From: "Stephen Izett stephen.izett@gmail.com" <flyrotary@lancaironline.net>
Received-SPF: pass
 receiver=logan.com; client-ip=209.85.215.181; envelope-from=stephen.izett@gmail.com
Content-Type: text/plain;
	charset=utf-8
Content-Transfer-Encoding: quoted-printable
Mime-Version: 1.0 (Mac OS X Mail 13.4 \(3608.120.23.2.4\))
Subject: Re: [FlyRotary] cooling modes
Date: Sun, 22 Nov 2020 06:08:09 +0800
References: <list-968622@logan.com>
To: Rotary motors in aircraft <flyrotary@lancaironline.net>
In-Reply-To: <list-968622@logan.com>
Message-Id: <3AFF3CD4-7979-4CA8-AEED-D5E984D1364A@gmail.com>

I=E2=80=99m with you Charlie.

Many variables!
If its cooling well enough after first flight then perhaps you could =
leave it as is.
But having faster aircraft and therefore the possibility of reducing =
cooling drag for speed or economy, being able to measure whats actually =
going on is very helpful.

Cheers
Steve Izett
> On 22 Nov 2020, at 1:10 am, Charlie England ceengland7@gmail.com =
<flyrotary@lancaironline.net> wrote:
>=20
> You're right, and I should have been more clear about that in my =
posts. I started out trying to use the streamline expansion formulas =
from the round inlet to the beginning of the 'wedge', but I'm not sure =
how much (if any) survived my tweaking to get flow across the full face =
of the core. What I was trying to give in my previous post was a mental =
image of how the back of the wedge isn't straight, but is more like the =
flatter section of a 'French curve', where the aft end almost touches =
the core. Quick&dirty drawing attached, which ignores actual inlet =
geometry.=20
>=20
> I suspect that what Steve and I have both attempted strays away from =
the idealized wedge diffusers that have been documented. The ones I've =
seen in the texts typically show a rectangular inlet that is equal to =
the core in width, and some percentage of the core depth in height. The =
same applies to the streamline diffusers. And I strongly suspect that =
the actual streamline equation depends on the resistance of the core at =
the end of the 'trumpet' shape, for it to function optimally. With no =
core resistance, a pure exponential horn (which is close, but not =
identical) would likely be more efficient.=20
>=20
> I struggle enough with 'good enough'; 'perfect' can easily become the =
enemy. :-)
>=20
> Charlie
>=20
>=20
>=20
> On Fri, Nov 20, 2020 at 11:04 PM Stephen Izett stephen.izett@gmail.com =
<flyrotary@lancaironline.net> wrote:
> Hi guys.
>=20
> I think we are talking about two different beasts.
> K&W Chapter 12 describes the Streamline (Trumpet) shape.
> However we also use Oblique Flow (Wedge) diffusers as descibed by =
London Clopfer and Wolf.
> And we use both together, a Streamline diffuser feeding an Oblique =
diffuser as in Tracy=E2=80=99s RV8 setup that many of us modelled from.
>=20
> Steve Izett
>=20
>=20
> > On 21 Nov 2020, at 11:34 am, Charlie England ceengland7@gmail.com =
<flyrotary@lancaironline.net> wrote:
> >=20
> > I agree, but I'd describe it more like the core was down the =
centerline of a streamline diffuser, and the air feeds from the wide =
end. Half the original diffuser disappears.
> >=20
> > On 11/20/2020 8:22 PM, David COOK hoursaway1@comcast.net wrote:
> >> Wedge is kind of a poor term for what is needed, more of a bell =
mouth like on a trumpet with the rad. being at the large end. DRC RV6A =
Rotary
> >>> On 11/20/2020 9:03 PM Charlie England ceengland7@gmail.com =
<flyrotary@lancaironline.net> wrote:
> >>> =20
> >>> =20
> >>> Hi Bruce,=20
> >>>=20
> >>> The driver for me was trying to remain within the constraints of =
money, a stock cowl with minimum 'surgery', with nothing hanging out of =
the cowl. Another driver was due to 'standing on the shoulders of =
giants'; Tracy Crook does really light weight installations, and his =
calcs showed that at least in his RV4, he needed the weight of the heat =
exchangers to be up front, to keep CG in the right place. My FWF weighs =
almost exactly the same as an O-320, so for now, I'll keep the  wedge in =
place. Please be aware that my setup may not survive to the 2nd flight. =
:-)=20
> >>>=20
> >>> In retrospect, I might well have done the same as Steve mentioned, =
and at least try to tuck the rad into the space behind/below the engine =
and feed it with a duct up to the prop.=20
> >>>=20
> >>> Charlie=20
> >>>=20
> >>> On 11/20/2020 6:59 PM, Bruce Cosgrove bcosgrove@mac.com wrote:
> >>>> Hi Steve, Charlie, Matt, Roy and others
> >>>> I am in the process making up a  cooling system for a Tango 2 and =
was also planning to use the wedge shaped diffuser design, however the =
information that I have been reading from the homebuilt airplanes site =
discussing alternate approaches with less drag and improved efficiency? =
has me re thinking my design towards a belly mounted radiator.  The =
comments under the heading  Successful Auto Conversion Trials, =
Tribulations and Tips, I found particularly interesting - in particular =
input from Russell and RJ6ejGUY who both have devised systems that are =
working well for them  Belly radiators may not be everyones idea of how =
to proceed but I like the idea of relatively unimpeded air flow and =
freeing up space in the cowling.  any feedback would be much =
appreciated.
> >>>> =20
> >>>> Bruce
> >>>>=20
> >>>> On 21 November 2020 at 10:00:57, Stephen Izett =
stephen.izett@gmail.com (flyrotary@lancaironline.net) wrote:
> >>>>=20
> >>>>> Nice work Charlie.=20
> >>>>>=20
> >>>>> Steve=20
> >>>>>=20
> >>>>> > On 21 Nov 2020, at 6:50 am, Charlie England =
ceengland7@gmail.com <flyrotary@lancaironline.net> wrote:=20
> >>>>> >=20
> >>>>> > My (untested in combat) version for my radiator looks similar, =
but because the inlet is offset, I had to use a turning vane to get flow =
into the lower front of the core. The outside doesn't give a good idea =
on how 'tight' the inner surface is. I had to use expanding foam to fill =
the inside, then re-shape, then glass over the foam. As I said, the back =
1/4-1/3 of the diffuser tapers tightly to almost touch the core at the =
back.=20
> >>>>> >=20
> >>>>> > After observing both rotary and piston installations that =
work, I'm beginning to think that turning vanes may be the 'good enough' =
vs fighting for perfection in diffuser shape. My oil cooler core has a =
much more elaborate vane setup, driven by the very short duct length and =
even greater offset to the bottom edge of the core.=20
> >>>>> >=20
> >>>>> > Charlie=20
> >>>>> >=20
> >>>>> >=20
> >>>>> > On 11/20/2020 4:22 PM, Stephen Izett stephen.izett@gmail.com =
wrote:=20
> >>>>> >> Hi Roy=20
> >>>>> >>=20
> >>>>> >> Here is a photo of our Mocal 44 row oil cooler with diffuser. =
A K&M Streamline into a wedge.=20
> >>>>> >> Not sure of the dimensions of hand. The Mocal web site will =
have specs so you could get a rough idea of size.=20
> >>>>> >> I tested this one recently with a leaf blower approximating =
100 knot air flow and in that test it was very even.=20
> >>>>> >>=20
> >>>>> >> Having said that, I tested our water radiator wedge diffuser =
when I built it years ago and had it reasonably even=20
> >>>>> >> measuring pressures across the exit face, but now =
instrumenting it in flight its flow is vastly uneven.=20
> >>>>> >>=20
> >>>>> >> I have found that the flow is dependant not only on getting =
the diffuser shape right according to the various papers (eg. K&W, =
London, Oblique Flow Diffusers - Streamline )=20
> >>>>> >> But importantly the environment it ends up in. The papers =
share data for ideal conditions with for example unobstructed exit air =
paths.=20
> >>>>> >> My thoughts now are that this has a significant impact on the =
performance and therefore the advantage of instrumenting in flight and =
reshaping for actual conditions.=20
> >>>>> >>=20
> >>>>> >> The data I collected yesterday of our H2O diffuser reveals =
that the air flow in various points across the face are very different.=20=

