X-Virus-Scanned: clean according to Sophos on Logan.com From: "Stephen Izett" Received: from mail-pg0-f43.google.com ([74.125.83.43] verified) by logan.com (CommuniGate Pro SMTP 6.2c1) with ESMTPS id 9610522 for flyrotary@lancaironline.net; Sun, 26 Mar 2017 09:16:58 -0400 Received-SPF: pass receiver=logan.com; client-ip=74.125.83.43; envelope-from=steveize@gmail.com Received: by mail-pg0-f43.google.com with SMTP id g2so16504565pge.3 for ; Sun, 26 Mar 2017 06:16:58 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=from:mime-version:subject:date:references:to:in-reply-to:message-id; bh=Az2WScrFVofFIjiAP0uhxRlO/m3OYaaICc86HvebGTk=; b=Xw+25993RyIhdpKPUZ0t7lB44SWxXLnYPRaXaG3lI/Wx7u/RbAWjrh9vxn8FEra/87 G04m4frskt3KyjPurUGCbQrgJxBg69QYN6tnGmJYtRh4HfPyFm4FW/V8rKJp7bI1Koic fr61Xy54KvUkDUvDGQg4g41P9JoeB9a5VrndjQGg+cmYbWZ7nMWscQqLk72sZso461dK bHgX4J6itVQ7tgKS3uvVHVPmIHJ6ffhmwJmQqYVmOP7d/Vw6bYx/xpNaMyXvqEqoK8yQ oNnmpgV5fLx5OzrvPS4fWHL/y+hO0HtgsGp3mVMjbz00gv4a7s3dJIQDUHtdFjiYx05A ZDGQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:mime-version:subject:date:references:to :in-reply-to:message-id; bh=Az2WScrFVofFIjiAP0uhxRlO/m3OYaaICc86HvebGTk=; b=QH78Yo9Xt3/5KKlbR1b5tP3HDxXQcKlxJzOPZ5Vs8WiGybly1j7UC2R41vcnhaaKj5 uvJzp8/Ws/kcaJ+Rgo0wIW89j+HyBr9M/8x0B/m9cNpGKqIxy5WVggrqx2EZ40/SDXpB Swng0+1RLasjHzUA1e/wimw38VGM7dPCIYpkw9eFsScelFF1TB5lOgjsfAjVE1NmVnpi xqqO7sYoAR68smIuT2lBka6fIRR4XozJuDrxsgPo8BacRQzZWjverS0BSN7Lsqd3R9Ov Ha1uLiBQGUMcdXx9t9u6h78HiKEwz0zT1k+CGEkBrZ6dPd+J0Zv//KWqkvuWlDNYsvFb ymuA== X-Gm-Message-State: AFeK/H2VCdlujKasOVTLUwJ3abYdL3fdysaOKfcQD32YYtSPTwR/vq/BaTWL/5276hvQTQ== X-Received: by 10.99.138.68 with SMTP id y65mr19094126pgd.73.1490534199668; Sun, 26 Mar 2017 06:16:39 -0700 (PDT) Return-Path: Received: from [10.1.1.6] ([124.169.71.123]) by smtp.gmail.com with ESMTPSA id 23sm15233868pfw.94.2017.03.26.06.16.36 for (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Sun, 26 Mar 2017 06:16:38 -0700 (PDT) X-Google-Original-From: Stephen Izett Content-Type: multipart/alternative; boundary="Apple-Mail=_18AADCE4-C1C2-4198-98CC-DF5BCA90CED9" Mime-Version: 1.0 (Mac OS X Mail 10.2 \(3259\)) Subject: Re: [FlyRotary] Fuel Pump current and pressures Date: Sun, 26 Mar 2017 21:16:33 +0800 References: To: Rotary motors in aircraft In-Reply-To: Message-Id: X-Mailer: Apple Mail (2.3259) --Apple-Mail=_18AADCE4-C1C2-4198-98CC-DF5BCA90CED9 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=utf-8 Hi Charlie In retrospect I=E2=80=99d seriously consider building the 18 Gal header = tank to suit a couple of submerged pumps. As my pumps live below the header tank perhaps I could use turbine = pumps, but does anyone make them for inline use. Cheers Steve Izett > On 26 Mar 2017, at 3:52 am, Charlie England = wrote: >=20 > I'd say the Walbro is the 'gold standard', but there are plenty of = others out there. If you could get into your header tank, or have room = for an additional small 'sump', you could try this type: > = http://www.ebay.com/itm/360843503639?