

Rotax : Well, from what I read, similar issue with voltages going up with unloaded. The regulator is a bit slow. The 25,000uF capacitor slows everything down to some degree, and most certainly smooths out the high frequency components. but that cap is a BAD design fix  if the regulator to battery circuit is switched, because if it is switched, when the cap  gets connected across the battery, it will (almost) weld the contacts of that switch, be it a tooggle switch or a contactor/relay

Moat certainly, use a 105deg C rated capacitor. Not a 85 deg C one...

Glen when you write your recommendation for the Jabiru charging system, perhaps you could dumb it down a bit for blokes like me.

 

I have a fuse on one feed wire from the alternator, a PowerMate regulator, a Lithium battery and there is also a battery isolator. The maker of PowerMates seems to have retired and all I can find is the blurb below, taken from the Spruce cattle dog.

 

My understanding is that the PowerMate limits charge to 8A (about 100w) and the sort of spikes found with other regulators are ironed out, protecting sensitive instruments.

 

In the few years I’ve had this regulator installed with my LiFePO4 battery I’ve done a few longer trips without apparent damage. In recent times, no flights longer than an hour. The Battery voltage seems to max out at 13.7v.

 

I presume during those flights most of the 100 watts being delivered to the battery is used up by instruments, strobes, wig wags, recharging iPad, radio, etc. If so, my Li battery should be safe from “overeating”.

 

Now I need to know how to prevent damage to my Jab alternator in the event that, in flight, it gets disconnected from the battery.

 

 

[RFguy]

Old K, I probably need to do some drawings to make it easier to understand, and I also likely need to study some GA certified aircraft so I do not get too far away from canon.

 

The first point I should make is that there are voltage spikes and current spikes.

 

Current is something that has to be held in check on an average basis.

This is because things take time to heat up.. A high current for a short period might be just fine if the time averaged current is below the maximum.

 

Voltage spikes are what damages electronics instantly.

 

I dont know anything about the Power Mate but it sounds like it is doing the right thing- limiting the voltage to about 13.8V and 8A for your application . 8A might not be enough for some ops. Powermate  says "14.5V" so if you system is only getting to 13.8V ever, probably not fully charging the LIFEPO4 (14.4) . but it can.

 

The 13.8V is good for LIFEPO4 batteries but for lead acid, the temperature compensation thing rears its head and all aircraft seem to fail here- they dont have temperature compensation for the lead acid battery, which leads to overheated and damaged batteries ALWAYS.

 

Your voltmeter will read the average voltage, but will not tell you about high voltage spikes ("excursions") on the bus.  The battery is supposed to be the BIG capacitor. If it is doing its job then anexernal capacitor is not required. This comes down to enforcing all power to go via the battery. The alternator charging goes to the batteris + and - terminals.

and the rest of the aircraft goes also from the + and - terminals

 

this way, the power is forced to go via the battery- IE via the capacitor which is the battery in this case. 

 

 

 

 

 

 

 

Edited December 26, 2020 by RFguy
added 14.5V info on power mate

[Blueadventures]

Hi RF these are some past forum comments about the Capacator by other Forum members.

 

My old Skyfox Ca21 had a capacitor that went u/s at 490 hoursRotax UL80 hp)  and had the brown stuff leaking and a noisy radio, replaced the Cap like for like and radio quite again.

 

Past member comment that may assist;-

 

"What surprises me also is that most of the RA aircraft do not have any spike protection on the starter solenoid (only needs a simple 1N 5400 diode), & in the case of Jabbys especially, do not have a battery master solenoid as well"

 

 

"Electrolytic Capacitor which should be on the bus output wire from the regulator, that will clean up the dirty ripple current from the alternator & make life better for radio ops - all the Rotax engines have them."

 

"The Rotax manual it says we recommend a 22,000 uf cap. I have seen quite a few rotax installs without it. It also protects the regulator if you turn off the master before the mags...apparently you can take out the regulator easily if you don't have the cap there."

 

Cheers.

[RFguy]

Warning- using capacitor in Rotax-

for those who connected up their 22,000 uF capacitor across the charge regulator output-

A warning -  this is  a big, combustible, smoking bomb across your system without a fuse.

