separating fiction from reality... a couple of days on the DYNO
For what it's worth, my car has ID1000 injectors in all 4 ports on the Xcessive LIM, with the original primary injector ports blocked off. The idle and throttle response aren't a problem, but the car has stock ports and a different ECU so your experience may be different.
Part of the reason I changed to the current fuel system was because one of those little plastic diffusers crumbled when trying to remove the injectors to have them cleaned.
Part of the reason I changed to the current fuel system was because one of those little plastic diffusers crumbled when trying to remove the injectors to have them cleaned.
FWIW, I always remove the diffusers, and run higher fuel pressures to increase atomization as well as flow. On my 20B I run 65PSI base pressure. Theoretically this is hard on injectors, but I have not had one fail yet.
If someone was motivated to try and improve their idle efficiency, they could modify their manifold to include a venturi at the injection point, but I don't see the relevance to this thread.
If someone was motivated to try and improve their idle efficiency, they could modify their manifold to include a venturi at the injection point, but I don't see the relevance to this thread.
modest beginnings
Joined: Jan 2012
Posts: 50
Likes: 0
From: springfield
nsk1, go back and look at the picture in post 677 and you'll see the diffuser in the left port. The diffuser is a plastic piece under the primary injectors that helps spread out, diffuse, the fuel spray from the injector into the air flow. You don't want to tear your motor apart to remove them!
here is what a normal diffuser is supposed to look like (pic front atkins rotary's website)
you want them if you care about low-load driveability, fuel efficiency, and dont want your fuel shooting straight across onto the wall of your intake.
maybe in high boost situations, you dont have to worry about that due to the speed of the air flowing by, but i do not know for sure if that is true
you want them if you care about low-load driveability, fuel efficiency, and dont want your fuel shooting straight across onto the wall of your intake.
maybe in high boost situations, you dont have to worry about that due to the speed of the air flowing by, but i do not know for sure if that is true
Howard asked me to comment on the diffuser issue so I am going to speak more to Howard's particular situation.
A short answer/reply is that I am confident the benefits of removing the failure point outweigh potential drawbacks of removing the diffusers. This is partly because we don't have the resources to fully understand the interaction between a non-stock injector spray pattern and the diffusers. For all we know, the diffusers are helpful with some injectors and useless/counter-productive with others.
Injector spray patterns, injector targeting, and injection timing are very complicated things during the actual engine development cycle. It gets even more complicated on direct injected engines (especially diesels), a technology the rotary has not used in mass production yet.
Modern port injectors (Bosch EV14, Rx-8 style injectors) come in a variety of spray patterns and have finer atomization than what was possible with technology from 20-30 years ago. So here's a Bosch spec sheet from the EV14 if you haven't seen something similar:
The alpha, beta, gamma, and delta angles are part of the main specifications used in the design of multi-hole injectors, either for conventional port injection or for gasoline direct injection.
Besides combustion efficiency (for performance) a lot of the spray pattern stuff has to do with HC emissions and oil dilution. Qualitatively speaking, with port injection the injector sprays and part of the fuel forms a film along the manifold/port wall or back of the valve on a piston engine. This is subject to lots of dynamic effects having to do with port/valve timing and port shape. There are two tools used to understand this:
1. computer modeling (computational fluid dynamics) which requires expensive software, an engineering team, and proprietary information. I know this isn't port injection (much less effort is going into developing that now) but here is an example. It is computer model visualization of three different injector configurations evaluated during the development of GM's LLT 3.6 engine, the direct injected motor found in the V6 Camaro and Cadillac CTS:
You can see that these different injectors had a different number of holes and different spray angle characteristics which then affected the mixture distribution in the cylinder near TDC. Direct injected engines are more sensitive to spray patterns than port injected engines but you get the idea. This modeling stuff wasn't used nearly as much 20 years ago because the computers did not have the processing power. An older method is to look inside the cylinder itself.
2. an optical engine where high-speed photography and statistical methods are used to understand injector spray patterns, fuel film formation, mixture formation, etc. We know for sure that Mazda has rotary versions of these; there are references in the [very outdated] literature. They're also used by the other major OEM's in the form of single-cylinder lab engines. Here is a photo of the single-cylinder quartz optical engine supplied by AVL and used in Ford's flex-fuel Ecoboost prototype program of the past few years
Here is an example of the information ultimately generated by an optical engine. You get photographs of the combustion chamber and statistical information on fuel distribution taken from multiple tests:
That's what's really involved in the minutiae of spray pattern optimization. None of us has millions of dollar to do this kind of analysis. And then there are a gazillion other factors that go into it like spark timing, port shape, etc.