_trksid=3Dp2060353.m1438.l2649&ssPage= Name=3DSTRK%3AMEBIDX%3AIT = >=20 > (Not saying to use a $12 pump; just this style of pump.) >=20 > I considered this style for my in-tank pumps, but they must be mounted = vertically, and would have interfered with the float of the fuel level = sender. >=20 > Variations on this are in almost all new cars, but in the gas tank. = That one is ~1.5 inches diameter & ~4.5 inches tall. Supposed to draw = around 4 amps, so likely to be lower flow than any of the gerotor pumps. = Walbro makes one that's about the same size, for typically around US = $80-$90. They're turbine pumps, so the bottom must be in fuel to prime. = No easy way to attach an input, except the purpose-made filter sock. >=20 > There are complete after-market assemblies (single pump, filter sock, = surge housing, 40-240 ohm fuel level sender, and cover plate with both = fuel & electrical connections, for around US $30-$50. >=20 > Charlie >=20 > On Sat, Mar 25, 2017 at 7:31 AM, steve Izett = > = wrote: > Thanks Steve > Really appreciate your helping me with this. > Is there an option other than the Walbro GSL393 or is this the = recommended inline pump at present? >=20 > Thanks > Steve >=20 >=20 >=20 >> On 25 Mar 2017, at 3:11 pm, Steven W. Boese = > = wrote: >>=20 >> Steve, >>=20 >> It is quite likely that getting your fuel system to behave would not = involve looking at more than can be seen in the photo of the pumps that = you posted. >>=20 >> Your item 1 would most likely be resolved if pumps of lower flow = ratings were used. The stock RX7 regulator is very similar to the one = you are using and has worked well with those smaller pumps. >>=20 >> Item 2 may be a result of the pump with the higher current draw = having more internal friction causing a load on the motor in addition to = the load from pumping fuel.=20 >> The additional load would slow the motor down resulting in less flow = and lower pressure. It is possible that the decreased flow allows the = regulator to control the pressure at 44 psi. Otherwise there may be a = constriction either within the pump or in the flow path somewhere before = the point where the flow of the two pumps are combined. >>=20 >> Item A is unlikely to be the case since if the seat in the regulator = hadn't been broken, there would be no flow, thus stalling the pump and = causing a current draw of over 20 amps. >>=20 >> In item B, you are likely correct that the regulator cannot support = the flow from the high volume pump, thus causing the rail pressure to be = above the regulator's pressure rating. It seems unlikely, however, that = the pressure would drop with the engine running since there is so much = more fuel circulating through the system (~70 gal/hr) than what the = engine could ever use (~20 gal/hr) even at full throttle. >>=20 >> All this discussion is directed toward the pumps or the system = downstream of them. Consider also the system between the tank and the = pumps. Trying to draw 140 gal/hr with the two pumps through a single -6 = line, and also through a 20 to 40 micron screen if one is installed as = required by the pump manufacturer, would have to involve a very low = pressure at the inlet to the pumps. This seems to be a very good recipe = for vapor lock. >>=20 >> Steve Boese =20 >>=20 >>=20 >> From: Rotary