X-Virus-Scanned: clean according to Sophos on Logan.com From: Received: from omr-m005e.mx.aol.com ([204.29.186.5] verified) by logan.com (CommuniGate Pro SMTP 6.1.10) with ESMTPS id 8609731 for flyrotary@lancaironline.net; Mon, 25 Apr 2016 10:08:07 -0400 Received-SPF: pass receiver=logan.com; client-ip=204.29.186.5; envelope-from=Lehanover@aol.com Received: from mtaomg-mce01.mx.aol.com (mtaomg-mce01.mx.aol.com [172.29.27.207]) by omr-m005e.mx.aol.com (Outbound Mail Relay) with ESMTP id 097CB380009F for ; Mon, 25 Apr 2016 10:07:49 -0400 (EDT) Received: from core-moa05a.mail.aol.com (core-moa05.mail.aol.com [172.27.97.15]) by mtaomg-mce01.mx.aol.com (OMAG/Core Interface) with ESMTP id ECE633800008A for ; Mon, 25 Apr 2016 10:07:47 -0400 (EDT) Full-name: Lehanover Message-ID: <31d4f2.425cb526.444f7eb2@aol.com> Date: Mon, 25 Apr 2016 10:07:46 -0400 Subject: Re: [FlyRotary] Re: Not the water jacktet, Corner seal buttons? To: flyrotary@lancaironline.net MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="part1_31d4f2.425cb526.444f7eb2_boundary" X-Mailer: AOL 9.8 sub 2019 X-Originating-IP: [74.140.106.236] x-aol-global-disposition: G DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=mx.aol.com; s=20150623; t=1461593269; bh=SBLEPKgSP76UG+5+OmGOtYOO4uVDcHUDdWL2PoFA6ik=; h=From:To:Subject:Message-ID:Date:MIME-Version:Content-Type; b=75LmfvD8Q5/CK8Lbajzde1kPguoN/HwjWtnbPGGs6wDWVNcjrDoenUWrOJL6h9SSW PZedE56XcEpVqnx2U0+UQGWRkjXM76Ga+JH+6jT0rFh5bpA2jC0e9zSZFWxyAjToOj OauKAAfa/SY7oQuEoHwxDWlK82iWn5wv5P9Qp3gI= x-aol-sid: 3039ac1d1bcf571e24b361fb --part1_31d4f2.425cb526.444f7eb2_boundary Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Content-Language: en The object of a restrictor is to prevent Cavitating water in the pump or= =20 pump inlet. It may also have the benefit of slightly raising the pressure= =20 inside the block. Other attempts include a smaller crank pulley to slow= the=20 pump a bit. In normal operation the radiator is the restrictor. There is= a=20 need to keep the inlet side of the pump at a high enough pressure to=20 prevent cavitation. In my case even with the tiny crank pulley the vanes= on the=20 pumps became rounded and less effective. So I added the restrictor. Just= =20 the thermostat outer ring at first and then worked my way down to 5/8".= It=20 is welded over the end of the water outlet so it is impossible to forget.= =20 What combination will work perfectly at 6,000 to 6,500 RPM is unknown to= me.=20 How restrictive is the radiator core. What size hoses and fittings. How lo= ng=20 are the runs and so-on. If the run from the outlet to the rad inlet is 1"= =20 hose and it is long enough to produce the required drag, then no restrict= or=20 is needed. No two installations will be the same. On every teardown, inspect the water pump vanes. They should stay as new= in=20 normal operation. =20 Lynn E. Hanover =20 =20 =20 =20 In a message dated 4/24/2016 6:52:36 P.M. Eastern Daylight Time, =20 flyrotary@lancaironline.net writes: =20 Lynn, Thanks for all the info even if it is from Dave=E2=80=99s grief.= The=20 restrictor on the water outlet sound very feasible. 5/8 seems a little= too=20 small, but you obviously see it work, would a 1 inch be too large?? Neil= . =20 =20 =20 From: _Lehanover_ (mailto:flyrotary@lancaironline.net) =20 Sent: Monday, April 25, 2016 12:37 AM To: _Rotary motors in aircraft_ (mailto:flyrotary@lancaironline.net) =20 Subject: [FlyRotary] Re: Not the water jacktet, Corner seal buttons? =20 =20 =20 The water temperature tells us the average of many temperatures collected= =20 inside the engine. Nothing specific and not the highest temperature =20 encountered by the coolant. The highest may be between the spark plugs or= below=20 the lower plug. The highest oil temperature is probably in the drainback= in=20 the center iron. That