Return-Path: Received: from mailout1.pacific.net.au ([61.8.0.84] verified) by logan.com (CommuniGate Pro SMTP 4.2.5) with ESMTP id 573876 for flyrotary@lancaironline.net; Mon, 20 Dec 2004 07:00:37 -0500 Received-SPF: none receiver=logan.com; client-ip=61.8.0.84; envelope-from=peon@pacific.net.au Received: from mailproxy2.pacific.net.au (mailproxy2.pacific.net.au [61.8.0.87]) by mailout1.pacific.net.au (8.12.3/8.12.3/Debian-7.1) with ESMTP id iBKC04Gx024342 for ; Mon, 20 Dec 2004 23:00:04 +1100 Received: from ar1 (ppp2CF1.dyn.pacific.net.au [61.8.44.241]) by mailproxy2.pacific.net.au (8.12.3/8.12.3/Debian-7.1) with SMTP id iBKBxm7L007586 for ; Mon, 20 Dec 2004 22:59:52 +1100 Message-ID: <001801c4e68b$66b88d70$f12c083d@ar1> From: "Leon" To: "Rotary motors in aircraft" References: Subject: Side Housing Machining & Nitriding Info was Re: [FlyRotary] Re: Engine Failure modes etc Date: Mon, 20 Dec 2004 22:59:02 +1100 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0015_01C4E6E7.7FE27D40" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2800.1409 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1409 This is a multi-part message in MIME format. ------=_NextPart_000_0015_01C4E6E7.7FE27D40 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Re: [FlyRotary] Engine Failure modes was Re: [FlyRotary] EHi Ken, =20 SIDE HOUISNG WARPAGE Just have a look at say the centre housing in particular. The water = gallery is about 1.5" away from the edge of the hole where the E-Shaft = goes through. It is also not of constant thickness. So the main = cooling on this area is the spill of oil out of the rotors. Lose the = cooling spill of oil from the rotor and as I mentioned, they buckle = like a belleville washer as this part gets hot, and the outer part in = contact with the coolant remains at the constant temp of the coolant. =20 When the engine is stripped, most of the scuffing is close around the = hole. This is because the rotor is slightly thicker close to the = eccentric shaft. The rotor gear sits slightly proud of the rotor side = face and there is an equivalent raised land on the other side of the = rotor. In fact, according to Mazda, standard clearance between the = rotor and the side housing is only 0.15 mm, which is about 6 thou for = you SAE trained people. So when the rotor gets hot and expands, and = the side housings buckle, it's "good night nurse!". MACHINING Over the years, I have had various people machine my side housings. A = Blanchard is ideal, but few engineering shops here in OZ have such an = animal, let alone know what one is!!. Alternatively, for some time, = I had a saw blade specialist do them for me on a vertical "SNO" grinder = (Vertical spindle, and a horizontal oscillating magnetic table). This = guy went to a lot of trouble to get me a really fine surface finish, = purchasing special fine grit stones just for my job. and setting the = spindle angle as close to vertical for the final cut so that the surface = wasn't dished. His table was huge, and he used to do several housings = at a time. Indeed, from the witness marks on factory housings, I'm = sure that this is the way the factory do theirs. However, most of the engineers over here just use a conventional = horizontal spindle surface grinder with an automatic oscillating = horizontal magnetic table. Several machine shops in Sydney do this on a = regular basis. There is also no reason that they couldn't be machined = in a lathe with a grinding attachment fitted to the tool post. I have = seen it done that way. Messy, but with the right setup, it can be = done. Obviously it needs a rather big lathe to swing the housings. In all cases, to do it accurately, this does requires special care. It = is imperative that the machined surfaces are at exactly 90 degrees to = the bore of the housings. . So for competition engines, I use a = special set of jigs to hold the end housings, and are set up off the = stationary gear bore holes. Mostly, it will work fine by assuming the = bore is square to the back face. (Obviously it doesn't matter on the = centre housing as the hole is only a clearance hole). However, one = bit of dirt, or a bruise, or a piece of gasket can upset the whole = shebang. The back faces should therefore be carefully stoned with a = large flat whetstone to make sure that they are both clean and dead = flat. In particular, ALL the front cover gasket needs to be = scrupulously removed from the front housing. Obviously, if the bore is not DEAD square to the working face, then = you end up with a bent motor, and the eccentric shaft will run at an = angle to the bearing bore. The rotors will also run at an angle to the = E-Shaft journals. The motor will then always be slightly tight, not = perform properly, and will forever burn oil and just be a right bitch. = I speak from personal experience!! So each housing needs to have the = bore-to-working-face angle checked after machining. I have had some = personal experience with engineers who are either not scrupulously clean = in their working habits, and/or just rely on the bore being square to = the back faces. Sometimes they are NOT. I've even had the odd factory = brand new housings that have been faulty in this respect. Sometimes you = can see this from the uneven wear on the stationary gear bearings. The = babbitt metal shows taper wear. So assume NOTHING, trust NOBODY, = check EVERYTHING!! NITRIDING Yes, it is always advisable to re-nitride once the surfaces have been = machined. The factory nitriding case is probably only about 6 - 8 thou = and machining will remove most if not all of the case. I've been = successfully nitriding my housings since the early '70s, even back in = the twin dizzy engine days, and well before Mazda started doing it in = 1978. I must give credit to a New Zealand bloke by the name of Billy = Shiels who introduced me to the process in 1974. It gives a really = obdurate surface that is virtually immune to wear, and also ends up = slightly porous, which holds the oil very well. =20 *** One word of warning. ALWAYS replace ALL the frost plugs and ALL the oil = gallery plugs after nitriding. The heat causes them to lose their = tension. Otherwise, you will end up with a milkshake in your sump, = and it will leak out of all the oil galleries!! Yes, it HAS happened = to me - that's how I know!! *** The best process is "Gas Nitiriding" where the housings are heated to = cherry red in an atmosphere of ammonia and allowed to cool slowly over a = couple of days. My current heat treater only does it on weekends. The = housings go into the oven on Friday, and they are allowed to cool = slowly over ther weekend.. It is important that the housings be placed = vertically. If they are lain down flat, the working faces will end up = like great auntie's washboard! Final preparation is to give them a good hand lap with some Clover = lapping paste. I normally lap two faces together rather than use a = dedicated lapping plate. Make sure you scrupulously wash off the lapping = paste and residue!! Hot soapy water, a nail brush and some bottle = brushes will do the job. Make sure all the oil galleries are = scrupulously clean. You will also find that the surfaces grow slightly when nitrided, so it = is imperative to clean all threaded holes out with a quality tap. MY = current heat treater has some sort of paint that he paints on to stop = most of this, but it's a bit hard to get right down inside the through = bolt hole threads. So always scrupulously clean out ALL threaded holes. = Otherwise, you will have some REAL fun getting correct torque = settings! Enjoy, Leon ----- Original Message -----=20 From: Ken Welter=20 To: Rotary motors in aircraft=20 Sent: Monday, December 20, 2004 5:10 PM Subject: [FlyRotary] Re: Engine Failure modes was Re: [FlyRotary] = Engine damage thoughts My experience has been that there are 2 basic failure modes when oil = pressure is lost. Hey leon I never thought about the side housings warping and causing the = seizure but it makes sense, how and who machines the side housings, do = you blanchard grind them or lap them and do you have to re nitride the = housings after machining them??? I got them kind of housings stacked like cord wood. Ken Welter ------=_NextPart_000_0015_01C4E6E7.7FE27D40 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Re: [FlyRotary] Engine Failure modes was Re: = [FlyRotary] E
Hi Ken, =20
 
