What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Mixed Battery Chemistry, Common Buss?

Toobuilder

Toobuilder

Well Known Member
I?ve been looking at various schemes to provide backup power to my electrically dependent airplane. I?ve considered everything from airliner style dual busses so I can press on with 100% capability in a failure, to dual alternators with somewhat reduced capability, and considering my mission, have settled on a simple backup battery that handles ONLY my engine needs for an hour or so. I?m not here to debate risk posture, so let?s try keep that in check.

I am exploring the technical drawbacks of mixing battery chemistry on the same buss however. My thinking is that I?m going to run the airplane with a single alternator and battery just like we?ve been doing for decades. However, I?d like to add a smallish Li-Ion (or other suitable chemistry) battery in reserve to power the fuel pumps, ignition and engine CPU if the ship?s main system goes Tango Uniform. Think of this as a scaled up EFIS backup battery. I?d rather not have the various auto isolation methods taking care of tending the battery though ? I want a stone reliable toggle switch that connects the battery to the engine buss through a single circuit. I?m not opposed to charging the battery once a month on the ground and keeping the backup strictly isolated from the ship?s main buss, but I think it would be beneficial to have a ?charge? position on the toggle so that when skies are blue and all is well, I can use the ship to keep the backup topped off. Most of the time the switch would be in a ?NORM? position ? backup battery completely disconnected.

I am only considering the more recent, exotic chemistry because this battery will be 100% dead weight until an emergency so I want it to be as small as possible while meeting the power requirements. I have only scratched the surface of the various battery offerings and know only that some of these batteries have different charging requirements WRT input voltage and amperage.

So long story comes down to a simple, broad question: Is anybody running mixed battery chemistry on a common buss, and what are the cautions/concerns/mitigating actions?
 
Michael,

I've been doing some related work but it's sort of a back burner project right now since I'm trying to focus on building my fuselage.

I designed a small circuit board that includes a battery charger circuit and ORing diodes which have been implemented using MOSFETs for higher efficiency. I had planned to use a gel cell (like a PS-1290) for my backup battery. My plan is to have both batteries isolated by the ORing diodes and the backup battery charged using my charging circuit. My charger design is limited to a few amps of charging current so it will only work for smaller batteries. I wasn't planning on lithium but that may be a possible future extension.

One other benefit of my charging circuit is it is a buck-boost configuration so it will try to supply charging current even if the input voltage is less than the battery voltage. The idea being to use as much power as is available for these emergency backup operations when the power demand has been reduced.

When I'll have my circuit working is TBD but I haven't been working on it lately unfortunately. My charger was getting a little warmer than I liked so I need to track that down. Otherwise it seems to work as expected.
 
I posed a similar question to Robert at EFII regarding use of mixed chemistry and the Bus Manager. He told me it was fine to have for instance a PC680 as primary and an EarthX as secondary. They are kept isolated except for those instances when you choose to use both batteries for starting which would not cause a problem as they would be isolated again as soon as the engine starts.

Without the Bus Manager I am sure there is a way to do this with Diodes and I am attracted to keeping it simple also. Interested in hearing from others that have done this...
 
I think you do need to be careful mixing battery types with a simple system, due to small but important differences. For example, a conventional wet cell lead acid battery works fine in a system with the alternator set at 13.6-13.8 volts. OTOH Odyesse recommends alternator voltage be set above 14.0 for their batteries.
 
While I would like the option of adding the backup battery to assist with the engine start duties on occasion, its the charging that has me the most worried. Despite Robert's assurances, my research indicates that there can be significant differences in input voltage and charge rates between chemistry types. That is what I'm trying to nail down.
 
One method:

I have read that the batteries should not be mixed on the same buss - that could extend to same systems too? It might work if you have both hooked up thru their own contactors so they are never on line at the same time.

Changing from one to the other would require that the entire elec system be shut down and then 'rebooted' - but isn't that the situation you want to be able to recognize and rectify with the #2 battery? In other words, you would use #2 only if the main battery/system fails?

I'm setting up a 2 battery system in my next build, but it will use 2ea EarthX units with two contactors. I can use the #2 for checking the AWOS and getting my flight plan set up etc, and then turning on #1 for start-up. The architecture will not allow for the #2 to help with starting...so far...