oil was cooling the rotors. It mixes with oil in the= =20 sump to hide its true temperature. Oil temperature reads directly from the= =20 throttle setting. As though the throttle was linked to the gage.Convention= al =20 oils do not absorb heat nor give it up readily. Racing synthetics do a muc= h=20 better job on that. I use 40 weight RedLIne racing synthetic in the sump,= =20 and RedLine racing 2 cycle oil in the fuel. The 40 weight sounds like too= =20 high a number but it pours like lighter fluid and has very low drag. Low= er=20 oil temps help lower the coolant temps. =20 Keep in mind that the return line from the radiator to the water pump =20 should be bigger than the pressure=20 line(s) to the radiator. Because the water pump generates low pressure in= =20 its inlet side. The reason for high pressure radiator caps is to prevent= =20 boiling in front of the pump. Notice that the lower hose in your car has= a =20 wire spring in it to keep it from being sucked flat closed when you rev th= e =20 engine. The boiling temperature of the coolant is a function of the actual= =20 temperature, and the actual pressure.=20 =20 Another option is to install a restrictor in the outlet hose leaving the= =20 engine. I have been told by so many that this will not help, that I quit= =20 mentioning restrictors. Mine is 5/8" and is a disc welded over the outlet= =20 casting (Norice) where the top hose leaves the engine. It has not been wor= king =20 I am told since 1980. So don't try that. (It works great) =20 The corner seals with the rubber plugs are to get through the California= =20 (CARB) tests for sales in California. They are fragile and lack a bit of stiffness. Use early corner seals that= =20 are solid.=20 =20 Lynn E. Hanover=20 =20 =20 =20 =20 In a message dated 4/24/2016 7:36:53 A.M. Eastern Daylight Time, =20 flyrotary@lancaironline.net writes: This isn't likely to be the problem you had but since the failure happene= d=20 with the same exact sequence, I thought I'd mention it. When returning= =20 from a flyin in Texas, i had checked my coolant level and when replacing= the=20 fill cap the little rubber shim I used to boost the cap pressure got lodg= ed=20 under the check valve in the cap. I climbed to 5000 ft. With all=20 parameters looking good except coolant pressure which looked low. Less= than 30=20 seconds after my last instrument scan the high coolant temp alarm went of= f=20 and I glided under idle power to nearest airport 8 miles away. =20 =20 Engine had boiled almost all coolant out past the check valve. Removed= =20 the shim under cap , Refilled coolant and all was well on way back to=20 Florida. =20 Sent from my iPad On Apr 23, 2016, at 12:54, David Leonard <_flyrotary@lancaironline.net_=20 (mailto:flyrotary@lancaironline.net) > wrote: Thanks for the Great review Lynn. As usual it should be saved in the=20 archives forever. =20 Thanks for all the comments everyone else too. I guess I should give a= =20 little more background on what I am seeing. Keep in mind that I am not= =20 figuring this out for the first time. My plane now has almost 900 hrs on= it. =20 Formation team, coast to coast, Reno racer, blah blah.... =20 I am not talking about a small coolant leak somewhere either. Rather,=20 everything is working fine and I am flying along and for no particular re= ason=20 it suddenly shits all the coolant out the overflow past the 20# cap,=20 leading to a deadstick landing at my home field. Afterward it still pass= ed a=20 leak test and will hold coolant pressure indefinitely. A short test flig= ht=20 before the re-build showed exhaust gasses gathering in the coolant system= . =20 It always does this to a small degee, but now it is much much worse - mak= ing=20 cross country travel unsafe. =20 Thinking the problem is an o-ring that finally got tired (last rebuild wa= s=20 about 250 hrs prior), I set upon a quick rebuild to just replace the=20 o-rings and other rubber bits. To address Lynn's other points: =20 Bleed holes at the top of the engine - check =20 Covering the exhaust holes in the center iron - check, I think. The holes= =20 in the center housing are covered by the intake manifold. At the first= =20 rebuild I put potter's clay in the (dead end) passages, but it has long= since=20 disappeared. I don't think I ever had nozzles on my turbo rotor housings= . =20 I will be more careful about this issue on my next rebuild. =20 Coolant - I am back to running 50/50 since I don't have overheating =20 issues, my coolant temp runs 140 to 170 and peaks at 190 in the worst of= =20 conditions. But perhaps I would be better off with a higher water percen= tage to=20 help prevent nucleate boiling and/or carry more heat away.. =20 Silicone in the gasket groves. I used hylomar on my first rebuild but it= =20 was a pain. I have been just using Vaseline to hold the o-rings in place= =20 for the build. Seemed to work in the past but I Lynn says it, I will use= GE=20 Silicone to create a lasting seal. =20 Side seal clearance. I did not replace the side seals this last rebuild= =20 and I could tell the clearance was not zero. It is definitely a contribu= tor=20 to my oil blow-by problem. Note to self - Just replace the seals when=20 rebuilding. But all that stuff is getting more and more expensive lately= . =20 =20 Lapping the Irons - I suspect that this is really my main problem both in= =20 terms of coolant and oil blow by. Wish we could just buy new ones. Once= =20 lapped I assume the Teflon o-rings become too fat and we have to go back= to=20 the stock ones, or even custom? =20 sealing the legs of the rotor housing to prevent the seeping of hot oil..= =20 this one bit me this time. I didn't do it and have seeping oil. Good=20 thing I am about to do another rebuild anyway.... =20 Another thing I didnt do was install the rubber buttons in the corner =20 seals. Every time I tear down an engine they are long gone (especially= the=20 ones from RWS) so I figured they are not particularly needed. Since they= are=20 a pain, I skipped it. But now me thinks that there was a reason that I= had=20 to tear down the engine, and a reason Mazda included them in the first=20 place. Lynn? =20 Thanks for all the help everyone! =20 David Leonard Turbo Rotary RV-6 N4VY =20 =20 =20 On Fri, Apr 22, 2016 at 11:58 PM, Lehanover <_flyrotary@lancaironline.net_= =20 (mailto:flyrotary@lancaironline.net) > wrote: There are a number of methods used to blow the coolant out of a Mazda=20 engine. The most popular is actually overheating the engine by not removi= ng=20 every spec of air from the cooling system. Many bubbles form in the top= of the=20 cooling jacket. They do not move unless the engine RPM produces ample=20 coolant velocity. As in a take off or taxi test. Then they form up un front of the high=20 mounted water pump and cavitate the pump to a standstill. Seconds later= there=20 is no coolant to be found. Drill a hole in the top of the center iron and= =20 install a bleed valve. Can be a Shrader valve with a metal cap. This to= leave=20 open while filling with coolant so as to let said air out of the very top= =20 of the engine. =20 =20 Coolant means distilled water. 10% glycol. One teaspoonful of Dawn =20 dishwashing soap or Redline Water wetter per bottle instructions. A bottl= e of=20 Barsleak. Yes that brown water with the Rat poop in it. If it is good eno= ugh=20 for a $100,000.00 Cosworth V-8, then its good enough for a Mazda. A 22= pound=20 Stant pressure cap and a recovery bottle so you can see where the fluids= =20 went.=20 =20 Note that the rotor housings act as though they are made of hard rubber.