motors in aircraft > on behalf of steve Izett = > >> Sent: Friday, March 24, 2017 6:19:23 PM >> To: Rotary motors in aircraft >> Subject: [FlyRotary] Re: Fuel Pump current and pressures >> =20 >> I=E2=80=99ll remove the pumps to confirm the model and do some = testing. >> In old notes I see I thought they were GSL392=E2=80=99s. >> The real obvious is as you show in your calcs, the volume of fuel = heading back to the tank with both pumps running is huge. >> With a piddly little 150hp stock unit that probably has quite a small = orifice for bypassing regulation, quite large pressures are going to = build up. >> I have the two problems though. >> 1. Flow capacity of the regulator if I want to be able to run both = pumps simultaneously. >> 2. The difference between the pumps currently. Why does one draw more = current and produce less pressure? >>=20 >> What is causing the 4psi pressure differential: >> One of the pumps draws more current but the pressure has dropped from = 48 to 44psi. Now either one of two things could bring this about in my = thinking. >> A. The higher current pump is delivering very low fuel flow and so = measures a rail pressure drop, meaning not enough fuel pressure/flow to = break the regulator seat. But I think I can remember hearing fuel = returning to the tank when either pump is running. OR >> B. The other pump is delivering at idle more fuel pressure/flow than = the the regulator can cope with so is higher than the regulation = pressure. So as the engine begins to use fuel the rail pressure would = drop back to the regulator pressure. (I cant remember seeing this during = engine testing, but I wasn=E2=80=99t looking for it and Ive deleted some = video of engine monitoring). >>=20 >> Have I missed something? >>=20 >> Steve >>> On 25 Mar 2017, at 3:11 am, Steven W. Boese = > = wrote: >>>=20 >>> Steve, >>>=20 >>> Examining the Walbro pump specifications at: >>> = https://walbrofuelpumps.com/walbro-gsl-series-universal-inline-fuel-pumps = and looking at the data for the individual pumps enables the following = speculation. It is only speculation. >>>=20 >>> If you have GSL392 pumps, as are almost all of the pumps listed now = on Ebay, they have a flow rating of 255 L/hr (70 gal/hr) at 40 psi and 8 = amps. If your engine could use 20 gal/hr, that would still leave 50 = gal/hr returning to the tank. With two of those pumps running at 65 psi = =20 >>> they should be moving 60 gal/hr each with each drawing 10 amps. If = your engine was using 20 gal/hr, 100 gal/hr would be returning to the = tank. This would seem to be moving much more fuel than necessary. = Depending on where you are measuring the fuel pressure and the details = of the fuel flow path, the pressure at the pump outlet (and the current = draw) could be much higher with flow rates this high.