I am neither an EE nor a person who stayed at a Holiday Inn last night. My ideas are highly suspect!

Carry on!
Mark
 
Mike, I've been thinking about the same stuff, as I'll install a second EI this winter. My conclusion is (1) go old school, or (2) go whole hog.

Old school is a main bus PC925 and a 2.5 to 4 AH AGM battery, diode isolated, dedicated to a single ignition. Whole hog is the same scheme, but using blue batteries.

Here's the thing...when discussing main batteries, the blue battery saves a lot of weight. However, when adding a small battery dedicated to a single EI, there isn't a lot of weight to be saved.

I like the 925; zero maintenance, lots of reserve, and at 28 lbs, good for CG with the 390 and a Hartzell. No flashing lights, no special care. I'm inclined to keep it.

A 2.5AH AGM Motobatt MTB4BB is 2.31 lbs, while a 3.8AH MB3U is 2.95 lbs. They're $32 and $38 on Ebay.

Bet I can tell, the smallest EarthX is an ETX12A, rated at 4AH. It weighs 1.3 lbs. They are $169.

If I was willing to pair an AGM with a lithium, I'd only save 1.65 lbs (2.95 - 1.3). If I drop back to a 2.5AH battery the difference is one pound. 2.5AH isn't crazy; the EI draws less than an amp at cruise power. With an alternator out, I'd shut it down and hold it in reserve anyway.

So, I don't see much value in mixing battery chemistry to drive an EI alone. However, I assume you need more AH. What is the combined draw for the EI, EFI, and fuel pump?
 
Total current draw for our EFI/EI driving 8-12 plugs, 4-6 injectors with a single Walbro pump running is around 10-12 amps at 2500 rpm.

I run a PC680 main battery and an 18 amp hour AGM battery aft of the baggage bay, dedicated to backup power for the engine electrics.

The AGM is charged monthly on the ground, voltage tested prior to startup, load tested every 6 months and tied to the essential buss through a 30 amp ATO fuse and 10 gauge wire/ heavy duty toggle switch. I can isolate from the main battery and alternator by turning off the master. This should give me a solid 45-60 minutes if the main battery takes a dump or about double that if just the alternator dies.

Many ways to approach backup power and we find few people agree on the same solution...
 
Last edited:
Here is a circuit that will prevent brownout during engine start. A fuel pump and ignition could also be powered by the ebus for an electrically dependent engine. The second fuel pump and ignition could be connected to the main battery.
EmhpOPk6KkEleo9IJj9Hz9kBT2-oXi_HGovXxvoktVxBOwT0hmAzqaYkiEhlMdSZpczf9zndwHvs=w972-h1040-no
 
Last edited:
I was thinking I needed something along the lines of a PC680 just for the backup. Like Dan, I'm running a 925 for the main.

I was hoping to bring the 925 forward to the spar to shorten up the heavy guage battery cable run, but that is going to only add to my forward CG problem. Maybe a 680 way back in the tail will make use of the extra weight and simplify my charging scheme at the same time.

I guess what I really need is a backup battery that can deliver 12 amps for at least an hour (preferrably two hours), is feather weight, and can tolerate riding on the main buss with a 40 amp and 14.5 v input.
 
So if you limit the charging arrangement to directly on the alternator then you will be limiting the battery options to something that can handle that charging configuration. That may be a reasonable option though.

I don't know what the EarthX batteries have internally but they must have some sort of charge conditioning. Lithium batteries require protection circuitry to prevent over and under voltage conditions, and current limiting. Many bsttereies limt the charging current to some percentage of the capacity. You will see specs like C/10. I'm mostly familiar with smaller batteries that are used in electronic though.

As an aside, I have a friend at work who had a home project where she used lithium cells. They have extremely high capacity, but even these cells have a small internal protection circuit that you might not even realize is there.

Going back to your original idea about mixing chemistries on the same bus I don't think it's a good idea. There several factors involved that can be dealt with by an intermediate circuit.
 
One item worthy of mention is the amp-hour rating of batteries of different chemistry. We know from first-hand reports on this forum the LiFePo batteries crank an engine like crazy. That means they are good for high-amperage, short duration work.

But what about long-duration, low amperage discharge?