Getting back to Rx-7's: so if I take the diffuser out of my intake port on a non-stock injector, does it hurt anything for this kind of mostly racing application? umm maybe? If you are using the diffusers with an old-style pintle injector, then I'd say that the diffusers are likely helpful. I say that because the FB and FC had a pintle injector and they had a diffuser, although their diffusers may be designed a little differently than what's found on the FD.
If removing the diffusers doesn't show up unequivocally in timeslips or maybe controlled chassis dyno runs how could we really know of the drawbacks from removing the diffusers? And if I can somehow demonstate consistently a loss of say 2% horsepower, how much does that matter in the scheme of things? If removing them hurts driveability initially, maybe that can be mitigated with tuning. And just because you take out the diffuser doesn't mean you have to automatically cap off the atomization/air bleed port too.
It's highly likely that plastic diffusors were never designed to handle the kind of heat cycling you get from significant methanol injection. We shouldn't be surprised if a brand new one fails in Howard's circumstances.
If I'm going to choose between the unknown (and in some cases unknowable) drawback of removing the diffusors versus the known risk of keeping them in, I would just take them out.
Hope that helps
Raymond
A short answer/reply is that I am confident the benefits of removing the failure point outweigh potential drawbacks of removing the diffusers. This is partly because we don't have the resources to fully understand the interaction between a non-stock injector spray pattern and the diffusers. For all we know, the diffusers are helpful with some injectors and useless/counter-productive with others.
Injector spray patterns, injector targeting, and injection timing are very complicated things during the actual engine development cycle. It gets even more complicated on direct injected engines (especially diesels), a technology the rotary has not used in mass production yet.
Modern port injectors (Bosch EV14, Rx-8 style injectors) come in a variety of spray patterns and have finer atomization than what was possible with technology from 20-30 years ago. So here's a Bosch spec sheet from the EV14 if you haven't seen something similar:
The alpha, beta, gamma, and delta angles are part of the main specifications used in the design of multi-hole injectors, either for conventional port injection or for gasoline direct injection.
Besides combustion efficiency (for performance) a lot of the spray pattern stuff has to do with HC emissions and oil dilution. Qualitatively speaking, with port injection the injector sprays and part of the fuel forms a film along the manifold/port wall or back of the valve on a piston engine. This is subject to lots of dynamic effects having to do with port/valve timing and port shape. There are two tools used to understand this:
1. computer modeling (computational fluid dynamics) which requires expensive software, an engineering team, and proprietary information. I know this isn't port injection (much less effort is going into developing that now) but here is an example. It is computer model visualization of three different injector configurations evaluated during the development of GM's LLT 3.6 engine, the direct injected motor found in the V6 Camaro and Cadillac CTS:
You can see that these different injectors had a different number of holes and different spray angle characteristics which then affected the mixture distribution in the cylinder near TDC. Direct injected engines are more sensitive to spray patterns than port injected engines but you get the idea. This modeling stuff wasn't used nearly as much 20 years ago because the computers did not have the processing power. An older method is to look inside the cylinder itself.
2. an optical engine where high-speed photography and statistical methods are used to understand injector spray patterns, fuel film formation, mixture formation, etc. We know for sure that Mazda has rotary versions of these; there are references in the [very outdated] literature. They're also used by the other major OEM's in the form of single-cylinder lab engines. Here is a photo of the single-cylinder quartz optical engine supplied by AVL and used in Ford's flex-fuel Ecoboost prototype program of the past few years
Here is an example of the information ultimately generated by an optical engine. You get photographs of the combustion chamber and statistical information on fuel distribution taken from multiple tests:
That's what's really involved in the minutiae of spray pattern optimization. None of us has millions of dollar to do this kind of analysis. And then there are a gazillion other factors that go into it like spark timing, port shape, etc.
