X-Virus-Scanned: clean according to Sophos on Logan.com X-SpamCatcher-Score: 64 [XX] (100%) OBFUSCATED_WORD1_ONLINE Return-Path: Sender: To: lml@lancaironline.net Date: Sun, 25 Mar 2007 12:57:01 -0400 Message-ID: X-Original-Return-Path: Received: from mta15.adelphia.net ([68.168.78.77] verified) by logan.com (CommuniGate Pro SMTP 5.1.7) with ESMTP id 1942046 for lml@lancaironline.net; Sun, 25 Mar 2007 08:49:39 -0400 Received-SPF: pass receiver=logan.com; client-ip=68.168.78.77; envelope-from=glcasey@adelphia.net Received: from [75.82.254.207] by mta15.adelphia.net (InterMail vM.6.01.05.04 201-2131-123-105-20051025) with ESMTP id <20070325124850.KQZT9378.mta15.adelphia.net@[75.82.254.207]> for ; Sun, 25 Mar 2007 08:48:50 -0400 Mime-Version: 1.0 (Apple Message framework v752.2) In-Reply-To: References: Content-Type: multipart/alternative; boundary=Apple-Mail-11--1048764781 X-Original-Message-Id: <04EEEA0E-D875-4600-8FC8-5E6637A2306A@adelphia.net> From: Gary Casey Subject: Re: Legacy Battery Set-up X-Original-Date: Sun, 25 Mar 2007 05:48:50 -0700 X-Original-To: "Lancair Mailing List" X-Mailer: Apple Mail (2.752.2) --Apple-Mail-11--1048764781 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=WINDOWS-1252; delsp=yes; format=flowed Rick, I would certainly take the side of Keeping It Simple - and simple =20 especially means keeping the decisions to be made in case of failure =20 as simple as possible. I would leave the diodes feeding the e-buss =20 and avoid any relays or switches in that area (no decision to be made =20= - the important stuff keeps working as long as there is either side =20 functioning). There is no such thing as a "cranking spike" - the =20 starter just pulls the voltage in the system down, potentially =20 causing brown-outs or re-boots in electronic equipment. Shouldn't be =20= a problem for any competently-designed hardware. I would not =20 activate the alternators through oil pressure switches - just more =20 complexity for little or no benefit. One thing that should be taken =20 into account is that an alternator should not be allowed to become =20 disconnected from a battery. Even though it might be able to =20 continue to supply current it will not be able to follow load =20 fluctuations without producing "load dump" spikes. Therefore, the =20 master switch for each battery should also switch the power for the =20 field for that alternator. Unfortunately, that means that each =20 alternator is dedicated to its battery and can't "cross-feed" without =20= the batteries being connected (masters on). I'm ambivalent about the =20= location of the shunts as I would leave them out entirely and just =20 monitor voltage. If I had a current shunt I would put it on the =20 battery as you suggest, monitoring the charge current, not the =20 alternator load or the aircraft load. In my case if the voltage is =20 above 13 everything is okay; below 13 not okay and time to reduce =20 electrical load and land. In your case if the system voltage drops =20 you can turn on the cross-feed and land - you just have more time to =20 land. The decision process is the same. KIS Gary > > Also note, in my simplified sketch, I have an essential/endurance =20 > bus which is supplied from both the =93A=94 and =93B=94 side via a = diode on =20 > each. Thus, the essential bus gets power even if the A-side fails =20 > and I haven=92t yet activated the cross tie. Thus, the EFIS doesn=92t = =20 > quit and have to re-boot. This also provides some additional EFIS =20 > power redundancy in the event of a simultaneous A side and cross-=20 > tie failure. Normally, my =93A=94 side voltage regulator will be set =20= > to slightly higher than the B side (~0.1 volt). Thus, power will =20 > normally flow into the essential bus via the A side. =20 > Alternatively, I could have a manual switch on the B-side lead or a =20= > normally closed relay on the B side lead which is dis-engaged by =20 > the A side (open then A is hot). I opted for the diodes as KISS in =20= > this case and am seeking to minimize the number of switches. =20 > However, a manual switch would create some additional starting =20 > spike isolation. Perhaps a normally closed relay which is powered =20 > (dis-engaged, opened) by the starter relay. Perhaps the B side =20 > essential feed should be tied to the B-side alternator switch, =20 > since it would normally be off during starting (providing isolation). > > > > I=92m also considering activating the alternators based on an oil =20 > pressure switch (one for each side as redundancy). That way the =20 > alternators com on-line automatically with oil pressure (just after =20= > engine start). Without oil pressure the alternators won=92t work =20 > long