If we dig further into the archives on this site we'll find discussions around battery ratings, and, in particular, discussions around how LiFePo batteries are often rated in "Pb equivalent" amp-hours. In short, many of the LiFePo batteries simply do not have the capacity to provide long-endurance sustainment at the low current draws discussed in this thread. If considering using a LiFePo as your backup battery for a critical system such as an engine controller and fuel pump, it would be a wise thing to actually TEST the duration for which a fully-charged battery can sustain operation of this critical equipment.

As for how to implement a second battery, I went nearly as stone-simple as the method suggested by Ross and others and am happy I did. Simple is good, and, generally, light.
 
Toobuilder said:
I guess what I really need is a backup battery that can deliver 12 amps for at least an hour (preferrably two hours), is feather weight, and can tolerate riding on the main buss with a 40 amp and 14.5 v input.
Click to expand...

Given the 12 amp hour requirement (wow), something like the EarthX ETX680 starts to make sense; expensive, but it only weighs 4 lbs.

EarthX says you can't parallel AGM and Lithium...but can they be run in an isolated fashion? Perhaps a system which allows power draw from one or the other, but never both at the same time?

No deep research here, but at first glance it looks like a lithium backup could be charged via a schottky diode, just like any other isolated backup battery. The charge voltage requirements are similar; Odyssey says 14.1 to 14.7 for full charge, while EarthX says 13.9 to 14.6. It would seem that typical charging voltage would allow the diode drop without starving the EarthX. Comments?
 
Dan,

I don't have the specs on the EarthX batteries so I'll limit my response to general comments. If you look at the available capacity of most batteries the capacity changes significantly with only small changes in battery voltage as you approach fully charged. This is relevant since when combining battery charging via your diode suggestion you need to be aware of this fact. But given the fully charged voltage specs are in agreement or the fully charged battery 1 voltage equals fully charged battery 2 + diode voltage then you should be ok. One small detail is that diodes at low current have lower voltage drops and attention to this detail would be advised.
 
The fuel pump is the biggest draw at around 4.5 amps at typical 40psi pressures.

4 cylinder systems draw around 10 amps, sixes around 12 (2 more injectors and 4 more plugs to fire).

The ECUs only draw around 0.1 amps each.

Some recent experience on the Rocket at Reno with Lithium batteries suggests they may not be so good for the purpose of supplying that 10-12 amps for an hour or two unless upsized quite a bit.

Look at the voltage vs. capacity ratings of these vs AGM batteries. They are impressive at high current for short periods but the curve drops off more sharply than lead acid stuff soon after.

AGMs often have an amp/hr rating to 9 volts threshold and this is what you need to use for comparison IMO. The ECU will function down to about 7.3V.

There is a lot of useful info here: http://www.power-sonic.com/images/powersonic/technical/1277751263_20100627-TechManual-Lo.pdf

Don't forget age and temperature affect reserve capacity too.
 
rapid_ascent said:
If you look at the available capacity of most batteries the capacity changes significantly with only small changes in battery voltage as you approach fully charged. This is relevant since when combining battery charging via your diode suggestion you need to be aware of this fact.
Click to expand...

Yep, needs examination.

But given the fully charged voltage specs are in agreement or the fully charged battery 1 voltage equals fully charged battery 2 + diode voltage then you should be ok.
Click to expand...

Yep again. I'm spit-balling, but why couldn't the basics be this simple?



One small detail is that diodes at low current have lower voltage drops and attention to this detail would be advised.
Click to expand...

Save me some book-digging. Would diodes in parallel reduce the voltage drop with increasing amperage?

rv6ejguy said:
Some recent experience on the Rocket at Reno with Lithium batteries suggests they may not be so good for the purpose of supplying that 10-12 amps for an hour or two unless upsized quite a bit.
Click to expand...

Gee, that would make the whole mixed-chemistry question moot. What did you hear from Reno?
 
Last edited by a moderator:
DanH said:
Yep, needs examination.



Yep again. I'm spit-balling, but why couldn't the basics be this simple?

Click to expand...



Perfect! the Beauty of Simplicity:)

With this solution you would not require individual backup batteries for your AV and at the same time you'll have the backup power to power other items that typical don't have dedicated battery backup options. And it's all automatic no pilot intervention wow.
 