= =20 If one is only slightly warped it goes flat when the stack is torqued up.= No=20 problem. If a iron is warped it is scrap. You can lap facing surfaces of= =20 Irons against each other. This where minimal wear has been observed. For= =20 heavy wear you can=20 have an automotive shop take off a few thousandths, then lap the piece=20 against its partner to remove most tooling marks. You can also do a great= job=20 with a cleaning tank using kerosene and an air powered random orbital=20 sander. Glue on some 320 wet or dry paper and sand the wear surface under= a=20 stream of Kerosene. The little circles are great for holding oil. Once clean check for cracks= .=20 =20 =20 And now into the weeds.......Many schemes have been tried in the pollutio= n=20 control area. Most involve exhaust gasses being loose inside the rotor=20 housings where compressed air from an air pump was fed into the exhaust= stream=20 through two steel nozzles that project into the exhaust port. This air=20 came from a port in the center iron that is covered by the intake manifol= d.=20 (In most every case). Those nozzles must be plugged. The supply port in the center iron must be covered. The nozzles do not fi= t=20 tight and may leak fresh air into the exhaust stream and supply a crackin= g=20 and popping sound that confuses the tuner and Oxygen sensor. =20 The center iron should match the coolant openings pattern of both rotor= =20 housings. On assembly add a thin coating of GE 100% silicone tub and tile= =20 sealant to the faces of the land areas that hold the "O" rings. Works gre= at to=20 cure the tiny leaks and be sure to put it on the down legs clear to the= =20 pan rail to keep oil from seeping out when hot. I use the same GE silicon= e to=20 seal the exhaust flanges to the engine. Stays leak free until you pry off= =20 the exhaust system for service. If there is no gasket, it cannot fail. I= =20 have used a copy of the RX-2 RX-3 cooling system since 1980 Never an=20 overheated engine. I can publish that drawing if anyone wants it.=20 =20 Set the side seal end clearance to zero, not .002" to 004" as seen in som= e=20 books. If both side seals and the gives you as close to advertised=20 compression as is possible. Quick starts even when hot. If the side seals= and the=20 corner seal pop back up after being compressed, then it is perfect. Good= to=20 9,600 RPM. =20 More on bullet proof Rotaries as soon as I remember it.......The last=20 racer is available and the spare (250 HP) engine is available. =20 Lynn E. Hanover =20 =20 =3D --part1_31d4f2.425cb526.444f7eb2_boundary Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Content-Language: en
The object of a restrictor is to prevent Cavitating water in the= pump=20 or pump inlet. It may also have the benefit of slightly raising the= =20 pressure inside the block.  Other attempts include a smaller cra= nk=20 pulley to slow the pump a bit. In normal operation the radiator is the=20 restrictor.  There is a need to keep the inlet side of the pump at a= high=20 enough pressure to prevent cavitation.  In my case even with the tiny= crank=20 pulley the vanes on the pumps became rounded and less effective. So I adde= d the=20 restrictor. Just the thermostat outer ring at first and then worked my way= down=20 to 5/8". It is welded over the end of the water outlet so it is impossible= to=20 forget. What combination will work perfectly at 6,000 to 6,500 RPM is unkn= own to=20 me. How restrictive is the radiator core. What size hoses and fittings. Ho= w long=20 are the runs and so-on. If the run from the outlet to the rad inlet is 1"= hose=20 and it is long enough to produce the required drag, then no restrictor is= =20 needed. No two installations will be the same.
On every teardown, inspect the water pump vanes. They should stay as= new in=20 normal operation.   
Lynn E. Hanover
 