anyway {g}. Each could also be turned off by pulling the =20 > circuit breaker on the respective voltage regulator. I=92m undecided =20= > on that, but I digress a bit=85 > > > > 2) If I understand your second question correctly=85 I have a shunt =20= > and ammeter between the battery and the bus/alternator (on both the =20= > A and B side). These are not shown on my prior sketch. On the B =20 > side it is past the starter so the =93starting current=94 does not go =20= > though the shunt (a special high current event). Otherwise, both =20 > sides are essentially setup the same. Thus, the ammeter(s) show =20 > the current flowing into (or out of) each battery. With the =20 > alternator on, this current shown is normally very low (just any =20 > battery recharging). With the alternator off, the ammeters will =20 > show the =93drain=94 being placed by the electrical equipment that=92s = on =20 > at the time. This setup dose not directly show the total load on =20 > the alternator. However, it is indirectly available. If you want =20 > to know (approximately) how much total load is being placed on the =20 > alternator at any point in time, you can look at the ammeter with =20 > the alternator on (the current going into the battery) and then =20 > momentarily turn the alternator off and see how much current flows =20 > from the battery to the electrical stuff. The current flowing out =20 > from the battery to the other stuff, was previously being supplied =20 > by the alternator. So, the total alternator load is (was) the sum =20 > of the two. Note: the needle (flow) will shift from positive to =20 > negative, which just indicates current direction relative to the =20 > battery. The total current being supplied by the alternator is =20 > (was) the sum of the absolute values (disregarding the direction =20 > into/out of the battery). > > > > Some folks (manufacturers) favor putting the shunt in the =20 > alternator B lead (between the bus/battery and the alternator). =20 > That setup shows total alternator load directly, but is useless in =20 > an alternator failure scenario, when you might want to really know =20 > total current draw to preserve the remaining battery power. It =20 > also provides no mechanism for determining whether the battery is =20 > charging (health) or draining (unhealthy). Others favor putting =20 > the shunt at the head of the bus(es) i.e.before all the electrical =20 > stuff. That setup shows total equipment load directly, but =20 > provides no mechanism for determining total alternator load, nor =20 > whether the battery is charging (health) or draining (unhealthy). =20 > Thus, the std battery ammeter setup seems the best approach (as I =20 > understand it). > > > > Rick --Apple-Mail-11--1048764781 Content-Transfer-Encoding: quoted-printable Content-Type: text/html; charset=WINDOWS-1252 Rick,
I would certainly take = the side of Keeping It Simple - and simple especially means keeping the = decisions to be made in case of failure as simple as possible.=A0 I = would leave the diodes feeding the e-buss and avoid any relays or = switches in that area (no decision to be made - the important stuff = keeps working as long as there is either side functioning).=A0 There is = no such thing as a "cranking spike" - the starter just pulls the voltage = in the system down, potentially causing brown-outs or re-boots in = electronic equipment.=A0 Shouldn't be a problem for any = competently-designed hardware.=A0 I would not activate the alternators = through oil pressure switches - just more complexity for little or no = benefit.=A0 One thing that should be taken into account is that an = alternator should not be allowed to become disconnected from a battery.=A0= Even though it might be able to continue to supply current it will not = be able to follow load fluctuations without producing "load dump" = spikes.=A0 Therefore, the master switch for each battery should also = switch the power for the field for that alternator.=A0 Unfortunately, = that means that each alternator is dedicated to its battery and can't = "cross-feed" without the batteries being connected (masters on).=A0 = I'm=A0ambivalent about the location of the shunts as I would leave them = out entirely and just monitor voltage.=A0 If I had a current shunt I = would put it on the battery as you suggest, monitoring the charge = current, not the alternator load or the aircraft load.=A0 In my case if = the voltage is above 13 everything is okay; below 13 not okay and time = to reduce electrical load and land. =A0In your case if the system = voltage drops you can turn on the cross-feed and land - you just have = more time to land. =A0The decision process is the same. = =A0KIS