Last edited by a moderator:
Save me some book-digging. Would diodes in parallel reduce the voltage drop with increasing amperage?
Click to expand...
Below from the internet, not from a book. It makes sense though.
Only if the diodes are well matched, and preferably
attached thermally. A mismatch in impedance of the
diodes will simply cause one diode to take slightly
more current, which causes it to heat up, which in turn
causes it to conduct even more current. This is caused
by the diodes having a positive temperature
coefficient.
When one diode is heating up and conducting even more,
this results in a positive feedback that causes more
and more of the current to flow through that diode.
This is called thermal runaway, and leads to failure if
the total current is greater than capacity.
When the first diode fails, the second will soon fail
as well as it in turn has to take up the total current.
This is called cascade failure.
So remember: diodes in parallel is a bad idea because
positive temperature coefficient causes thermal runaway,
which leads to cascade failure.
Thermally bonding the diodes reduces this effect as the
heating of one diode will cause the second diode to
heat up as well, reducing (but not eliminating) the
effect of thermal runaway. For this reason, diodes in
parallel never double the current carrying capacity.
If you do need to run diodes in parallel for whatever
reason, you'll need to think about means to prevent
thermal runaway, such as thermal bonding, or adding
components with negative thermal coefficients.
Click to expand...
 
With regards to Reno, the Griffin Rocket from here had a forced landing at the races after a charging system problem. The lithium battery did not keep it running for very long. Granted, there were other electrical loads on there from ADI pumps and such.

Perhaps Ralph Inkster or Bill Beaton will pipe up here and comment.
 
Dan,

Not sure I fully understand your drawing unless you are trying to show the two diodes in parallel. Both of the diodes originate at the Main Bus and terminate on a common line. Maybe I'm interpreting you drawing wrong.

I would avoid diodes in parallel. They are normally not perfectly matched so one will have higher current and lower drop the other lower current and higher drop. The imbalance has to do with how closely matched they are.

Diode technology is still advancing and there are high current low voltage drop diodes available for the currents that would be needed for our application. That would be a more desirable configuration. These parts will need good heat sinking but otherwise can do the job.

I did some investigation into these types of diodes a while back. In my desired configuration I want both of my batteries to supply my Essential Bus. I was looking for maximum efficiency or lack of heat so I decided to take another path, but part are available for this configuration. I'd have to dig up my notes to provide an actual part number.
 
rapid_ascent said:
Dan,
Not sure I fully understand your drawing unless you are trying to show the two diodes in parallel. Both of the diodes originate at the Main Bus and terminate on a common line. Maybe I'm interpreting you drawing wrong.
Click to expand...

You have it right, two diodes in parallel, thus my question about less voltage drop, like resistors in parallel. And a single diode failure doesn't shut down the engine.

I would avoid diodes in parallel. They are normally not perfectly matched so one will have higher current and lower drop the other lower current and higher drop. The imbalance has to do with how closely matched they are.
Click to expand...

I was assuming each diode individually would have more than enough capacity to drive Mike's EI/EFI/pump load, without being pushed very hard. If either can carry the entire load without overheating, thermal runaway can't happen.
 
DanH said:
Gee, that would make the whole mixed-chemistry question moot...
Click to expand...

So I guess this really is the salient point.

Is the newer, exotic chemistry really suitable for this relatively long duration, deep cycle type load? If the Pb chemistry is more suitable, then life just got much easier (aside from the weight).
 
I have a setup more or less similar to what Dan had drawn, but minus the diode pointing downward.

The idea is to let the charge voltage access the back up battery, but prevent the back up battery from feeding into the main buss, while letting either battery feed the essential stuff.

Works fine-------but both batteries are Odyssey.
 
DanH said:
You have it right, two diodes in parallel, thus my question about less voltage drop, like resistors in parallel. And a single diode failure doesn't shut down the engine...
Click to expand...

OK, so concerns about chemistry aside for a moment, why not replace the diode(s) with a SPST toggle? "NORM" is closed and everything on the same buss, "EMER" is open and the backup battery and engine buss is completely isolated. First sign of trouble, simply isolate the engine buss with the flip of a switch and then figure out what is happening with the ship. Seems like a very simple and robust circuit.

Opens the door to an insidious failure like the undetected loss of alternator and drain of both batteries, but OTOH, the reliability of a toggle should be higher than a diode (or is it?). If not for reliability, the toggle does allow very positive severing of the two buses - there's some points for utility there.
 