Getting back to Rx-7's: so if I take the diffuser out of my intake port on a non-stock injector, does it hurt anything for this kind of mostly racing application? umm maybe? If you are using the diffusers with an old-style pintle injector, then I'd say that the diffusers are likely helpful. I say that because the FB and FC had a pintle injector and they had a diffuser, although their diffusers may be designed a little differently than what's found on the FD.
If removing the diffusers doesn't show up unequivocally in timeslips or maybe controlled chassis dyno runs how could we really know of the drawbacks from removing the diffusers? And if I can somehow demonstate consistently a loss of say 2% horsepower, how much does that matter in the scheme of things? If removing them hurts driveability initially, maybe that can be mitigated with tuning. And just because you take out the diffuser doesn't mean you have to automatically cap off the atomization/air bleed port too.
It's highly likely that plastic diffusors were never designed to handle the kind of heat cycling you get from significant methanol injection. We shouldn't be surprised if a brand new one fails in Howard's circumstances.
If I'm going to choose between the unknown (and in some cases unknowable) drawback of removing the diffusors versus the known risk of keeping them in, I would just take them out.
Hope that helps
Raymond
Joined: Mar 2001
Posts: 31,837
Likes: 3,234
From: https://www2.mazda.com/en/100th/
I've heard of and seen several primary diffuser failures over the last two years. Mostly in higher boost applications. Seen it with all types of fuel abd think it is just an issue with quality from Mazda. All of these failures have happened in the last few years in engines that had brand new diffusers installed.
Thread Starter
Joined: Oct 2001
Posts: 6,279
Likes: 728
From: Florence, Alabama
it would be great if any additional people who have removed their diffusers would check in as to idle quality and any other observations.
especially if you are in the 300-450 hp group.
i sense a fair amount of unease about them given my two failures. probably if you aren't running 100% meth as AI injectant they aren't a problem. i had no problems from 99 til 2010 when i upped the meth.
of course they are plastic and many are vintage 1993...
if, there isn't any material difference diffuserless maybe they all should be ashcanned if you have your motor apart.
hc
especially if you are in the 300-450 hp group.
i sense a fair amount of unease about them given my two failures. probably if you aren't running 100% meth as AI injectant they aren't a problem. i had no problems from 99 til 2010 when i upped the meth.
of course they are plastic and many are vintage 1993...
if, there isn't any material difference diffuserless maybe they all should be ashcanned if you have your motor apart.
hc
talking head
Joined: Apr 2008
Posts: 2,775
Likes: 15
From: Perth, WA, OZ
345,346, 347, 348, 349
i have made a habit of removing these things for many years for the "sleep at night factor"
i deal mostly with FC 's with aftermarket ECU , so quantifying the changes to stock OEM is difficult
note most of the time i do these engines i also do a coolant to manifold crossover delete
on FC engines the OEM the primary difuser is 2 stage , and i replace each section of the difuser with an extra mazda injector base O rings ( square sided rubber donut )
even with these relatively "wet" setup manifolds there is no idle or stumble issues with top feed 550's ( pintle ) or with marren/rochy multech 2 1000's ( disc, with inbuilt difuser )
idle quality even with a crappy microtech and large extendports with 4 x 1000 rochys is very good, have been able to achieve 13 AFRs and as much as 18 occasional 19 flicker inch Hg idle @1000 rpm,
i have made a habit of removing these things for many years for the "sleep at night factor"
i deal mostly with FC 's with aftermarket ECU , so quantifying the changes to stock OEM is difficult
note most of the time i do these engines i also do a coolant to manifold crossover delete
on FC engines the OEM the primary difuser is 2 stage , and i replace each section of the difuser with an extra mazda injector base O rings ( square sided rubber donut )
even with these relatively "wet" setup manifolds there is no idle or stumble issues with top feed 550's ( pintle ) or with marren/rochy multech 2 1000's ( disc, with inbuilt difuser )
idle quality even with a crappy microtech and large extendports with 4 x 1000 rochys is very good, have been able to achieve 13 AFRs and as much as 18 occasional 19 flicker inch Hg idle @1000 rpm,
Full Member
Joined: Dec 2008
Posts: 213
Likes: 0
From: Rennes, Britanny - France
Hi,
4x850cc, mild street port, 12.5 at idle ~1100 RPM (no iacv).
I removed the diffusers because when my previous engine died, one was missing and I wasn't sure about it's role. It was a stock engine with 3 pieces apex seals, I assume the diffusers were as old as the engine.