 
 
In a message dated 4/24/2016 6:52:36 P.M. Eastern Daylight Time,=20 flyrotary@lancaironline.net writes:
Lynn,   Thanks for all the info even if it is from Dave= =E2=80=99s=20 grief.  The restrictor on the water outlet sound very feasible.&nbs= p; 5/8=20 seems a little too small, but you obviously see it work, would a 1 inch= be too=20 large??  Neil.
 
The corner seals with the rubber plugs are to get through the Calif= ornia=20 (CARB) tests for sales in California.
They are fragile and lack a bit of stiffness. Use early corner seal= s that=20 are solid.
 
Lynn E. Hanover
 
 
 
In a message dated 4/24/2016 7:36:53 A.M. Eastern Daylight Time,=20 flyrotary@lancaironline.net writes:
This isn't likely to be the problem you had but since the failure= =20 happened with the same exact sequence, I thought I'd mention it. = When=20 returning from a flyin in Texas, i had checked my coolant level and wh= en=20 replacing the fill cap the little rubber shim I used to boost the cap= =20 pressure got lodged under the check valve in the cap.   I cl= imbed=20 to 5000 ft. With all parameters looking good except coolant pressure= which=20 looked low.  Less than 30 seconds after my last instrument scan= the=20 high coolant temp alarm went off and  I glided under idle power= to=20 nearest airport 8 miles away. 
 
Engine had boiled almost all coolant out= past the=20 check valve.  Removed the shim under cap , Refilled coolant and= all was=20 well on way back to Florida.  

Sent from my iPad

On Apr 23, 2016, at 12:54, David Leonard <flyrotary@lancaironline.ne= t>=20 wrote:

Thanks for the Great review Lynn.  As usual it= should be=20 saved in the archives forever.  =20
 
Thanks for all the comments everyone else too.  I guess I= should=20 give a little more background on what I am seeing.  Keep in min= d that=20 I am not figuring this out for the first time.  My plane now ha= s=20 almost 900 hrs on it.  Formation team, coast to coast, Reno=20 racer,  blah blah....
 
I am not talking about a small coolant leak somewhere either.&n= bsp;=20 Rather, everything is working fine and I am flying along and for no= =20 particular reason it suddenly shits all the coolant out the overflow= past=20 the 20# cap, leading to a deadstick landing at my home field. = =20 Afterward it still passed a leak test and will hold coolant pressure= =20 indefinitely.  A short test flight before the re-build showed= exhaust=20 gasses gathering in the coolant system.  It always does this to= a=20 small degee, but now it is much much worse - making cross country tr= avel=20 unsafe.
 
Thinking the problem is an o-ring that finally got tired (last= =20 rebuild was about 250 hrs prior), I set upon a quick rebuild to just= =20 replace the o-rings and other rubber bits.  To address Lynn's= other=20 points:
 
Bleed holes at the top of the engine - check
 
Covering the exhaust holes in the center iron - check, I think.= The=20 holes in the center housing are covered by the intake manifold. = ; At=20 the first rebuild I put potter's clay in the (dead end) passages, bu= t it=20 has long since disappeared.  I don't think I ever had nozzles= on my=20 turbo rotor housings.  I will be more careful about this issue= on my=20 next rebuild.
 
Coolant - I am back to running 50/50 since I don't have overhea= ting=20 issues, my coolant temp runs 140 to 170 and peaks at 190 in the wors= t of=20 conditions.  But perhaps I would be better off with a higher wa= ter=20 percentage to help prevent nucleate boiling and/or carry more heat= =20 away..
 
Silicone in the gasket groves.  I used hylomar on my first= =20 rebuild but it was a pain.  I have been just using Vaseline to= hold=20 the o-rings in place for the build.  Seemed to work in the past= but I=20 Lynn says it, I will use GE Silicone to create a lasting seal.
 
Side seal clearance.  I did not replace the side seals thi= s last=20 rebuild and I could tell the clearance was not zero.  It is=20 definitely a contributor to my oil blow-by problem.  Note to se= lf -=20 Just replace the seals when rebuilding.  But all that stuff is= =20 getting more and more expensive lately. 
 
Lapping the Irons - I suspect that this is really my main probl= em=20 both in terms of coolant and oil blow by.  Wish we could just= buy new=20 ones.  Once lapped I assume the Teflon o-rings become too fat= and we=20 have to go back to the stock ones, or even custom?
 
sealing the legs of the rotor housing to prevent the seeping of= hot=20 oil..  this one bit me this time.  I didn't do it and have= =20 seeping oil.  Good thing I am about to do another rebuild=20 anyway....
 
Another thing I didnt do was install the rubber buttons in the= corner=20 seals.  Every time I tear down an engine they are long gone=20 (especially the ones from RWS) so I figured they are not particularl= y=20 needed.  Since they are a pain, I skipped it.  But now me= thinks=20 that there was a reason that I had to tear down the engine, and a re= ason=20 Mazda included them in the first place.  Lynn?
 