Last edited:
Fair enough. In this extremely limited usage the toggle has essentially an infinite lifespan, so if a diode is an improvement, who cares? Both devices far exceed the requirement, so it comes down to desired utility.

So back to the question about suitability of exotic chemistry for this application...
 
Toobuilder said:
Fair enough. In this extremely limited usage the toggle has essentially an infinite lifespan, so if a diode is an improvement, who cares? Both devices far exceed the requirement, so it comes down to desired utility.
Click to expand...

The diode approach requires no problem identification, and no pilot correction. And it's never left parked in the wrong position.

Now, what "utility" trumps those points?

Seriously, does it make sense to build a sport airplane that requires special understanding of unique systems, when we can build one that requires nothing of the pilot in order to remain running?
 
The "utility" is the option to totally divorce the engine buss from the ship if desired.

And to be honest I'm not even sure that's an advantage at this point... It just seems like a desirable trait.
 
DanH said:
Seriously, does it make sense to build a sport airplane that requires special understanding of unique systems, when we can build one that requires nothing of the pilot in order to remain running?
Click to expand...

I dunno; ask almost every US aircraft manufacturer.
 
Deep cycle LiPo battery? Not really..

The Rocket problem at Reno was that the LiPo battery shuts down at 12.5V, and the backup alternator was set to turn on at something less that 12.5V...so it did not kick in when the #1 alt went TU, so the battery drained to it's set limit. Very surprising to hear about this! Bill was not sure how long he ran on the battery alone, but things did get very quiet for him. He was on a downwind leg more or less, and simply landed NORDO.

So - do not expect your big buck LiPo battery to run till it's exhausted (deep cycle)- it will only go to 12.5V and then shut off. If your plan includes running the battery till it dies, then your choice is a lead acid type - or maybe a gang of 10 Eveready 1.5V D cells?

Yes - my next build will have two of the big buck blue units on board, along with two magnetos (until Ross gets his TCM setup ready to go). What could possibly go wrong?:eek:

Carry on!
Mark
 
F1Boss said:
So - do not expect your big buck LiPo battery to run till it's exhausted (deep cycle)- it will only go to 12.5V and then shut off. If your plan includes running the battery till it dies, then your choice is a lead acid type - or maybe a gang of 10 Eveready 1.5V D cells?
Click to expand...

For one thing, LiPo is different than LiFePo. Second, it sounds like in this case the battery may well have run till it was exhausted--just for lithium-type batteries, you don't define that by the same voltage levels that you do a "traditional" battery. The discharge curve of a LiFePo battery is pretty flat with a sharp drop at the end, rather than the gradual drop of a lead-acid battery--and if you try to go too far into that sharp drop, bad things happen. That's why the battery cuts off before that.

If you take a lithium-type battery and a lead-acid battery of the same useful capacity at the same discharge rate, the lithium one will cut off at a higher voltage--but you will still get the same energy out of it.

In other words, just because a lithium battery doesn't act like a lead-acid one doesn't mean one won't work. It means you need to understand how the battery operates and set your electrical system up accordingly.
 
Last edited:
DanH said:
The diode approach requires no problem identification, and no pilot correction. And it's never left parked in the wrong position.

Now, what "utility" trumps those points?

Seriously, does it make sense to build a sport airplane that requires special understanding of unique systems, when we can build one that requires nothing of the pilot in order to remain running?
Click to expand...

Ok, a compromise. The added cost is one toggle which should never have to be moved out of the "NORM" position.

If things are going fine, the switch stays in the "NORM" position at all times and the diode takes care of everything.

If desired, the "TIE" position allows the emer battery to help out with the ship or can be used in the unlikely event the diode goes TU. And the "ISO" position is also available if you want to jump off the ships buss entirely.

Note that this assumes compatable batteries for charge/discharge - which I don't think we've established yet for "mixed" chemistry.

qou4c0.jpg
 
Last edited:
Toobuilder said:
Ok, a compromise. The added cost is one toggle which should never have to be moved out of the "NORM" position.
Click to expand...

Which makes it a useless "what if" functionality...as well as an source of potential pilot error.

KIS, brother, KIS...
 
Ok, point taken.

System architecture aside for a moment, it appears that this battery is a drop in replacement for the conventional Pb chemistry.