> >>>>> >> Simplified, if I average the pressures at the five points I =
measured, then comparatively the back is up 20%, the centre up 25% and =
the front down 60% from that average figure.=20
> >>>>> >> I=E2=80=99m figuring a 20% difference is probably OK, but my =
front to back/middle difference of ~80% down in the front top of that =
radiator may hold significant room for improvement.=20
> >>>>> >> Further, my data suggests that as IAS rises those differences =
are amplified.=20
> >>>>> >> So I am now targeting the modifications for our worst case =
conditions of WOT sea level ~200hp and initial climb numbers of 115KIAS.=20=

> >>>>> >>=20
> >>>>> >> Cheers.=20
> >>>>> >>=20
> >>>>> >> Steve Izett=20
> >>>>> >>=20
> >>>>> >>=20
> >>>>> >>=20
> >>>>> >>> On 21 Nov 2020, at 3:37 am, ROY GLENN rglenn14@sbcglobal.net =
<flyrotary@lancaironline.net> wrote:=20
> >>>>> >>>=20
> >>>>> >>> Could some ne post a photo of their wedge shape and =
dimensions.=20
> >>>>> >>> Thanks Roy=20
>=20
>=20
> 	Virus-free. www.avast.com
> <wedge (Small).jpg>--
> Homepage:  http://www.flyrotary.com/
> Archive and UnSub:   =
http://mail.lancaironline.net:81/lists/flyrotary/List.html