SIDE HOUISNG WARPAGE
 
Just have a look at say the centre = housing in=20 particular.  The water gallery is about 1.5" away from the edge of = the hole=20 where the E-Shaft goes through.  It is also not of constant=20 thickness.  So the main cooling on this area is the spill of oil = out of the=20 rotors.  Lose the cooling spill of oil from the rotor and as I=20 mentioned,  they buckle like a belleville washer as this part gets=20 hot,  and the outer part in contact with the coolant remains at the = constant temp of the coolant. 
 
When the engine is stripped,  most = of the=20 scuffing is close around the hole.  This is because the rotor = is=20 slightly thicker close to the eccentric shaft.  The rotor gear sits = slightly proud of the rotor side face and there is an equivalent raised = land on=20 the other side of the rotor.  In = fact, =20 according to Mazda,  standard clearance between the rotor and the = side=20 housing is only 0.15 mm,  which is about 6 thou for you SAE trained = people.  So when the rotor gets hot and expands,  and the side = housings buckle,  it's "good night nurse!".
 
MACHINING
 
Over the years,  I have had = various people=20 machine my side housings.  A Blanchard is ideal,  but few = engineering shops here in OZ have such an animal,  let alone = know what=20 one is!!.  Alternatively,  for some time,  I had a saw = blade=20 specialist do them for me on a vertical "SNO" grinder (Vertical spindle, = and a=20 horizontal oscillating magnetic table).  This guy went to a lot of = trouble=20 to get me a really fine surface finish,  purchasing special fine = grit=20 stones just for my job.  and setting the spindle angle as close to = vertical=20 for the final cut so that the surface wasn't dished.  His table was = huge,  and he used to do several housings at a time.  = Indeed, =20 from the witness marks on factory housings,  I'm sure that = this is the=20 way the factory do theirs.
 
However,  most of the engineers = over here just=20 use a conventional horizontal spindle surface grinder with an automatic=20 oscillating horizontal magnetic table.  Several machine shops in = Sydney do=20 this on a regular basis.  There is also no reason that they = couldn't=20 be machined in a lathe with a grinding attachment fitted to the = tool=20 post.  I have seen it done that way.  Messy,  but with = the right=20 setup,  it can be done.  Obviously it needs a rather big lathe = to=20 swing the housings.
 