No mention of any special charge requirements, can be deep cycled, etc. Expensive, but at a bit over 4 pounds it seems to pack a punch.
 
Before using Lithium tech for backup, be very familiar with what the specs mean (as others have mentioned).

That one says 104 watt-hours of energy. That's 8.5 amp-hours, and it doesn't mention a discharge rate to get the full 8.5 AH. If you draw 12 amps, that's less than 45 minutes, *if* it can supply 12 amps continuously and still deliver the full 8.5 AH of *energy*. (Not likely at all.) It could be much less; possibly as short as 10-20 minutes.

Power & energy are not the same thing. A Corvette will go 150 mph. If it only has 2 gallons of gas in the tank, it won't go very far.

Charlie
 
So have we established that Pb chemistry is the optimum choice for this application?

In my feeble mind it seems like the exotics are great for a high discharge like cranking a starter or minimal discharge like Ross' CPI, but a medium draw like I need is a whole different ballgame?
 
The various lithium techs will still likely have more energy per pound; that's why they're in all our portable devices. It's just not likely to be the 10-1 ratio vs lead that everyone seems to believe. For instance, you could use two of the ones you mentioned, and have a bit more capacity than a PC680 & still have only 9 lbs of weight. And you'll carry a lot less money-weight around in your wallet while you're flying. A win-win! :)

The EarthX product now seems to have similar energy content to a PC680 lead battery, and they are publishing at least some discharge curve specs now. But they still have some unexplained, seemingly nonsensical restrictions, like limiting alternator size, even though they are supposed to have battery mgmt systems built in.

The *lightest* (not simplest) path to dependable electrons is probably a small lithium iron starting battery and two alternators, each capable of supporting all continuous loads plus some charging capacity. (Things like landing lights, gear motors, etc are so intermittent that they don't need to be counted at full value.) Two 7 lb alternators, one 4 lb battery=unlimited electrons.

Charlie
 
rv7charlie said:
That one says 104 watt-hours of energy. That's 8.5 amp-hours, and it doesn't mention a discharge rate to get the full 8.5 AH. If you draw 12 amps, that's less than 45 minutes, *if* it can supply 12 amps continuously and still deliver the full 8.5 AH of *energy*. (Not likely at all.) It could be much less; possibly as short as 10-20 minutes.
Click to expand...

All batteries act like that, not just lithium-based ones.

Toobuilder said:
So have we established that Pb chemistry is the optimum choice for this application?

In my feeble mind it seems like the exotics are great for a high discharge like cranking a starter or minimal discharge like Ross' CPI, but a medium draw like I need is a whole different ballgame?
Click to expand...

I think it means you need hard data from the battery manufacturer showing usable energy at your desired discharge rate. Figure out how many amps you need and how long you need them, then find a battery that can meet those numbers.

A lithium-based battery will definitely be lighter than a lead-acid battery sized to the same current-time requirement. The lithium battery may have a higher amp-hour rating on paper, it will probably be more expensive, and it will operate differently than a lead-acid one.
 
Mike,
Played around a bit last night. This diagram assumes the VFR airplane is already wired conventionally, and the goal is to add a backup battery system dedicated solely to keeping an EFI/EI working if the primary electrical system goes down...due to failure, or because the pilot found it necessary to kill the master.

It is also simple, requiring little in terms of special knowledge to fly it safely. It's got a master switch; on or off has nothing to do with keeping the engine running, just like a magneto airplane. It's got one or two ignition switches, depending on single (left diagram) or dual EFI/EI controllers (right diagram). It's got a fuel pump switch, which the pilot must turn on or the engine won't start. The pump switch only allows running one pump at a time, important with EFI. A pump failure requires a switch action from the pilot, but any current EFIS/EIS takes care of problem identification. The pilot gets a low fuel pressure warning, and reaches for a pump switch, again very conventional. BTW, that too can be made automatic.

I've drawn in fusible links for the EFI/EI battery feeds, bulletproof when fabricated properly.

The two batteries are isolated, so a lithium aux battery is possible, assuming the answers to other issues are positive.