=20 >>>=20 >>> At these flow rates, your whole fuel system could be a constriction = and any fittings (tee's etc) may have significant effects. >>>=20 >>> That is why it would be good to establish which pumps you are using = either by identifying them or measuring their flow rates. GSL393 pumps = (45 gal/hr at 40 psi and 5 amps) might be more appropriate for your = installation if your present pumps have high flow ratings. >>>=20 >>> Steve Boese >>>=20 >>> =20 >>> From: Rotary motors in aircraft > on behalf of Charlie England = > >>> Sent: Friday, March 24, 2017 12:24 PM >>> To: Rotary motors in aircraft >>> Subject: [FlyRotary] Re: Fuel Pump current and pressures >>> =20 >>> Hey Steve,=20 >>>=20 >>> Just re-read your post with the measurements, & I'm not sure which = had higher current draw. It was the bottom pump, right? >>>=20 >>> Thanks, >>>=20 >>> Charlie >>>=20 >>> On 3/24/2017 8:06 AM, Stephen Izett wrote: >>>> Good idea Charlie.=20 >>>> That would clear it up as far as the manifold flow characteristics. >>>> Just a pain in the but to get to. >>>> Thanks >>>> Steve >>>>=20 >>>>=20 >>>>=20 >>>>> On 24 Mar 2017, at 8:56 pm, Charlie England = > = wrote: >>>>>=20 >>>>> Have you tried swapping the position of the pumps & making the = same measurements? It's hard to imagine it making that much difference, = but the bottom pump does have a tight right turn and then a sharp edged = 'T' turn to the left. I couldn't guess how much, but that would account = for at least some pressure increase.=20 >>>>>=20 >>>>> Charlie >>>>>=20 >>>>> On Fri, Mar 24, 2017 at 4:14 AM, steve Izett = > = wrote: >>>>> Hi Guys >>>>> Hi Peoples >>>>>=20 >>>>> Here is a photo of our parallel pumps plumbed with 3/8 aluminium = lines feed from a 28 gallon header tank above to the right. >>>>> Fuel then passes through the firewall and race filter before = feeding the rail and returning via the pressure regulator (4cyl toyota = reg)=20 >>>>> back through the firewall to the header tank, again in 3/8. >>>>>=20 >>>>> I did further measurements today.=20 >>>>>=20 >>>>> 1. Bottom Pump only ~8A and 48psi - Turning both pumps on this = pump draws 9.9A and produces a rail pressure of 65psi >>>>> 2. Bottom Pump only ~9.9A and 44psi - Turning both pumps on this = pump draws 14.8A and produces the same rail pressure of 65psi >>>>>=20 >>>>> So bottom pump goes from 8.0 -> 9.9A (1.9A increase) under higher = head pressure >>>>> Top pump goes from 9.9A -> 14.8A (5.1A increase) under same head!! >>>>>=20 >>>>> Clearly Pressure Reg bypass capacity is inadequate for both pumps, = perhaps with even one pump running.=20 >>>>> (I don=E2=80=99t have data of fuel pressure under load to see if = it drops as power is applied) >>>>> I think I modelled the pump layout/manifold after seeing someone = else=E2=80=99s and not sure if it causes any problems? >>>>> The pumps came from eBay stating that they were genuine Walbro and = now I=E2=80=99m wondering. >>>>>=20 >>>>> Cheers >>>>>=20 >>>>> Steve Izett >>>>> Glasair Super II RG Renesis 4 port RD1C EC3 EM3=20 >>>>>=20 >>>>>=20 >>>>>=20 >>>>>=20 >>>>>=20 >>>>> >=20 >=20 --Apple-Mail=_18AADCE4-C1C2-4198-98CC-DF5BCA90CED9 Content-Transfer-Encoding: quoted-printable Content-Type: text/html; charset=utf-8 Hi Charlie
In retrospect I=E2=80=99d seriously = consider building the 18 Gal header tank to suit a couple of submerged = pumps.
As my pumps live below the header tank = perhaps I could use turbine pumps, but does anyone make them for inline = use.