In all cases,  to do it = accurately, this does=20 requires special care.  It is imperative that the machined = surfaces=20 are at exactly 90 degrees to the bore of the housings.  .  So = for=20 competition engines, I use  a special set of jigs to hold the end=20 housings,  and are set up off the stationary gear bore holes.  = Mostly,  it will work fine by assuming the bore is square to the = back=20 face. (Obviously it doesn't matter on the centre housing as the = hole is=20 only a clearance hole).   However,  one bit of = dirt,  or a=20 bruise,  or a piece of gasket can upset the whole shebang.  = The back=20 faces should therefore be carefully stoned with a large flat = whetstone to=20 make sure that they are both clean and dead flat.  In = particular,  ALL=20 the front cover gasket needs to be scrupulously removed from the front=20 housing.
 
Obviously,  if the bore is not = DEAD square to=20 the working face,  then you end up with a bent motor,  and the = eccentric shaft will run at an angle to the bearing bore.  The = rotors will=20 also run at an angle to the E-Shaft journals.  The motor will then = always=20 be slightly tight, not perform properly,  and will forever = burn oil=20 and just be a right bitch.  I speak from personal = experience!!  So=20 each housing needs to have the bore-to-working-face angle checked after=20 machining.  I have had some personal experience with engineers who = are=20 either not scrupulously clean in their working habits,  and/or = just=20 rely on the bore being square to the back faces. Sometimes they are = NOT.=20  I've even had the odd factory brand new housings that have been = faulty in=20 this respect.  Sometimes you can see this from the uneven wear on = the=20 stationary gear bearings. The babbitt metal shows taper wear.  = So  assume NOTHING,  trust NOBODY,  check=20 EVERYTHING!!
 
NITRIDING
 
Yes,  it is always advisable to = re-nitride=20 once the surfaces have been machined.  The factory nitriding case = is=20 probably only about 6 - 8 thou and machining will remove most if not all = of the=20 case.  I've been successfully nitriding my housings since the early = '70s,  even back in the twin dizzy engine days,  and well = before Mazda=20 started doing it in 1978. I must give credit to a New Zealand bloke = by the=20 name of Billy Shiels who introduced me to the process in 1974.  It = gives a=20 really obdurate surface that is virtually immune to wear,  and also = ends up=20 slightly porous,  which holds the oil very well.  =
 
***
One word of warning.  ALWAYS = replace ALL the=20 frost plugs and ALL the oil gallery plugs after nitriding.  The = heat causes=20 them to lose their tension.  Otherwise,  you will end up with = a=20 milkshake in your sump,  and it will leak out of all the oil=20 galleries!!  Yes,  it HAS happened to me - that's how I=20 know!!
***
 
The best process is "Gas = Nitiriding"  where=20 the housings are heated to cherry red in an atmosphere of ammonia = and=20 allowed to cool slowly over a couple of days.  My current heat = treater only=20 does it on weekends.  The housings go into the oven on = Friday,  and=20 they are allowed to cool slowly over ther weekend..  It is = important that=20 the housings be placed vertically.  If they are lain down = flat,  the=20 working faces will end up like great auntie's washboard!
 
Final preparation is to give them a = good=20 hand lap with some Clover lapping paste.  I normally lap two = faces=20 together rather than use a dedicated lapping plate. Make sure = you=20 scrupulously wash off the lapping paste and residue!!  Hot soapy=20 water,  a nail brush and some bottle brushes will do the = job. =20 Make sure all the oil galleries are scrupulously clean.
 
You will also find that the surfaces = grow slightly=20 when nitrided,  so it is imperative to clean all threaded = holes out=20 with a quality tap.  MY current heat treater has some sort of paint = that he=20 paints on to stop most of this,  but it's a bit hard to get right = down=20 inside the through bolt hole threads.  So always scrupulously clean = out ALL=20 threaded holes.  Otherwise,  you will have some REAL fun = getting=20 correct torque settings!
 
Enjoy,
 
Leon
----- Original Message -----
From:=20 Ken=20 Welter
Sent: Monday, December 20, 2004 = 5:10=20 PM
Subject: [FlyRotary] Re: Engine = Failure=20 modes was Re: [FlyRotary] Engine damage thoughts

 
My = experience has=20 been that there are 2 basic failure modes when oil pressure is=20 lost.



  Hey leon

  I never thought about the side = housings=20 warping and causing the seizure  but it makes sense, how and who = machines=20 the side housings, do you blanchard grind them or lap them and do you = have to=20 re nitride the housings after machining them???
 I got them kind of housings stacked = like cord=20 wood.

  Ken=20 Welter
------=_NextPart_000_0015_01C4E6E7.7FE27D40--