One of these might work well to feed the single EFI bus:

http://www.periheliondesign.com/powerschottkydiodesfiles/Power_Deuce_Schottky_ Manual.pdf

 
Last edited:
Looks good Dan, my put is that the engine-pumps and ignition- is controlled by a single switch. Simply a "off, run, start" deal just like a car or a piece of aerospace ground support equipment. Pumps are either A or B, and part of the run up check is to switch from one to the other and verify its working. Simply start on whatever pump is selected, and switch sometime before flight. This alternates each pump each flight and should even out their use, ensuring there is no "favorite".

And yes, I am a fan of the fusable link for those systems that HAVE to work.
 
Last edited:
I've been going though many electrical schemes also as I am about to convert back to an electrically dependant airplane, after having same for several years of Subaru experimentation. This time it will be with my IO-375 and all of Ross's latest offerings.This time I decided I want to have the confidence to fly right past good landing spots to a place of my choice where I will repair or replace my main alternator when it fails. To get to this comfort level I will use a B&C backup alternator running on the vacuum pad. I like that it will be waiting to see the voltage drop down and automatically come in as needed. This feature is dependant on whether the busses are kept separate. If kept separate, I can adjust the regulator on the backup alternator to a charging voltage the backup battery likes, but keeping the busses separate requires the second battery.

I echo Dan's thoughts in that the system needs to be simple enough that I could send an experienced pilot up in my plane with a simple briefing of the system. He should not have to study for a long time to understand how to react in various failure situations.

Dan, your dual EFI scheme is very close to what I have been coming up with. I keep comparing this simple approach to the use of the Bus Manager that provides the automatic switchover to backup fuel pump, and an emergency power switch that bypasses the "box" and connects both batteries to the ebus.

I do not understand the comment about not running both fuel pumps on an EFI system. Some people would routinely turn on the backup fuel pump when taking off or landing, much like turning one on as a backup to a mechanical pump. As far as I know, the fuel pressure regulator adjusts accordingly. The fuel rail pressure might come up a little and cause a slightly richer mixture but with no associated problems that I am aware of.

I want a system that uses the tried and true AGM PC680 as primary battery with main and backup alternators. Adding a second AGM is a weight penalty I do not want to pay, but adding a 4 pound Lith Iron batt is appealing as it allows a scheme (Bus Manager) that keeps the busses separate except in two instances: If you choose to use both batteries during start operation, and if you need to engage the emergency power switch to keep her running.

Robert says it is acceptable to have the mixed battery chemistries in this situation but i am looking for more reassurance.

I am surprised no one has jumped in here with experience with use of both battery types in a single system. Seems like a good mix of reliability and light weight if it works properly.

For those considering the EarthX, take a look on their web site at the discharge charts. It clearly shows how the voltage stays relatively flat until a sharp drop off where the BMS will kick in and shut it down. This feature rules it out as primary use in my opinion, and in my situation, at least until I get some real world experience with one.
 
I have to go back to my notes to refresh my memory but I do recall that there is a problem in this subject area - dealing with mixing batteries - due to a phenomena called "circular currents."

Just on a break from re-organizing the shop, with that being said, I ask assistance from those with more knowledge-experience and smarts! :^)

mjb
 
Randy said:
Dan, your dual EFI scheme is very close to what I have been coming up with. I keep comparing this simple approach to the use of the Bus Manager that provides the automatic switchover to backup fuel pump, and an emergency power switch that bypasses the "box" and connects both batteries to the ebus.
Click to expand...

Automating the backup fuel pump engagement is easy. I drew manual switch control for both primary and backup pumps. In reality, your ECU may control the primary pump through a relay; you would switch it only as a test function.

Bus Manager power circuits:



The backup power switch is necessary because of the other architecture; a failure of the primary switch, power relays, or any of the control wiring shuts down the buses. The backup switch is merely a relay bypass.

The BM combines the ship's power requirements with the engine management power requirement. The wiring offered in post 39 puts engine management power on isolated, battery-direct circuits, not shared with any other power requirement. Ships power can be single bus, main and essential, whatever you want. Any or all of the ships power can be shut down, and the big fan keeps turning. For the pilot, it's like flying a Cherokee.

I do not understand the comment about not running both fuel pumps on an EFI system. Some people would routinely turn on the backup fuel pump when taking off or landing, much like turning one on as a backup to a mechanical pump. As far as I know, the fuel pressure regulator adjusts accordingly. The fuel rail pressure might come up a little and cause a slightly richer mixture but with no associated problems that I am aware of.
Click to expand...