Gary


Also note, in my simplified sketch, = I have an essential/endurance bus which is supplied from both the =93A=94 = and =93B=94 side via a diode on each.=A0 Thus, the essential bus gets = power even if the A-side fails and I haven=92t yet activated the cross = tie.=A0 Thus, the EFIS doesn=92t quit and have to re-boot.=A0 This also = provides some additional EFIS power redundancy in the event of a = simultaneous A side and cross-tie failure.=A0 Normally, my =93A=94 side = voltage regulator will be set to slightly higher than the B side (~0.1 = volt).=A0 Thus, power will normally flow into the essential bus via the = A side.=A0 Alternatively, I could have a manual switch on the B-side = lead or a normally closed relay on the B side lead which is dis-engaged = by the A side (open then A is hot).=A0 I opted for the diodes as KISS in = this case and am seeking to minimize the number of switches. =A0However, = a manual switch would create some additional starting spike isolation.=A0 = Perhaps a normally closed relay which is powered (dis-engaged, opened) = by the starter relay.=A0 Perhaps the B side essential feed should be = tied to the B-side alternator switch, since it would normally be off = during starting (providing isolation).=A0 =

=A0

I=92m also = considering activating the alternators based on an oil pressure switch = (one for each side as redundancy).=A0 That way the alternators com = on-line automatically with oil pressure (just after engine start).=A0 = Without oil pressure the alternators won=92t work long anyway {g}.=A0 = Each could also be turned off by pulling the circuit breaker on the = respective voltage regulator. =A0I=92m undecided on that, but I digress = a bit=85

=A0

2) If I = understand your second question correctly=85 I have a shunt and ammeter = between the battery and the bus/alternator (on both the A and B side).=A0 = These are not shown on my prior sketch.=A0 On the B side it is past the = starter so the =93starting current=94 does not go though the shunt (a = special high current event).=A0 Otherwise, both sides are essentially = setup the same.=A0 Thus, the ammeter(s) show the current flowing into = (or out of) each battery.=A0 With the alternator on, this current shown = is normally very low (just any battery recharging).=A0 With the = alternator off, the ammeters will show the =93drain=94 being placed by = the electrical equipment that=92s on at the time.=A0 This setup dose not = directly show the total load on the alternator.=A0 However, it is = indirectly available.=A0 If you want to know (approximately) how much = total load is being placed on the alternator at any point in time, you = can look at the ammeter with the alternator on (the current going into = the battery) and then momentarily turn the alternator off and see how = much current flows from the battery to the electrical stuff.=A0 The = current flowing out from the battery to the other stuff, was previously = being supplied by the alternator.=A0 So, the total alternator load is = (was) the sum of the two.=A0 Note: the needle (flow) will shift from = positive to negative, which just indicates current direction relative to = the battery.=A0 The total current being supplied by the alternator is = (was) the sum of the absolute values (disregarding the direction = into/out of the battery).

=A0

Some folks = (manufacturers) favor putting the shunt in the alternator B lead = (between the bus/battery and the alternator).=A0 That setup shows total = alternator load directly, but is useless in an alternator failure = scenario, when you might want to really know total current draw to = preserve the remaining battery power.=A0 It also provides no mechanism = for determining whether the battery is charging (health) or draining = (unhealthy).=A0 Others favor putting the shunt at the head of the = bus(es) i.e.before all the electrical stuff.=A0 That setup shows total = equipment load directly, but provides no mechanism for determining total = alternator load, nor whether the battery is charging (health) or = draining (unhealthy).=A0 Thus, the std battery ammeter setup seems the = best approach (as I understand it).

=A0

Rick

= --Apple-Mail-11--1048764781--