Cheers
Steve Izett




On = 26 Mar 2017, at 3:52 am, Charlie England <flyrotary@lancaironline.net> wrote:

I'd say the Walbro is the 'gold standard', but there are = plenty of others out there. If you could get into your header tank, or = have room for an additional small 'sump', you could try this type:
http://www.ebay.com/itm/360843503639?_trksid=3Dp2060353.m1438.l= 2649&ssPageName=3DSTRK%3AMEBIDX%3AIT

(Not saying to use a $12 = pump; just this style of pump.)

I considered this style for my in-tank = pumps, but they must be mounted vertically, and would have interfered = with the float of the fuel level sender.

Variations on this are in almost all = new cars, but in the gas tank. That one is ~1.5 inches diameter & = ~4.5 inches tall. Supposed to draw around 4 amps, so likely to be lower = flow than any of the gerotor pumps. Walbro makes one that's about the = same size, for typically around US $80-$90. They're turbine pumps, so = the bottom must be in fuel to prime. No easy way to attach an input, = except the purpose-made filter sock.

There are complete after-market = assemblies (single pump, filter sock, surge housing, 40-240 ohm fuel = level sender, and cover plate with both fuel & electrical = connections, for around US $30-$50.

Charlie

On Sat, = Mar 25, 2017 at 7:31 AM, steve Izett <flyrotary@lancaironline.net> wrote:
Thanks Steve
Really appreciate your helping me with this.
Is there an option other than the Walbro GSL393 or is this = the recommended inline pump at present?

Thanks
Steve



On 25 = Mar 2017, at 3:11 pm, Steven W. Boese <flyrotary@lancaironline.net> wrote:

Steve,

It is quite likely that getting your fuel system to behave = would not involve looking at more than can be seen in the photo of the = pumps that you posted.

Your item 1 would most likely be resolved if pumps of lower = flow ratings were used.  The stock RX7 regulator is very similar to = the one you are using and has worked well with those smaller = pumps.
Item 2 may be a result of the pump with the higher current = draw having more internal friction causing a load on the motor in = addition to the load from pumping fuel. 
The additional = load would slow the motor down resulting in less flow and lower = pressure.  It is possible that the decreased flow allows the = regulator to control the pressure at 44 psi.   Otherwise = there may be a constriction either within the pump or in the flow path = somewhere before the point where the flow of the two pumps are = combined.

Item A is unlikely = to be the case since if the seat in the regulator hadn't been broken, = there would be no flow, thus stalling the pump and causing a current = draw of over 20 amps.

In item = B, you are likely correct that the regulator cannot support = the flow from the high volume pump, thus causing the rail pressure to be = above the regulator's pressure rating.  It seems unlikely, = however, that the pressure would drop with the engine running since = there is so much more fuel circulating through the system (~70 gal/hr) = than what the engine could ever use (~20 gal/hr) even at full = throttle.

All this = discussion is directed toward the pumps or the system downstream of = them.  Consider also the system between the tank and the = pumps.  Trying to draw 140 gal/hr with the two pumps through a = single -6 line, and also through a 20 to 40 micron screen if one is = installed as required by the pump manufacturer, would have to involve a = very low pressure at the inlet to the pumps.  This seems to be a = very good recipe for vapor lock.

Steve Boese     


From: Rotary = motors in aircraft <flyrotary@lancaironline.net> on = behalf of steve Izett <flyrotary@lancaironline.net>
Sent: Friday, = March 24, 2017 6:19:23 PM
To: Rotary = motors in aircraft
Subject: [FlyRota= ry] Re: Fuel Pump current and pressures
 
I=E2=80=99ll remove the pumps to confirm the model and do = some testing.
In old notes I see I thought they were = GSL392=E2=80=99s.
The real obvious is as you show = in your calcs, the volume of fuel heading back to the tank with both = pumps running is huge.
With a piddly little 150hp = stock unit that probably has quite a small orifice for bypassing = regulation, quite large pressures are going to build up.
I have the two problems though.
1. Flow = capacity of the regulator if I want to be able to run both pumps = simultaneously.
2. The difference between the pumps = currently. Why does one draw more current and produce less = pressure?

What = is causing the 4psi pressure differential:
One of = the pumps draws more current but the pressure has dropped from 48 to = 44psi. Now either one of two things could bring this about in my = thinking.
A. The higher current pump is delivering = very low fuel flow and so measures a rail pressure drop, meaning not = enough fuel pressure/flow to break the regulator seat. But I think I can = remember hearing fuel returning to the tank when either pump is running. = OR
B. The other pump is delivering at idle more = fuel pressure/flow than the the regulator can cope with so is higher = than the regulation pressure. So as the engine begins to use fuel the = rail pressure would drop back to the regulator pressure. (I cant = remember seeing this during engine testing, but I wasn=E2=80=99t looking = for it and Ive deleted some video of engine monitoring).

Have I missed = something?

Steve
On 25 Mar 2017, at 3:11 am, Steven W. Boese = <flyrotary@lancaironline.net> wrote:

Steve,

Examining the Walbro pump specifications at:
https://walbrofuelpumps.com/walbro-gsl-series-universal-inline-fuel-pumps
and looking at the data for the = individual pumps enables the following speculation.  It is only = speculation.