Should be in the 350hp range, not dynoed yet.
4x850cc, mild street port, 12.5 at idle ~1100 RPM (no iacv).
I removed the diffusers because when my previous engine died, one was missing and I wasn't sure about it's role. It was a stock engine with 3 pieces apex seals, I assume the diffusers were as old as the engine.
Should be in the 350hp range, not dynoed yet.
Thread Starter
Joined: Oct 2001
Posts: 6,279
Likes: 728
From: Florence, Alabama
Driveway and Diffusers
as you look at this picture taken 30 minutes ago your mind immediately thinks:
DIFFUSERS!
notice part of the driveway thinks it is Miami and the other half Alaska.
we had an inch of snow last night which i just shoveled and we have a 10 MPH North wind.
note trees block the wind North of the Miami driveway and no trees block the wind North of the Alaska section.
so the Miami area has been sheltered from the North wind and the Alaska area has had a 10 MPH breeze.
both areas have equal exposure from the sun.
one are is soaking wet and the other area if you licked it w your tongue you would have to wait til spring to get disconnected.
only diff? 10 MPH airspeed.
so if 10 MPH significantly reduces heat what happens at 500 mph in our intake runners?
our diffusers were designed from a material that can fully cope w the 500 MPH back and forth of the intake runners combined w the heat removal of atomising gasoline which is 952 BTUs per gallon.
they were (probably) not designed for methanol which removes 3136 BTUs.
or ethanol at 2398... (better think about this you E85 guys)
alot of people run water as AI injectant and are probably wondering about their situation w re to diffusers. fortunately water doesn't "flash"/atomise as alcohol so it doesn't remove much heat from the intake charge. it removes heat in the combustion chamber... no problem w water and diffusers.
howard
as you look at this picture taken 30 minutes ago your mind immediately thinks:
DIFFUSERS!
notice part of the driveway thinks it is Miami and the other half Alaska.
we had an inch of snow last night which i just shoveled and we have a 10 MPH North wind.
note trees block the wind North of the Miami driveway and no trees block the wind North of the Alaska section.
so the Miami area has been sheltered from the North wind and the Alaska area has had a 10 MPH breeze.
both areas have equal exposure from the sun.
one are is soaking wet and the other area if you licked it w your tongue you would have to wait til spring to get disconnected.
only diff? 10 MPH airspeed.
so if 10 MPH significantly reduces heat what happens at 500 mph in our intake runners?
our diffusers were designed from a material that can fully cope w the 500 MPH back and forth of the intake runners combined w the heat removal of atomising gasoline which is 952 BTUs per gallon.
they were (probably) not designed for methanol which removes 3136 BTUs.
or ethanol at 2398... (better think about this you E85 guys)
alot of people run water as AI injectant and are probably wondering about their situation w re to diffusers. fortunately water doesn't "flash"/atomise as alcohol so it doesn't remove much heat from the intake charge. it removes heat in the combustion chamber... no problem w water and diffusers.
howard
Seriously, do you even read what others post? How do you explain all the other failures on cars that dont run meth? This has only been a reoccurring problem over the last couple years. I know of two cars with this failure that were NOT running any form of meth. Stop over thinking it and remove them. Anyone running more than 12-14psi should do the same as well.
Driveway and Diffusers
as you look at this picture taken 30 minutes ago your mind immediately thinks:
DIFFUSERS!
notice part of the driveway thinks it is Miami and the other half Alaska.
we had an inch of snow last night which i just shoveled and we have a 10 MPH North wind.
note trees block the wind North of the Miami driveway and no trees block the wind North of the Alaska section.
so the Miami area has been sheltered from the North wind and the Alaska area has had a 10 MPH breeze.
both areas have equal exposure from the sun.
one are is soaking wet and the other area if you licked it w your tongue you would have to wait til spring to get disconnected.
only diff? 10 MPH airspeed.
so if 10 MPH significantly reduces heat what happens at 500 mph in our intake runners?
our diffusers were designed from a material that can fully cope w the 500 MPH back and forth of the intake runners combined w the heat removal of atomising gasoline which is 952 BTUs per gallon.
they were (probably) not designed for methanol which removes 3136 BTUs.