Thanks for all the help everyone!
 
David Leonard
Turbo Rotary RV-6 N4VY
 
 
 
On Fri, Apr 22, 2016 at 11:58 PM, Lehanover= <flyrotary@lancaironline.net> wrote:
 
There are a number of methods used to blow the coolant out of= a=20 Mazda engine. The most popular is actually overheating the engine= by not=20 removing every spec of air from the cooling system. Many bubbles= form in=20 the top of the cooling jacket. They do not move unless the engine= RPM=20 produces ample coolant velocity.
As in a take off or taxi test. Then they form up un front of= the=20 high mounted water pump and cavitate the pump to a standstill. Sec= onds=20 later there is no coolant to be found. Drill a hole in the top of= the=20 center iron and install a bleed valve. Can be a Shrader valve with= a=20 metal cap. This to leave open while filling with coolant so as to= let=20 said air out of the very top of the engine. 
 
Coolant means distilled water. 10% glycol. One teaspoonful of= Dawn=20 dishwashing soap or Redline Water wetter per bottle instructions.= A=20 bottle of Barsleak. Yes that brown water with the Rat poop in it.= If it=20 is good enough for a $100,000.00 Cosworth V-8, then its good enoug= h for=20 a Mazda.  A 22 pound Stant pressure cap and a recovery bottle= so=20 you can see where the fluids went.
 
Note that the rotor housings act as though they are made of= hard=20 rubber. If one is only slightly warped it goes flat when the stack= is=20 torqued up. No problem. If a iron is warped it is scrap. You can= lap=20 facing surfaces of Irons against each other. This where minimal we= ar has=20 been observed. For heavy wear you can
have an automotive shop take off a few thousandths, then lap= the=20 piece against its partner to remove most tooling marks. You can al= so do=20 a great job with a cleaning tank using kerosene and an air powered= =20 random orbital sander. Glue on some 320 wet or dry paper and sand= the=20 wear surface under a stream of Kerosene.
The little circles are great for holding oil. Once clean chec= k for=20 cracks.    
 
And now into the weeds.......Many schemes have been tried in= the=20 pollution control area. Most involve exhaust gasses being loose in= side=20 the rotor housings where compressed air from an air pump was fed= into=20 the exhaust stream through two steel nozzles that project into the= =20 exhaust port. This air came from a port in the center iron that is= =20 covered by the intake manifold. (In most every case). Those nozzle= s must=20 be plugged.
The supply port in the center iron must be covered. The nozzl= es do=20 not fit tight and may leak fresh air into the exhaust stream and= supply=20 a cracking and popping sound that confuses the tuner and Oxygen=20 sensor.
 
The center iron should match the coolant openings pattern of= both=20 rotor housings. On assembly add a thin coating of GE 100% silicone= tub=20 and tile sealant to the faces of the land areas that hold the "O"= rings.=20 Works great to cure the tiny leaks and be sure to put it on the do= wn=20 legs clear to the pan rail to keep oil from seeping out when hot.= I use=20 the same GE silicone to seal the exhaust flanges to the engine. St= ays=20 leak free until you pry off the exhaust system for service. If the= re is=20 no gasket, it cannot fail. I have used a copy of the RX-2 RX-3 coo= ling=20 system since 1980 Never an overheated engine. I can publish that= drawing=20 if anyone wants it.
 
Set the side seal end clearance to zero, not .002" to 004" as= seen=20 in some books. If both side seals and the gives you as close to=20 advertised compression as is possible. Quick starts even when hot.= If=20 the side seals and the corner seal pop back up after being compres= sed,=20 then it is perfect. Good to 9,600 RPM.
 
More on bullet proof Rotaries as soon as I remember it.......= The=20 last racer is available and the spare (250 HP) engine is available= .=20
 
Lynn E. Hanover
 
 
 
=3D
=
--part1_31d4f2.425cb526.444f7eb2_boundary--