If you have = GSL392 pumps, as are almost all of the pumps listed now on Ebay, they = have a flow rating of 255 L/hr (70 gal/hr) at 40 psi and 8 amps.  = If your engine could use 20 gal/hr, that would still leave 50 gal/hr = returning to the tank.  With two of those pumps running at 65 psi =  
they should be moving 60 = gal/hr each with each drawing 10 amps.  If your engine was using 20 = gal/hr, 100 gal/hr would be returning to the tank.  This would seem = to be moving much more fuel than necessary.  Depending on where you = are measuring the fuel pressure and the details of the fuel flow path, = the pressure at the pump outlet (and the current draw) could be much = higher with flow rates this high. 

At these flow rates, your whole fuel = system could be a constriction and any fittings (tee's etc) may = have significant effects.

That is why it would be good to establish which pumps you are = using either by identifying them or measuring their flow rates.  = GSL393  pumps (45 gal/hr at 40 psi and 5 amps) might = be more appropriate for your installation if your present pumps = have high flow ratings.

Steve Boese

 
From: Rotary = motors in aircraft <flyrotary@lancaironline.net> on = behalf of Charlie England <flyrotary@lancaironline.net>
Sent: Friday, = March 24, 2017 12:24 PM
To: Rotary = motors in aircraft
Subject: [FlyRota= ry] Re: Fuel Pump current and pressures
 
Hey Steve, 

Just re-read your post with the measurements, = & I'm not sure which had higher current draw. It was the bottom = pump, right?

Thanks,

Charlie

On 3/24/2017 8:06 AM, = Stephen Izett wrote:
Good idea Charlie. 
That would clear it = up as far as the manifold flow characteristics.
Just a = pain in the but to get to.
Thanks
Steve



On 24 = Mar 2017, at 8:56 pm, Charlie England <flyrotary@lancaironline.net> wrote:

Have you tried swapping the = position of the pumps & making the same measurements? It's hard to = imagine it making that much difference, but the bottom pump does have a = tight right turn and then a sharp edged 'T' turn to the left. I couldn't = guess how much, but that would account for at least some pressure = increase. 

Charlie

On Fri, Mar 24, 2017 at 4:14 AM, = steve Izett <flyrotary@lancaironline.net> wrote:
Hi Guys
Hi Peoples

Here is a photo of our parallel pumps = plumbed with 3/8 aluminium lines feed from a 28 gallon header tank above = to the right.
Fuel then passes through the firewall = and race filter before feeding the rail and returning via the pressure = regulator (4cyl toyota reg) 
back through the = firewall to the header tank, again in 3/8.

I did further measurements = today. 

1. = Bottom Pump only ~8A and 48psi - Turning both pumps on this pump = draws 9.9A and produces a rail pressure of 65psi
2. = Bottom Pump only ~9.9A and 44psi - Turning both pumps on this pump = draws 14.8A and produces the same rail pressure of 65psi

So bottom pump goes from = 8.0 -> 9.9A (1.9A increase) under higher head pressure
Top pump goes from 9.9A -> 14.8A (5.1A increase) under = same head!!

Clearly Pressure Reg bypass capacity is inadequate for both = pumps, perhaps with even one pump running. 
(I = don=E2=80=99t have data of fuel pressure under load to see if it drops = as power is applied)
I think I modelled the pump = layout/manifold after seeing someone else=E2=80=99s and not sure if it = causes any problems?
The pumps came from eBay = stating that they were genuine Walbro and now I=E2=80=99m = wondering.

Cheers

Steve Izett
Glasair Super II RG Renesis = 4 port RD1C EC3 EM3 





<Fuel Pres = Pumps.JPG>



= --Apple-Mail=_18AADCE4-C1C2-4198-98CC-DF5BCA90CED9--