or ethanol at 2398... (better think about this you E85 guys)
alot of people run water as AI injectant and are probably wondering about their situation w re to diffusers. fortunately water doesn't "flash"/atomise as alcohol so it doesn't remove much heat from the intake charge. it removes heat in the combustion chamber... no problem w water and diffusers.
howard
as you look at this picture taken 30 minutes ago your mind immediately thinks:
DIFFUSERS!
notice part of the driveway thinks it is Miami and the other half Alaska.
we had an inch of snow last night which i just shoveled and we have a 10 MPH North wind.
note trees block the wind North of the Miami driveway and no trees block the wind North of the Alaska section.
so the Miami area has been sheltered from the North wind and the Alaska area has had a 10 MPH breeze.
both areas have equal exposure from the sun.
one are is soaking wet and the other area if you licked it w your tongue you would have to wait til spring to get disconnected.
only diff? 10 MPH airspeed.
so if 10 MPH significantly reduces heat what happens at 500 mph in our intake runners?
our diffusers were designed from a material that can fully cope w the 500 MPH back and forth of the intake runners combined w the heat removal of atomising gasoline which is 952 BTUs per gallon.
they were (probably) not designed for methanol which removes 3136 BTUs.
or ethanol at 2398... (better think about this you E85 guys)
alot of people run water as AI injectant and are probably wondering about their situation w re to diffusers. fortunately water doesn't "flash"/atomise as alcohol so it doesn't remove much heat from the intake charge. it removes heat in the combustion chamber... no problem w water and diffusers.
howard
Thread Starter
Joined: Oct 2001
Posts: 6,279
Likes: 728
From: Florence, Alabama
thanks for your feedback, if you read my recent posts you would know that i did decide to remove them. the current conversation relates to those not running meth. i, and other readers, appreciate you suggestion which is for most to remove them.
howard
howard
I have a few close friends that have been running 100% meth with pump gas for quite a few years with no issues to the diffusers. I also ran 100% meth before swapping over to ethanol and no issues. There are a number of guys that run E85 and no issues either. Like DJ mentioned, try not to over think it too much and just remove them, after I raised the boost I removed them as well.
When you say you need a rotor and a housing before the motor is opened, is this what you needed the first time it happened?
Any apex seal damage the first time around or this time?
Also can you please post some pics with the damage to the rotor and the housing? (Tape the ports, I'd just like to see what kind of damage these diffusers cause).
Thanks Howard, hope things get back on track soon.
Anthony
Thread Starter
Joined: Oct 2001
Posts: 6,279
Likes: 728
From: Florence, Alabama
plastic breaks ceramic which breaks motor...
crunched
pinched
mostly missing
rotor housing
as one who truly loves the rotary i really feel bad looking at this. this is actually the first rotary i have destroyed. much of the carnage is unfortunately due to the ceramic seals which work great until they break. when i lost the diffuser 2010 November i lost compression and the turbine wheel on my 4094 but no housings rotors etc. so i don't consider that motor destroyed/blowed up. i was running steel seals.
ceramics have some advantages but they come at significant risk.
there is a significant amount of ceramic powder that circulated thru the motor's housings. it got into the rear housing and seems to have taken a bit of the heat treat off the surfaces of the irons. i will have to look closely at them. in addition ceramic bits made there way to the rear of the motor as there are a few pieces embedded in the rotor face.
as to the original cause of the destruction... i do not have a firm answer ATM. on the last two dyno runs the Datalogit recorded one single datapoint which is a maxxed MAP sensor reading. you can see it on the MAPS i posted earlier. the boost which was at 24.2 jumped to max (29.5 or higher) and then the next data point was approx 24. the dyno boost plot shows no spike.
Luke thinks that, if it actually occurred, it might have been a backfire into the UIM. it happened at 7629 RPM on run 26 and 7309 in run 27. all engine metrics were fine leading to the spike reading. here are the charts.
any ideas as to the reading? it does appear that the engine failed right after the spike. knock was normal throughout.
howard
crunched
pinched
mostly missing
rotor housing
as one who truly loves the rotary i really feel bad looking at this. this is actually the first rotary i have destroyed. much of the carnage is unfortunately due to the ceramic seals which work great until they break. when i lost the diffuser 2010 November i lost compression and the turbine wheel on my 4094 but no housings rotors etc. so i don't consider that motor destroyed/blowed up. i was running steel seals.