When that thing lets go, it will put out more white smoke than a tear gas cannister.

 

If you cannot rewire your aircraft to have correct wiring  method to avoid need of the capacitor- 

- the capacitor  can be wired inline with the regulator output

 

this way the current is forced past the capacitor so it can do its job.

The fuses going each way protect the aircraft and the cap.

 I will do a drawing

 

 

[RFguy]

drawing and new topic at

 

 

[RFguy]

Hi Mike

I'll go through these one by one : 

 

"My old Skyfox Ca21 had a capacitor that went u/s at 490 hoursRotax UL80 hp)  and had the brown stuff leaking and a noisy radio, replaced the Cap like for like and radio quite again."

GE : capacitor likely under rated for the job. quite a hard job for the capacitor. Suggest 105 deg C rated capacitor, and use two capacitors in parallel, each of half the value to do the job for best life.  Or use a 'ripple rated' capacitor. 

 

""What surprises me also is that most of the RA aircraft do not have any spike protection on the starter solenoid (only needs a simple 1N 5400 diode), & in the case of Jabbys especially, do not have a battery master solenoid as well""

GE : 1N4004 , 1N5404 etc  is certainly better than nothing, spikes from the solenoids certainly damaged electronics... but the simple diode on the coil however can increase solenoid release times by x10, so best to have a rectifier diode in series with a 15V TVS or zener diode to permit the contacts to open fast, otherwise contacts will be progressively damaged 

 

"Electrolytic Capacitor which should be on the bus output wire from the regulator, that will clean up the dirty ripple current from the alternator & make life better for radio ops - all the Rotax engines have them."

GE : Well sort of. see my drawing at the new topic post.  Needs to be fuses to and from, and needs to be ripple current rated,  and temperature rated, at least 105C, because an engine bay is a very tough place for an electrolytic capacitor. 

 

"The Rotax manual it says we recommend a 22,000 uf cap. I have seen quite a few rotax installs without it. It also protects the regulator if you turn off the master before the mags...apparently you can take out the regulator easily if you don't have the cap there."

GE : It DOES NOT protect the regulator if you turn off master before the mags !!!!  Not true. It may reduce the voltage spike amplitude depending on the speed that the engine shuts down, and the speed of the switches and switching contacts.  Certainly it will reduce the ripple voltage on the bus if the wiring has not been done correctly (perhap by design) to avoid common bus charging currents....It will likely damage the charging system switch (if there is one) as it is like putting a short circuit across the contacts -  the huge  capacitor looks like a super duper short circuit before it gets charged up.....

 

[RFguy]

anyway, time for everyone to read

 

"AC 21-99 Aircraft Wiring and Bonding"

 

[Yenn]

Why do you need a master solenoid. I have built two aircraft and deliberately did not use a master solenoid?

How do you get a voltage spike from the starter motor? I would have thought that the massive drain would have removed any chance of a high voltage. There is never 12V in the circuit when a 12V starter is engaged.

[RFguy]

Hi Yenn

As you know, (so I am writing for others)  the idea is to disconnect the battery as close to the battery as possible. In the event of twisted metal, reducing the number of components that are compromised and also connected to B+ is a priority.

Circuit breakers should take care of any downstream over current situation. The usual rule applies of course- circuit breaker must be rated at less than the smoking result of any intermediate wiring.  If the wire goes through the firewall direct from the battery , and a short occurs as it passes through the firewall, OR underneath the instrument panel somewhere, there would be no way to stop the ensuing fire..  Hence the battery isolation switch. 

 

Most certified aircraft have a single battery isolator for everything.

I've also seen two solenoids- one for MASTER and the other that goes to the starter motor.  I feel the one that goes to the starter motor via the starter solenoid does the same job- as long as the distance to that solenoid is as short as practical to reduce the vulnerability of the wire. Sometimes battery fuses are used at the terminal. Having the starter solenoid not through the master solenoid certainly reduces the load requirements of that contactor.

 

As for the starter causing spikes- the starter solenoid, actually.

 

This one depends on the interruption capability of the starter pushbutton switch. Many switches cannot interrupt the solenoid coil current and voltage that is produced when the coil energy is removed, and few hundred volts can jump across the switch onto the main bus .... 

 

-glen