Rotary vane pumps require flow for cooling. A standard backup pump for constant flow injection has a built-in bypass and pressure regulator, so fuel always circulates through the pump, even if none flows downstream. For example, you can run the pump with the mixture knob in idle cutoff.

The EFI pumps have no ability to recirculate. They're just in and out; pressure regulation is at the far end of the fuel rail. If you plumb two in parallel and one has a higher output pressure than the other, the weak sister will have limited flow, and it will get hot. Or so I'm told; I have not personally conducted that experiment. I do note that engine manufacturers with standard EFI are pretty clear about it. This example is from the ULPower installation manual:

Important !
Provision should be made for a separate switch on the dashboard to interrupt power supply to the second (backup) pump. Experience has shown that leaving both pumps running simultaneously can cause one of the pumps to overheat due to insufficient fuel flow through the pump. This can result in a pump failure, eliminating backup if/when it is needed. Install an “or-or” switch to be sure that both pumps never can run together. Before take-off , make a test to be sure both pumps are working.
 
Last edited:
That note from Protec[edit: should be UL Power; I thought Dan was referring to the same company as the drawing] says to introduce a single point of failure in the 'redundant' fuel delivery system. Pump switching might be better left to the pilot to manage, unless you're willing to accept the complexity of a pressure sensing auto-on circuit for the backup pump.

BTW, for those who don't know, just about everything being discussed here has previously been hashed out on the Aeroelectric list under the guidance of Bob Nuckolls (one of the brains behind most of B&C's products). You can buy the book, or download the PDF version for free. It's a really valuable source for anyone wiring a homebuilt, whether it's 'conventional' or any variation of electrically dependent. There are diagrams for just about anything we can dream up.

Charlie
 
Last edited:
Toobuilder said:
So have we established that Pb chemistry is the optimum choice for this application?

In my feeble mind it seems like the exotics are great for a high discharge like cranking a starter or minimal discharge like Ross' CPI, but a medium draw like I need is a whole different ballgame?
Click to expand...

Mike, maybe the best bet (as usual in our game) is to measure it yourself.

Buy an EarthX ETX680 (12.4 ah rating).

Discharge it with some reasonable load until the internal cell-saver circuitry shuts it down.

Recharge it at the bus voltage you expect to supply when installed in the airplane. I would actually do it in the airplane; strap it in the back seat, run a power and ground, complete with a schottky diode for isolation, as it slightly reduces charging voltage.

Now put it on the bench, rig a 12 amp load, and start a timer.

If you like the results, do a permanent installation. If you don't, the ETX680 will easily resell here. Plenty of builders want one as a conventionally-installed primary battery, and the cell-saver circuits mean you can't hurt it with this test.
 
Yep, testing is the way to cut through the BS. I like testing. Thanks for being the voice of reason (again).

However:

The EarthX is 4 pounds and $400 bucks

The Odessy is 7 pounds and $100 bucks.

The weight savings are compelling, but at a cost of $100 bucks per pound... I'm going to have to think on that one.
 
Dan,

Thanks for that information regarding running two electric fuel pumps at the same time. I and I suspect many others was not aware of the potential for the limited flow causing overheating of the weaker pump. I will plan on a switch that only allows one or the other to run. Saves a switch position too!
Randall
 
If we're talking about the newest turbine style pumps, then the issue of no flow from the weaker pump could be an issue. However, I'm not aware of anyone selling a turbine style pump for an aviation application (turbine style pumps must have 'head' on the input; they won't lift to self-prime). All the models I've seen are using either gerotor or roller-vane pumps. Both styles are positive displacement pumps (a bit like an engine oil pump). If there's fuel at the input, I'm pretty sure there's going to be fuel coming out, unless there's a very strong cap on the output. :)

RE: switches, see my post #44.

Charlie
 
We've seen no issues running two Walbro or Bosch pumps simultaneously. Not sure what UL is doing or what type of pumps they use. That being said, there is no reason to run both pumps in cruise.
 
Bus Manager

The isolation of the Bus Manager allows the mixed battery voltages to work fine together. We have many customers running one Odyssey and one EarthX.
The batteries are not tied together except if and when they are both turning the starter motor. At that operating point the voltage difference doesn't matter.

Robert
 
You must log in or register to reply here.
Back
Top