ceramics have some advantages but they come at significant risk.
there is a significant amount of ceramic powder that circulated thru the motor's housings. it got into the rear housing and seems to have taken a bit of the heat treat off the surfaces of the irons. i will have to look closely at them. in addition ceramic bits made there way to the rear of the motor as there are a few pieces embedded in the rotor face.
as to the original cause of the destruction... i do not have a firm answer ATM. on the last two dyno runs the Datalogit recorded one single datapoint which is a maxxed MAP sensor reading. you can see it on the MAPS i posted earlier. the boost which was at 24.2 jumped to max (29.5 or higher) and then the next data point was approx 24. the dyno boost plot shows no spike.
Luke thinks that, if it actually occurred, it might have been a backfire into the UIM. it happened at 7629 RPM on run 26 and 7309 in run 27. all engine metrics were fine leading to the spike reading. here are the charts.
any ideas as to the reading? it does appear that the engine failed right after the spike. knock was normal throughout.
howard
RX 4BPT
Joined: Feb 2011
Posts: 293
Likes: 0
From: Huddersfield, England
I have seen some bad housings over the last few years but that has to take the number 1 spot!!
I bet the turbine on the BW is prety bad by the looks of those bits.
Why did you choose ceramic seals over normal Mazda seals?
I bet the turbine on the BW is prety bad by the looks of those bits.
Why did you choose ceramic seals over normal Mazda seals?
GTX35R @ 15psi (maybe 17-18 with water inj) stock primaries / Bosch 2000 (ev14) secondaries, diffuser or no diffuser?
Did Mazda have diffusers in their 787B, 792P or any other race cars?
Did Mazda have diffusers in their 787B, 792P or any other race cars?
I know people are tired of me talking about ALS seals and how they basically cure cancer for rotary engines, but I saw an engine with ALS suck a diffuser through. While it did cause the car to run on one rotor(plastic melted all over the housing front housing and melted the seals in place in the apex grooves), the rotors and housings were fine onced cleaned and I even reused the apex seals in a personal engine later on with no issues.
I NO LONGER SELL ALS SEALS, but will continue to preach how great they are until something better comes along.
Any pics from the first time you sucked a diffuser through, pretty sure you were using ALS seals then? Would be nice to compare data as this is what this thread is about in general.
I NO LONGER SELL ALS SEALS, but will continue to preach how great they are until something better comes along.
Any pics from the first time you sucked a diffuser through, pretty sure you were using ALS seals then? Would be nice to compare data as this is what this thread is about in general.
Last edited by djseven; Jan 30, 2012 at 03:47 PM.
Thread Starter
Joined: Oct 2001
Posts: 6,279
Likes: 728
From: Florence, Alabama
"pretty sure you were using ALS seals then? Would be nice to compare data as this is what this thread is about in general."
"While it did cause the car to run on one rotor(plastic melted all over the housing front housing and melted the seals in place in the apex grooves), the rotors and housings were fine onced cleaned and I even reused the apex seals in a personal engine later on with no issues."
yes they were ALS and other than having a bunch of plastic from the diffuser on them they probably are OK. i will clean them and see if they are still straight...
there is nothing that happened here as far as the seals that i didn't already know...
ceramics stay dead straight, are wonderful on the housings, weigh nothing but are a train wreck if they break. steel are harder on the housings, warp if overheated and the shoulders degrade over time. they are, however, alot easier on the motor when bad things happen.
i will be running steel w the new motor. i will post a few pics of the ALS seals w the diffuser fused and will check to see if they are still straight.
hc
"While it did cause the car to run on one rotor(plastic melted all over the housing front housing and melted the seals in place in the apex grooves), the rotors and housings were fine onced cleaned and I even reused the apex seals in a personal engine later on with no issues."
yes they were ALS and other than having a bunch of plastic from the diffuser on them they probably are OK. i will clean them and see if they are still straight...
there is nothing that happened here as far as the seals that i didn't already know...
ceramics stay dead straight, are wonderful on the housings, weigh nothing but are a train wreck if they break. steel are harder on the housings, warp if overheated and the shoulders degrade over time. they are, however, alot easier on the motor when bad things happen.
i will be running steel w the new motor. i will post a few pics of the ALS seals w the diffuser fused and will check to see if they are still straight.
hc